INTERNATIONAL COURT OF JUSTICE
CASE CONCERNING
AERIAL HERBICIDE SPRAYING
ECUADOR
v.
COLOMBIA
REPLY OF ECUADOR
VOLUME I
31 JANUARY 2011 Table of Contents
Page
CHAPTER 1. INTRODUCTION............................................................................1 ▯
Section I. Structure of the Reply..........................................................................8 ▯
CHAPTER 2. COLOMBIA’S MISREPRESENTATIONS REGARDING
THE SPRAY PROGRAMME...............................................................................27 ▯
Section I. Colombia’s Misrepresentations Regarding the Spray Mixture.........36 ▯
A. ▯The Glyphosate-Based Product...............................................................37 ▯
1.▯ Roundup Export and Roundup Ultra................................................38 ▯
2.▯ Roundup SL and GLY-41 ................................................................44 ▯
B. ▯Other Chemicals......................................................................................50▯
1.▯ POEA ............................................................................................▯..50
2.▯ Additional Chemicals.......................................................................53 ▯
3.▯ Cosmo-Flux 411F.............................................................................56 ▯
4.▯ Colombia’s Assertion That “No Other IngredientsŽ Have
Been Used ........................................................................................59
Section II. Colombia’s Misrepresentations Regarding Spraying
Operations..........................................................................................................63
A. ▯The Counter-Memorial’s Claims Regarding Colombia’s
Compliance With Operational Requirements to Prevent Spray
Drift .................................................................................................▯.......70
B. ▯Ecuador’s Receipt of Data From the Government of the United
States Regarding Colombia’s Aerial Spraying Programme....................76 ▯
C. ▯Aircraft Speed and Height of Spray Release...........................................79 ▯
1.▯ Speed............................................................................................▯...79
2.▯ Height...........................................................................................▯...86
3.▯ Reasons for Exceeding the Speed and Height Requirements ..........92 ▯
D. ▯Droplet Size...........................................................................................▯09
E. ▯ Application Rate....................................................................................113▯
F. ▯ Time of Day...........................................................................................1▯5
G. ▯Temperature, Humidity and Wind Conditions......................................120 ▯
i Section III. Colombia’s Prior Misrepresentations to Ecuador and Others ......123 ▯
Section IV. Colombia’s Failure to Stop Spraying in Buffer Zones and
Other Protected Areas......................................................................................128 ▯
A. ▯Colombia’s Spraying in Buffer Zones Established to Protect
Ecuador From Spray Drift.....................................................................128 ▯
1.▯ Colombia Has Violated the 2.7-3.0 Kilometre “Safety
MarginŽ It Previously Claimed Was Sufficient to Protect
Ecuador from Spray Drift...............................................................128 ▯
2.▯ Colombia Has Violated the 10 Kilometre Buffer Zone It
Now Claims to Respect..................................................................129 ▯
B. ▯Colombia’s Aerial Spraying in Close Proximity to Ecuador’s
Indigenous Reserves..............................................................................133 ▯
1.▯ Cofán-Bermejo Ecological Reserve...............................................135 ▯
2.▯ Awá Indigenous and Forest Reserve..............................................137 ▯
C. ▯Colombia’s Spraying in Close Proximity to Ecuador’s Non-
Indigenous Border Communities...........................................................138 ▯
Section V. The Spray Mixture Reaches Ecuador in Quantities Sufficient
to Cause Serious Harm ....................................................................................139 ▯
A. ▯Colombia’s Misrepresentations Regarding the Extent of Spray
Drift .................................................................................................▯.....140
B. ▯Spray Drift Modeling ............................................................................143 ▯
1.▯ Flaws in Colombia’s Modeling......................................................145 ▯
2.▯ Corrected Modeling........................................................................152 ▯
CHAPTER 3. THE EVIDENCE OF HARM IN ECUADOR.............................159 ▯
Section I. The Mutually Corroborating Evidence of Harm in Ecuador...........166 ▯
A. ▯Sucumbíos 2001 ....................................................................................166 ▯
B. ▯Sucumbíos 2002 ....................................................................................191 ▯
1.▯ The Kichwa and Cofán of Sucumbíos............................................206 ▯
C. ▯Esmeraldas 2000....................................................................................214 ▯
D. ▯Esmeraldas 2007....................................................................................225 ▯
1.▯ The Awá of Esmeraldas .................................................................229 ▯
Section II. The Evidentiary Value of Witness Statements...............................235 ▯
ii Section III. The Evidence of Harm Caused in Colombia Corroborates the
Evidence of Harm Caused in Ecuador.............................................................257 ▯
CHAPTER 4. COLOMBIA’S FAILURE TO CONDUCT AN
ENVIRONMENTAL IMPACT ASSESSMENT, AND OTHER
FAILURES TO PREVENT, OR EVEN ASSESS, HARM TO ECUADOR......277 ▯
Section I. Colombia’s Failure to Assess Environmental Impacts Before
the Aerial Spraying Programme Was Commenced.........................................286 ▯
Section II. Colombia’s Failure to Assess Environmental Impacts After
the Aerial Spraying Programme Commenced.................................................304 ▯
A. ▯The Studies Invoked in the Counter-Memorial.....................................304 ▯
B. ▯The Colombian Environment Ministry’s Ongoing Failure to
Secure an EIA........................................................................................314
C. ▯The Failed Efforts of Colombia’s Courts to Secure an EIA..................330 ▯
Section III. The Spray Programme Violates Other Critical Operational
and Safety Requirements.................................................................................33▯
A. ▯Roundup SL.......................................................................................▯...340
B. ▯GLY-41...........................................................................................▯......343
Section IV. Colombia’s Spray Programme Would Not Be Allowed
Elsewhere.................................................................................................▯.......346
A. ▯Europe...........................................................................................▯........346
B. ▯Individual States................................................................................▯...351
CHAPTER 5. VIOLATION OF TERRITORIAL SOVEREIGNTY..................361 ▯
CHAPTER 6. VIOLATION OF INTERNATIONAL ENVIRONMENTAL
LAW AND RESPONSIBILITY FOR TRANSBOUNDARY HARM...............381 ▯
Section I. Introduction ..................................................................................▯..383
Section II. The Applicable Law.......................................................................388 ▯
A. ▯General International Law.....................................................................388▯
B. ▯The 1988 UN Convention on Narcotic Drugs.......................................390 ▯
Section III. Failure to Carry Out a Prior Environmental Impact
Assessment................................................................................................▯......405
A. ▯Duty to Carry Out a Transboundary EIA ..............................................405 ▯
B. ▯Colombia Did Not Carry Out a Transboundary EIA.............................412 ▯
C. ▯An EMP is Not an EIA..........................................................................417▯
iii Section IV. Failure to Act with Due Diligence to Prevent Transboundary
Harm .........................................................................................................▯......421
A. Colombia’s Failure to Prevent Significant Harm..................................421 ▯
B. ▯Due Diligence in the Pulp Mills Case ...................................................427 ▯
C. ▯Application of the Precautionary Principle ...........................................430 ▯
D. No Prior Notification.............................................................................434▯
E. ▯Failure to Comply with the Inadequate EMP........................................436 ▯
F. ▯Buffer Zones.........................................................................................▯441
G. Failure to Cooperate and Monitor .........................................................444 ▯
CHAPTER 7. VIOLATIONS OF HUMAN RIGHTS AND INDIGENOUS
PEOPLES’ RIGHTS............................................................................................455 ▯
Section I. Overview .........................................................................................4▯7
Section II. Colombia’s Violations of Human Rights.......................................461 ▯
A. Aerial Spraying Has Seriously Injured Affected Individuals................461 ▯
B. ▯Aerial Spraying Violates the Fundamental Human Rights of
Affected Individuals..............................................................................477▯
C. ▯Targeting............................................................................................▯...480
Section III. The “TerritorialityŽ of Human Rights Obligations.......................483 ▯
A. The So-called “TerritorialityŽ of Human Rights Obligations ...............485 ▯
B. ▯The Concepts of a Common Legal Space and an International
Public Order of Human Rights at the Regional Level...........................490 ▯
1. ▯ The Inter-American Declaration of Human Rights........................492 ▯
2. ▯ Case Law of the Inter-American Court of Human Rights .............495 ▯
3. ▯ The Legal Consequences Flowing from the Concepts of
Common Legal Space and Public Order of Human Rights ...........497 ▯
Section IV. Breaches of the Rights of Indigenous Peoples .............................501 ▯
A. The Special Status of Indigenous Peoples under International
Law..................................................................................................▯.....502
B. ▯The Evidence of Harm...........................................................................508 ▯
C. ▯Targeting of Indigenous Peoples...........................................................514 ▯
D. Territoriality of Obligations Towards Indigenous Peoples...................516 ▯
E. ▯Consultation and Notification................................................................518 ▯
iv Section V. Relationship Between the Protection of the Environment and
the Protection of Human and Indigenous Rights.............................................527 ▯
CHAPTER 8.REMEDIES...................................................................................535 ▯
Section I. The Purported Impropriety of Ecuador’s Approach to Relief.........539 ▯
Section II. Ecuador Has Not Attempted to Restrict Colombia’s
Sovereignty Over Its Territory.........................................................................545 ▯
SUBMISSIONS...................................................................................................549 ▯
LIST OF FIGURES .............................................................................................555 ▯
LIST OF ANNEXES ...........................................................................................557 ▯
v CHAPTER 1.
INTRODUCTION1.1 The Applicant instituted these proceedings before the International Court
of Justice (“the CourtŽ) on 31 March 2008. In accordance with an Order of the
Court, the Applicant filed its Memorial on 28 April 2009, and the Respondent
filed its Counter-Memorial on 29 March 2010. By Order dated 25 June 2010, the
Court authorized the submission of a Reply by the Applicant and a Rejoinder by
the Respondent, and fixed 31 January 2011 as the time limit for the filing of the
Reply. This Reply is submitted in accordance with that Order, together with
accompanying Annexes.
1.2 This Reply supplements the submissions and arguments on law and
evidence put forward in the Memorial, all of which are maintained in full. As
provided by Article 49(3) of the Rules of the Court, Ecuador’s Reply addresses
the issues that continue to divide the Parties.
1.3 In its Counter-Memorial, Colombia has asked the Court to “adjudge and
declare that the claims of Ecuador, as set out in the Memorial of 28 April 2009,
are rejectedŽ . In support of that submission, Colombia has given a selective and
misleading presentation of the facts and the law. As described in more detail in
this Reply, the facts are materially different than those on which Colombia
purports to rely, and the Counter-Memorial is especially noteworthy for its failure
to put forward evidence that must be within Colombia’s possession but that it has
1Counter-Memorial of Colombia, Vol. I, Submissions (29 Mar. 2010) (hereinafter “CCMŽ).
3chosen not to tender. Ecuador notes the economy with which Colombia treats
facts that are within its knowledge, in particular the quality and composition of
the herbicidal spray and the manner of its delivery. Two points stand out and
bear careful focus by the Court: (1) the herbicidal mixture that is aerially sprayed
along the border with Ecuador is significantly more toxic than Colombia admits;
and (2) the conduct of the spraying has resulted in tens of thousands of occasions
on which Colombia has acted in a reckless manner in blatant disregard of its own
legal and administrative requirements for preventing spray drift into Ecuador. In
particular, Colombia has allowed the spraying to take place using inappropriate
aircraft that fly at excessive speeds and heights, and that make use of application
rates at times and under meteorological conditions that allow a much higher
occurrence of spray drift across the border. Ecuador invites the Court to draw all
appropriate inferences from Colombia’s failure to invoke certain evidence.
1.4 As regards the law, it is plain that certain matters are not disputed by the
Parties. In particular, there is no dispute between the Parties that when Colombia
began spraying along the border with Ecuador in January 2000 it had not carried
out any sort of prior environmental impact assessment, at the local or national
levels or in respect of transboundary impacts. It is also clear that Colombia had
not carried out any kind of assessment on the effects of the spraying on
indigenous peoples, as required by its own domestic law and ILO Convention No.
169. Ten years into the spraying, it has still not carried out any sort of
4environmental impact assessment, and the only actions it has taken are in the
form of an “Environmental Management PlanŽ, which is intended to manage the
environmental effects, not predict them. Moreover, the evidence before the Court
readily establishes that in carrying out its aerial spraying operations close to the
border, Colombia has not followed the guidelines for controlling spray drift as set
forth in its “Environmental Management PlanŽ, and that it has routinely allowed
spraying to occur in a manner that disregards the legally binding instructions
contained on the applicable product labels. This has resulted in an increased risk
of harm to the people and natural environment of Ecuador, and also actual harm
as set forth in the evidence.
1.5 The Parties are also not in dispute as to the rules of international law that
bind the Parties, although they disagree as to their application to the facts. They
agree that distinct legal obligations exist in relation to: the obligation to respect
Ecuador’s territorial sovereignty; the protection of the environment; respect for
fundamental human rights; and the protection of indigenous peoples on both sides
of the border. Each of these obligations exists independently and gives rise to its
own cause of action, though Colombia adopts a curiously minimalist approach to
the scope of those obligations.
51.6 These are important points of difference that separate Ecuador and
Colombia, but there is also much on which the Parties agree. The Parties agree,
for example, on the following matters:
(1) that the spray mixture is toxic and causes harm to people,
animals and plants, although they disagree as to the extent of
2
the harm ;
(2) that the composition of the spray mixture utilized by Colombia
3
has changed over time ;
(3) that Colombia has an obligation of due diligence to prevent or
4
mitigate transboundary harm ;
(4) that Colombia did not carry out any environmental impact
5
assessment before authorizing the aerial spraying ;
(5) that drift is inherent and unavoidable in aerial spraying ; 6
(6) that Colombia’s conclusions about the extent of drift are
premised upon the assumption that Colombia follows strict
7
operational parameters .
2
See, e.g., Memorial of Ecuador, Vol. I, Chap. 5, paras. 5.4-5.72 (28 Apr. 2009) (hereinafter
“EMŽ); CCM, Chap. 4, para. 4.43; Toxicological Opinion Nº 0685, regarding the toxicological
classification of the mix Glyphosate + POEA + Cosmo-Flux (1%), Colombian Health Ministry, 8
Oct. 2001. CCM, Vol. II, Annex 44; CCM, Chap. 4, para. 4.52.
3 See, e.g., EM, Chap. 5, paras. 5.33-5.34; United States Department of State, Bureau for
International Narcotics and Law Enforcement Affairs, Report on Issues Related to the Aerial
Eradication of Illicit Coca in Colombia: Updated Report on Chemicals Used in the Aerial
Eradication Program (Dec. 2003). EM, Vol. III, Annex 148; CCM, Chap. 4, para. 4.50.
4
See, e.g., EM, Chap. 8, paras. 8.10-8.37; CCM, Chaps. 1 and 8, paras. 1.40, 8.23, 8.50-8.52,
8.59.
5
See, e.g., EM, Chaps. 3, 5 and 8, paras. 3.7, 3.28-3.42, 5.25-5.26, 5.71-5.72, 8.41-8.42; CCM,
Chap. 6, paras. 6.23-6.24, 6.26.
6
See, e.g., EM, Chap. 5, paras. 5.74-5.83, 5.84-5.90; CCM, Chaps. 4 and 7, paras. 4.67-4.68,
7.17-7.29, 7.170.
6 (7) that Ecuador’s border region is poor and isolated, with limited
8
sanitation and health and other public services ;
(8) that Ecuador’s border region is rich in biodiversity and natural
9
resources ;
(9) that vulnerable areas, including human settlements, indigenous
and environmental reserves, and bodies of water require
10
heightened precautionary measures ;
11
(10) that a buffer zone is required ; and
(11) that Colombia has no right to overflight in Ecuador and that
12
overflight is a breach Ecuador’s territorial sovereignty .
They concur also about the factors that are likely to enhance the risk of harmful
spray drift, including excessive flight speed, dispersion of the herbicides at too
great an altitude, excessive spray application rate, and spraying at night time and
7
See, e.g., EM, Chaps. 3 and 5, paras. 3.15, 5.91-5.99; CCM, Chaps. 4, 7 and 8, paras. 4.23-4.29,
4.57-4.70, 4.80, 7.17, 7.31-7.32, 7.171-7.173, 8.60.
8
See, e.g., EM, Chap. 2, paras. 2.17-2.18, 2.22-2.28; CCM, Chaps. 2 and 7, paras. 2.4, 2.13, 2.15-
2.16, 2.20-2.23, 2.29-2.30, 2.32-2.35, 7.37, 7.183-7.184.
9
See, e.g., EM, Chap. 2, paras. 2.11-2.16; CCM, Chap. 3, paras. 3.20-3.23, 3.25.
10
See, e.g., EM, Chaps. 2, 6, 8 and 9, paras. 2.11, 2.16, 2.19, 2.22, 5.48, 5.80, 6.49, 8.17, 8.29-
8.30, 9.70-9.74; CCM, Chaps. 4, 7 and 10, paras. 4.59, 7.5, 7.17, 7.32, 7.82, 7.89, 7.176, 7.187,
10.11.
11
See, e.g., EM, Chap. 3, paras. 3.14, 3.59, 3.74-3.80; CCM, Chaps. 4, 5, 7, 8 and 10, paras. 4.59,
4.63, 5.62-5.63, 5.67, 5.109, 7.5, 7.17, 7.32, 7.82, 7.89, 7.176, 7.187, 8.60(4), 10.9.
12
See, e.g., EM, Chap. 7, paras. 7.16-7.17; CCM, Chaps. 1, 5 and 8, paras. 1.41(4), 5.101, 8.42.
7 13
in adverse meteorological circumstances . They also agree on the scientific
14
model for predicting the extent of spray drift .
1.7 These significant points of agreement make the Court’s task that much
easier. Although the differences that remain are important, they are narrow and
relate to matters of degree, not of principle. In Ecuador’s view, the Parties’
differences are easily resolved by reference to the evidence presented in the
Memorial and in this Reply, as will be shown in the Chapters that follow.
Section I. Structure of the Reply
1.8 This Reply consists of eight Chapters, followed by Ecuador’s
Submissions, together with Annexes. Chapter 2 begins Ecuador’s substantive
response to the Counter-Memorial by addressing certain issues of fact. It
establishes that Colombia’s description of the aerial spray programme is partial
and incomplete and provides an inaccurate account of the activities that Colombia
has authorized. The Counter-Memorial fails to provide a complete and accurate
description of the spray ingredients, their chemical composition and their toxicity.
It also fails to provide an accurate account of the manner in which the spraying
has been carried out, and mis-states Colombia’s compliance with its own
mandatory operational parameters that were imposed to prevent spray drift.
13
See, e.g., EM, Chap. 5, paras. 5.84-5.99; CCM, Chap. 4, para. 4.68.
14See infra Chap. 2, paras. 2.189-2.190.
8Moreover, and contrary to the description set forth in the Counter-Memorial,
Colombia sprays in areas that it has itself stated should be off-limits to spraying
in order to protect human settlements, indigenous peoples and ecologically
sensitive areas. Colombia’s failure to provide an accurate account of its spraying
activities confirms its inability to take all necessary measures to prevent risk and
harm.
1.9 The first part of Chapter 2 responds to Colombia’s claims regarding the
chemical composition and properties of the spray mixture. Whilst the parties
agree that the herbicidal mixtures used by Colombia are incapable of
discriminating between plants and destroy and seriously harm plant-life even in
very small quantities 1, they disagree as to the impacts on human health and
animals. The Parties are also in manifest disagreement as to the question of
whether Colombia has fully disclosed the chemical composition of the herbicide
spray: the evidence shows that it has not, and this is a matter of fact that may
easily be established by the Court. The Counter-Memorial asserts that Colombia
has sprayed only two herbicidal formulations: Roundup SL and, after 2004, GLY-
41 1. In fact, the evidence before the Court, including Colombia’s diplomatic
correspondence, demonstrates that Colombia has used two additional herbicidal
formulations that are highly toxic, not least for their propensity to cause serious
15EM, Chaps. 5 and 8, paras. 5.7-5.11, 8.28; CCM, Chap. 7, para. 7.87.
16CCM, Chap. 4, para. 4.50.
9 17
and irreversible eye damage . Moreover, the two formulations that Colombia
admits to using are also hazardous: indeed, Colombia has had to discontinue the
use of Roundup SL because of the dangers it poses to human eyes, and GLY-41’s
label expressly warns against allowing it to come into contact with human skin or
18
to be inhaled or ingested . Colombia does not disclose … in the Counter-
Memorial or voluminous annexes … the chemical formula or complete list of
ingredients for any of the various herbicidal mixtures it has sprayed along the
19
border with Ecuador . In short, the Counter-Memorial materially misrepresents
the herbicidal mixtures in ways that seriously understate their ability to cause
damage in Ecuador.
1.10 The misrepresentations do not end there. As shown in the second part of
Chapter 2, they also, and especially, concern the manner in which the spray
programme is conducted. The Counter-Memorial claims that Colombia complies
strictly with all of the operational requirements that the Parties agree are
necessary to prevent spray drift into Ecuador. Colombia seeks to assure the Court
that Ecuador is protected against spray drift because the spray planes are
equipped with sophisticated technology that automatically and instantaneously
record … for each and every flight … all the relevant data, including the flight
17See supra Chap. 2, paras. 2.18-2.30.
18
See supra Chap. 2, paras. 2.35, 2.38-2.40.
19
See supra Chap. 2, paras. 2.48-2.50.
10speed and altitude at the time of spray dispersion, the application rate and the
20
time of day . According to Colombia, these data have been subjected to regular
reviews and audits, on at least a quarterly basis, and they confirm Colombia’s
strict compliance with every operational requirement designed to prevent spray
21
drift . Yet it is notable that Colombia has not provided the Court with any of this
data or the audit reports on which its conclusions are said to rely. Nor does
Colombia disclose any information about the precise locations or dates of its
spraying activities, or the climatic conditions that pertained on each spraying
occasion. In effect, Colombia asks the Court to take on faith its use of harmless
substances and its strict compliance with all operational requirements designed to
prevent them from drifting across the border into Ecuador.
1.11 Ecuador notes that the Court has developed a well-established practise in
dealing with the evidence before it, rather than on bald assertions of fact. The
Reply sets forth compelling new evidence that was not available to Ecuador when
it submitted its Memorial. Ecuador has obtained and relies upon new evidence
obtained through a Freedom of Information Act Request directed to the United
States Department of State (which has funded and supported some of Colombia’s
aerial spraying operations): this includes detailed data that was generated
automatically by Colombia’s spray aircraft and recorded by the relevant data
20CCM, Chaps. 4 and 7, paras. 4.64 and 7.17.
21Ibid., Chap. 4, paras. 4.28-4.29.
11collection systems. This data provides incontrovertible evidence that undermines
Colombia’s factual assertions. Ecuador has put before the Court the totality of
the detailed information that it has obtained on the records of more than 100,000
spray flights between 2000 and 2008. Each of these spraying activities occurred
within 10 kilometres of Colombia’s border with Ecuador, and it shows the actual
locations, times, flight speeds and altitudes at the time of dispersion, and the
spray application rates (in litres per hectare). This data is contained on a CD-
ROM that accompanies the Reply, and it destroys Colombia’s contention that its
aerial spraying has been conducted in a manner designed to minimise risk and
harm by avoiding the drift of the herbicidal mixtures onto the territory of
Ecuador. The evidence before the Court proves that tens of thousands of spray
flights violated the operational requirements imposed by Colombia and that the
Parties agree are necessary to prevent spray drift. Specifically, Colombia’s spray
planes, within 10 kilometres of the border:
▯ on 75,841 occasions sprayed at a speed greater than the 165 miles per
hour speed limit imposed by Colombia (as was claimed in the
22
Counter-Memorial to have been strictly observed );
▯ on 16,143 occasions sprayed while flying higher than the 50 metre
height limit (as claimed in the Counter-Memorial to have been strictly
observed 23);
22CCM, Chap. 4, para. 4.62.
23Ibid., Chap. 4, para. 4.62.
12 ▯ on 27,139 occasions sprayed in excess of the application rate of 23.65
litres per hectare (as claimed in the Counter-Memorial to have been
strictly observed ); and
▯ on 24,540 occasions sprayed in darkness and at night (a practise that
the Counter-Memorial asserted was prohibited, because night time
25
spraying is more conducive to long-distance spray drift ).
1.12 Colombia had access to all this data … and more … when it filed its
Counter-Memorial, yet chose not to make it available to the Court. Newly
available to Ecuador, having been obtained from the U.S. Department of State by
means of a Freedom of Information Act request, the data make clear that
Colombia’s Counter-Memorial was inaccurate, and misleadingly so. It proves,
inter alia, that:
(i) Colombia has used airplanes that are ill-suited to controlling spray
drift to carry out its spraying operations;
(ii) the pilots … employees of a private U.S. company contracted to
carry out the spraying operations … are poorly trained and
undisciplined, and pay little heed to the operating requirements,
which are haphazardly enforced in any event; and
(iii) spraying has been routinely carried out within the spray-free buffer
zones (originally set at three kilometres and then increased to 10
kilometres) that Colombia unilaterally established in order to
24Ibid., Chap. 4, para. 4.62.
25R. John Hansman, Ph.D. & Carlos F. Mena, Ph.D., Analysis of Aerial Eradication Spray Events
in the Vicinity of the Border Between Colombia and Ecuador from 2000 to 2008, pp. 1, 23 (Jan.
2011). ER, Vol. II, Annex 1.
13 minimize harm to human settlements and ecologically-sensitive
areas, including those in Ecuador.
In short, despite its unsupported assertions to the contrary, the evidence before the
Court shows that Colombia has carried out its aerial spraying programme along
the border in a manner that virtually assures the deposit of highly toxic herbicidal
mixtures deep inside Ecuador.
1.13 Chapter 3 responds to Colombia’s efforts to cast doubt upon the
probative accuracy and reliability of the evidence presented by Colombia in the
Memorial, with regard to the risks and harms to people (including indigenous
peoples), plants and animals that have been inflicted in Ecuador. Contrary to the
Counter-Memorial’s claim that the Memorial’s evidence of harm is unreliable
26
and uncorroborated , Ecuador shows the opposite. In fact, the newly acquired
spray flight data from the U.S. State Department (which Ecuador did not receive
until after the Memorial was submitted) confirms the near-perfect correlation
between the times and locations of spraying near the Ecuadorian border and the
complaints of spraying and harm, including destruction and injury to vital crops
for subsistence farmers. Such reports, which are memorialized in witness
testimony, are corroborated by contemporaneous medical records, press reports
and independent field missions. The UN Special Rapporteur on the Right to
26CCM, Chaps. 1, 5, 7 and 9, paras.1.32, 5.43, 7.30, 7.82, 7.107, 7.121, 7.128, 7.134, 7.141,
7.143, 7.148, 7.151, 7.156, 7.180-7.182, 7.186, 9.123, 9.131-9.132.
14Health has described these reports as “credibleŽ and “reliableŽ, thereby
27
confirming their probative authority . Moreover, the harms to human health
reported by the witnesses … including injuries to the eyes, skin, respiratory and
digestive systems … correspond to the known harms caused by the two principal
elements in all of Colombia’s herbicidal spray mixtures (glyphosate and POEA),
as widely reported in authoritative scientific literature and as specified in explicit
warning labels prepared for these products by the manufacturers. Further, the
witnesses’ descriptions of a pattern of strikingly similar and simultaneous harm to
multiple plant species is consistent with exposure to the use of indiscriminate
herbicides rather than to disease or insect infestation, which would typically only
attack a particular plant species. Colombia has failed to provide any evidence
whatsoever for any other plausible cause for the damage to plants and the
environment and the infirmities reported by the residents of Ecuador’s border
regions … including indigenous peoples … in close proximity to repeated aerial
sprayings.
1.14 In Chapter 4, Ecuador shows that Colombia has disregarded its
international legal obligations with respect to transboundary environmental
impact assessment (“EIAŽ) and due diligence. In this regard, Ecuador has taken
27
Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest
Attainable Standard of Physical and Mental Health, Paul Hunt: Preliminary Note on Mission to
Ecuador and Colombia, Addendum, A/HRC/7/11/Add.3, para. 17 (4 Mar. 2007). EM, Vol. II,
Annex 31.
15careful account of the approach taken by the Court in its recent judgment in the
Pulp Mills case, which confirms the obligation under general international law to
28
carry out such an assessment . Against this background, it is no surprise that the
Parties agree that Colombia was required by international law to carry out a
transboundary EIA, before it began its aerial spraying near the border with
29
Ecuador . The Court has two significant questions to address: the first is a legal
question, namely what are the requirements for such an EIA; the second is a
simple question of fact, namely were those requirements met? The arguments
and evidence before the Court make it abundantly clear that the minimum
requirements under international law for an EIA were not met by Colombia
before it authorised aerial spraying, and they have not been met subsequently.
Similarly, Colombia has never assessed the effects of the spraying operations on
indigenous peoples, as required by ILO Convention No. 169.
1.15 Indeed, Colombia concedes that no EIA (transboundary or otherwise) was
carried out either before or after spraying began along the border with Ecuador.
Colombia justifies that omission on the grounds that its domestic law exempts the
spraying programme from being subject to any EIA requirement. The argument
lacks merit, if only because on this approach any State would be able to
circumvent its international obligations by invoking inadequate domestic laws.
28Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 60, para. 204.
29See infra Chap. 4, para. 4.1 (citing CCM, para. 1.40).
16Colombia is not entitled to be able to claim to circumvent its international legal
obligations by invoking its domestic laws, particularly when it has consciously
structured and applied those laws in order to avoid having to conduct an EIA that
would expose the spray programme’s danger to human health and the
environment. This is all the more so in respect of the clear and express
obligations of assessment that apply in relation to indigenous peoples. The
manifest inadequacy of Colombia’s argument here is made clear by the fact that
the urgent need for an EIA was recognized by Colombia’s own Ministry of
Environment, the government body charged with overseeing and enforcing
environmental regulations. For several years this Ministry repeatedly ordered the
agency responsible for the aerial spraying to carry out, and present for approval,
required environmental impact studies. Each of these orders was ignored. The
Ministry of Environment then imposed sanctions, but these too failed to secure
compliance with the obligation to carry out the required impact studies. Similar
orders and appeals for impact assessments by Colombia’s courts, Comptroller
General and Office of the Ombudsman were all disregarded. In short,
Colombia’s spraying of over 1.2 million litres of herbicide within 10 kilometres
of Ecuador has been conducted without ever having been subjected to a prior
impact assessment to determine its environmental and human health
consequences, including the effects on indigenous peoples.
171.16 In Chapters 5 to 8, Ecuador addresses Colombia’s legal arguments,
which are no more persuasive than its manifestly inadequate treatment of the
facts. There is no dispute between the Parties as to the Court’s jurisdiction to
apply any of the international legal obligations incumbent upon Colombia that
have been invoked by Ecuador, including under the Pact of Bogota and the 1988
UN Convention against Illicit Traffic in Narcotic Drugs and Psychotropic
Substances (“1988 Narcotics ConventionŽ). As regards the 1988 Narcotics
Convention, Ecuador notes that Colombia has not challenged jurisdiction as such,
but rather raised issues as to the meaning and legal effects of that instrument, an
issue that goes to the merits and not jurisdiction. Colombia has entirely
misunderstood the 1988 Narcotics Convention, and its proper meaning and effect
are addressed in Chapter 6 30.
1.17 Chapter 5 responds to Colombia’s case on territorial sovereignty. It
makes clear that the Parties are in agreement that respect for a State’s territorial
sovereignty is a fundamental obligation under both general international law and
treaties applicable as between the Parties, and that breach of that obligation gives
rise to a distinct and separately actionable claim. The Chapter shows that
Colombia has violated Ecuador’s territorial sovereignty by causing toxic amounts
of herbicide to be deposited in Ecuador in a manner that fails to respect Ecuador’s
30See infra Chap. 6, paras. 6.9-6.28.
18sovereignty under international law, including by imposing Colombia’s standards
on that territory.
1.18 Chapter 6 refutes Colombia’s arguments on international environmental
law. Applying the facts to the law, the Chapter establishes that Colombia has
failed to meet its minimum obligations with respect to the obligation to prevent
transboundary risks and harm; has failed to carry out any transboundary
environmental impact assessment prior to spraying near Ecuador; has failed to
cooperate with Ecuador; and has not applied a precautionary approach. In
particular, Ecuador shows that the existence of a real and demonstrable risk that
the herbicidal mixtures would drift into Ecuador and cause harm was known to
Colombia, and that international law required Colombia to carry out a
transboundary EIA prior to engaging in that risky and hazardous activity.
However, none was done, either before or after the commencement of spraying,
31
as Colombia recognizes ; it may have required an Environmental Management
Plan, but that plan did not amount to an environmental assessment, and was in
any event applied in a manifestly inadequate manner. Moreover, Colombia has
breached its duty of exercising due diligence in authorizing and conducting the
spraying programme by, among other things, allowing the use of inappropriate
chemicals; failing to ensure that its aircraft carry out spraying operations in ways
that minimize drift, including at appropriate height, speed, application rate and
31CCM, Chap. 6, paras. 6.23-6.24, 6.26.
19time of day; by failing to give warning before spraying; and by spraying within
10 kilometres of the border with Ecuador. Further, Colombia has breached its
duty to cooperate with Ecuador in good faith, as required by general international
law, by the 1988 Narcotics Convention, by ILO Convention No. 169 and by the
1992 UN Convention on Biological Diversity. Colombia has manifestly failed to
consult with Ecuador before spraying near the border, failed to notify Ecuador of
the chemical compositions of the various spray mixtures, and failed to undertake
joint monitoring of the impact of the spraying programme.
1.19 In Chapter 7, Ecuador responds to the arguments Colombia has made in
relation to its breaches of human rights law and the particular obligations that
international law imposed upon it to take specified measures to protect
indigenous peoples. In so doing, it refutes Colombia’s contention that its actions
32
have not affected the human rights of those living in Ecuador ; that Colombia’s
human rights obligations are geographically restricted to the territory of
Colombia, notwithstanding the Parties’ shared legal space 33; and that the
indigenous people of Ecuador have not suffered special, separately cognizable
34
harm under international law .
32
CCM, Chap. 9, paras. 9.75-9.79, 9.93-9.94, 9.101-9.109, 9.118, 9.123-9.125, 9.130-9.132,
9.140-9.143, 9.152.
33
Ibid., Chap. 9, paras. 9.10-9.50, 9.64, 9.72, 9.85, 9.114.
34
Ibid., Chap. 9, paras. 9.153-9.169.
201.20 In Chapter 8, Ecuador refutes Colombia’s arguments regarding the
remedies to which Ecuador is entitled for the breaches of international law
described in the preceding Chapters. Ecuador shows that its approach to
remedies is based upon and fully consistent with the practise previously adopted
by the Court, by which the quantification of monetary damages is deferred to a
subsequent phase of proceedings. In regard to the cessation of Colombia’s
unlawful acts, Ecuador seeks, inter alia, an order from the Court that Colombia
permanently cease and desist from carrying out aerial spraying operations within
10 kilometres of the Ecuador/Colombia border … a spray-free buffer zone that
Colombia itself claims to have implemented since 2007 (on a voluntary and non-
permanent basis) to prevent harms to Ecuador.
1.21 The Reply concludes with Ecuador’s Submissions. The Submissions set
forth in the Memorial are maintained in full, subject to only one difference: in the
Reply Ecuador has clarified its request by seeking a specific order from the Court
that Colombia shall refrain from aerial spraying within 10 kilometres of the
border between the two countries. Ecuador considers that this is the minimum
spray-free buffer zone necessary to protect Ecuador from further harms, and notes
that Colombia itself has accepted, albeit on a temporary and non-binding basis, a
buffer zone extending 10 kilometres from the border with Ecuador in which it
says it has not conducted aerial spraying operations since 2007. Ecuador notes
that Colombia’s declaration of a 10 kilometre spray-free zone was reconfirmed in
21 35
a public statement by the Ministry of Foreign Relations on 11 November 2010 .
Ecuador considers this, and previous Colombian declarations of a similar nature,
to constitute a recognition of the reasonableness of a 10 kilometre buffer zone
required for prevention of damage to Ecuador. What it seeks from the Court is an
order making permanent and binding what Colombia regards as merely temporary
and voluntary.
1.22 Ecuador’s Reply consists of 5 volumes. Volume I is the main text.
Volumes II-V contain supporting Annexes. In particular, Volume II contains
expert reports, which are presented in the following order:
▯ Annex 1 is a report by R. John Hansman, Ph.D. (Professor of
Aeronautics and Astronautics at the Massachusetts Institute of
Technology and an expert in aviation) and Dr. Carlos F. Mena, Ph.D.
(Professor of Geography and Ecology in the School of Life and
Environmental Sciences at the Universidad San Francisco de Quito
and an expert in geographical information systems). The Hansman &
Mena Report evaluates the flight data that were recorded by the spray
planes’ on-board instruments and obtained by Ecuador from the U.S.
Department of State. It shows that, on tens of thousands of occasions,
the spray programme violated the operational requirements most
relevant to the prevention of spray drift, including, inter alia, aircraft
35Republic of Colombia, Ministry of Foreign Affairs, Press Release (11 Nov. 2010). ER, Vol. V,
Annex 156.
22 speed, altitude of herbicide dispersion, application rate, and time of
day of spraying.
▯ Annex 2 is a report by Dr. Durham K. Giles, Ph.D. (Professor of
Biological and Agricultural Engineering at the University of
California, Davis and an expert in pesticide drift modeling). The Giles
Report applies the internationally accepted model for predicting drift
of aerially applied pesticides (predicting grams of herbicide deposited
per hectare downwind), using data that reflect actual flight conditions
as recorded in the Colombian flight data obtained from the U.S.
Department of State. Dr. Giles shows that significantly more
herbicide is deposited at distances as far as 10 kilometres from the site
of application than was appreciated by the modeling commissioned by
Colombia, which relied upon inaccurate assumptions regarding
compliance with the spray programme’s operational parameters.
▯ Annex 3 is a report by Dr. Stephen C. Weller, Ph.D. (Professor of
Weed Science at Purdue University in West Lafayette, Indiana and an
expert in the dose-response of plants to glyphosate). The Weller
Report compares the downwind deposition predictions generated by
Dr. Giles’s drift modeling with known toxicity thresholds for plants.
Dr. Weller shows that the amount of herbicide deposited at distances
at least 10 kilometres from the site of application is enough to cause
significant harm to plants, including food crops.
▯ Annex 4 is a report by Dr. Henrik Balslev, Ph.D. (Professor of
Biological Sciences at Aarhus University in Denmark and an expert in
the ecology of Ecuador). Dr. Balslev’s report describes the
extraordinary biodiversity of the area around Ecuador’s border with
Colombia and explains the vulnerability of its multiple ecosystems to
23 perturbations, including those caused by exposure to chemical
herbicides.
▯ Annex 5 is a report co-authored by Dr. Norman E. Whitten, Ph.D.
(Professor Emeritus of Anthropology and Latin American Studies at
the University of Illinois at Urbana-Champaign); Dr. William T.
Vickers, Ph.D. (Professor Emeritus of Anthropology at Florida
International University); and Dr. Michael Cepek (Assistant Professor
of Anthropology at the University of Texas at San Antonio). The
three co-authors are experts in the anthropology of northern Ecuador,
including the indigenous peoples, Afro-Ecuadorians and non-
indigenous farmers that inhabit the region. The Whitten et al. Report
explains that these people, many of whom live on the margin of
subsistence, are acutely vulnerable to damage to their health and to the
plant and animal life upon which they depend.
▯ Annex 6 is a report by Charles A. Menzie, Ph.D. and Pieter N. Booth,
M.S. The co-authors previously submitted a report that was annexed
to the Memorial. The present report responds to criticisms made in
the report of Stuart Dobson, Ph.D., which was appended to the
Counter-Memorial. The Menzie & Booth Report shows that the
conclusions in their original report have been validated by the
subsequent flight data evaluation, drift modeling and dose-response
analysis that is presented in Annexes 1-3. They further show that the
appropriate risk management strategy is to implement a buffer zone of
sufficient breadth to protect the vulnerable ecologies and human
communities in Ecuador from harm caused by spray drift, and that the
10-kilometre buffer zone sought by Ecuador is consistent with
international standards.
24 ▯ Annex 7 is a report by Reinhard Joas, Ph.D, who is an expert on
chemicals regulation and served as the technical advisor to the
European Commission in developing the Directive that prohibits in the
European Union aerial spraying as a means for dispersing pesticides.
The Joas Report describes the reasoning behind the EU’s decision to
ban aerial spraying, and shows that Colombia’s programme would not
be permitted in the EU.
▯ Annex 8 is a report by Ms. Claudia Rojas Quiñonez, Esq., a
Colombian lawyer and Lecturer at the Universidad Externado de
Colombia, where she specializes in Colombian environmental law.
The Rojas Report shows that Colombia, in carrying out its aerial
spraying programme, has breached its municipal law by, among other
things, failing to carry out an environmental impact assessment,
failing to comply with the terms of its Environmental Management
Plan, and failing to comply with applicable laws regulating the use of
pesticides.
Volumes III-V contain the remaining Annexes, which are presented in the
following order: (i) Regulations and Technical Reports; (ii) Verification and
Observation Reports; (iii) United States Government Documents; (iv) News
Articles; (v) Multilateral Organisation Documents; (vi) Other Documents; (vii)
Colombian Government Documents.
25 CHAPTER 2.
COLOMBIA’S MISREPRESENTATIONS REGARDING THE SPRAY
PROGRAMME2.1 In this Chapter, Ecuador shows that the Counter-Memorial’s portrayal of
the spray programme is a complete misrepresentation of the actual facts.
Colombia entirely distorts reality in order to hide the true danger the spray
programme poses to the people, plants, animals and ecology of Ecuador.
2.2 According to the Counter-Memorial, the spray programme consists of
spraying limited quantities of an essentially non-toxic mixture of chemicals in
strict conformance with a world-class environmental management plan in a
manner that assures no drift across the border into Ecuador. Nothing could be
further from the truth.
2.3 The grand deception begins with false assertions that the chemical
composition of the various spray mixtures that Colombia has used are both
publicly known and non-toxic. Contrary to the statements in the Counter-
Memorial, Colombia has consistently refused to disclose the formula of the
mixtures, identify all of their contents, or specify the proportions of each element;
it has repeatedly denied Ecuador’s explicit requests for this information; and,
quite remarkably, Colombia has even withheld it from the Court, which can put a
magnifying lens to the entire 513-page Counter-Memorial, and the 1,117 pages of
annexes, without finding the complete formula for the spray mixtures that are at
the centre of this dispute.
292.4 Even without knowledge of the precise formula of the mixture, or the
identification of all of its elements, the toxicity of the secret brew is well
established. It is, indeed, designed and intended to kill every plant that it touches;
and its toxicity to humans and animals is undeniable. True, its effects on people
might not necessarily be fatal, but the evidence that it harms the skin, the eyes,
the respiratory system and the digestive system is not only overwhelming, but
also admitted, expressly, by the manufacturers of the mixture’s main components,
by objective scientists, and by governments of third States.
2.5 Colombia’s pantomime continues with what turns out to be the biggest
falsehood of all: that Colombia’s aerial spraying programme is carried out in a
manner that fully complies with all of the operational parameters touted in the
Counter-Memorial, and in Colombia’s so-called “Environmental Management
PlanŽ, including supposedly stringent restrictions on aircraft speed, height of
spray release, droplet size, application rate and time of day spraying is done … all
of which, the Counter-Memorial boasts, is to ensure that the spray mixture falls
only on legitimate targets (i.e., coca plants) and has minimal, if any, off-target
effects, let alone impacts in Ecuador. Tellingly, although Colombia emphasizes
that it closely monitors the spray flights and records all of this information about
them, and that it has quarterly and semi-annual reports affirming that all of these
operational parameters were consistently complied with over the seven-year
30period when flights were conducted along or close to the border with Ecuador, it
has supplied none of these actual reports to the Court.
2.6 Now we know why.
2.7 Subsequent to the filing of its Memorial, Ecuador was able to obtain all of
these data from a different source: the government of the United States of
America, which finances the aerial spraying programme and receives all of the
operational data about the flights … speed, altitude, application rate, time of day,
etc. … directly from the U.S. contractor that furnishes the pilots and conducts the
spraying on behalf of the government of Colombia. The data show conclusively
that the parameters Colombia itself asserts are essential to maintain the safety of
the programme, and avoid off-target spray drift, were systematically ignored on a
vast scale involving literally tens of thousands of noncompliant flights along or
near the border with Ecuador, virtually guaranteeing that the toxic spray mixture
would drift into Ecuadorian territory and impact humans, animals, crops, forests,
rivers and streams in Ecuador. The data show that the spray planes regularly …
tens of thousands of times … flew too fast and too high, dropped too much spray
in dangerously small droplets (which are more prone to drift) and at prohibited
times of day, violating every safeguard Colombia allegedly required to assure the
accuracy of the spraying and prevent the mixture from drifting off target or into
Ecuador. Coupled with the testimonies of victims of the spraying inside Ecuador,
31medical inquest reports, findings by UN Special Rapporteurs, and studies by
experienced non-governmental organizations, the evidence is irrefutable that
Colombia systematically violated its own regulations and repeatedly sprayed
toxic chemicals that drifted into Ecuador between 2000 and 2007, causing
substantial harm to humans, crops, flora and fauna, and to vulnerable ecosystems
and the indigenous peoples and local residents who depend on them for their
daily existence.
2.8 When Colombia’s misrepresentations are revealed and set aside, what
remains is overwhelming and incontestable proof that in carrying out its aerial
spraying programme along and close to the border with Ecuador, Colombia
violated Ecuador’s sovereignty, and caused grievous harm and risk of harm to
Ecuadorian nationals, including indigenous peoples, and their livelihoods, as well
as the environment.
2.9 Section I of Chapter 2 addresses the contents of the spray mixture itself.
It begins by demonstrating that, contrary to the claims in the Counter-Memorial,
Colombia has never disclosed the full contents of the herbicide spray. The
Counter-Memorial’s claim that only two glyphosate-based products have been
used (Roundup SL and GLY-41) as the principal components of the spray is
proven false. In fact, Colombia used a different, more dangerous product,
Roundup Export, and discontinued its use only after the United States
32Environmental Protection Agency (“U.S. EPAŽ) gave it the highest possible
toxicity rating and determined, among other things, that it causes permanent eye
damage to humans. Moreover, the Counter-Memorial identifies only one of the
multiple additional surfactants used in these products … POEA … but even there
provides only the most general description of a class of toxic chemicals.
Colombia’s admission that POEA is used says nothing about the toxicity of the
specific form of POEA included in the herbicide. Nor does the Counter-
Memorial acknowledge the scientific consensus that all forms of POEA are
highly toxic. The evidence shows that even the two herbicides that Colombia
admits in the Counter-Memorial to using as the main ingredients in its spray are
harmful, including to human health and the environment.
2.10 The same is true with respect to Colombia’s addition of Cosmo-Flux 411F
to the spray … a chemical that is produced and used only in Colombia. Although
Colombia has identified the name of the product, that entitles it to no award for
transparency. The composition of Cosmo-Flux 411F, itself a mixture, is kept
confidential. Further, the Counter-Memorial does not inform Ecuador or the
Court about the identity or toxicity of any of the other chemicals that Colombian
and U.S. government agencies have admitted are or may be present in the
mixture. These include, at the very least, formaldehyde, one or more anti-
foaming agents, dioxin, 1, 4 dioxane and COSMO IN D. All of these chemicals
pose significant danger to human health and the environment.
332.11 In Section II, Ecuador shows that the Counter-Memorial misrepresents
Colombia’s compliance with its self-imposed operational parameters, which are
intended (the Court is told) to avoid spray drift and the resulting danger to non-
target humans, animals and plants. These parameters include restrictions on
aircraft speed, height of spray application, droplet size, application rate and time
of day when spraying in permitted. Colombia’s demonstrably false
representations about its “strict complianceŽ with these parameters … its main
defense to Ecuador’s charge that it deposited harmful sprays over Ecuador …
permeate the Counter-Memorial, which invokes the parameters as a shield against
Ecuador’s claims no fewer than 30 times. They also pervade Colombia’s
diplomatic communications to Ecuador and international organizations, including
the Inter-American Commission on Human Rights, which Colombia falsely
assured of its faithful compliance with its operational requirements.
2.12 Colombia’s own flight data reveal all of these statements to be manifestly
untrue. Between November 2009 and March 2010, Ecuador obtained electronic
flight databases from the U.S. Department of State, which provides operational
support and financing for Colombia’s spray programme. The flight information,
recorded by the spray planes’ on-board data systems, and obtained by Ecuador
36
pursuant to the U.S. Freedom of Information Act , includes the airspeed, altitude,
application rate, time of day, type of aircraft, and precise geographic location
36United States, Freedom of Information Act, 55 U.S.C. § 552.
34(longitude and latitude coordinates) at the time of spraying, among other data.
The records, which include 114,525 spray lines within 10 kilometres of Ecuador’s
border, reveal Colombia’s pervasive disregard for all of its own operational
parameters. The records prove that on tens of thousands of occasions Colombia
violated its self-imposed operational restrictions on, among other things, aircraft
speed, and height of spray release, application rate, and time of day. Colombia’s
wholesale failure to abide by its own standards for preventing spray drift not only
eviscerates Colombia’s effort to prove that the toxic spray mixture did not fall on
Ecuador, but assures that it did.
2.13 Section III of Chapter 2 shows that the Counter-Memorial also
misrepresents Colombia’s alleged respect for buffer zones and other areas it
promised would be off-limits to spraying. To the contrary, Colombia has
routinely sprayed in close proximity to sensitive areas that have been set aside for
use by indigenous peoples. Colombia has also sprayed over or near human
settlements. This is demonstrated by the flight data obtained from the U.S.
Department of State.
2.14 Finally, in Section IV, Ecuador proves that when the flight data is used in
the internationally accepted drift model that Colombia itself relies upon in the
Counter-Memorial, the results demonstrate that the amount of spray deposited in
Ecuador far exceeds what is needed to kill or significantly injure non-target
35plants, including the crops that sustain the livelihood of the local population, and
is capable of inflicting significant harm on humans and animals, as well.
Section I. Colombia’s Misrepresentations Regarding the Spray Mixture
2.15 In the Counter-Memorial, Colombia claims that it has disclosed the
complete chemical composition of the spray mixture. It asserts, for example, that
the formula has been “a matter of public knowledgeŽ since 2001 , and states at
paragraph 6.3 that it has never “sought to conceal or keep secret the formula of
the sprayed mixŽ. Neither of these assertions is true. Although Colombia
identifies the main chemical components of some (but not all) of the mixtures it
has sprayed, it has never revealed their precise formulations or the identities of all
of the additives. This information is conspicuously absent from the Counter-
Memorial.
2.16 The Counter-Memorial states that, in addition to water, the herbicide
spray has three main components: (1) a glyphosate-based product (Roundup SL
until it was replaced by GLY-41, both of which contain glyphosate as the “active
ingredientŽ); (2) a surfactant called POEA; and (3) an adjuvant called Cosmo-
37Counter-Memorial of Colombia, Vol. I, Chaps. 4, 5, 6 and 7, paras. 4.42, 5.96, 5.109, 6.2-6.3,
6.5-6.6, 6.10-6.11, 7.99 (29 Mar. 2010) (hereinafter “CCMŽ).
36Flux 411F 38. With respect to the risk posed to human health and the environment,
the Counter-Memorial claims that none of these chemicals is harmful. It asserts
at paragraph 7.187, for example, that “[t]here is no scientific evidence of serious
39
threats to human and animal healthŽ posed by the spray mixture . In this section,
Ecuador demonstrates that these are blatant misrepresentations, which drastically
understate the spray’s ability to harm Ecuador’s people and natural environment.
A. T HE G LYPHOSATE -B ASED P RODUCT
2.17 In asserting that Colombia has never “conceal[ed] or ke[pt] secret the
formula of the sprayed mixŽ 40, the Counter-Memorial is specific in identifying
the only two glyphosate-based products that it claimed Colombia has ever used.
It states: “since 2000, only two glyphosate-based formulated products have been
41
used, namely, Roundup SL and GLY41Ž . As described below, the evidence
contradicts the Counter-Memorial. Colombia has used additional highly toxic
glyphosate-based formulations. Moreover, the Counter-Memorial understates the
toxicity of Roundup SL and GLY-41.
38Ibid., Chap. 4, paras. 4.42-4.56. See also Report by the Anti-Narcotics Directorate of the
Colombian National Police (DIRAN), p. 306 (8 Feb. 2010) (hereinafter “DIRAN ReportŽ). CCM,
Vol. II, Annex 67.
39See also, e.g., CCM, Chaps. 1 and 7, paras. 1.34(2), 7.10, 7.13.
40CCM, Chap. 6, para. 6.3.
41Ibid., Chap. 4, para. 4.50 (emphasis added); see also Report by the Colombian Agriculture and
Livestock Institute (ICA), p. 288 (29 Jan. 2010). CCM, Vol. II, Annex 65.
37 1. Roundup Export and Roundup Ultra
(a) Roundup Export
2.18 According to correspondence between the U.S. EPA and the Monsanto
Company, the manufacturer of the herbicide, the spray that Colombia actually
42
used was Roundup Export . Colombia’s use of Roundup Export has been
confirmed by the U.S. EPA 43.
2.19 Roundup Export is a highly toxic herbicide that, as its label makes clear,
is “Not Registered for use in the United States of AmericaŽ . Under the heading
“Hazards to Humans and Domestic AnimalsŽ, it warns:
“Keep out of reach of children.
DANGER!
CAUSES IRREVERSIBLE EYE DAMAGE.
HARMFUL IF SWALLOWED OR INHALED.
MAY CAUSE SKIN IRRITATION.
Do not get in eyes, on skin or on clothing.
Avoid breathing vapor or spray mist.
Wash thoroughly with soap and water after handling.
42
Freedom of Information Act Correspondence With United States Environmental Protection
Agency Producing Email from Stephen J. Wratten, Monsanto Company, to Jay Ellenberger,
United States Environmental Protection Agency (30 Oct. 2001). ER, Vol. IV, Annex 118.
43
Freedom of Information Act Correspondence With United States Environmental Protection
Agency Producing Roundup Export Label (3 Mar. 2009). ER, Vol. IV, Annex 112.
44
Ibid.
38 45
Remove contaminated clothing and wash before reuseŽ .
2.20 Further reflecting the danger that Roundup Export poses for human
health, its label provides the following medical instructions:
“FIRST AID: IF IN EYES, immediately flush with plenty of
water for at least 15 minutes. Get medical attention.
IF ON SKIN, immediately flush with plenty of water. Remove
contaminated clothing. Wash clothing before reuse.
IF SWALLOWED, this product will cause gastrointestinal tract
irritation. Immediately dilute by swallowing water or milk. Get
medical attention.
NOTE TO PHYSICIANS: Probable mucosal damage may
contraindicate the use of gastric lavage.
IF INHALED, remove individual to fresh air. Get medical
46
attention if breathing difficulty developsŽ .
2.21 In fact, when the U.S. EPA reviewed the formulation in 2002, it
concluded that the “product used in the coca eradication programŽ, Roundup
47
Export, warranted the highest possible toxicity rating: “toxicity category IŽ .
45Ibid.
46Freedom of Information Act Correspondence With United States Environmental Protection
Agency Producing Roundup Export Label (3 Mar. 2009). ER, Vol. IV, Annex 112.
47United States Environmental Protection Agency, Office of Prevention, Pesticides and Toxic
Substances, Report on Issues Related to the Aerial Eradication of Illicit Coca in Colombia,
Response from EPA Assistant Administrator Johnson to Secretary of State, p. 8 (19 Aug. 2002)
(hereinafter “EPA 2002 AnalysisŽ). ER, Vol. III, Annex 45.
39Among the reasons for concluding that the spray was so dangerous was the fact
48
that it causes “irreversible eye damageŽ .
2.22 Given the danger posed by spraying such a highly toxic substance, the
U.S. EPA determined that it was far too toxic for use in Colombia’s aerial
spraying programme. It therefore recommended in August 2002 that a less
49
harmful alternative with a “lower potential for acute toxicityŽ be used . This
change was necessary, the agency determined, “due to the acute eye irritation
caused by the concentrated glyphosate formulated productŽ 5.
2.23 Nevertheless, Colombia appears to have continued to use Roundup
Export. Thus, Colombia regularly sprayed within 10 kilometres of Ecuador a
product with the highest possible toxicity rating that causes “irreversible eye
damageŽ and “skin irritationŽ and is “harmful if swallowed or inhaledŽ . The 51
only change in formulation acknowledged in the Counter-Memorial did not occur
until 2005. It is no wonder that during the period when Roundup Export was
used, many of Ecuador’s border residents reported serious eye damage, skin
48 Ibid.; Freedom of Information Act Correspondence With United States Environmental
Protection Agency Producing Roundup Export Label (3 Mar. 2009) (indicating on the Roundup
Export label that the product “CAUSES IRREVERSIBLE EYE DAMAGEŽ). ER, Vol. IV,
Annex 112.
49EPA 2002 Analysis, op. cit., p. 8. ER, Vol. III, Annex 45; see also ibid., pp. 18-19.
50Ibid.
51Freedom of Information Act Correspondence With United States Environmental Protection
Agency Producing Roundup Export Label (3 Mar. 2009). ER, Vol. IV, Annex 112.
40irritation and respiratory difficulty following spraying events 52. As Ecuador
described in the Memorial and further elaborates in Chapter 3 of this Reply, large
numbers of Ecuadorians exhibited these precise symptoms upon exposure to
Colombia’s spray mixture during this period.
(b) Roundup Ultra
2.24 In addition to the troubling evidence regarding the use of Roundup
Export, the Counter-Memorial fails to acknowledge the use of another
glyphosate-based product. Contrary to the representations made in the Counter-
Memorial, Colombia has on several occasions informed Ecuador that it used
Roundup Ultra. For example, in a diplomatic note dated 14 July 2001,
Colombia stated:
“The herbicide used by the Program of Eradication of Illicit Crops
- PECI- is a commercial formulation made with glyphosate, which
has the registered name of Roundup Ultra, manufactured by the
53
company Monsanto Inc.Ž .
2.25 Colombia’s diplomatic note represented that Roundup Ultra “is at the
bottom of the universally accepted toxicity scaleŽ and assured Ecuador that “this
productŽ is “less irritatingŽ than “dish detergentsŽ, and its toxicity is comparable
52See infra Chap. 3, Section I(A-C).
53
Diplomatic Note DM/AL No. 25009, sent from the Ministry of Foreign Affairs of Colombia to
the Ministry of Foreign Affairs of Ecuador (14 July 2001). EM, Vol. II, Annex 42.
41to “baby shampooŽ 54. (As explained in paragraph 2.21, at the time of that note
the U.S. EPA was reporting that the product being sprayed had the highest
possible toxicity rating).
2.26 Colombia’s representation that it used Roundup Ultra was repeated at a
bilateral meeting with Ecuador from 13 to 15 February 2002. Colombia then
stated that the formulation “currentlyŽ in use was “Roundup UltraŽ, and that “[i]n
55
the futureŽ Colombia would “use Roundup SLŽ . Thus, on at least two
occasions … in July 2001 and February 2002 … Colombia represented to Ecuador
that it was using Roundup Ultra. These representations cannot be squared with
the Counter-Memorial, where Colombia admits to using only Roundup SL and
GLY-41. Nor are they consistent with the U.S. government’s statements that the
actual product used by Colombia at that time was Roundup Export.
2.27 Assuming quod non that Colombia was accurate when it said it used
Roundup Ultra, Colombia was still wrong when it told Ecuador that Roundup
Ultra is no more dangerous than “baby shampooŽ. Roundup Ultra is also a
dangerous product. For example, the Roundup Ultra label contains the following
warnings:
54Ibid.
55
Republic of Ecuador, Ministry of Environment, Joint Report from the Workshop: Eradication of
Illicit Crops, Bogotá, Colombia, p. 10, para. 14 (13-15 Feb. 2002) (emphasis added). EM, Vol.
IV, Annex 163.
42 “Keep out of reach of children.
CAUTION!
CAUSES EYE IRRITATION.
Avoid contact with eyes or clothing.
FIRST AID: IF IN EYES, flush with plenty of water. Get
medical attention if irritation persistsŽ6.
2.28 Ecuador doubts that the authors of the Counter-Memorial … or anyone
else, for that matter … would consider shampooing their babies with this product.
And not only babies are advised to avoid it. The label warns: “Do not apply this
product in a way that will contact workers or other persons, either directly or
through drift. Only protected handlers may be in the area during applicationŽ . 57
2.29 The Roundup Ultra label also makes clear that it is dangerous for animals,
warning that:
“ingestion of this product or large amounts of freshly sprayed
vegetation may result in temporary gastrointestinal irritation
(vomiting, diarrhea, colic, etc.). If such symptoms are observed,
provide the animal with plenty of fluids to prevent dehydration.
Call a veterinarian if symptoms persist for more than 24 hoursŽ . 58
56
United States Roundup Ultra Label. ER, Vol. III, Annex 27. The Advisor on the Plan
Colombia Illicit Eradication Program stated that the spraying programme used a formulation that
is “commercialized in the United States under the name Roundup UltraŽ. See Republic of
Colombia, Advisor on the Plan Colombia Illicit Crop Eradication Program, Certain Toxicological
and Technical Considerations For Aerial Spraying With Glyphosate on Illicit Crops, Bogotá,
Colombia, p. 2 (9 July 2001). ER, Vol. V, Annex 138.
57
United States Roundup Ultra Label. ER, Vol. III, Annex 27.
58Ibid.
432.30 Accordingly, even if Colombia did use Roundup Ultra (contrary to its
assertions in the Counter-Memorial and the U.S. government’s reports), it still
sprayed along the border with Ecuador a highly toxic product that is dangerous to
human health and the environment.
2. Roundup SL and GLY-41
2.31 The two glyphosate-based products that Colombia admits to spraying …
Roundup SL and GLY-41 … are no less dangerous to humans, animals, crops or
the environment in Ecuador than Roundup Export and Roundup Ultra.
(a) Roundup SL
2.32 The danger of Roundup SL to human health is confirmed by the product
label, which emphasizes its hazardous nature, particularly with respect to human
eyes and skin:
“Avoid contact with eyes and skin. Causes irritation. Upon
completion of work, change clothes and wash with plenty of soap
and waterŽ 59.
2.33 The label further instructs: “In the event of contact with eyes, wash them
immediately with plenty of water for 15 minutesŽ and “[i]f it falls on the skin,
59Colombia Roundup SL Label. EM, Vol. III, Annex 115.
44 60
wash with plenty [of] soap and waterŽ . These warnings are similarly
emphasized by the Roundup SL Technical Data Sheet, which states that the
product causes “moderateŽ to “severeŽ eye irritation and is irritating to the skin . 61
2.34 The Roundup SL label also warns against ingesting the product, stating:
“If ingested, drink water to dilute it. Call a doctor immediately or bring the
62
patient to a doctor and show them a copy of this labelŽ . Similarly, the
63
Technical Data Sheet unambiguously warns that it is “harmful if ingestedŽ . It
also warns that the product is of “middle toxicity to rainbow troutŽ, thus making
it hazardous to fish, and that it “must not be poured near channels, drains, nor
64
running water or water reservoirsŽ . This directive is repeated by the label,
which states: “Do not contaminate water sources. Do not apply or pour surplus
product directly over water bodiesŽ . 65
2.35 Although the Counter-Memorial brushes aside these warnings and claims
that Roundup SL is incapable of causing harm, in fact, Colombia was forced to
60
Ibid.
61
Safety Data Sheet for Roundup SL, p. 424. CCM, Vol. III, Annex 133.
62Colombia Roundup SL Label. EM, Vol. III, Annex 115.
63Safety Data Sheet for Roundup SL, p. 423. CCM, Vol. III, Annex 133.
64Ibid.
65Colombia Roundup SL Label. EM, Vol. III, Annex 115. The label also makes clear that the
spray will kill or seriously injure crops and other beneficial plants: “During application, avoid
allowing the product to fall on leaves or green parts of the stems of cropsŽ. Colombia Roundup
SL Label. EM, Vol. III, Annex 115.
45abandon spraying of the product precisely because of its harmfulness. This is
clear from footnote 312, which admits that “[a]s of 2005, with the purpose of
avoiding the possible effects associated with the use of POEA surfactant
contained in Roundup SLŽ, specifically the “the risk of eye irritationŽ, the
“Colombian Government decided to use a new glyphosate-based formulated
product, called GLY41Ž 6.
2.36 Thus, the Counter-Memorial leaves no doubt that for at least the first five
years of spraying adjacent to and near Ecuador … i.e., between 2000 and 2005 …
Colombia sprayed at least one (and possibly more) glyphosate formulation that
had to be discontinued due to the danger it posed to human health. There can be
no truth to Colombia’s assertion that Roundup SL is not harmful. The Counter-
Memorial itself admits otherwise.
(b) GLY-41
2.37 The Counter-Memorial also ignores the dangers posed by GLY-41, the
herbicide with which Colombia replaced Roundup SL in 2005, even though the
label for GLY-41 cautions that the mixture is harmful to human health:
66CCM, Chap. 4, para. 4.50, n. 312. Notably, however, reports by the U.S. Government that
describe the spray formulation during this time period do not reflect or otherwise acknowledge
any such change.
46 “CAUTION: THIS PRODUCT CAUSES EYE IRRITATION. Avoid
contact with eyes and clothesŽ 6.
2.38 The Counter-Memorial also ignores that the Safety Data Sheet for GLY-
41, which expressly warns against allowing the product to come into contact with
human skin:
“In case of skin contact: Wash the skin immediately with water
and soap. If possible, remove the shoes and the clothes of the
patient. Contaminated clothes must be washed separately before
wearing it againŽ .8
2.39 The danger of inhaling GLY-41 is also stressed by the Safety Data Sheet:
“In case of inhaling: Take the person to a fresh air area; if he/she is
not breathing, provide artificial breathing 69d oxygen if necessary.
LOOK FOR MEDICAL ASSISTANCEŽ .
2.40 The GLY-41 label leaves no doubt about the risk that it poses to animals,
as well as humans:
“Ingestion of this product or large amounts of freshly sprayed
vegetation may result in temporary gastrointestinal irritation
(vomiting, diarrhea, colic, etc.). If such symptoms are observed,
provide the animal with plenty of fluids to prevent dehydration.
70
Call a veterinarian if symptoms persist for more than 24 hoursŽ .
67
Label and Safety Data Sheet for GLY-41. CCM, Vol. III, Annex 134.
68Ibid.
69Ibid.
70
Ibid., p. 428. CCM, Vol. III, Annex 134. The label also makes clear that the chemical will
destroy crops and other beneficial plants: “AVOID CONTACT OF HERBICIDE WITH
FOLIAGE, GREEN STEMS, EXPOSED ROOTS OR FRUIT OF CROPS, DESIRABLE
472.41 In keeping with its pattern of ignoring or excising evidence of the dangers
posed by its spray products, and then extolling its studious reliance on substances
as innocent as “baby shampooŽ, Colombia redacts from the English translation of
the product label for GLY-41 (included in an annex to the Counter-Memorial) the
pictorial warnings that graphically represent the danger the product poses to
71
human health and the environment .
72
Figure 2.1. Excerpt From GLY41 Label
According to the official Andean Community Technical Manual on Registration
and Control of Pesticides, these warning symbols indicate that the applicator
should use gloves and boots for protection, as well as a respirator mask. The
applicator should also wash his or her body after using the product 73.
PLANTS AND TREES, BECAUSE SEVERE INJURY OR DESTRUCTION MAY RESULTŽ.
Ibid.
71Colombia GYL-41 SL Label and Safety Data Sheet. ER, Vol. III, Annex 28.
72
Ibid.
73
Andean Community, Resolution 630, Andean Technical Manual for the Registration and
Control of Chemical Pesticides for Agricultural Use, p. 129 (25 June 2002). EM, Vol. II, Annex
17.
48Figure 2.2. Warning Symbols From GLY-41 Label … Human Health
With respect to animals and the environment, the warnings dictate “do not
74
contaminate watercoursesŽ and “do not permit animals in the treated areaŽ .
Figure 2.3. Warning Symbols From GLY-41 Label … Animals & Environment
74Ibid.
492.42 In short, GLY-41 bears no resemblance to baby shampoo. It is a
dangerous product that poses serious risks for human health and the environment.
B. OTHER C HEMICALS
1. POEA
2.43 All glyphosate-based herbicides, including Roundup Export, Roundup
Ultra, Roundup SL and GLY-41, use glyphosate as their “active ingredientŽ,
meaning that this chemical does the main work of killing plants. But glyphosate
is not the only chemical in these glyphosate-based products, which are also called
glyphosate-based “formulationsŽ. Such herbicides include additional chemicals
75
known as “formulantsŽ . One class of “formulantsŽ is composed of surfactants,
which are designed to increase lethality by improving the glyphosate’s ability to
penetrate the plant . Not only do these surfactants magnify the formulation’s
effect on plants, but they can be even more toxic to human health and the
environment than glyphosate itself.
75Formulants may also be called “inert ingredientsŽ or “adjuvantsŽ. See Stephen C. Weller,
Ph.D., Glyphosate-Based Herbicides and Potential for Damage to Non-Target Plants Under
Conditions of Application in Colombia, p. 7 (Jan. 2011) (hereinafter “Weller ReportŽ). ER, Vol.
II, Annex 3.
76Keith R. Solomon et al., Environmental and Human Health Assessment of the Aerial Spray
program for Coca and Poppy Control in Colombia , p. 23 (31 Mar. 2005) (hereinafter “Solomon
et al., 2005Ž). CCM, Vol. III, Annex 116; EPA 2002 Analysis, op. cit., p. 11. ER, Vol. III,
Annex 45.
502.44 The only formulant that Colombia admits in the Counter-Memorial to
77
using in its various spray mixtures is a surfactant called POEA . But that
admission reveals little, other than that Colombia sprays a hazardous product.
POEA is not a single chemical, but instead describes an entire category of
chemicals with a range of toxicities and effects 78. Therefore, unless Colombia
provides specific information about which POEA surfactant it uses, it is
impossible for Ecuador (or the Court) to determine the precise toxicity of the
POEA that has been sprayed.
2.45 Nevertheless, this much is clear: all chemicals that are classified as POEA
are harmful to human health and to the environment. As discussed in the
Memorial, POEA causes severe eye damage, among other serious health and
environmental effects 7. Even Colombia’s expert report acknowledges that
POEA is a “worst case for surfactant toxicityŽ and “has been identified as a major
77 POEA is an abbreviation for polyethoxylated tallow amine, also referred to as
polyethoxyethylene alkylamine.
78Federal Republic of Germany, Federal Institute of Risk Assessment, Health Assessment Report
POE-tallowamines, p. 6 (6 Sep. 2010). ER, Vol. III, Annex 26; Weller Report, op. cit., p. 15. ER,
Vol. II, Annex 3; Charles A. Menzie, Ph.D. & Pieter N. Booth, M.S., Response to: “Critique of
Evaluation of Chemicals Used in Colombia’s Aerial Spraying Program, and Hazards Presented
to People, Plants, Animals and the Environment in Ecuador,” As Presented in: Counter-
Memorial of the Republic of Colombia, Appendix, p. 11 (Jan. 2011) (hereinafter “Menzie & Booth
ReportŽ). ER, Vol. II, Annex 6.
79See EM, Chap. 5, paras. 5.18-5.19; CCM, Chap. 4, para. 4.50, n. 312; EPA Analysis, 2002, op.
cit., p. 10. ER, Vol. III, Annex 45.
51 80
contributor to the aquatic toxicity of glyphosate formulationsŽ . According to
81
the U.S. EPA, POEA causes severe skin irritation and is corrosive to the eyes .
A study of POEA published by Germany’s Federal Institute of Risk Assessment
(BfR) concluded that POEA is a skin irritant and skin sensitizer and is strongly
irritating to the eyes, noting that POEA is a “surface-active substanceŽ for which
82
“certain ability to penetrate through biological membranes can be assumedŽ .
The study also recommended that POEA be labelled for acute oral toxicity due to
“strong mucosal irritationŽ and the likely occurrence of “systemic effects after
83
ingestion or inhalationŽ . Colombia’s own National Health Institute has stated
that POEA may cause gastrointestinal damage, kidney and liver damage, affect
the central nervous system, destroy red blood cells and induce breathing
84
difficulties . POEA has also been linked to problems with pregnancy and even
80
Dr. Stuart Dobson, Critique of “Evaluation of Chemicals Used in Colombia’s Aerial Spraying
Program, and Hazards Presented to People, Plants, Animals and the Environment in Ecuador”
Menzie et al., pp. 521, 528 (2009) (hereinafter “Dobson ReportŽ). CCM, Vol. I, Appendix.
81United States Environmental Protection Agency, Office of Pesticide Programs, Details of the
2003 Consultation for the Department of State: Use of Pesticide for Coca and Poppy Eradication
Program in Colombia, p. 13 (June 2003) (hereinafter “EPA 2003 AnalysisŽ). EM, Vol. III,
Annex 146.
82
Federal Republic of Germany, Federal Institute of Risk Assessment, Health Assessment Report
POE-tallowamines, pp. 4, 8, 20, 21 (6 Sep. 2010). ER, Vol. III, Annex 26.
83
Ibid.
84
Government of Colombia National Health Institute, Evaluation of Effects of Glyphosate on
Human Health in Illicit Crop Eradication Program Influence Zones, p. 5 (2003). EM, Vol. II,
Annex 96.
52 85 86
cancer . These effects were detailed in the Memorial , but the Counter-
Memorial completely ignores them.
2.46 Indeed, the high level of POEA in Roundup SL was one of the reasons
Colombia switched to GLY-41 in 2005. As the Counter-Memorial concedes,
Roundup SL was replaced by GLY-41 to “avoid[] the possible effects associated
with the use of the POEA surfactant . . . in particular the risk of eye irritationŽ . 87
Thus, based on this admission alone, the Court can conclude that Colombia used
an herbicide with an unacceptably high level of POEA for at least four years
88
while spraying in close proximity to Ecuador . It remains for Colombia to
demonstrate that the POEA it used after 2005 is less harmful (if, in fact, it
switched to a different and less harmful type of POEA).
2. Additional Chemicals
2.47 POEA is not the only formulant added by Colombia to its glyphosate-
based spray mixtures. The scientific reports relied upon by Colombia, including
its own expert reports, show that other formulants are also present in the toxic
cocktail sprayed along the border with Ecuador. For example, the Solomon
85
Ibid.; EPA 2003 Analysis, op. cit., p. 13. EM, Vol. III, Annex 146.
86
EM, Chap. 5, para. 5.18.
87CCM, Chap. 4, para. 4.50, n.312.
88In fact, the amount of POEA in GLY-41, the formulation that replaced Roundup SL in 2005, is
unacceptably high as well.
53study, which was prepared at Colombia’s request in 2005 and which features
prominently in the Counter-Memorial, states that: “The glyphosate formulation
89
used in Colombia includes several formulantsŽ . However, none of these
chemical components of the spray mixture have been identified, other than
POEA. Nor are they identified in Colombia’s expert report by Dr. Stuart Dobson
submitted with the Counter-Memorial, despite the fact that the Dobson Report
90
also acknowledges that POEA is not the only added formulant . Specifically, the
Dobson Report states that the “formulation uses predominantly POEA as its
surfactantŽ . In other words, there are other surfactants as well, none of which
Colombia has identified, and which are still unknown to Ecuador and the Court.
2.48 Reports by the U.S. government confirm that Colombia has not fully
disclosed all the chemicals in the spray mixture. For example, multiple reports by
the U.S. EPA state that the spray contains an unnamed ingredient for which
“information [is] not included as it may be entitled to confidential treatmentŽ . 92
Similarly, the U.S. Department of State reported that the spray contains a
89Solomon et al., 2005, op. cit., p. 23, 24 (emphasis added). CCM, Vol. III, Annex 116.
90
Dobson Report, op. cit., p. 521. CCM, Vol. I, Appendix.
91
Ibid. (emphasis added).
92EPA 2002 Analysis, op. cit., p. 10 (listing the “components of the glyphosate productŽ to
include Polyoxyethylene alkylamine (“POEAŽ) and another unnamed ingredient for which
“information [is] not included as it may be entitled to confidential treatmentŽ). ER, Vol. III,
Annex 45. See also EPA 2003 Analysis, op. cit., p. 13. EM, Vol. III, Annex 146.
54 93
“surfactant blendŽ . In other words, POEA is just one of the surfactants in a
“blendŽ; the other ingredients are not disclosed.
2.49 Colombia’s persistent refusal to identify the other chemicals in the spray
mixture is of serious concern, because these chemicals significantly contribute to
the mixture’s toxicity, and the consequent risks to humans, animals and plants
that are exposed to it. As the Dobson Report explains, “[s]urfactants play the
principal role in the environmental toxicity of formulations of glyphosate and the
94
Colombian spray mix; they are probably key in symptoms to humansŽ . This is
confirmed by the 2005 Solomon study, which states that “it is the surfactants that
determine the toxicity of the formulationŽ, and by the U.S. EPA, which explains
that “formulations of glyphosate products that EPA has registeredŽ are “more
toxic than glyphosate aloneŽ . 95
2.50 Yet Colombia has thus far refused to identify all of the formulants it uses
in the spray mixture. Not even its own experts appear to know what they are,
although they know enough to state that whatever is used is more toxic than the
93
United States Department of State, Bureau for International Narcotics and Law Enforcement
Affairs, Report on Issues Related to the Aerial Eradication of Illicit Coca in Colombia: Chemicals
Used in the Aerial Eradication of Illicit Coca in Colombia and Conditions of Application, p. 1
(Sep. 2002) (hereinafter “Chemicals UsedŽ). EM, Vol. III, Annex 144.
94Dobson Report, op. cit., p. 546. CCM, Vol. I, Appendix.
95Solomon et al., 2005, op. cit., p. 93. EM, Vol. III, Annex 151; EPA 2002 Analysis, op. cit., p.
32. ER, Vol. III, Annex 45; see also EM, Chaps. 2 and 5, paras. 2.41, 5.16-5.18; Weller Report,
op. cit., pp. 7-8. ER, Vol. II, Annex 3.
55 96
glyphosate itself . Nevertheless, as if Colombia failed to read what its experts
wrote … or chose to ignore them … the Counter-Memorial brazenly represents that
the chemical composition of the spray mixture is “a matter of public
knowledgeŽ . 97
3. Cosmo-Flux 411F
2.51 The Counter-Memorial states that Colombia adds an adjuvant known as
Cosmo-Flux 411F (“Cosmo-FluxŽ) to the glyphosate-based spray mixtures that it
admits using, i.e., Roundup SL and GLY-41 . 98 The Counter-Memorial then
falsely asserts that the chemical composition of Cosmo-Flux is known and non-
toxic.
2.52 Colombia has never revealed the composition of Cosmo-Flux, a mixture
of chemicals that is only produced and used in Colombia. (As Ecuador noted in
the Memorial, and as Colombia did not dispute in the Counter-Memorial, a
British chemical manufacturer refused to provide Colombia with the chemicals
96Dobson Report, op. cit., p. 546. CCM, Vol. I, Appendix; see also Solomon et al., 2005, op. cit.,
p. 93. EM, Vol. III, Annex 151.
97CCM, Chap. 5, para. 5.96; see also CCM, Chaps. 4, 6 and 8, paras. 4.45, 6.36, 8.60.
98Ibid., Chap. 4, paras. 4.51-4.56. According to the Solomon Report (2005), an “adjuvantŽ is a
chemical added to the formulated product at the time of application to increase efficacy and ease
of use. Solomon et al., 2005, op. cit., p. 23. EM, Vol. III, Annex 151.
56used to produce Cosmo-Flux, explaining that the spray mixture “had not properly
99
been testedŽ for use in aerial spraying .).
2.53 The specific ingredients contained in Cosmo-Flux are labelled
“confidentialŽ in reports by the U.S. government. For example, a 2002 report by
the U.S. EPA states:
“The Cosmo-Flux 411F adjuvant used in the glyphosate tank mix
is produced by a Colombian company and is not sold in the U.S. . .
. Cosmo-Flux 411F consists mainly of (information not included
as it may be entitled to confidential treatment) with a non-ionic
surfactant blend primarily composed of (information not included
as it may be entitled to confidential treatment)Ž 100.
2.54 The secret nature of Cosmo-Flux is alluded to in the 2005 Solomon study,
which states only that the active ingredient in Cosmo-Flux is a “mixtureŽ
containing certain “linear and aryl polyethoxylatesŽ. But it does not provide any
101
details about the “mixtureŽ .
2.55 Although Colombia’s failure to disclose the chemical composition of
Cosmo-Flux makes it impossible for Ecuador (or the Court) to evaluate the full
extent of its toxicity, there is no doubt that Cosmo-Flux makes the spray mixture
more toxic. To be sure, the Counter-Memorial blithely claims that Cosmo-Flux
99EM, Chap. 5, para. 5.24.
100EPA 2002 Analysis, op. cit., p. 7 (emphasis in original). ER, Vol. III, Annex 45; see also EPA
2003 Analysis, op. cit., p. 14. EM, Vol. III, Annex 146.
101Solomon et al., 2005, op. cit., p. 24. EM, Vol. III, Annex 151.
57 102
“does not increase the toxicity of the mixtureŽ . That assertion is impossible to
reconcile with the 2005 Solomon study, however, which concludes that “the
toxicity of the mixture of glyphosate and Cosmo-FluxŽ is “greater than that
reported for formulated glyphosate itselfŽ 103. How much more toxic, we cannot
know … until Colombia reveals the chemical composition of Cosmo-Flux. What
we can … and do … know, is that, based on Colombia’s own expert reports, its
statement that Cosmo-Flux “does not increase the toxicity of the mixtureŽ is
another misrepresentation.
2.56 Indeed, Colombia uses Cosmo-Flux precisely because of its ability to
104
enhance the killing power of glyphosate . As Colombia’s expert, Dr. Dobson,
has admitted, “the addition of extra surfactant Cosmo-Flux . . . increases the
potency of the glyphosate formulation to coca plants fourfoldŽ and that “plants
other than cocaŽ will also be “more susceptible to the herbicide spray enhanced
105
with the adjuvantŽ . Cosmo-Flux’s Colombian manufacturer also
acknowledges that the product “substantially modifies the biological activity of
102CCM, Chap. 4, para. 4.53.
103
Solomon et al., 2005, op. cit., p. 86. EM, Vol. III, Annex 151; see also ibid., p. 69.
104
Ronald T. Collins & Charles S. Helling, Surfactant-Enhanced Control of Two Erythroxylum
Species by Glyphosate, Weed Technology, Vol. 16, p. 851 (2002) (identifying adjuvants that
increased glyphosate phytotoxicity fourfold and explaining that “[i]n consequence, the glyphosate
mixture used in Colombia for coca eradication was modified with substantially improved
resultsŽ). EM, Vol. III, Annex 141; CCM, Chap. 4, para. 4.53 (noting that Cosmo-Flux was
selected as a result of research by Collins and Helling); see also Chemicals Used, op. cit., p. 1.
EM, Vol. III, Annex 144.
105Dobson Report, op. cit., p. 538. CCM, Vol. I, Appendix.
58 106
agrochemicalsŽ . Thus, Colombia’s denial that Cosmo-Flux makes the spray
107
mixture more toxic cannot be believed .
4. Colombia’s Assertion That “No Other Ingredients” Have Been Used
108
2.57 Notwithstanding the Counter-Memorial’s assertion to the contrary ,
Cosmo-Flux is not the only chemical that is added to the glyphosate-POEA
mixture. The Counter-Memorial’s claim is “supportedŽ by the Colombian
National Police’s Anti-Narcotics Directorate, in a report dated 8 February 2010,
which appears to have been prepared for the present case. The report states
109
without exception that “no other ingredients have been used in the programŽ .
Once again, the falsity of Colombia’s statements is exposed by the evidence,
including official reports of the U.S. government and Colombia’s diplomatic
communications to Ecuador.
2.58 To cite a few examples, in its 14 July 2001 diplomatic note to Ecuador,
Colombia stated that the spray mixture contains dioxin 110, a chemical associated
with significant human health problems, including reproductive and
106
Cosmoagro, S.A., Cosmo-Flux 411F. EM, Vol. III, Annex 112.
107As explained in Ecuador’s Memorial, Cosmo-Flux also poses risks to human health. See EM,
Chap. 5, para. 5.22.
108CCM, Chap. 4, para. 4.42-4.56.
109DIRAN Report, op. cit., p. 307 (emphasis added). CCM, Vol. II, Annex 67.
110CCM, Chap. 6, para. 6.15, quoting Diplomatic Note DM/AL No. 25009, sent from the Ministry
of Foreign Affairs of Colombia to the Ministry of Foreign Affairs of Ecuador (14 July 2001).
EM, Vol. II, Annex 42.
59developmental problems, damage to the immune system, interference with
hormones, and cancer; dioxin is also a long-lasting environmental contaminant 111.
2.59 The following year, at a meeting from 13 to 15 February 2002, Colombia
112
informed Ecuador that the spray mixture also contains 1,4 Dioxane . Colombia
has acknowledged that this chemical is “demonstrated to have a carcinogenic
capability in animals and to cause harm to human livers and kidneysŽ 113.
According to the World Health Organization, 1,4 Dioxane has genotoxic potential
and is regulated as a drinking water contaminant 114.
2.60 Ecuador’s Memorial noted that the U.S. Department of Agriculture
(USDA), which was tasked with evaluating the aerial spraying programme, stated
in 2001 (that is, over a year after the spraying began alongside Ecuador) that the
spray mixture contains formaldehyde. Because this substance poses significant
111United Nations World Health Organization, Dioxins and Their Effects on Human Health,
available at http://www.who.int/mediacentre/factsheets/fs225/en/index.html(last visited 14 Jan.
2011).
112
Republic of Ecuador, Ministry of Environment, Joint Report from the Workshop: Eradication
of Illicit Crops, Bogotá, Colombia, p. 10, para. 14 (13-15 Feb. 2002). EM, Vol. IV, Annex 163.
See also Republic of Colombia, Ministry of Environment, Division of Environmental Licenses,
Order No. 599, p. 5 (23 Dec. 1999). ER, Vol. V, Annex 132.
113Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 599, p. 16 (23 Dec. 1999). ER, Vol. V, Annex 132.
114United Nations World Health Organization, Guidelines for Drinking Water Quality, § 12,54(a)
1,4 Dioxane, available at
http://www.who.int/water_sanitation_health/dwq/chemicals/gdwq366_366a.p… (last visited 2
Dec. 2010).
60 115
human health risks, the USDA recommended that Colombia cease using it . In
the Counter-Memorial, Colombia does not attempt to explain the USDA’s
statement that formaldehyde is or was included in the spray mixture. Indeed,
Colombia’s Dobson Report confirms that an unnamed “preservativeŽ is “addedŽ
to the spray mixture and notes that “[p]reservatives may also play a significant
116
role in specific effects on humansŽ . Although the Dobson Report does not
117
identify this “preservativeŽ, it observes that it could be “formaldehydeŽ .
Colombia is so secretive about what it puts into its spray mixture that even its
own experts, hired to provide scientific support for the aerial spraying
programme, are left to guess about some of the ingredients.
2.61 In addition to dioxin, 1,4 Dioxane and formaldehyde, there is evidence
that other chemicals are added as well. In that regard, the U.S. Government’s
Congressional Research Service reported, based on statements by the U.S.
Department of State, that the spray includes an “anti-foaming additiveŽ called
118
COSMO IN D . This is especially troubling because the Colombian Ministry
of Health describes COSMO IN D as “extremely toxic due to its severe eye
115
EM, Chap. 5, para. 5.28.
116
Dobson Report, op. cit., pp. 521, 546. CCM, Vol. I, Appendix.
117Ibid.
118United States Congressional Research Service, Andean Regional Initiative (ARI): FY2002
Assistance for Colombia and Neighbors, p. 25 (14 Dec. 2001). ER, Vol. III, Annex 43.
61irritation effectsŽ19. To the same effect, a separate report by the USDA states
that Colombia includes a “second additive to minimize foaming of the spray
120
tankŽ. The report does not identify this anti-foaming agent .
*
2.62 In sum, the Counter-Memorial misrepresents both the contents of
Colombia’s aerial spray mixture, and its toxicity to humans, animals, plants and
the environment. Colombia’s assertion that the chemical contents of the spray
mixture are “publicly knownŽ would be laughable, if it were not for the
dangerously high level of toxicity of the mixture and the serious risks of harm it
poses (and has caused) to people, animals, plants and the environment in
Ecuador. Colombia’s own description of the mixture’s contents, and resulting
toxicity, bears scant resemblance to the truth.
2.63 As shown, Colombia sprayed Roundup Export, a highly toxic herbicide
formulation that the U.S. EPA determined had to be discontinued given its
propensity to cause permanent eye damage. During the same time period that
Roundup Export was used, Colombia represented to Ecuador that it was using
another product, Roundup Ultra, which it likewise fails to mention in the
119
Toxicological Opinion Nº 0685, regarding the toxicological classification of the mix
Glyphosate + POEA + Cosmo-Flux (1%), Colombian Health Ministry, 8 Oct. 2001. CCM, Vol.
II, Annex 44.
120
Charles S. Helling & Mary J. Camp, United States Department of Agriculture, Verifying Coca
Eradication Effectiveness in Colombia, pp. 10-11 (date unknown). EM, Vol. III, Annex 160.
62Counter-Memorial, and which is also highly toxic. The products that Colombia
admits to using … Roundup SL and GLY-41 … are no better: among other safety
directives, their labels expressly warn that these products cause eye irritation and
instruct against allowing inhalation or contact with human skin. Worse yet, the
spray mixture contains other dangerous chemicals, including POEA and Cosmo-
Flux, as well as others that the Counter-Memorial fails to acknowledge, including
unidentified anti-foaming agents and preservatives. All of these chemicals are
widely recognized as harmful to human health and the environment, including by
Colombia’s own experts.
2.64 The following section addresses the reckless and irresponsible manner in
which Colombia has sprayed this secret and highly toxic chemical stew …
including on people, animals, plants, forests and water bodies in Ecuador.
Section II. Colombia’s Misrepresentations Regarding Spraying Operations
2.65 The Counter-Memorial relies heavily on the alleged rigor with which
Colombia supposedly enforces the spray programme’s operational parameters.
Colombia claims that, by its full and faithful compliance with these requirements
… which strictly regulate the manner in which spraying operations are allegedly
conducted, including flight speed, altitude, spray application rate, droplet size,
time of day, and other parameters … it prevents any spray from reaching Ecuador,
63making it impossible for Ecuador to have been harmed by the aerial spraying
programme. Apart from denying the capacity of the spray itself to cause harm (as
addressed in the previous section), Colombia’s defense in this case rests on its
argument that spraying operations are conducted in a scrupulously careful manner
that precludes any possibility that the spray mixture reaches across the border into
Ecuador or affects people, animals, plants or the environment there. In one of the
many places where Colombia makes this argument, paragraph 4.69 states that
Colombia “set minimum and maximum figures in the Environmental
Management Plan for the parameters upon which drift is contingent, with the
purpose of reducing it as much as possible. These parameters are strictly
121
observed by the personnel involved in spraying operationsŽ . Based on
Colombia’s purportedly rigorous compliance with these “strict technical
parametersŽ and the resultant reduced drift, the Counter-Memorial concludes that
122
“no damage could have occurred in Ecuadorian territoryŽ .
2.66 For this self-serving conclusion, Colombia relies on a report from its very
own National Narcotics Directorate (“DNEŽ), dated 4 February 2010, that was
obviously prepared for this litigation. This “unbiasedŽ report claims the aerial
spraying is a “highly technical program that is carried out pursuant to the strictest
parameters in order to ensure that its implementation poses no risks to human
121CCM, Chap. 4, para. 4.69 (emphasis added); see also CCM, Chap. 7, para. 7.17.
122Ibid., Chap. 7, para. 7.5.
64 123
health or the environmentŽ . It assures that the programme “complies with all
the environmental parameters set out for its implementation and to assess impacts
on the environment, human health and farming activitiesŽ 124. Unsurprisingly, a
similar for-purposes-of-litigation report (also dated 4 February 2010) has been
provided by Colombia’s equally “unbiasedŽ Anti-Narcotics Direction of the
Colombian National Police (“DIRANŽ). This report, too, declares that “the
operational parameters foreseen in the Environmental Management Plan, such as
flight altitude, maximum wind speed and herbicide dosage, are fully observedŽ 125.
2.67 The adequacy of Colombia’s “Environmental Management PlanŽ
(“EMPŽ) and its “technical parametersŽ for preventing spray drift and avoiding
harm in Ecuador are addressed in Chapter 4. In that Chapter, Ecuador shows,
inter alia, that Colombia’s EMP, upon which so much of its argument stands, was
prepared without the benefit of an environmental impact assessment, violates
Colombia’s own environmental laws, and sets standards for controlling drift in
aerial spraying operations that are far more permissive than those in the rest of the
world. In this Chapter, Ecuador shows that, notwithstanding the self-serving
reports recently prepared by the anti-narcotics agencies responsible for execution
123Report by the National Narcotics Directorate (DNE), 4 Feb. 2010. CCM, Vol. II, Annex 66.
See also ibid. (“The program is implemented in compliance with the legislation in force and with
standardized protocols and procedures; therefore, it is carried out exactly the same way regarding
operational parameters, dose and ingredients in the spray mixture, etc. All over the national
territory where illicit crops are detected, including the border zone with EcuadorŽ.).
124
Ibid., Appendix.
125
DIRAN Report, op. cit., para. 2.2.3 (emphasis added). CCM, Vol. II, Annex 67.
65of the aerial spraying programme, Colombia failed to meet even the excessively
low standards set in the EMP for carrying out its aerial spraying operations: data
contemporaneously recorded by the spray planes themselves show that literally
on tens of thousands of flights along or near the Ecuadorian border the pilots
employed by Colombia recklessly and by a wide margin violated all of the
conditions required by the EMP to prevent spray drift, virtually assuring that it
would reach into Ecuador.
2.68 The Parties are in agreement on the factors that, if not adequately
controlled, cause aerial spray to drift, including over long distances. These
factors are, among others, the speed of the aircraft; the altitude at which the spray
is released; the spray application rate (in litres per hectare); the size of the spray
droplets; the time of day; and meteorological conditions, including wind speed
and direction, temperature and humidity. Ecuador agrees with Colombia that:
“spray drift depends essentially on wind speed and direction, as
well as on a number of other atmospheric factors including
temperature, relative humidity and atmospheric stability. It is also
dependent on the altitude at which spraying takes place and the air
speed of the spraying aircraft, as well as the calibration of the
spraying equipment, the de126ty of the spray mix and the initial
size of the spray dropletsŽ .
2.69 The Counter-Memorial asserts that Colombia has taken all of these factors
into account and has adopted strict regulations in regard to each factor.
126CCM, Chap. 7, para. 7.17.
66According to Colombia, there are “strict parametersŽ that set the “minimum and
maximum figuresŽ for each factor affecting “driftŽ, including those regulating
“aircraft speed, heightŽ, “wind speedŽ and “temperatureŽ 12. The Counter-
Memorial also claims that Colombia enforces operational limits for application
rate and droplet size, as well as a prohibition on spraying at night, when
temperature and other climatic conditions are more conducive to spray drift 128.
At paragraph 4.34, Colombia calls the EMP, which contains the operational
parameters for the programme, a “set of rules and procedures that must be
129
followed and observedŽ .
2.70 Paragraph 4.74 of the Counter-Memorial reaffirms the mandatory nature
of the operational parameters set forth in the EMP, and specifically acknowledges
that they are necessary to ensure the protection of people, plants and animals in
Ecuador; Colombia states that it:
“enacted rules and standards governing the aerial spraying
program in order to ensure that standards relating to the protection
of human health and the environment were consistently followed
and applied throughout its territory. When aerial spraying
operations started in the zone near the border with Ecuador, the
same rules were applied with no modifications. These rules were
127
Ibid., Chaps. 7 and 8, paras. 7.32, 8.60.
128
Ibid., Chaps. 4 and 7, paras. 4.62, 7.27; Resolution Nº1054 of 30 September 2003 of the
Ministry for the Environment of Colombia, p. 173. CCM, Vol. II, Annex 50.
129CCM, Chap. 4, para. 4.23 (referencing Environmental Management Plan in Resolution Nº
1054). Resolution Nº 1054 of 30 September 2003 of the Ministry for the Environment of
Colombia. CCM, Vol. II, Annex 50.
67 embodied in the Environmental Management Plan that governs
130
eradication program up to the present timeŽ .
2.71 The Counter-Memorial claims that Colombia has not only promulgated
restrictions on the relevant operational parameters, but that it has strictly enforced
them as well. It claims to have rigorously monitored the spray programme to
assure that all flights are in compliance with the requirements of the EMP in all
131
respects . It claims to have regularly analysed the fight data automatically
generated by the spray planes (the same data subsequently obtained by Ecuador
and submitted with this Reply), and to have prepared quarterly and semi-annual
reports assessing compliance with all operational parameters 132. Although no
such reports are presented to the Court with the Counter-Memorial, Colombia
avers that every such report confirms that the aerial spraying has been conducted
in strict compliance with the requirements of the EMP. It admits to no deviations
from these standards; nor does it concede that any violations have occurred.
Paragraph 4.69 in the Counter-Memorial asserts that “[t]hese parameters are
strictly observed by the personnel involved in spraying operationsŽ 133.
130CCM, Chap. 4, para. 4.74.
131
See infra Chap. 2, paras. 2.75-2.80.
132
Ibid., Chap. 2, paras. 2.81-2.82.
133CCM, Chap. 4, para. 4.69 (emphasis added). See also ibid., Chap. 7, para. 7.17 (“The PECIG’s
Environmental Management Plan has taken into account all these factors and set minimum and
maximum figures for the parameters upon which drift is contingent, with the purpose of reducing
it as much as possible . . . . These parameters are strictly observed by the personnel involved in
spraying operationsŽ.).
682.72 As shown below, these representations by Colombia about its execution of
the aerial spraying programme are impossible to reconcile with the flight data
recorded by the spray planes themselves, which were furnished not only to
Colombia but also to the U. S. Department of State by the private U.S. company
that conducts the spray flights on behalf of Colombia (with financing from the
U.S. government). As indicated previously, it is from the U.S. Department of
State that Ecuador obtained all of the raw data generated by the spray planes
134
during flight . They show unequivocally that Colombia has grossly
misrepresented the facts concerning the conduct of the spray programme. Indeed,
the Counter-Memorial is wrong in all material respects concerning the spray
flights, including especially whether they have complied with the requirements to
prevent spray drift, such as: maximum flight speed; maximum altitude for
dispersion of spray; maximum application rate; minimum droplet size; and
prohibition of night time spraying. Failure to comply with any one of these
requirements increases the potential for spray to reach Ecuador and cause harm
there. But, contrary to the representations in the Counter-Memorial, the evidence
shows that Colombia has failed to comply with all of them … on literally tens of
thousands of spray flights along or near the border with Ecuador.
134Supra Chap. 2, para. 2.7.
692.73 Specifically, the evidence … described in more detail in the following
pages … shows that, in regard to spray flights within 10 kilometres of border with
Ecuador between 2000 and 2008, there were:
▯ At least 75,841 flights when spray was dispersed at speeds higher
than the 165 mile per hour (“mphŽ) limit asserted in Colombia’s
Counter-Memorial.
▯ At least 16,143 flights when spray was dispersed at altitudes
higher than the 50 metre limit set in the EMP.
▯ At least 27,429 flights when the spray application rate was above
the 23.65 litre per hectare limit allowed by the EMP.
▯ At least 24,540 flights at night, contrary to the Counter-
Memorial’s claim that spraying only occurs during the daytime.
A. THE COUNTER -MEMORIAL’SC LAIMSR EGARDING COLOMBIA ’S
C OMPLIANCE W ITHO PERATIONALR EQUIREMENTS TO PREVENT SPRAY D RIFT
2.74 Colombia states in the Counter-Memorial that the “extentŽ of “spray driftŽ
is a “central issueŽ in this . Ecuador agrees: the measures that Colombia
has, or as the case may be, has not, taken to minimize spray drift are fundamental
to assessing Colombia’s international responsibility.
2.75 Colombia gives itself high marks for the prudence with which it claims to
have conducted the aerial spraying. For example, it avers that it has always taken
13CCM, Chap. 7, para. 7.16.
70“every careŽ to “ensure that spraying occurs only on Colombian territoryŽ and
that, as a result, no harm could be caused to Ecuador as a result of spray drift 136.
2.76 Colombia claims that it ensures against spray drift by strict compliance
with the operational requirements boasted of in the Counter-Memorial and set
forth in its EMP. Colombia fastens its sail in this case to the mast of these
requirements. Colombia’s compliance with them, according to the Counter-
Memorial, negates the possibility of spray drift into Ecuador. Colombia even
considers itself legally bound, under Colombia law, to comply with them.
Resolution 1054 of Colombia’s Ministry for the Environment, which approved
the EMP, describes it as having the same legal status as “environmental
regulationsŽ:
“Each and every activity and component of this program should be
set with within the context of current environmental regulations so
that constitutional assumptions can be complied with, which
impose a duty on the State to protect environmental diversity and
integrity, preserve areas of special ecological importance, 137
stimulate education so that these goals may be attainedŽ .
2.77 Colombia has long claimed that the safety of the aerial spraying
programme, including the prevention of drift, depends on strict compliance with
136Ibid., Chap. 7, paras. 7.5, 7.16.
137Resolution Nº 1054 of 30 September 2003 of the Ministry for the Environment of Colombia.
CCM, Vol. II, Annex 50. See also Claudia Rojas Quiñonez, Esq., The Aerial Spray Program and
Violations of Colombia’s Domestic Laws Regarding the Environment and the Rights of
Indigenous Peoples, paras. 15-17, 39, 78, 84, 89-90, and 124 (Jan. 2011) (discussing Colombia’s
legal obligation to comply with the spray programme’s Environmental Management Plan). ER,
Vol. II, Annex 8.
71the operational requirements/regulations. As early as 1993, the Colombian
Health Minister informed the Director of Colombia’s DNE of the critical
importance of complying with the spraying programme’s operational
138
parameters . This view was repeated by the General Manager of Colombia’s
National Institute for Renewable Natural Resources and the Environment, who
emphasized the “importance that must be given to compliance with the specific
and technical parametersŽ 139.
2.78 Colombia’s Agriculture Institute, which is charged with the regulation of
the chemicals in the spray mixture, reiterated the need to comply with the
operational parameters in January 2003, in correspondence with the Director of
Colombia’s DNE. The Director was reminded that “in carrying out these
sprayings all technical measures continue to be applied regarding nozzles
calibration, proper mixture preparationŽ and that “[t]hese sprayings shall be
carried out in accordance with environmental parameters for this type of
sprayingŽ 140.
138Note from the Colombian Health Minister to the Director of the National Narcotics
Directorate, 11 Oct. 1993 (stating that operational parameters “must be keptŽ). CCM, Vol. II,
Annex 36.
139
Resolution No. 001 of 11 February 1994 of National Narcotics Council of Colombia. CCM,
Vol. II, Annex 37.
140Note No. 00500 from the Assistant Manager for Agricultural Protection and Regulation of the
Colombian Agriculture and Livestock Institute to the Director of the National Narcotics
Directorate, 28 Jan. 2003. CCM, Vol. II, Annex 47.
722.79 The highest administrative tribunal in Colombia, the Council of State,
issued an order in 2004 requiring the Colombian authorities to observe strict
compliance with the requirements of the EMP in carrying out the aerial spraying
programme, as well as the resolutions of the Ministry of the Environment
regarding the programme, allowing “not even the slightest deviationŽ:
“the guidelines stated by the environmental authorities should be
followed when illicit crops are being sprayed, and not even the
slightest deviation from these should be permitted, which means
that it is therefore necessary for permanent controls to be
undertaken, with continuous evaluations, of any effects which
141
might begin to appearŽ .
2.80 The Counter-Memorial describes the monitoring programme Colombia
has put in place to assure that all aerial spraying is carried out in strict accordance
with the operational requirements of the EMP, as ordered by the Council of State.
The monitoring programme includes elaborate and sophisticated data collection
techniques for obtaining the data from each and every flight pertaining to the
operational requirements. Paragraph 7.172 of the Counter-Memorial states that:
“All spray aircraft are equipped with a satellite monitoring system
which guarantees the accuracy of the operations and ensures that
the sprayings are carried out on the areas targeted. Wind
conditions are constantly monitored by the aircraft and if they are
not within the parameters allowed, the mission is annulled or
postponed. Each operation is recorded in detail, including the
place, hour, number of hectares sprayed, spraying locations, etc.
141State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca (19 Oct. 2004) (emphasis added). ER, Vol. V, Annex
151.
73 The amount of products used in the operation (glyphosate and
142
adjuvant) is also stated in the operational recordsŽ .
According to Colombia, spray events are further documented in detailed post-
operational reporting:
“Once daily spraying operations are finished, a detailed report of
the day’s operations is prepared on the basis of the computerized
system which records each spraying operation with its respective
route, geo-referenced areas of application and the amount of spray
mix released per minute. This allows verification of the location
of the places where the operations took place and quantification of
the hectares sprayed. A record is signed by the Base Commander
143
and the personnel involved in the operationŽ .
2.81 The Counter-Memorial describes how the data collected by these means
are reported, reviewed and analyzed. It says, for example, at paragraph 4.26 that
the “Ministry for the Environment oversees the implementation of the EMP and
verifies compliance with the guidelines and duties foreseen in itŽ, and that “[t]wo
reports per year are submitted to the Ministry, which may issue rulings on the
activities carried out by the agencies involved in the implementation of the [aerial
144
spraying] programmeŽ . Further, according to the Counter-Memorial, the data
regarding the spray programme are reviewed by “an external technical audit,
contracted yearly through public tender with resources provided by the National
Narcotics CouncilŽ that “submits quarterly reports of its evaluations and
142CCM, Chap. 7, para. 7.172.
143Ibid., Chap. 4, para. 4.64.
144
Ibid., Chap. 4, para. 4.26.
74assessments, with recommendations for any outstanding corrective actions to be
145
takenŽ .
2.82 Notably, Colombia did not provide the Court with any of the data
regarding the execution of the aerial spray programme that the Counter-Memorial
says has been recorded. Nor did Colombia submit to the Court any of the
evaluative reports or audits that it claims the data have been subjected to (i.e., the
twice-yearly reports to the Ministry of the Environment, or the quarterly reports
of the technical auditors contracted by the National Narcotics Council). These
are conspicuous omissions. Colombia, in essence, asks the Court to take it on
faith that the aerial spraying programme has been consistently carried out along
the border with Ecuador since 2000 in strict compliance with the operational
requirements of the EMP, and that the data collected about the spray flights and
the evaluative reports all confirm this. Where are these data and reports? The
Court’s case law has made clear that a State’s failure to present records in its
possession to substantiate factual claims, particularly when they relate to alleged
compliance with legal requirements, warrants a cautious approach to these claims
since “[a] public authority is generally able to demonstrate that it has followed the
appropriate procedures and applied the guarantees required by law … if such was
145Ibid., Chap. 4, paras. 4.28-4.29. See also ibid., para. 4.74 (“the program is overseen by a
permanent external auditŽ).
75the case … by producing documentary evidence of the actions that were carried
outŽ14.
B. E CUADOR ’S RECEIPT OFD ATA FROM THE G OVERNMENT OF THE U NITED
S TATES REGARDING C OLOMBIA ’SA ERIAL SPRAYING P ROGRAMME
2.83 Just as Colombia failed to provide the Court with the data recorded by its
spray planes, or with its quarterly and semi-annual reports on compliance with the
operational requirements of the EMP, it also refused to provide the information
147
and documents to Ecuador, despite repeated requests over several years .
Colombia’s persistent refusals led Ecuador to look elsewhere for the data on the
spray flights. Ecuador turned to the only other known source for these data: the
U.S. Department of State. It is public knowledge that the United States provides
financial and operational support for Colombia’s aerial spraying programme,
including the purchase of chemicals and aircraft, and by contracting with
DynCorp International LLC to provide the pilots and ground personnel to carry
out the spray missions and service the equipment8. As part of its cooperation
with Colombia, the State Department receives the data recorded by the spray
planes. A State Department Report from 2002 explains:
146Case Concerning Ahmadou Sadio Diallo (Republic of Guinea v. Democratic Republic of the
Congo), Judgment, I.C.J. Reports 2010, p. 24, para. 65.
147
See e.g., EM, Chap. 3, paras. 3.1-3.3, 3.9, 3.17, 3.21, 3.28-3.30.
148
See CCM, Chap. 1, para. 1.35, 4.63; DIRAN Report, op. cit., p. 301. CCM, Vol. II, Annex 67.
76 “Onboard computer and digital global positioning systems
(D/GPS)-driven equipment (SATLOC and Del Norte)
automatically record each aircraft’s actual flight parameters,
including differential-GPS track, airspeed, altitude (mean sea
level), application rate, and precise geographic location (longitude
and latitude coordinates) at the time of aspersionŽ 14.
2.84 In a formal request under the U.S. Freedom of Information Act 150made on
3 April 2009, Ecuador sought from the U.S. Department of State “records of
aircraft flight parameters recorded during missions to eradicate illicit coca in
ColombiaŽ for “aerial spraying operations conducted in the Putumayo and Nariño
Departments of Colombia within 20 kilometres of the international border with
EcuadorŽ 151. In a response dated 13 November 2009, the Department of State
provided 11 documents from its International Narcotics and Law Enforcement
Affairs Bureau (a set of ten maps depicting the location of spray events between
2000 and 2008, and a set of screen shots from the data recording system), and a
152
computer disk containing “the detailed underlying data to the documentsŽ . In
153
response to a follow-up request , on 12 March 2010 the U.S. Department of
149
Chemicals Used, op. cit., p. 3. EM, Vol. III, Annex 144.
150United States, Freedom of Information Act, 55 U.S.C. § 552.
151Letter from Rebecca L. Puskas, Counsel to Government of Ecuador to United States
Department of State Office of Information Programs and Services (3 Apr. 2009). ER, Vol. IV,
Annex 113.
152Letter from Margaret P. Grafeld, Director, Office of Information Programs and Services,
United States Department of State to Rebecca L. Puskas, Counsel to Government of Ecuador (13
Nov. 2009). ER, Vol. IV, Annex 114. A copy of the data CD obtained by Ecuador from the U.S.
Department of State has been deposited with the Registry.
153
Letter from Rebecca L. Puskas, Counsel to Government of Ecuador to A. Harold (Hal) Eisner,
United States Department of State Office of Information Programs and Services (19 Feb. 2010).
ER, Vol. IV, Annex 115.
77State provided a description of each parameter reflected in the flight data and the
154
units associated with the data .
2.85 The data that Ecuador received from the U.S. Department of State provide
information about 247,977 spray events within 20 kilometres of Ecuador that
were flown between 2000 and 2008, 114,525 of which occurred within 10
155
kilometres of Ecuador . During this time, Colombia’s spray planes deposited a
total of 326,658 gallons (1,236,535 litres) of the toxic herbicide spray mixture
within 10 kilometres of Ecuador’s border 156.
2.86 In light of the Counter-Memorial’s failure to present any information
about the location and conditions of aerial spraying near Ecuador’s border, it falls
to Ecuador to bring this evidence to the Court’s attention. As discussed below,
the data make a mockery of Colombia’s assertions about the prudence with which
it allegedly conducts the aerial spraying programme, and the supposed “strict
complianceŽ with the operational requirements of the EMP. As recorded by the
154
Fax from A. Harold Eisner, Office of Information Programs and Services, United States
Department of State to Rebecca L. Puskas, Counsel to Government of Ecuador (12 Mar. 2010).
ER, Vol. IV, Annex 116.
155R. John Hansman, Ph.D. & Carlos F. Mena, Ph.D., Analysis of Aerial Eradication Spray
Events in the Vicinity of the Border Between Colombia and Ecuador from 2000 to 2008, p. 11
(Jan. 2011) (hereinafter “Hansman & Mena ReportŽ). ER, Vol. II, Annex 1. Ecuador received
information regarding flight paths located within 20 kilometres of the international border but has
focused its analysis on the flight paths within 10 kilometres of the frontier. Ibid. As discussed in
the Hansman & Mena Report, the records represent aircraft ground tracks when chemicals were
being sprayed. Ibid., p. 6. Ecuador has not obtained information regarding the flight paths of
Colombia’s spray planes when the chemicals were not being sprayed.
156Hansman & Mena Report, op. cit., p. 11. ER, Vol. II, Annex 1.
78spray planes, the data thoroughly contradict the representations about the aerial
spraying programme provided in the Counter-Memorial, especially in relation to
Colombia’s alleged compliance with the operational requirements in regard to
flight speed, altitude of spray release, application rate, droplet size, time of day,
and other pertinent factors that affect spray drift, including type of spray planes
used, competence of pilots, and avoidance of no-spray zones and buffer zones
intended to protect Ecuador and sensitive areas. In sum, what the data show are
pervasive violations of every requirement designed to prevent spray drift.
C. AIRCRAFT SPEED AND H EIGHT OFS PRAY RELEASE
2.87 Among the most important factors that influence spray drift are the speed
the aircraft is travelling and the height at which the spray is released. The Parties
agree that the faster a plane is travelling, and the higher a spray mixture is
released, the more likely it is to drift long distances.
1. Speed
2.88 Ecuador and Colombia agree that aircraft speed is a principal factor that
determines whether there will be long-distance drift. In that regard, Colombia
79acknowledges in paragraph 4.68 of the Counter-Memorial that “drift dependsŽ on
157
the “air speed of the spraying aircraftŽ .
2.89 Colombia further asserts in the Counter-Memorial that to prevent long-
range drift, its planes are forbidden from spraying while travelling in excess of
165 miles per hour (265 km/hr). For instance, Colombia states at paragraph 7.32
that its spray planes fly at a “maximum operation speed of 165 miles per hourŽ 158.
2.90 In its submission to the Inter-American Commission on Human Rights
(IACHR), Colombia claimed an even more restrictive speed limit, informing the
Commission that its aircraft are barred from flying faster than 140 mph (225
159
km/hr) .
2.91 Colombia’s representations … that its spray planes never fly faster than
165 mph (as it asserts in the Counter-Memorial), or 140 mph (as it claimed to the
IACHR) … are false. In fact, Colombia’s aircraft routinely exceed these speed
limits by huge margins. This is confirmed, for example, in the testimony of a
pilot with extensive experience flying spray missions for Colombia, who testified
157
CCM, Chap. 4, para. 4.68. See also ibid., Chap. 7, para. 7.17.
158CCM, Chap. 7, para. 7.32.
159Diplomatic Note No. DDH 58003 from the Colombian Foreign Ministry to the Executive
Secretary of the Inter-American Commission on Human Rights, 18 Sept. 2005, p. 26. ER, Vol. V,
Annex 154.
80that the aircraft travel “hundreds of miles per hourŽ 160. And it is corroborated by
reports authored by the U.S. Department of State that evaluate the aerial spraying
programme, which repeatedly criticize the “pilotsŽ for “flying too fastŽ on “spray
161 162
passesŽ and “faster than prescribed airspeedsŽ .
2.92 The State Department’s observation that the pilots fly “too fastŽ is well
warranted. Colombia’s aircraft routinely spray near the border with Ecuador
while flying much faster than 165 mph (the limit asserted in the Counter-
Memorial), let alone 140 mph (the limit that Colombia represented to the
IACHR).
2.93 In fact, according to the spray operation data, Colombia’s planes sprayed
toxic herbicides within 10 kilometres of the Ecuadorian border while travelling
163
faster than 165 mph on at least 75,841 occasions between 2000 and 2008 . That
164
is over 69 percent of all spray flights in the region . The pervasive violations
of the speed limit near the border are illustrated in Figure 2.4, which depicts
160
United States District Court of the District of Colombia, Arias, et al. v. Dyncorp, et al.,
Quinteros, et al. v. Dyncorp, et al., Declaration of Redacted Witness (2 June 2010). ER, Vol. IV,
Annex 117.
161Memorandum from Tim Doty, COR, INL/RM/AS to Dyncorp, PSD Manager, p. 3 (28 May
1997). ER, Vol. III, Annex 37.
162United States Department of State, Evaluation Summary Technical Operations (Undated). ER,
Vol. III, Annex 50.
163Hansman & Mena Report, op. cit., p. 20. ER, Vol. II, Annex 1.
164Ibid.
81violations along Ecuador’s Sucumbíos Province. Indeed, in 2002 alone,
Colombia’s planes sprayed while travelling faster than 165 mph no fewer than
165 166
19,594 times . In 2006, they did so 12,855 times .
2.94 Colombia’s violations of the 140 mph speed limit represented to the
IACHR were, of course, even worse. In fact, 108,563 flights … 98 percent of all
flights within 10 kilometres of Ecuador’s border between 2000 and 2008 …
167
exceeded 140 mph . When Colombia made that claim in September 2005, it
had already violated the limit over 75,000 times 168.
2.95 Nor were Colombia’s violations of the speed limit trivial. To the contrary,
they were often far above it. The Counter-Memorial describes 333 kilometres per
hour … equivalent to 207 miles per hour … as the “worst case scenarioŽ 169. In
other words, even though the aircraft were required to fly slower than the speed
limit (165 or 140 mph), Colombia claims that in no circumstances would it be
possible for them to fly faster than 207 mph. Assuming for a moment that
Colombia’s “worst case scenarioŽ is accurate, this is over 40 mph faster than the
165Ibid., Appendix 3, p. 14.
166
Ibid.
167
Hansman & Mena Report, op. cit., p. 20. ER, Vol. II, Annex 1.
168Ibid., Appendix 3, p. 27.
169CCM, Chap. 7, para. 7.25.
82speed limit asserted in the Counter-Memorial and over 65 mph faster than the one
represented to IACHR.
2.96 But the Counter-Memorial is wrong. In fact, the “worstŽ flying speed
within 10 kilometres of Ecuador’s border was 246.9 mph (or 397.3 km/hr), which
170
occurred in 2001 . That is 82 mph faster than the 165 mph speed limit claimed
in the Counter-Memorial and 107 mph faster than the one represented to the
IACHR. Flights in other years were not much slower. In 2000, planes sprayed
near the Ecuadorian border while travelling up to 237 mph 171. In 2002, they flew
as fast as 244 mph, and in 2004 they reached 231 mph 172. In 2005 and 2006,
planes flew up to 231 mph and 220 mph, respectively 173. Indeed, there were at
least 11,113 spray flights … 10 percent of the total … that were faster than what the
174
Counter-Memorial describes as the “worst case scenarioŽ . In 2002 alone, the
putative worst case was exceeded 5,992 times (16 percent of all flights within 10
175
kilometres of the border that year) . Figure 2.4 depicts spray lines flown faster
than the Counter-Memorial’s “worst caseŽ near Ecuador’s Sucumbíos Province.
170Hansman & Mena Report, op. cit., Appendix 3, p. 17. ER, Vol. II, Annex 1.
171Ibid.
172Ibid.
173Ibid. In 2003, Colombia’s planes reached 197 mph. In 2007 and 2008, they flew up to 213
and 185 mph, respectively. Ibid.
174Ibid., p. 20.
175
Hansman & Mena Report, op. cit., Appendix 3, p. 15. ER, Vol. II, Annex 1.
83 0°30'0"N
km
Figure 2.4
River
Putumayo
! Palma Seca
River
San Miguel
76°30'0"W !Chone 2Chone 1 76°30'0"W
!
!
Playeratal 0 5 10 15 20 25
!
±
Puertoo
Corazón Orense
Puert! MestanzaAgosto
Puerto EscondidoLojana
! !
C O L O M P u t u m a y o
!
! ! S u c u m b í o s
Monterrey
! E C U A D O R
Dios Peña
! Santa La Charapa
Marianita
! !
San Miguel
Farfán
General
!
La Condor
! !
0 - 114-615> 20707 !
10 de Agosto
Sucumbíos ProvSpeed (mph)- 2008) !
! San SFarannFcrisacnoci2sco 1
77°0'0"W 77°0'0"W
Salinas Virgen del Cisne
Flight Speed of Spray Events Within 10 Kilometres of Ecuador’s
Peru
River
San Miguel Colombia
International boundary
Protected areas international boundary
Ecuador
Legend
N a r i ñ o
Cofán Reserve
0°30'0"N2.97 Such violations of the speed limit have a dramatic impact on spray drift,
as Colombia itself concedes 176. The Hansman & Mena Report gives the technical
reasons for this: “[i]f the spray aircraft airspeed is too high, the droplets from the
spray nozzle will explode into much smaller droplets due to aerodynamic forces
177
as they hit the high relative windŽ . This principle, depicted graphically in
Figure 2.5, has also been recognized by regulatory authorities, including
Colombia’s own Agriculture Institute, which explained in a 1999 technical report
that “emission of drops of small and medium size, coming from the spraying
nozzlesŽ which is “influenced by a high speed of operation and turbulence, cause
larger breakage of dropsŽ 178. Likewise, Australia’s Operating Principles in
Relation to Spray Drift Risk recognize that “high airspeeds can cause excessive
fragmentation of droplets delivered by the nozzle system when the droplets are
179
impacted by fast moving air flowing relative to the wing boomŽ . This will
180
“shatterŽ the “dropletsŽ, making them “more drift proneŽ . The U.S. EPA
likewise explains that “[l]arge droplets released into the turbulence created by an
aircraft travelling in excess of 120 mph tend to break into smaller more driftable
176
See CCM, Chaps. 4 and 7, paras. 4.61, 4.68, 7.17-7.18, 7.32; CCM, Appendix, p. 522-523;
CCM, Annex 131-B.
177
Hansman & Mena Report, op. cit., p. 20, n. 7. ER, Vol. II, Annex 1; see also Durham K.
Giles, Ph.D., Spray Drift Modeling of Conditions of Application for Coca Crops in Colombia, pp.
6, 20 (Jan. 2011) (hereinafter “Giles ReportŽ). ER, Vol. II, Annex 2.
178Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 599, p. 13 (23 Dec. 1999). ER, Vol. V, Annex 132.
179 Australian Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating
Principles in Relation to Spray Drift Risk, p. 20 (15 July 2008). ER, Vol. III, Annex 22.
180Ibid., p. 9.
84dropletsŽ due to “shearing effectsŽ caused by high airspeeds 181. Over 110,000
spray flights were flown at speeds faster than the speed the U.S. EPA said could
cause “driftable dropletsŽ 182.
183
Figure 2.5. Breakup of Spray Droplets as They Encounter Wind Turbulence
181EPA 2002 Analysis, op. cit., p. 34. ER, Vol. III, Annex 45.
182Hansman & Mena Report, op. cit., Appendix 3, p. 16. ER, Vol. II, Annex 1.
183Giles Report, op. cit., p. 6. ER, Vol. II, Annex 2.
85 2. Height
2.98 Flying too fast is not the only factor that Ecuador and Colombia concur
increases spray drift. The Parties also agree that a spray mixture is prone to drift
longer distances if it is released at too high an altitude. As Colombia states at
paragraph 4.68 of the Counter-Memorial, “[s]pray driftŽ is “dependent on the
184
altitude at which spraying takes placeŽ . However, just as Colombia
misrepresents its compliance with restrictions on aircraft speed, the Counter-
Memorial also falsely claims that Colombia has complied with the EMP’s
requirements regarding the altitude from which the spray is released.
2.99 On paper at least, Colombia has imposed limits regarding the height from
which the spray mixture may be released, in order to prevent it from drifting. For
example, Colombia informed Ecuador in April 2004, that is, over four years into
the programme, that the “[h]ighest release heightŽ is “25 meters in compliance
with technical parametersŽ 185. Colombia said the same thing in September 2005
when it represented to the Inter-American Commission on Human Rights that the
186
“flight altitudeŽ for its “spraying operationsŽ is “not above 25 metresŽ . By the
184CCM, Chap. 4, para. 4.88. See also ibid., Chap. 7, para. 7.17.
185Note No. SARE -142 from the Director of the National Narcotics Directorate of Colombia to
the President of the Scientific and Technical Commission of Ecuador, 14 Apr. 2004. CCM, Vol.
II, Annex 13.
186Diplomatic Note No. DDH 58003 from the Colombia Foreign Ministry to the Executive
Secretary of the Inter-American Commission on Human Rights, p. 52, 18 Sep. 2005. CCM, Vol.
II, Annex 19.
86time Colombia made that representation, spraying near the Ecuadorian border had
187
been ongoing for over five-and-a-half years .
2.100 Colombia appears to have recognized that it was on shaky ground in
claiming that it complies with the height limitation. Thus, notwithstanding its
representations to Ecuador and the IACHR that its spray planes were forbidden
from spraying above 25 metres, Colombia’s current EMP doubles the height limit
to 50 metres so that it can more comfortably claim that the standard is not
exceeded. 188 In the Counter-Memorial, Colombia states that its aerial spraying
“usuallyŽ takes place at a height of 30 metres 189. No data are presented, however,
in support of this representation, or any of its other claims about the altitude at
which it releases the herbicide.
2.101 Yet again, Colombia’s representations are contradicted by the evidence.
The data collected by the spray planes demonstrate that Colombia routinely
exceeds the allowable altitude for releasing the chemical spray, regardless of
187 Moreover, at the time Colombia made these representations, it had an Environmental
Management Plan in place … since September 30, 2003 … which provided that the maximum
application height was 50 metres. Resolution Nº 1054 of 30 September 2003 of the Ministry for
the Environment of Colombia, p. 173. CCM, Vol. II, Annex 50. Thus, Colombia was allowing
its spray planes to operate twice as high as it admitted to Ecuador and the IACHR.
188
Resolution No. 1054 of 30 September 2003 of the Ministry of Environment of Colombia.
CCM, Vol. II, Annex 50, p. 173.
189
CCM, Chap. 4, para. 4.62 (“The Environmental Management Plan foresees a maximum flight
altitude of 50 meters when spraying, subject to geographical features or obstacles so as to avoid
risks to the pilotsŽ.). See also CCM, Chap. 7, para. 7.32 (“aircraft fly at low altitudes, usually 30
metersŽ).
87whether the permitted ceiling is 25 metres, as Colombia represented to Ecuador
and the IACHR, or 50 metres, as Colombia claims in the Counter-Memorial. As
the U.S. Department of State concluded in one of its evaluations, Colombia’s
“pilotsŽ are “flying . . . too high on spray passesŽ 190. The USDA came to the
same conclusion after conducting a field verification mission in 2001, reporting
191
“decreased efficiencyŽ due to “sprayingŽ at “too high an altitudeŽ .
2.102 The U.S. Departments of State and Agriculture were right: Colombia’s
planes do fly “too highŽ and at “too high an altitudeŽ. Indeed, they frequently
spray at altitudes far above the prescribed limit. In fact, nearly all of Colombia’s
spraying near the Ecuadorian border … 96 percent … has been done in violation of
the 25 metre height restriction it once told Ecuador and the IACHR was the
192 193
allowable limit . The total number of flights in excess of 25 metres is 89,124 .
In 2002 alone, Colombia released the chemical spray at altitudes higher than 25
194
metres 37,293 times .
190Memorandum from Tim Doty, COR, INL/RM/AS to Dyncorp, PSD Manager, p. 3 (28 May
1997). ER, Vol. III, Annex 37.
191U.S. Department of Agriculture, Agricultural Research Service, Colombia Coca Verification
Mission April-May 2001, p. 2 (7 July 2001). ER, Vol. III, Annex 41.
192Hansman & Mena Report, op. cit., p. 19. ER, Vol. II, Annex 1.
193Ibid.
194
Ibid., Appendix 3, p. 5.
882.103 Nor were these violations of the 25 metre limit trivial. In fact, Colombia
released the chemical spray from a height over 50 metres, that is, double the 25
metre limit, at least 16,143 times between 2000 and 2008 195. This represents 17
percent of all spraying near the Ecuadorian border 196. The altitude violations in
197
Ecuador’s Sucumbíos Province are depicted in Figure 2.6 .
2.104 As noted above, the Counter-Memorial acknowledges that releasing the
spray mixture at an excessive height increases drift 198. The Giles Report
explains:
“A critical parameter in the potential for spray drift is the aircraft
altitude or height of the application above the underlying canopy.
When spray droplets are released, they must travel from the point
of release downward to the intended target. Any cross wind that
affects the droplets during their downward trajectory will displace
them downwind. As the height of release increases, the travel time
195Hansman & Mena Report, op. cit., p. 19. ER, Vol. II, Annex 1. For example, in 2000,
Colombia sprayed the chemical herbicide at heights that reached 196 metres. Hansman & Mena
Report, op. cit., Appendix 3, p. 8. ER, Vol. II, Annex 1. That is more than 7 times higher than the
25 metre height restriction that Colombia claimed to Ecuador and the IACHR, and more than 3
times the 50 metre limit claimed in the Counter-Memorial. There were many spray missions that
year that flew far in excess of the height restriction. For example, at least nine other flights
exceeded 170 metres in altitude. Ibid. In 2006, more than 10 flights were higher than 120 metres.
Ibid. 2007 was even worse: Colombia sprayed as high as 170 metres, and at least 10 flights were
over 125 metres. Ibid.
196
Ibid., p. 19.
197Hansman & Mena Report, op. cit., pp. 7-10. ER, Vol. II, Annex 1.
198CCM, Chaps. 4 and 7, paras. 4.68, 7.17, 7.32, Appendix, paras. 22-25; A.J. Hewitt et al.,
“Spray Droplet Size, Drift Potential, and Risks to Nontarget Organisms from Aerially Applied
Glyphosate for Coca Control in ColombiaŽ, in Journal of Toxicology and Environmental Health,
Part A, 72:930-936 (2009) (hereinafter “Hewitt et al., 2009Ž). CCM, Vol. II, Annex 131-B.
89 0°30'0"N
km
Figure 2.6
River
Putumayo
! Palma Seca
River
San Miguel
76°30'0"W !Chone 2Chone 1 76°30'0"W
!
!
Playeratal 0 5 10 15 20 25
!
±
Puertoo
Corazón Orense
Puert! MestanzaAgosto
Puerto EscondidoLojana
! !
C O L O M P u t u m a y o
!
! ! S u c u m b í o s
Monterrey
! E C U A D O R
Dios Peña
! Santa La Charapa
Marianita
!
San Miguel
Farfán
General
!
La Condor
! !
0 - 55 - 5>050 !
10 de Agosto
Sucumbíos Province (2000 - 2008) !
Altitude (meters)
! San SFarannFcrisacnoci2sco 1
77°0'0"W 77°0'0"W
Salinas Virgen del Cisne
Altitude of Spray Events Within 10 Kilometres of Ecuador’s
Peru
River
San Miguel Colombia
International boundary
Protected areas international boundary
Ecuador
Legend
N a r i ñ o
Cofán Reserve
0°30'0"N of the droplets increases correspondingly, allowing a greater
horizontal displacement of the droplets to occurŽ 19.
2.105 This is understood by regulatory authorities the world over. For example,
Australia’s Operating Principles in Relation to Spray Drift Risk observe that
“[s]pray release heightŽ is one of the “major factors affecting spray drift risk: the
higher the release height, the greater the potential for off target driftŽ 20.
2.106 In fact, few if any aerial spraying programmes have ever been conducted
at the heights flown by Colombia’s spray planes. Typically, aerial spraying of
pesticides for agricultural purposes is performed only a few metres above crop
level, to avoid drift away from the target site 201. In a 2004 study commissioned
by Colombia (which is not cited in the Counter-Memorial), Colombia’s own
technical consultants acknowledged after reviewing the available literature on
spray drift, that “there is no information regarding aerial spraying of illicit crops
from more than 20 meters of altitudeŽ; therefore, “the corresponding technical
parameters cannot be used as applicable references for the spraying done in
199
Giles Report, op. cit., p. 16. ER, Vol. II, Annex 2; see also Hansman & Mena Report, op. cit.,
p. 18, n. 6 (“The altitude above ground level impacts spray drift. The higher the initial spray
application, the more time the spray has to drift during its descent to the groundŽ.). ER, Vol. II,
Annex 1
200 Australian Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating
Principles in Relation to Spray Drift Risk, p. 24 (15 July 2008). ER, Vol. III, Annex 22. Ibid., p.
9 (“higher release heights add to spray drift riskŽ).
201See infra Chap. 4, para. 4.108; Giles Report, op. cit., p. 16. ER, Vol. II, Annex 2.
90ColombiaŽ 202. Accordingly, Colombia has no experiential basis for concluding
that its height limit of 25 metres, let alone of 50 metres, is sufficient to prevent
spray drift.
2.107 The risks of extensive spray drift caused by releasing the spray mixture at
higher-than-allowed altitudes are, of course, multiplied when the spray is also
released at higher-than-allowed flight speeds 203. The data generated by the spray
planes and obtained from the U.S. Department of State show that Colombia’s
violation of both altitude and speed limits simultaneously has been a common
occurrence. More than 92 percent of all recorded flights between 2000 and 2008
… 85,364 separate spray events within 10 kilometres of Ecuador’s border …
exceeded the 140 mph speed limit and the 25 metre altitude restrictions
simultaneously 20. The more lenient restrictions of 165 mph for speed and 50
metres for altitude were also frequently breached in tandem: 12,155 times to be
exact, which represents 13 percent of all flights 205. Colombia’s exceedances of
the parameters for both height and speed, and the corresponding implications for
202 Las Palmas Ltda., Technical Department, Glyphosate (10,4 l/ha) and Three Different
Adjuvants, For Illicit Coca Crop (Erythoxylum spp.) Control, Agronomic Efficacy Testing of
Doses of Glyphosate in Illicit Crops: Final Report, p. 12 (July 2004) (hereinafter “Las Palmas
ReportŽ). ER, Vol. III, Annex 15.
203
Giles Report, op. cit., pp. 11-15. ER, Vol. II, Annex 2.
204
Hansman & Mena Report, op. cit., p. 21. ER, Vol. II, Annex 1.
205Ibid. These values represent the number of flights with recorded data for both parameters
(speed and altitude). Ibid.
91spray drift, in comparison to typical aerial spraying operations, are shown in the
schematic drawing found at Figure 2.7.
3. Reasons for Exceeding the Speed and Height Requirements
(a) Colombia’s Spray Planes Fly High and Fast To Evade Hostile Gunfire
From the Ground
2.108 In the Memorial, Ecuador pointed out that the pilots responsible for flying
Colombia’s spray planes have an incentive to exceed the required height and
speed limits. The reason is simple: narco-traffickers protecting their illicit crops
206
attempt to shoot them down . Moreover, variable topography and towering
rainforest trees present physical hazards for Colombia’s spray pilots 20. It is self-
evident that these failures would cause the pilots to fly higher and faster than they
otherwise would, in violation of the spray programme’s mandatory operational
208
parameters .
2.109 The Counter-Memorial, however, denies that the spray planes are
subjected to hostile gunfire, and asserts that Colombia’s pilots have no incentive
206
EM, Chap. 5, para. 5.95.
207
Ibid., Chap. 5, paras. 5.93-5.94.
208
There are other reasons that would encourage the pilots to exceed these restrictions as well.
For example, when flying at night, without the benefit of being able to see the ground, the aircraft
are likely to fly high. Likewise, the height of the surrounding tree canopy can at times exceed 50
or even 65 metres. Henrik Balslev, Ph.D., The Vulnerability of the Ecuador-Colombia Border
Region to Ecological Harm, pp. 7, 18, 25 (Jan. 2011) (hereinafter “Balslev ReportŽ). ER, Vol. II,
Annex 4.
92 >50 Meters
Figure 2.7
>165 mph (266 km/hr)
Actual Caselshtbfve5r0man1rs5 mph
50 Meters
165 mph (266 km/hr)
CoiloCboia’trepreseortaltion
25 Meters
140 mph (225 km/hr)
CotlomIioe’-nmrupiraneCtomtmnis-
Effects of Altitude and Flight Speed on Spray Drift
5 Meters
120 mph (193 km/hr)
Typical Agricultural Sprayingto fly too high or too fast. However, that they do both is no longer disputable,
given the flight data obtained from the U.S. Department of State. The only
question is why. At paragraph 4.61, Colombia states that prior to spray missions
there is “an assessment of whether the minimum requisite securityŽ is “present in
the areas to be sprayedŽ and that “[a]ny alteration to these conditions entails the
immediate cancellation of the spraying missionŽ 209. Similarly, at paragraph 4.70,
Colombia asserts there is no incentive to fly high or fast because “no spraying
operations are authorized on plots that are assessed as being high risk until
military operations to guarantee appropriate security conditions are carried
210
outŽ . In other words, according to the Counter-Memorial, the planes do not
spray in areas where they are vulnerable to hostile gunfire and thus have no
incentive to exceed the height and speed limits touted in the Counter-Memorial.
2.110 Here, again, the evidence negates Colombia’s assertions. For example,
the shooting of spray planes has been frequently reported in the press, in articles
with titles like Anti-Drug Plane Shot Down, U.S. Says 21. The danger faced by
209
CCM, Chap. 4, para. 4.61.
210
Ibid., Chap. 4, para. 4.70. See also ibid., Chap. 7, para. 7.25 (“no spraying operations are
authorized on plots that are assessed as being high risk until military operations to guarantee
appropriate security conditions are carried out, and spraying missions are cancelled if the situation
changesŽ).
211See, e.g., “Anti-drug plane shot down, U.S. sHICAGOCTRIBUNE (Chicago, 23 Sept. 2003)
(quoting a Dyncorp spokesman as saying that an “aircraft was struck by hostile ground fireŽ). ER,
Vol. IV, Annex 77; Tod Robberson, “2 U.S. Pilots Die on Colombian Anti-Narcotics MissionŽ,
D ALLAS M ORNING NEWS (Dallas, 29 July 1998) (“Colombian and U.S. military sources said . . .
leftist guerrillas . . . frequently try to shoot down government aircraft on illicit-crop eradication
missionsŽ. “Their work has become so dangerous in recent months … due mainly to anti-aircraft
93Colombia’s spray pilots was also conveyed in an article in Soldier of Fortune
212
magazine entitled Pray and Spray .
2.111 Colombia’s claim that the planes do not spray areas where there is a risk
of hostile gunfire is further refuted by the testimony of a spray pilot who spent
years working in the aerial spraying programme. The pilot testified that he and
his colleagues are frequently subjected to armed attack, both from gunfire and
from improvised explosive devices:
“narco-terrorist and criminal groups oppose and attempt to disrupt
the drug-spraying missions that I and others fly in Colombia.
These opposition and disruption efforts include such conduct as
shooting at the low-flying aircraft with high calibre firearms from
the ground, rigging above-ground wires to snag and damage the
aircraft, and placing tall poles or trees (with limbs removed) that
cannot easily be seen from the fast-moving aircraft to create
obstacles. On several occasions, our aircraft have been rocked by
fire from guerrillas and other gunmen protecting illicit-crop fields and drug laboratories … that
earlier this year, they began conducting eradication missions at night to make their aircraft harder
to targetŽ.). ER, Vol. IV, Annex 53.
212Steve Salisbury, “Pray and Spray: SOF With Coke-Bustin’ BroncosŽ, OLDIER OF FORTUNE , p.
61 (July 1998) (describing the shooting down of spray planes and death and injuries caused to
pilots). ER, Vol. IV, Annex 52. The DNE, which is responsible for the spray program, was also
warned that spraying under these conditions leads to an increased risk of off-target drift. In a
letter sent to the DNE, Colombia’s Minister of Environment stated that “the greatest possibility of
spraying areas that are not the object of the program are related to errors on the spraying process,
which depend on conditions of public order and the difficulty of the operationŽ. Among other
risks, the Minister of Environment noted “deviation from the flight path due to attack or risk of an
accidentŽ and “invasion of buffer zones due to human errorŽ. Letter from Juan Mayr Maldonado,
Minister of the Environment, Republic of Colombia, to Gabriel Merchan Benevides, Director
General of the National Drug Directorate, Republic of Colombia (Undated). ER, Vol. V, Annex
140.
94 explosions from remote-controlled Improvised Explosive Devices
(‘IEDs’) installed in the tops of tall treesŽ 21.
2.112 Continuing, the pilot testified regarding the deaths and injuries that these
hazards have caused:
“I have personally witnessed all of these dangers. I have had
aircraft I was piloting seriously damaged by weapons fire to the
point where I had to abort my spraying mission and return to base.
I have seen obstacles intentionally placed as hazards to low-flying
spray planes. I have known fellow pilots who were killed during
spraying missions in plane crashes caused by these types of
hazardsŽ 214.
2.113 This testimony is confirmed by contemporaneous State Department
reports that record the extensive time and resources required to make bullet-
ridden aircraft serviceable again. For example, in March 2003, the State
Department reported that aircraft participating in the aerial spraying programme
experienced “16 incidents of hostile fire resulting in 47 individual small arms
hitsŽ215. These “battle damage incidentsŽ … to use the State Department’s words …
required the expenditure of “[a]pproximately 2,000 maintenance man-hoursŽ to
216
make the necessary repairs . The following month, damage from hostile fire
was even worse. “Colombia’s eight OV-10 recorded 1,989 depot hours (83 days)
213
United States District Court of the District of Colombia, Arias, et al. v. Dyncorp, et al.,
Quinteros, et al. v. Dyncorp, et al., Declaration of Redacted Witness, para. 6 (2 June 2010). ER,
Vol. IV, Annex 117.
214
Ibid.
215
Memorandum from Lowell Neese, Senior Aviation Advisor, DoS/INL/A (Colombia), to Paul
O’Sullivan, COR, DoS/INL/A, p. 6 (21 Apr. 2003). ER, Vol. III, Annex 46.
216
Ibid.
95for battle damage reportsŽ, equalling “almost 3,000 maintenance man-hoursŽ 217.
These statistics prompted the State Department to complain that the “3000+
218
maintenance man hours detracted from aircraft availabilityŽ .
2.114 Beyond the incentive to fly higher and faster that is provided by the risk of
attack and other hazards from the ground, the variable terrain in the border region
also compels pilots to operate spray missions at greater altitudes. The rainforest
canopy, which itself may reach up to 30 to 35 metres in height, and towering
emergent trees which reach higher still, present significant physical obstacles 219.
The DNE itself has indicated that it is “impossibleŽ to follow fixed altitude
parameters given the variable terrain and other factors:
“Regarding flight height. As is well known, illicit crops are located
in different landscapes, varying topographical conditions and
many plots present obstacles, in addition to the dangers of drug
traffickers, which make it impossible for the operation to follow
fixed parameters. It must be recalled that the spraying is not being
carried out on plots of rice, cotton or banana fields where
217
Memorandum from Lowell Neese, Senior Aviation Advisor, DoS/INL/A (Colombia), to Paul
O’Sullivan, DoS/INL/A, p. 2 (9 May 2003). ER, Vol. III, Annex 47.
218
Ibid.
219
Balslev Report, op. cit., pp. 7, 18, 25. ER, Vol. II, Annex 4. See also Republic of Colombia,
Ministry of Environment, Division of Environmental Licenses, Order No. 599, p. 12 (23 Dec.
1999); Charles A. Menzie, Ph.D., Pieter N. Booth, MS & Susan B. Kane Driscoll, Ph.D., with
contributions/advice from Angelina J. Duggan, Ph.D., Charlotte H. Edinboro, DVM, Ph.D., Anne
Fairbrother, DVM, Ph.D., Marion J. Fedoruk, MD, CIH, DABT, FACMT, Janice Chunn Lindsay,
Ph.D., Katherine Palmquist, Ph.D. & Brian J. Prince, MRQA, Evaluation of Chemicals Used in
Colombia’s Aerial Spraying Program and Hazards Presented to People, Plants, Animals, and the
Environment in Ecuador, p. 12 (Apr. 2009) (hereinafter “Menzie et al., 2009Ž). EM, Vol. III,
Annex 158.
96 conditions for spraying are uniform and do not present any
obstaclesŽ 220.
(b) Colombia Uses Aircraft Unsuited for Aerial Spraying
2.115 The Counter-Memorial misrepresents the type of aircraft Colombia uses
in the spray programme, falsely claiming to use only aircraft that are designed for
221
depositing chemical sprays with pinpoint accuracy . In fact, many of the spray
missions have been conducted with military aircraft ill-suited for aerial spraying.
This, too, increases the likelihood of extensive spray drift.
2.116 Colombia asserts that the only aircraft employed in the aerial spraying
programme is the AT-802, an aircraft specifically designed for aerial application
of pesticides 22. For example, at paragraph 4.63, Colombia states that “[t]he
aircraft used for spraying operations are AT-802 planes manufactured by Air
223
TractorŽ . A photograph of the AT-802 is shown at Figure 2.8. The Counter-
Memorial declares that since Colombia uses this model of aircraft exclusively, the
Court can be confident that spray does not drift into Ecuador because the AT-802
is “specially designed to operate with precision during those tasks and possesses a
220
Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 143, p. 4 (29 Mar. 2000). ER, Vol. V, Annex 133.
221
CCM, Chaps. 4 and 7, paras. 4.63, 7.32, 7.172.
222
Ibid., Chap. 4, para. 4.63; Hansman & Mena Report, op. cit., p. 25. ER, Vol. II, Annex 1.
223CCM, Chap. 4, para. 4.63.
97system of tanks, nozzles and pumps similar to those used for the spraying of
crops in other parts of the worldŽ 224.
225
Figure 2.8. Photograph of AT-802 Spray Plane
2.117 Here is another example of a misrepresentation by Colombia regarding the
spray programme. The AT-802 is not the only aircraft used in Colombia’s aerial
spraying programme. Nor is it even the predominant one. To the contrary, the
majority of aerial sprayings between 2000 and 2008 were carried out by aircraft
224Ibid.
225Hansman & Mena Report, op. cit., p. 25. ER, Vol. II, Annex 1.
98 226
other than the AT-802 . And the aircraft that Colombia has used are especially
prone to causing long-range spray drift.
2.118 Most significantly, Colombia has made extensive use of the OV-10, a
military aircraft neither designed nor suitable for use in aerial spraying. This
usage is clear from testimony provided by a pilot with many years of experience
flying aircraft in Colombia’s spraying programme, who testified that the “OV-10
BroncoŽ is “used for aerial spraying missionsŽ 227. Colombia’s widespread use of
this aircraft is also confirmed by evaluation reports produced by the U.S.
Department of State, and by reliable accounts published in the news media 228.
226
Hansman & Mena Report, op. cit., p. 24. ER, Vol. II, Annex 1; ibid., Appendix 3, p. 25.
227
United States District Court of the District of Colombia, Arias, et al. v. Dyncorp, et al.,
Quinteros, et al. v. Dyncorp, et al., Declaration of Redacted Witness (2 June 2010). ER, Vol. IV,
Annex 117.
228See Chemicals Used, op. cit., p. 3. EM, Vol. III, Annex 144; United States Department of
State, Bureau for International Narcotics and Law Enforcement Affairs, Report on Issues Related
to the Aerial Eradication of Illicit Coca in Colombia: Updated Report on Chemicals Used in the
Aerial Eradication Program, p. 3 (Dec. 2003). EM, Vol. III, Annex 148; Report by the Ministry
for the Environment, Housing and Territorial Development on the Program for the Eradication of
Illicit Crops by Aerial Spraying with Glyphosate Herbicide … PECIG, February 2010, Appendix 1:
Sample Report of a Verification Mission (Technical Report … 19th Verification Mission
concerning the spraying operations conducted between September 2008 and February 2009), p.
347. CCM, Vol. II, Annex 70; K.R. Solomon et al., “Human Health and Environmental Risks
from the Use of Glyphosate Formulations to Control the Production of Coca in Colombia:
Overview and ConclusionsŽ, in Journal of Toxicology and Environmental Health, Part A, 72:914-
920, p. 914 (2009). CCM, Vol. III, Annex 131-A; A.J. Hewitt et al., “Spray Droplet Size, Drift
Potential, and Risks to Nontarget Organisms from Aerially Applied Glyphosate for Coca Control
in ColombiaŽ, in Journal of Toxicology and Environmental Health, Part A, 72:921-929, 2009.
CCM, Vol. III, Annex 131-B; E.J.P. Marshall et al., “Coca (Erythroxylum coca) Control is
Affected by Glyphosate Formulations and AdjuvantsŽ, in Journal of Toxicology and
Environmental Health, Part A, 72:930-936 , p. 930 (2009). CCM, Vol. III, Annex 131-C; Steve
Salisbury, “Pray and Spray: SOF With Coke-Bustin’ BroncosŽ, S OLDIER OF F ORTUNE , p. 72 (July
992.119 The great extent to which Colombia has used the OV-10 is reflected in the
data generated by the planes’ on-board equipment, which record the type of
aircraft in use. These records make clear that the OV-10 has been employed on a
vast scale. Indeed, in the 10-kilometre area near the Ecuadorian border,
Colombia has used the OV-10 for aerial spraying at least 20,251 times 229. This
represents 18 percent of all the sprayings along or near the border. The volume of
230
spray mixture deposited by OV-10s is impressive: over 376,000 litres .
2.120 Colombia’s use of the OV-10 for aerial spraying dramatically increases
the likelihood of spray drifting into Ecuador. Unlike the AT-802 … the aircraft
Colombia falsely claims is the only plane used … the OV-10 was not designed for
aerial spraying or any other agricultural or forestry application 231. To the
contrary, the OV-10, shown in Figure 2.9, is an armed military reconnaissance
airplane used for observation and counterinsurgency missions 232. Among other
problems, it flies too fast for use in aerial spraying. The data show that it has
almost never sprayed while flying within the prescribed speed limit. Of the
20,251 times Colombia sprayed near the Ecuadorian border using an OV-10, it
1998) (quoting spray programme pilots as saying that “the OV-l0 may be good for reconnaissance
and being armed. But it isn’t as accurate as the Thrush for sprayingŽ.). ER, Vol. IV, Annex 52.
229
Hansman & Mena Report, op. cit., Appendix 3, p. 25. ER, Vol. II, Annex 1.
230Ibid., Appendix 3, p. 27.
231Ibid., p. 27.
232Ibid., p. 27. Janes All The World’s Aircraft, p. 421 (stating that the OV-10 aircraft was
designed as an “armed reconnaissance aeroplane . . . specifically suited for counterinsurgency
missionsŽ).
100flew faster than the 165 mph speed limit claimed in the Counter-Memorial more
than 99 percent of the time, and faster than Colombia’s self-described “worst case
233
scenarioŽ speed of 333 km/hr 56 percent of the time .
Figure 2.9. Photograph of OV-10 Aircraft. 234
2.121 As a consequence, chemical spray released from the OV-10 is particularly
235
prone to drift . This problem was recognized by one of Colombia’s own spray
pilots, who stated: “The OV-10s fly so fast that the air turbulence doesn’t allow
233
Ibid., Appendix 3, p. 18.
234Hansman & Mena Report, op. cit., p. 27. ER, Vol. II, Annex 1.
235Ibid. (“[T]he OV-10 was operated at a significantly higher airspeed than other aircraft type. . .
the high speed results in smaller droplets in the spray which will drift furtherŽ.).
101the herbicide to fulfill its purpose. The herbicide is blown into a scatter and
236
vanishesŽ . It was also acknowledged by the USDA which, after conducting a
verification mission in 2001, recommended that, “to avoid overspray[]Ž, the
programme should stop using “[the] OV-10Ž for small fields after finding
237
evidence of “[s]erious oversprayingŽ .
2.122 Perhaps Colombia will eventually explain why it has used the OV-10 for
aerial spraying when it is manifestly unsuited for that purpose (and why it
conceals these facts in the Counter-Memorial). Until it does so, the best
explanation might be the one published in Soldier of Fortune magazine, to the
effect that OV-10s are “cheapŽ because they are “obtainedŽ from “surplus
collecting dust in the bone yards of U.S. federal agencies in Maryland and
238
VirginiaŽ .
236Steve Salisbury, “Pray and Spray: SOF With Coke-Bustin’ BroncosŽ, OLDIER OF FORTUNE , p.
72 (July 1998). ER, Vol. IV, Annex 52. See also ibid. (“The OV-10 may be good for
reconnaissance and being armed. But it isn’t as accurate as the Thrush for sprayingŽ.).
237
U.S. Department of Agriculture, Agricultural Research Service, Colombia Coca Verification
Mission April-May 2001, p. 78 (7 July 2001). ER, Vol. III, Annex 41.
238
Steve Salisbury, “Pray and Spray: SOF With Coke-Bustin’ BroncosŽ, OLDIER OF FORTUNE , p.
72 (July 1998). ER, Vol. IV, Annex 52. Nor is the OV-10 the only other aircraft used by
Colombia in the aerial fumigation programme. Colombia also makes extensive use of still another
aircraft, the T-65, which was used for 52,025 spray events within 10 kilometres of Ecuador’s
border. Hansman & Mena Report, op. cit., Appendix 3, p. 25. ER, Vol. II, Annex 1. The T-65
also creates a significant risk of off-target drift. As explained by the Colombian Agricultural
Institute, “Turbo Thrush Commander aircrafts . . . are “high speedŽ aircraft. The use of these
planes is problematic because it “influences the aspersion cloud that is discharged, producing
smaller drops, compared to those produced by Helicopters, which are classified as low speedŽ.
Republic of Colombia, Colombian Agriculture and Livestock Institute, ICA Concepts Regarding
A Report Issued by the Environmental Audit Techeca Ltda., p. 4 (1994) (emphasis in original).
102(c) The Pilots of the Spray Planes Lack Proper Training and Discipline, and
Routinely Ignore Operational Requirements to Prevent Spray Drift
2.123 The Counter-Memorial represents that the spray pilots are well-trained
professionals with extensive aerial spraying experience. Paragraph 4.63 states:
“The spray personnel … pilots … are provided by DynCorp, Inc., a private
companyŽ and these “pilots are specifically certified, trained and experiencedŽ.
Similarly, paragraph 7.17 emphasizes that the “Aircrew are fully trained and
flights are monitored and recordedŽ. This is yet another example of how the
Counter-Memorial misrepresents key aspects of the aerial spraying programme.
2.124 The deficiencies in the management and execution of the programme,
including in the capability and responsibility of the pilots, was highlighted by the
U.S. Department of State, which observed that the “expansion of [aerial spraying]
operationsŽ during the 1990s “brought with it associated problems in Colombian
contract pilot capabilityŽ 239, and expressed concern about their “lack of
ER, Vol. V, Annex 122; see also ibid., p. 7. It appears that none of the aircraft used by Colombia
were properly evaluated prior to their use for aerial sprayingA report by the Colombian
Agricultural Institute in 1999 states that “[w]e consider that the authorization for equipment and
aircraft must first respond to a technical evaluation regarding its effectiveness. The ICA evaluated
the applications carried out via helicopter, positively conceptualizing on its use; however, it did
not do this for applications via airplane, despite the fact that applications are being carried out via
airplane. In this regard, the technical concepts have been issued in a timely manner but have not
been taken into accountŽ. Republic of Colombia, Ministry of Environment, Division of
Environmental Licenses, Order No. 599, p. 11 (23 Dec. 1999). ER, Vol. V, Annex 132.
239
Memorandum from Peter P. Trent, INL/RM/ASD, PSC Bogota, to Grant Harden,
INL/RM/ASD, COR, p. 2 (4 July 1996). ER, Vol. III, Annex 34.
103experienceŽ 240. The State Department cited as a particular problem the fact that
“managementŽ had failed to “adequately deal with the pilots (sic) lack of
241
disciplineŽ .
2.125 The same concerns about pilot capability and lack of experience continued
to be voiced by the State Department over the next 10 years. The State
Department criticized the slipshod manner in which pilots were trained, finding
that the programme was so deficient that, in a classic case of the blind leading the
blind, novice pilots were themselves placed in charge of training those with even
less experience:
The contractor developed pilot training program for the OV-10
does not yet meet FAR [Federal Aviation Regulations]
requirements. Since there is no FAA or DoS [Department of
State] approved pilot training program, pilots with less than 50
airframe hours are signed off as instructor pilots. These instructor
pilots are in turn training other instructor pilots and plans are to
sign them off as instructor pilots with the same minimum
training242.
2.126 Such deficiencies in pilot training and discipline were raised in many
State Department evaluations. One observed that the “pilots being trainedŽ were
“demonstrating lapses of pilot discipline and lack of ability to follow published
240Memorandum from David Johnson, INL/C/ASD, to Grant Harden, INL/C/ASD, p. 1 (14 Nov.
1996). ER, Vol. III, Annex 35.
241Ibid.
242
Memorandum from Tim Doty, COR, INL/RM/AD, to Dyncorp, PSD Manager, pp.1-2 (Aug.
1997). ER, Vol. III, Annex 38.
104guidanceŽ and that this “lack of discipline is singularly the most dangerous safety
of flight issueŽ 243. This was demonstrated by:
“clear examples that some spray pilots were using poor judgment.
This, combined with unacceptable techniques resulted in collateral
damage to legitimate crops and pastures. The overall damage to
the image of the program this month by these pilots was
significantŽ 24.
2.127 The State Department reported that although these “issuesŽ had been
brought to the attention of the programme’s “managementŽ, the managers still
“maintained their focus on meeting deadlines while sometimes jeopardizing
245
safetyŽ . The State Department therefore concluded that “Management is not
246
adequately supporting program safetyŽ .
2.128 Repeated attempts to address these problems fell on deaf ears. In 1999,
the State Department was forced to conclude that the “contractorŽ had “failed to
provide adequate OV-10 experience levelŽ and that “for almost six months the
243
Memorandum from David Johnson INL/C/ASD to Grant Harden, INL/C/ASD, p. 1 (12 Dec.
1996). ER, Vol. III, Annex 36.
244
Ibid.
245Ibid.
246Ibid. Another evaluation reported that the spray planes were:
“operating in Colombia without a base of reference to operate from and
are developing tasks as they need. Considering the hostile
environment and the harsh operating conditions, this seems to be the
least safe approach and dramatically increases the risk associated with
the mission. Contractor operations should have taken a more proactive
approach as directed by the DoS/AD Chief to prevent this situationŽ.
Memorandum from Tim Doty, COR, INL/RM/AS to Dyncorp, PSD Manager, p. 9 (28 May
1997). ER, Vol. III, Annex 37.
105contractor has not provided adequate OV-10 standardization oversight of the OV-
10 programŽ 24.
2.129 Another inspection, this time in 2000, observed serious “deficienciesŽ
that, the State Department determined, “requires immediate attention by
248
managementŽ . These problems included such basic issues as “the overall
organization and chain of commandŽ, which, the State Department concluded,
had “led to several complications and an overall confusion about duties and
249
responsibilitiesŽ . Among other criticisms leveled by the State Department
were the following, each of which identified fundamental problems with the spray
programme:
x “managersŽ had “no record of a job description or briefings as to
duties and responsibilitiesŽ;
x incoming personnel were “not provided any formal in-processingŽ and
that “procedures are not briefed, provided or discussedŽ;
x “[n]o self-inspection program exist[ed] except for a limited quality
control audit programŽ;
x “[m]anagersŽ were “unfamiliar with the contract and technical
directivesŽ;
247Memorandum from Stephen H. Harris, INL/RM/AD, to Dyncorp, p. 3 (21 June 1999). ER,
Vol. III, Annex 39.
248Aviation Resource Management Inspection of Air Wing Colombia Site, p. 1 (23 Mar. 2000).
ER, Vol. III, Annex 40.
249
Ibid.
106 x “[c]ompliance with proceduresŽ needed to be “improved to increase
both safety of personnel involved and effectivenessŽ of the “missionŽ;
and
x “[t]he Site Safety Manager in ColombiaŽ was “not conducting
250
required safety inspections, surveys and hazard analysisŽ .
2.130 In light of these pervasive problems, the State Department made the
following determination: “Overall complacency towards safety is a primary
concern and needs to be corrected immediatelyŽ 25.
2.131 Apparently, no such corrections were made. In February 2001, the State
Department complained about the “constant[]Ž need to “remind spray pilots of
what to avoid, for example villages, etcŽ 25. Several months later, the State
Department was forced to criticise the spray programme for the “observed
substandard condition of the aircraftŽ, which it determined was a “direct
253
reflection of being understaffedŽ . The State Department insisted that
254
“IMMEDIATE attention in this area is recommended!Ž .
250Aviation Resource Management Inspection of Air Wing Colombia Site, pp. 2-6, 28 (23
Mar.2000). ER, Vol. III, Annex 40.
251Ibid.
252
Memorandum from Michael J. Kenna, INL/RM/AD, Senior Aviation Advisor, to Steve Harris
(COR) and George Arzente, INL/RM/AD/COR (Undated). ER, Vol. III, Annex 51.
253
Memorandum from Stephen H. Harris, COR, DoS/INL/A, to Dyncorp, p. 4 (Oct. 2001). ER,
Vol. III, Annex 49.
254
Ibid. (emphasis in original).
1072.132 In March 2002, the State Department criticized the “systemic failure in
following of aircraft operational proceduresŽ and complained about the “lack of
importance placed on the safety programŽ 255. Frighteningly, the State
Department even found that the programme was “sufferingŽ from “incomplete
256
and inaccurate cartographic dataŽ .
2.133 In short, Colombia’s attempt to portray its aerial spraying programme as a
well-run and professional operation is completely contradicted by the evidence,
which shows that pilots are undisciplined, lack sufficient training and ignore
operational requirements, especially in regard to preventing spray drift. These
deficiencies provide further explanation for why they consistently release the
spray mixture at unsafe speeds and altitudes in violation of the
requirements/regulations described in the Counter-Memorial. As discussed
below, the evidence shows that Colombia’s “cowboyŽ pilots not only routinely
flout the limits on flight speed and altitude of spray release, but also the
requirements relating to droplet size, spray application rate and time of day when
spraying is permitted.
255Memorandum from Lowell E. Neese, SAA, DoS/INL/A (Bogota), to Stephen H. Harris, COR,
DoS/INL/A, p. 5 (13 Mar. 2002). ER, Vol. III, Annex 44.
256Memorandum from David A. Campbell, COR, DoS/INL/A, to Dyncorp, PSD Manager, p. 19
(Feb. 2004). ER, Vol. III, Annex 48.
108 D. D ROPLET SIZE
2.134 Aircraft speed and height of spray release are not the only factors that
Ecuador and Colombia agree significantly contribute to drift. The size of the
spray droplets is also a major contributor. Smaller droplets drift longer distances.
Colombia agrees. It states in the Counter-Memorial that the amount of spray drift
257
depends on the “initial size of the spray dropletsŽ . Indeed, Colombia’s experts
acknowledge that droplet size is one of the most important determinants of spray
drift58.
2.135 To defend the aerial spraying programme, Colombia has on numerous
occasions represented that the droplet size of the spray mixture is large, since
larger droplets are less likely to drift off-target. For example, an official
communication from the Director of Colombia’s DNE to the Ecuadorian
Scientific and Technical Commission on 14 April 2004 represented that the
“average dropŽ was 650 microns in size 259. In September 2005, Colombia
informed the IACHR that its technical parameters allowed spray droplets 300-
257
CCM, Chap. 4, para. 4.68. See also CCM, Chap. 7, para. 7.17.
258Dobson Report, op. cit., p. 523. CCM, Vol. I, Appendix. See also Hewitt et al., 2009, op. cit.,
pp. 921-922. CCM, Vol. III, Annex 131-B.
259Note SARE-142, sent from the National Directorate of Narcotics of the Ministry of Interior
and Justice of Colombia to the President of the Technical-Scientific Commission of Ecuador, p. 5
(14 Apr. 2004). EM, Vol. II, Annex 62.
1091,500 microns (0.3 to 1.5 mm) in size 260, a range that is considered larger than
“extremely coarseŽ, the largest size specified by internationally accepted droplet
size standards 261. Based on information provided to them by Colombia, the
authors of the 2005 Solomon study also reported that the size of the spray
262
droplets were in the range of 300-1,500 microns . Colombia’s EMP allows
263
droplets 300-1,000 microns in size .
2.136 However, the size of the droplets is much smaller than Colombia has
claimed. Subsequent analysis in 2009, paid for by the Colombian and U.S.
governments, dramatically revised the 2005 study’s estimate, determining that the
actual median droplet size is only 128-140 microns (0.138 to 0.140 mm) 264. In
other words, the median droplet size currently expected by Colombia is now half
the estimated size that the 2005 Solomon study assumed as the worst case when it
concluded that drift was unlikely to be a problem, and 50 percent smaller than the
260Diplomatic Note Nº DDH. 58003 from the Colombian Foreign Ministry to the Executive
Secretary of the Inter-American Commission on Human Rights, 18 Sept. 2005, p. 26. ER, Vol. V,
Annex 154.
261
Giles Report, op. cit., p. 22. ER, Vol. II, Annex 2; American Society of Agricultural and
Biological Engineers, Spray Nozzle Classification by Droplet Spectra, ANSI/ASAE S572.1 (Mar.
2009). ER, Vol. III, Annex 23.
262
Solomon et al., 2005, op. cit., p. 28. CCM, Vol. III, Annex 116.
263
Resolution Nº1054 of 30 September 2003 of the Ministry for the Environment of Colombia, p.
173. CCM, Annex 50.
264
Hewitt et al., 2009, op. cit., pp. 921. CCM, Vol. III, Annex 131-B. The Agricultural Institute
reported that Colombia has known its publicly cited droplet size range was inaccurate since at
least 1999, when the Colombian Institute rejected the suggestion that the droplet size could be
“between 300, 1000 and 1500 microsŽ. Republic of Colombia, Ministry of Environment,
Division of Environmental Licenses, Order No. 599, p. 13 (23 Dec. 1999). ER, Vol. V, Annex
132.
110smallest size permitted by the EMP. Moreover, the “fine to very fineŽ droplets
admittedly sprayed by Colombia are dispersed in a spectrum 265. In other words, a
full 50 percent of the droplets are even smaller than the 128-140 micron median
size. This is particularly problematic because droplets smaller than 150 microns
are considered by Colombia’s own experts to be particularly prone to spray
266
drift . As Dr. Giles explains, “[g]iven the importance of the smallest droplets in
contributing to spray drift . . . the drift estimates based on the Colombian EMP
droplet size values would severely under-predict actual driftŽ 267.
2.137 In fact, even Colombia’s 2009 estimate overstates the actual size of the
droplets. The 2009 study’s authors predicated their droplet size estimate on the
presumption (based on information provided to them by Colombia) that spraying
is carried out in strict compliance with the aircraft speed and height requirements
claimed by Colombia … a presumption that the flight data obtained from the U.S.
265Hewitt et al., op. cit., 2009, pp. 921, 923-925. CCM, Vol. III, Annex 131-B. See also
Republic of Colombia, Colombian Agriculture and Livestock Institute, ICA Concepts Regarding
A Report Issued by the Environmental Audit Techeca Ltda., p. 2 (1994). ER, Vol. V, Annex 122.
266
Hewitt et al., 2009, op. cit., p. 922. (“In this study, there was interest in the spray volume
contained in relatively small droplets, i.e., those with diameter below 150 m. This represents the
finer droplets in the spray, which might present more of an exposure risk for downwind spray drift
under unfavorable conditionsŽ.). CCM, Vol. III, Annex 131-B. See also Spray Drift Task Force,
A Summary of Aerial Application Studies, p. 2 (1997). ER, Vol. III, Annex 10. (“The cut-off point
of 141 microns or 150 microns has been established as a guide to indicate which droplet sizes are
most prone to drift. However, it is important to recognize that drift doesn’t start and stop at 141
microns. Drift potential continually increases as droplets get smaller than 141 microns, and
continually decreases as droplets get biggerŽ.)
267Giles Report, op. cit., p. 22. ER, Vol. II, Annex 2.
111 268
government prove unfounded ; since the actual flight speeds and altitudes of
spray release are very frequently much higher than allowed, the droplet size is
necessarily even smaller than Colombia’s experts presumed when they conducted
their study.
2.138 Colombia’s misrepresentation of droplet size is of fundamental
importance. As Australia’s Operating Principles in Relation to Spray Drift Risk
put it, “[s]pray droplet size . . . is the most important single factor in spray drift
risk. Smaller, lower mass droplets have greater potential for drifting off
269
targetŽ . This view is corroborated by the pesticide industry’s Spray Drift Task
Force, which assessed the “relative role of the factors that affect spray driftŽ and
determined that “[d]roplet size was . . . the most important factorŽ 270. The Giles
Report confirms that droplet size is one of the predominant factors influencing
spray drift and off-site deposition of Colombia’s spray mixture:
“The size of the spray droplets has important implications for
spray drift because it affects the droplet’s terminal velocity and
rate of evaporation. Smaller droplets have significantly slower
terminal velocities and are displaced greater distances by cross
winds than larger droplets. Likewise, smaller droplets have higher
surface area to mass ratios and the effect of evaporation is to
decrease their size more rapidly than larger droplets. As a result of
268
See supra Chap. 2, paras. 2.91-2.97, 2.101-2.107.
269
Australian Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating
Principles in Relation to Spray Drift Risk, p. 18 (15 July 2008). ER, Vol. III, Annex 22.
270
Spray Drift Task Force, A Summary of Aerial Application Studies, p. 1 (1997). ER, Vol. III,
Annex 10.
112 these two factors, smaller droplets are carried greater distances,
resulting in greater spray driftŽ27.
E. A PPLICATION RATE
2.139 Colombia also violates the EMP’s requirement governing the rate at
which the spray mixture may be applied, contrary to its representations in the
Counter-Memorial. In that regard, Colombia asserted that under no
circumstances does it apply the spray mixture at a rate greater than 23.65 litres
per hectare 27. Like its representations regarding aircraft speed, altitude of release
and droplet size, this one is false, too.
2.140 In fact, Colombia routinely exceeds the maximum allowable volume
sprayed per hectare. For example, in 2002, the threshold was violated 12,184
times in the area adjacent to Ecuador, representing 31 percent of all spraying
273
done that year in the border area . Nor was this the only year when Colombia’s
spraying exceeded the threshold with great regularity. It did so 5,267 times in
2006 (30 percent of all spraying in that year) and another 4,143 times in 2007 (38
percent of all spraying in that year)274.
271Giles Report, op. cit., p. 7. ER, Vol. II, Annex 2.
272
CCM, Chap. 4, para. 4.62.
273
Hansman & Mena Report, op. cit., Appendix 3, p. 22. ER, Vol. II, Annex 1.
274Ibid.
1132.141 In total, Colombia violated its own limits on application rate at least
275
27,429 times between 2000 and 2008 . This is 31 percent of the documented
spraying within 10 kilometres of Ecuador. (In fact, the number of violations is
almost certainly much higher since Colombia failed to record data on application
rates in 2000 and 2001 and the data was also unusable for 2004 276.)
2.142 Colombia has no plausible excuse for so frequently exceeding its own
application rate limit. The Counter-Memorial states that Colombia’s aircraft are
fitted with specialized nozzles from which the spray is released, and that these
nozzles “have an automatic calibration mechanism that determines the amount of
spray mix to be released in order for the number of litres discharged per hectare
to be kept constant at 23.65 litres per hectareŽ 277. Elsewhere, the Counter-
Memorial represents that “the spray mix is propagated through automatically
calibrated nozzles that release the same amount of mixŽ 278. Supposedly, the
application rate is tracked after each spray mission; the Counter-Memorial states
that a “detailed report of the day’s operationsŽ is:
“prepared on the basis of the computerized system which records
each spraying operation with its respective route, geo-referenced
areas of application and the amount of spray mix released per
275
Ibid., p. 23.
276The data reported for 2004 is recorded in both metric and English units without specifying
which units were used for any particular spray event, making it impossible to determine how
many flights that year exceeded the requirement for volume.
277
CCM, Chap. 4, para. 4.62.
278Ibid., Chap. 7, para. 7.32.
114 minute. This allows verification of the location of the places
where the operations took place and quantification of the hectares
sprayed. A record is signed by the Base Commander and the
279
personnel involved in the operationŽ .
2.143 Colombia made the same representations to the authors of the 2005
Solomon study, who stated, based on information provided by Colombia, that
“[t]he aircraft spray systems are electronically calibrated to disperse a specified
quantity of spray mix per hectare, compensating for variances in ground
speedŽ 28. Solomon et al. relied on Colombia’s assurances that sophisticated
equipment ensures that only the precise amount of spray, and no more, is
released: “These electronic spray controls are checked each day by technicians
and also during the pilot’s preflight inspectionŽ281. Evidently, the controls were
not checked very carefully (if they were checked at all). That is the only
conclusion when Colombia violated its own limits on application rate … dumping
more of the spray mixture on a given area along the border with Ecuador than its
EMP allowed … tens of thousands of times between 2000 and 2008.
F. TIME OF D AY
2.144 Atmospheric conditions that Colombia concedes are prevalent at night are
also conducive to wider spray drift. As a result, the Counter-Memorial’s
279Ibid., Chap. 4, para. 4.64.
280Solomon et al., 2005, op. cit., p. 28. EM, Vol. III, Annex 116.
281
Ibid.
115assertion that Colombia’s spray missions occur only during the daytime is a
significant misrepresentation that has important implications in regard to the
distances travelled by the spray mixture it disperses along and near the border
with Ecuador.
2.145 Specifically, Colombia represents at paragraph 7.27 that its “spray
operationsŽ occur only “after sunriseŽ 28. Similarly, the authors of the 2005
Solomon study, based on information provided to them by Colombia, state that
283
“[s]praying is only conducted in daylight hours before mid-afternoonŽ . Thus,
on at least two occasions … once in the Counter-Memorial and again to the
authors of the 2005 Solomon study … Colombia has claimed that its aerial
spraying is only done during the daytime.
2.146 Once again, Colombia’s representations regarding the conditions in which
spraying occurs are contradicted by the evidence. In fact, much of the aerial
spraying takes place at night. This is evident from the U.S. Department of State’s
reports, which describe Colombia’s “support and planning for night sprayingŽ 284.
282
CCM, Chap. 7, para. 7.27.
283
Solomon et al., 2005, op. cit., p. 30. EM, Vol. III, Annex 116.
284Memorandum from Stephen H. Harris, COR, DoS/INL/A, to Dyncorp, PSD Manager, p. 11
(Dec. 2001) (emphasis added). ER, Vol. III, Annex 42.
116 2.147 The data collected by the spray planes confirm Colombia’s night time
spraying on a massive scale. Between 8 p.m. and 4 a.m., Colombia aerially
dispersed the spray mixture at least 24,540 times within 10 kilometres of the
285
border with Ecuador, representing 22 percent of all spraying in the region . If
late afternoon spraying is included (4 p.m. to 8 p.m.) … which contradicts the 2005
Solomon study’s presumption that spraying is only done “before mid-afternoonŽ
… the number of violations rises to 47,262 and the percentage increases to 43
286
percent . The falsity of Colombia’s claim that aerial spraying only occurs “after
sunriseŽ is graphically depicted in Figure 2.10.
12000
Daytime (62,419 flights, 57%)
10000
Late Afternoon (22,722 flights, 21%)
Nighttime (24,540 flights, 22%)
8000
Prohibited (Solomon et al., 2005)
Prohibited (Colombia Counter-Memorial)
6000
4000
NUMBER OF SPRAY EVENTS
2000
0
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 00 1 2 3 4
TIME OF DAY (Local Time)
Figure 2.10. Time of Day of Spray Events
285Hansman & Mena Report, op. cit., p. 23. ER, Vol. II, Annex 1.
286
Ibid., p. 24.
1172.148 Colombia’s night time operations greatly increase the propensity for the
spray mixture to drift longer distances, including into Ecuador. According to the
Solomon study (2005) … the main scientific evidence relied upon by Colombia …
the purpose of the prohibition on spraying after “mid-afternoonŽ is “to ensure that
conditions are appropriate for applicationŽ 287. One reason for this is not difficult
to fathom: spraying during darkness increases the likelihood of depositing the
spray off-target because the pilot is unable to see the targeted area and has a
stronger incentive to fly higher than otherwise in order to avoid objects that
cannot be seen at night, such as unusually tall trees 288.
2.149 There is a technical reason why spraying at night increases spray drift: the
meteorological condition known as a “thermal inversionŽ, in which air at a higher
altitude is warmer than air closer to the ground 289. This condition, which
frequently occurs at night in the border area between Colombia and Ecuador, is
particularly favourable to spray drift. According to Australia’s Operating
Principles in Relation to Spray Drift Risk, the “potential for or presence of a
surface temperature inversion condition is a very important factor in spray drift
risk managementŽ 290. Because the “[n]ight-time hoursŽ are “often associated
287
Solomon et al., 2005, op. cit., p. 30. EM, Vol. III, Annex 116.
288
Hansman & Mena Report, op. cit., p. 23, n. 9. ER, Vol. II, Annex 1.
289Ibid. See also Menzie et al., 2009, op. cit., p. 14. EM, Vol. III, Annex 158.
290 Australian Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating
Principles in Relation to Spray Drift Risk, p. 24 (15 July 2008). ER, Vol. III, Annex 22.
118with surface temperature inversion[s]Ž, Australia’s Operating Principles explain,
291
“spray operations should not be conductedŽ at night . Hansman & Mena further
explain this phenomenon:
“[D]uring daylight the atmosphere has better mixing and it is less
likely that the spray will drift away from the intended target zone.
At night the temperature of the surface is often lower than the
warmer air above due to rational cooling. This results in a low
altitude temperature inversion with a cool lower layer of air often
only 10 or 20 meters thick. This phenomena will result in calm
winds at night and ground fog if there is sufficient moisture in the
air. The inversion stratifies the atmosphere and prevents mixing
between the layers. As a consequence if the aircraft sprays in the
warm layer above the inversion most of the spray (particularly the
smaller droplets) will not hit the target area but will drift with the
winds in the upper layer. This can result in advection or drift of
the spray significant distancesŽ 292.
The Counter-Memorial does not dispute that thermal inversions cause spray to
drift long distances, or that they frequently occur along the border with Ecuador.
But Colombia dismisses thermal inversions as a cause for concern because they
occur only “at nightŽ, and the “spraying operationsŽ take place exclusively “after
sunriseŽ 293. Except for the more than twenty-four thousand times Colombia
291
Australian Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating
Principles in Relation to Spray Drift Risk, p. 25 (15 July 2008). ER, Vol. III, Annex 22.
292
Hansman & Mena Report, op. cit., p. 23, n. 9. ER, Vol. II, Annex 1; see also Menzie et al.,
2009, p. 14. EM, Vol. III, Annex 158. Dr. Giles further explains that “[a]n extremely undesirable
meteorological condition in terms of increasing the risk of significant spray drift is when there is a
temperature inversion and associated light and variable wind. In a temperature inversion, vertical
dispersion of small spray droplets is inhibited because they remain trapped between layers of air.
Thus, spray droplets can remain aloft and often become highly concentrated in relatively small
packets or layers of air . . . . During the inversion or as the inversion weakens, the wind can easily
displace these high concentrations of suspended small droplets over significant distances and in
concentrations greatly exceeding those typical of normal, cross wind driven driftŽ. Giles Report,
op. cit., p. 43. ER, Vol. II, Annex 2.
293
CCM. Chap. 7, para. 7.27.
119 294
sprayed at night … all within 10 kilometres of the border with Ecuador ! To
make matters even worse, on those 24,000+ occasions, Colombia also violated its
limits on aircraft speed (165 mph) and altitude of dispersion (50 metres) more
than 2,431 times 295. Flying too fast and too high at night … a perfect trifecta of
violations … virtually ensures widespread spray drift, including into Ecuador.
G. TEMPERATURE , HUMIDITY AND W IND C ONDITIONS
2.150 Colombia acknowledges the importance of meteorological conditions for
spray drift. At paragraph 4.68 of the Counter-Memorial, Colombia states that
“[s]pray drift depends essentially on wind speed and direction, as well as on a
number of other atmospheric factors including temperature, relative humidity and
atmospheric stabilityŽ29.
2.151 Higher temperature and lower humidity lead to greater evaporation of the
spray droplets, reducing their size and making them more prone to drift 297. As
explained in the UN Food and Agriculture Organization Guidelines on Good
Practice for Aerial Application of Pesticides, “[i]n conventional (water-based)
spraying, high temperature, combined with low relative humidity will reduce
294
Hansman & Mena Report, op. cit., p. 23. ER, Vol. II, Annex 1. Night was defined as 8 pm to
4 a.m. local time. Ibid.
295
Hansman & Mena Report, op. cit., Appendix 3, p. 27. ER, Vol. II, Annex 1.
296CCM, Chap. 4, para. 4.68; see also CCM, Chap. 7, para. 7.17.
297Giles Report, op. cit., pp. 26, 29. ER, Vol. II, Annex 2; Menzie et al., 2009, op. cit., p. 15.
EM, Vol. III, Annex 158.
120droplet size through evaporation, which will increase the risk of driftŽ 298.
Colombia’s experts acknowledge this risk, reporting that “at least half of the
droplet volume could potentially be lost through evaporationŽ 299.
2.152 Wind speed and direction also play an important role in spray drift. As
explained in the Giles Report, strong winds may carry spray droplets …
particularly the small droplets dispersed by Colombia’s spray planes … great
distances, leading to deposition at 10 kilometres or more from the application
site300. Less self-evidently, even low wind speeds can be conducive to drift.
Australia’s Operating Principles in Relation to Spray Drift Risk observe that
“times of no wind (essentially below 3 km/hr) often precede or accompany
periods of highly stable air and surface temperature inversion conditions both of
which can greatly increase spray drift risk. Moreover, when wind resumes after
301
periods of calm, its direction is not predictableŽ . As a result, some regulatory
298
Food and Agriculture Organization of the United Nations, Guidelines on Good Practice for
Aerial Application of Pesticides, p. 23 (2001). ER, Vol. IV, Annex 98. See also Australian
Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating Principles in
Relation to Spray Drift Risk, p. 22 (15 July 2008) (“For water based tank mixes, humidity and
temperature affect droplet evaporation rates and can make a surprisingly large difference in drift
deposits at longer downwind distances due to shrinkage in droplet size (and therefore mass)Ž.).
ER, Vol. III, Annex 22.
299
Hewitt et al., 2009, op. cit., pp. 925-926. CCM, Vol. III, Annex 131-B.
300
Giles Report, op. cit., p. 28. ER, Vol. II, Annex 2.
301 Australian Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating
Principles in Relation to Spray Drift Risk, p. 21 (15 July 2008). ER, Vol. III, Annex 22.; see also
Giles Report, op. cit., pp. 27-28, 42-44. ER, Vol. II, Annex 2; Menzie et al., 2009, op. cit., p. 14.
EM, Vol. III, Annex 158.
121authorities establish both minimum and maximum wind speeds for aerial spraying
302
operations .
2.153 Recognizing the importance of these factors for affecting spray drift,
Colombia’s EMP includes specific parameters including a “maximum outside
temperature during applicationŽ of 35ºC and a “maximum wind velocityŽ of 5
knots 303. The Counter-Memorial assures the Court that “[t]hese parameters are
strictly observed by the personnel involved in spraying operationsŽ 304, and that
wind conditions are “constantly monitored by the aircraft and if they are not
305
within the parameters allowed, the mission is annulled or postponedŽ . In
paragraph 4.16, Colombia provides the blanket assurance that weather conditions
are evaluated prior to each spray mission and that no spraying occurs when the
meteorology favours spray drift:
“Following an assessment of whether the minimum requisite
security and weather conditions … including temperature, wind
direction and speed and relative humidity … are present in the areas
to be sprayed, the operations begin. Any alteration to these
conditions entails the immediate cancellation of the spraying
missionŽ.
302
See infra Chap. 4, paras. 4.110-4.112.
303Resolution Nº1054 of 30 September 2003 of the Ministry for the Environment of Colombia, §§
3.2.2.1. CCM, Vol. II, Annex 50. The EMP provides no minimum wind velocity, despite the
prevalence of low wind conditions and thermal inversions in the border region.
304
CCM, Chap. 7, para. 7.17.
305Ibid., Chap. 7, para. 7.172.
1222.154 Conspicuously, Colombia provides no data to demonstrate the truthfulness
of these representations. It does not appear that Colombia even records or
maintains data about temperature, wind or other meteorological conditions during
spray flights. At least no evidence of this has been provided in the Counter-
Memorial. Nor is any such data recorded by the spray planes themselves, or
included within the information supplied to the U.S. Department of State and
obtained by Ecuador 306. In light of the fact that every other representation by
Colombia about its alleged compliance with the operational requirements of the
spray programme … regarding flight speed, altitude, type of aircraft, pilot
capability and discipline, droplet size, application rate and night time spraying …
has been proven false, Ecuador believes that Colombia’s unsupported and self-
serving assertions about compliance with temperature, wind and other such
requirements do not merit the full faith and credit Colombia requests of the Court.
To the contrary, the absence of data from Colombia to demonstrate its
compliance with meteorological requirements constitutes an additional reason to
doubt that it has succeeded in preventing spray drift into Ecuador 307.
Section III. Colombia’s Prior Misrepresentations to Ecuador and Others
2.155 The Counter-Memorial, in asserting that Colombia never breaches the
rules against spraying too fast or too high, that the volume of spray is perfectly
306Hansman & Mena Report, op. cit., p. 30. ER, Vol. II, Annex 1.
307See supra Chap. 2, para. 2.82; Giles Report, op. cit., pp. 26-30, 42-44. ER, Vol. II, Annex 2.
123calibrated not to exceed a specified dose, and that the spray droplets are always
sufficiently large, repeats prior misrepresentations that Colombia repeatedly made
to Ecuador.
2.156 For example, on 14 April 2004, Colombia informed Ecuador via
diplomatic note that the aerial spray programme is “carried out under the
technical parameters established in the Environmental Management Plan for the
Program for the Eradication of Illicit Crops by Aerial Spraying with Glyphosate
308
(PECIG)Ž . Colombia explicitly told Ecuador that the programme “includes
operational parameters for spraying such as flight altitude, temperature, relative
humidity, cloudiness, rain fall, type of nozzle, droplet size, and dosage, among
othersŽ, and that “[i]f any of these parameters is not met, the spraying mission is
309
cancelledŽ . Colombia made similar representations in August 2004, when its
delegation to the Binational Scientific and Technical Commission declared that
“the technical conditions necessary to prevent the spray from reaching
Ecuadorian territory shall be guaranteedŽ 31. When Colombia made these
representations to Ecuador, it surely knew that they were untrue: by August 2004,
the speed limit of 165 mph had been violated at least 41,714 times, the height
308
Note NºSARE-142 from the Director of the National Narcotics Directorate of Colombia to the
President of the Scientific and Technical Commission of Ecuador, para. 2.2, 14 Apr. 2004. CCM,
Vol. II, Annex 13.
309
Ibid.
310Minutes of the Fourth Meeting of the Joint Scientific and Technical Commission (2 Aug.
2004). EM, Vol. II, Annex 64.
124limit of 50 metres had been breached 10,815 times, and the application rate limit
of 23.65 litres per hectare had been exceeded 14,654 times in sprayings along the
311
border with Ecuador . And Colombia itself says its Ministry of the
Environment was closely monitoring compliance with all of these parameters, as
it was ordered to do by the Colombian Council of State 312.
2.157 On 20 December 2006, Colombia’s Ministry of Foreign Affairs again
assured Ecuador’s Embassy in Colombia that the aerial spraying was being
conducted in accordance with the operational requirements, and thereby
preventing spray drift into Ecuador. On that occasion Colombia stated:
“The Ministry of Foreign Affairs respectfully ensures to the
Honourable Government of Ecuador that the Program for
Eradication of Illicit Crops with Glyphosate … PECIG … is
executed under the strictest technical measures which guarantee
the protection of the environment and human health, also
313
preventing the sprayed mixture to reach Ecuadorian territoryŽ .
311Hansman & Mena Report, op. cit., Appendix 3, p. 27. ER, Vol. II, Annex 1. Similarly, in
November 2004, the Colombian Minister of Foreign Affairs wrote to the Director-General of the
National Police of Colombia regarding Ecuador’s concerns over the spray’s transboundary impact
in Ecuador. The Foreign Minister stressed it was essential to “guarantee[] compliance with the
technical and operational conditions established for the environmental management plan for the
eradication program that prevent any impact that may be derived from the spraying operations on
non-target areasŽ. Note Nº 001727 from the General Director of the National Police of Colombia
to the Colombian Foreign Minister, 2 Nov. 2004. CCM, Vol. II, Annex 55. Colombia’s Ministry
of Foreign Affairs reiterated this view in September 2005 when it stated that “the spraying tasks
with glyphosate herbicide (PECIG) are regulated by the relevant environmental rules that are
strictly enforced in the Program’s implementationŽ. Aide-Mémoire “Aerial Spraying Issue with
EcuadorŽ, Ministry of Foreign Affairs of Colombia, Division of Multilateral Political Affairs,
Sub-division for Drug Affairs, Sept. 2005. CCM, Vol. II, Annex 56.
312See supra Chap. 2, paras 2.80; see also CCM, Chaps. 4 and 7, paras. 4.26, 4.64, 7.172.
313Diplomatic Note from the Colombian Foreign Ministry to the Ecuadorian Embassy in Bogotá,
20 Dec. 2006. CCM, Vol. II, Annex 26.
125Again, the data supplied by the spray planes say exactly the opposite.
2.158 Nevertheless, the empty assurances given to Ecuador were repeated by
President Álvaro Uribe of Colombia in a letter to the President of Ecuador on 21
December 2006, in which Colombia’s Head of State represented that the
resumption of aerial spraying near the Ecuadorian border would “comply[] with
314
all the technical requirementsŽ . In fact, by the time Colombia gave these
assurances to Ecuador, the reality was that in the border area it had already
violated the speed limit of 165 mph at least 54,336 times, exceeded the height
limit of 50 metres 13,114 times, and violated the application rate limit of 23.65
315
litres per hectare 18,871 times .
2.159 Colombia made these representations not just to Ecuador, but to
international organizations as well. It did so, for instance, to the IACHR.
Colombia informed the IACHR by note dated 18 September 2005 that the
programme “is carried out in accordance with the Environmental Management
Plan and in strict observance of environmental and human health care legislation
316
in forceŽ . Colombia further represented to the Commission that:
314
Note from the President of Colombia to the President of Ecuador, 21 Dec. 2006. CCM, Vol.
II, Annex 27.
315Hansman & Mena Report, op. cit., Appendix 3, p. 27. ER, Vol. II, Annex 1.
316Diplomatic Note Nº DDH. 58003 from the Colombian Foreign Ministry to the Executive
Secretary of the Inter-American Commission on Human Rights, 18 Sept. 2005, § IV. CCM, Vol.
II, Annex 19.
126 “[s]praying must comply with a series of technical parameters
(flight altitude, maximum herbicide release, droplet size, foreseen
drift, temperature, relative humidity, and maximum wind speed)
that guarantee that the mixture used is targeted exclusively
towards the targeted eradication vegetation, that is to say, the illicit
crops, therefore minimizing any possible effect due to driftŽ 317.
2.160 Based on Colombia’s representations regarding its compliance with the
operational conditions necessary to prevent spray drift, the IACHR declined to
318
indicate provisional measures requested to protect people in Ecuador .
Colombia never informed the IACHR that, in reality, at the time of its submission
to the Commission, the operational requirements it invoked in opposition to the
request for provisional measures had already been violated many thousands of
times. More specifically, by 18 September 2005 (the date of its submission) the
speed limit of 140 mph had been violated at least 74,984 times, the height limit of
25 metres had been breached 56,592 times, and the application rate limit of 23.65
319
litres per hectare had been exceeded 15,223 times .
2.161 In short, Colombia has a long history of misrepresenting its compliance
with the operational requirements for its aerial spraying programme. The
Counter-Memorial breaks no new ground in this regard.
317
Ibid., § IV(c).
318
Note Nº DSF40.1/3.1.3-4-00423 from the Executive Secretary of the Inter-American
Commission on Human Rights to the Colombian Foreign Minister, 18 Nov. 2005. CCM, Vol. II,
Annex 20.
319
Hansman & Mena Report, op. cit., Appendix 3, p. 27. ER, Vol. II, Annex 1.
127 Section IV. Colombia’s Failure to Stop Spraying in Buffer Zones and
Other Protected Areas
2.162 Colombia praises itself in the Counter-Memorial for scrupulously
respecting areas where it has agreed not to spray. In this section, Ecuador shows
that this claim is untrueThe evidence shows that Colombia: (i) has twice
identified “buffer zonesŽ adjacent to the border with Ecuador, which it offered to
treat as a “safety marginŽ in which no aerial spraying would be conducted in
order to protect Ecuador against spray drift; and (ii) has nevertheless routinely
carried out aerial spraying operations in those areas close to the border even when
spray flights in those areas were supposedly “suspendedŽ. Colombia has also
sprayed in very close proximity to border area reserves that Ecuador has set aside
for vulnerable indigenous peoples, particularly the Awá and the Cofán, who have
been especially impacted by Colombia’s deposition of the spray mixture over
their communities.
A. COLOMBIA ’SS PRAYING INBUFFER ZONES ESTABLISHED TO PROTECT
E CUADOR FROM SPRAY DRIFT
1. Colombia Has Violated the 2.7-3.0 Kilometre “Safety Margin” It Previously
Claimed Was Sufficient to Protect Ecuador from Spray Drift
2.163 The data recorded by the spray planes demonstrate that they routinely
spray very close to the border, including in areas that Colombia said would be
128 320
off-limits to spraying in order to protect Ecuador . On 14 July 2001, in
responding to Ecuador’s concerns about harm to Ecuadorian territory, Colombia
represented that a “safety margin of 2.7-3.0 kilometres is believed to be
enoughŽ 321. Ecuador did not then agree, nor has it ever subsequently, that a 3
kilometre buffer zone is sufficient to protect it from aerial spraying by Colombia.
The point here is that Colombia did not observe its own self-described (and
inadequate) “safety marginŽ. Instead, it has routinely sprayed within 3.0
kilometres of the border. A total of 29,057 spray events were recorded in this
322
zone between 2000 and 2008 . Clearly, this volume of spraying adjacent to
Ecuadorian territory could not have been accidental.
2. Colombia Has Violated the 10 Kilometre Buffer Zone It Now Claims to
Respect
2.164 Ecuador does not consider that a 3 kilometre buffer zone is sufficient to
protect its people, animals and plants from harm caused by Colombia’s spray
programme. In Ecuador’s view, the evidence supports a prohibition on aerial
320
In addition, the flight path records indicate that there have been at least 4 spray events
conducted over Ecuadorian territory. Hansman & Mena Report, op. cit., p. 13. ER, Vol. II,
Annex 1.
321
Diplomatic Note DM/AL No. 25009, sent from the Ministry of Foreign Affairs of Colombia to
the Ministry of Foreign Affairs of Ecuador, p. 3 (14 July 2001). EM, Vol. II, Annex 42.
322
Hansman & Mena Report, op. cit., p. 12. ER, Vol. II, Annex 1.
129spraying by Colombia within 10 kilometres of Ecuadorian territory, as Ecuador
323
has continually demanded of Colombia since 2001 .
2.165 Colombia has never formally or definitively committed to refrain from
spraying within 10 kilometres of the international border. To be sure, the
Counter-Memorial describes a series of voluntary, non-permanent suspensions of
spraying activity within that distance of the border beginning in 2005. From
Ecuador’s standpoint, these suspensions by Colombia: (i) constitute an implicit
acknowledgement that 10 kilometres is the appropriate width of a buffer zone
sufficient to protect Ecuador from spray drift; and (ii) show that a 10 kilometre
324
buffer zone will not cause unacceptable consequences for Colombia .
2.166 Colombia maintains that it suspended spraying operations in the border
province of Nariño (adjacent to Ecuador’s province of Esmeraldas) from 27
December 2005 to 17 December 2006, and from 15 January 2007 to the present.
In Putumayo (adjacent to Ecuador’s province of Sucumbíos), Colombia states
323
Diplomatic Note 55416/2001- GM/SOI/SSN, sent from the Ministry of Foreign Affairs of
Ecuador to the Ministry of Foreign Affairs of Colombia (2 July 2001). EM, Vol. II, Annex 41;
EM, Chap. 3, para. 3.14.
324Moreover, the 10 kilometre buffer zone has been endorsed by international observers. Soon
after the January 2007 suspension, the UN Special Rapporteur on the right of everyone to the
enjoyment of the highest attainable standard of physical and mental health applauded Colombia’s
decision. He “welcome[d] the fact that aerial spraying of coca crops in the 10-km border zone
had ceased in February 2007Ž and further welcomed the Colombian Vice President’s statements
“that manual eradication tends to be more effective than aerial sprayingŽ. Report of the Special
Rapporteur on the Right of Everyone to the Enjoyment of the Highest Attainable Standard of
Physical and Mental Health, Paul Hunt: Preliminary Note on Mission to Ecuador and Colombia,
Addendum, U.N. Doc. A/HRC/7/11/Add.3, para. 16 (4 Mar. 2007). EM, Vol. II, Annex 31.
130that it suspended spraying from 1 January 2005 to 24 September 2005, from 11
325
December 2005 to 11 December 2006 and from 22 January 2007 to the present .
Ecuador wishes that this were so.
2.167 Despite Colombia’s assurances, the flight data generated by the spray
planes and furnished to the U.S. Department of State show that aerial spraying
inside the 10 kilometre buffer zone was carried out thousands of times while the
suspensions described above were supposedly in effect. For example, paragraph
5.90 of the Counter-Memorial states that the suspensions beginning in January
2007 have been “continuously maintainedŽ 326. However, Colombia has
conducted at least 5,287 spray events within the 10 kilometre buffer zone in
Nariño province since that date. In fact, Colombia’s spray planes have deposited
at least 20,630 gallons (78,093 litres) of the spray mixture within 10 kilometres of
327
Ecuador’s border since 1 February 2007 . The total number of spray flights
during Colombia’s purported suspensions in both Nariño and Putumayo
provinces exceeds 6,046 328.
325CCM, Chaps. 5, 7 and 10, paras. 5.62-5.63, 5.67, 5.71, 5.90, 7.3, 10.9; Report by the Anti-
Narcotics Direction of the Colombian National Police (DIRAN), pp. 310-311, 8 Feb. 2010. CCM,
Vol. II, Annex 67.
326CCM, Chap. 5, para. 5.90; see also CCM, Chap. 7, para. 7.3.
327Hansman & Mena Report, op. cit., p. 13. ER, Vol. II, Annex 1. Because Ecuador does not
have the precise calendar dates associated with each spray event, these figures represent spraying
within 10 kilometres of Ecuador’s border beginning on 1 February 2007 to the present. Ibid., n. 4.
328Ibid., p. 13.
1312.168 Colombia’s misrepresentations regarding the suspension of spraying
within the 10 kilometre buffer zone do not end there. On 11 November 2010, in
response to allegations by an indigenous organization that were published by
329
several news outlets , the Colombian Ministry of Foreign Affairs issued a Press
Release that stated:
“With respect to the information published today in the
Ecuadorian newspaper La Hora regarding alleged aerial aspersions
over illicit crops in the boundary area with Colombia, the National
Government allows itself to specify:
1. Since the month of January 2006, it has strictly complied with
the matters agreed by the Ministries of Foreign Affairs of
Colombia and Ecuador on 7 December 2005 in Quito, with regard
to suspending the aspersions in the boundary area with
330
EcuadorŽ .
2.169 This statement by the Colombian Ministry of Foreign Affairs … that
spraying within the 10 kilometre buffer zone has been suspended since January
2006 … directly contradicts the Counter-Memorial, which acknowledges that the 7
December 2005 agreement made by the Colombian Minister of Foreign Affairs as
a “gesture of goodwill towards EcuadorŽ was broken by a resumption in spraying
in both Nariño and Putumayo provinces between December 2006 and January
329
“Colombia Fumigates AgainŽ, LA HORA (Quito, 11 Nov. 2010). ER, Vol. IV, Annex 93;
“Colombian Government Violates Pact and Fumigates with GlyphosateŽ, VOCES.ORG (San
Salvador, 10 Nov. 2010). ER, Vol. IV, Annex 92.
330
Republic of Colombia, Ministry of Foreign Affairs, Press Release (11 Nov. 2ER, Vol.
V, Annex 156.
132 331
2007 . In fact, the flight data show that the period between December 2006 and
January 2007 was a period of heavy spraying in the border region: a total of
22,555 spray events were conducted within 10 kilometres of Ecuador’s border 332.
As discussed in Chapter 8, these serious discrepancies between Colombia’s words
and deeds demonstrate the need for a binding Court order requiring Colombia not
333
to spray within 10 kilometres of Ecuadorian territory.
B. C OLOMBIA ’SA ERIAL SPRAYING IN CLOSE PROXIMITY TO ECUADOR S
NDIGENOUS R ESERVES
2.170 Colombia has not only sprayed within its self-described buffer zones, it
has done so in very close proximity to highly sensitive areas in Ecuador,
specifically reserves set aside to protect the vulnerable communities of the Awá
and Cofán indigenous peoples living on their traditional lands. Colombia has
done so despite its imposition of a putative no-spray zone around sensitive areas
such as these.
2.171 In November 2001, Colombia’s Environment Ministry imposed a 2,000
metre buffer zone around national parks, in view of the likelihood of harm were
331
CCM, Chap. 5, para. 5.109(6). See also ibid., Chaps. 2 and 5, paras. 2.41, 5.62, 5.71; EM,
Chap. 3, paras. 3.3, 3.53-3.54, 3.64.
332Hansman & Mena Report, op. cit., Appendix 3, p. 27. ER, Vol. II, Annex 1. Ecuador cannot
verify Colombia’s statements regarding compliance with the 10 kilometre buffer zone in
November 2010 because the data in Ecuador’s possession ends in January 2009.
333
See infra Chap. 8, paras. 8.3, 8.14-8.15, 8.19.
133 334
spraying to occur in closer proximity . The Environment Ministry explained
that it was necessary to protect such sensitive environments “given the strategic
ecosystem’s importance, as well as environmental characteristicsŽ and “based on
preventative action principleŽ 33. The Ministry further explained:
“From the functionality perspective of the eco-systems found in
the parks, it should be taken into account that their bordering
areas, more than artificial borders are borders that interrelate with
organisms (flora and fauna) and natural resources contained within
and without its borders, therefore impacts sustained by the
surrounding areas may be harmful to ecosystems, resources and
species located within the natural parks considered of value, not
336
only for the country, but also for humanityŽ .
2.172 The Ministry of Environment further observed that it “based its
determination concerning the 2000 meter security zone surrounding the Natural
National Park system on the risks associated with the drift effects of the
Glyphosate when it is sprayed over illicit crops, as well as the fragility of the eco-
systems found in these areas, and their strategic importance from a social,
337
economic and ecological standpointŽ . For Ecuador, a 2 kilometre buffer zone
is terribly insufficient, as discussed above. The point here is that even the
334Republic of Colombia, Ministry of Environment, Resolution No. 1065, art. 5(d) (26 Nov.
2001). EM, Vol. II, Annex 15. In so ruling, the Ministry rejected a 1,000 metre buffer zone that
had been proposed by the DNE.
335Republic of Colombia, Ministry of Environment, Resolution No. 108, p. 5 (31 Jan. 2002). ER,
Vol. V, Annex 141. (“As for the parks’ protected zones under the scope of the Natural National
Park system’s Special Administrative Unit, this Ministry considered a 2,000 meter long zoneŽ.)
336Republic of Colombia, Ministry of Environment, Resolution No. 108, p. 5 (31 Jan. 2002). ER,
Vol. V, Annex 141.
337Ibid.
134inadequate 2 kilometre safety zone decreed by Colombia in regard to ecologically
sensitive areas was ignored, with particular consequence for the indigenous
communities living in remote areas along the Ecuador-Colombia border.
1. Cofán-Bermejo Ecological Reserve
2.173 The Cofán-Bermejo Ecological Reserve covers 55,451 hectares of tropical
338
lowland rainforest in northeastern Ecuador along the border with Colombia . It
is managed by the Cofán people themselves, in cooperation with the Government
of Ecuador, and is intended to protect their unique cultural and biological
heritage 339. Approximately one-fourth of Ecuador’s indigenous Cofán population
… about 320 people … reside in the Cofán-Bermejo Ecological Reserve, making it
340
a critical area for the protection of this vulnerable indigenous group .
2.174 The Cofán people’s stewardship of the Reserve has caused its well-
preserved forests to retain exceptionally high levels of plant and animal
341
diversity . For example, the forest is estimated to contain upwards of 2,000
plant species, including at least 15 endemic plants (i.e., species that exist nowhere
338Balslev Report, op. cit., p. 32. ER, Vol. II, Annex 4.
339Ibid.; Norman E. Whitten, Jr., Ph.D., Dr. William T. Vickers, Ph.D. & Michael Cepek, Ph.D.,
Tropical Forest Cultural Ecology and Social Adaptation in the Ecuadorian Border Region with
Colombia, pp. 19-20 (Jan. 2011) (hereinafter “Whitten et al. ReportŽ). ER, Vol. II, Annex 5.
340Balslev Report, op. cit., p. 32. ER, Vol. II, Annex 4.
341Balslev Report, op. cit., pp. 32-33. ER, Vol. II, Annex 4; Whitten et al. Report, op. cit., pp. 19-
20. ER, Vol. II, Annex 5.
135else on earth) 342. As explained by botanical and anthropological experts
intimately familiar with the area, apart from the biological significance of these
species, many are used by the Cofán for cultural and/or medicinal purposes 343.
The Reserve is also rich in animal life; a single scientific survey conducted in
2002 recorded 399 bird species, 42 species of large mammals and 31 species of
344
amphibians and reptiles .
2.175 The spray flight data demonstrate that Colombia has frequently conducted
aerial spraying within 2 kilometres of this sensitive area. In fact, 1,021 spray
lines were recorded within 2 kilometres of the Cofán-Bermejo Reserve between
2000 and 2008 345. Within 10 kilometres of the Reserve … a more suitable safety
346
zone … Colombia sprayed more than 12,398 times during the same period .
2.176 In Chapter 3, Ecuador describes in detail how the Cofán people have been
harmed by Colombia’s spraying in areas adjacent to the Cofán-Bermejo
Reserve 347.
342
Balslev Report, op. cit., p. 33-34. ER, Vol. II, Annex 4; Whitten et al. Report, op. cit., p. 19.
ER, Vol. II, Annex 5.
343
Balslev Report, op. cit., p. 33. ER, Vol. II, Annex 4; Whitten et al. Report, op. cit., p. 22. ER,
Vol. II, Annex 5.
344
Balslev Report, op. cit., p. 34. ER, Vol. II, Annex 4.
345
Hansman & Mena Report, op. cit., p. 14. ER, Vol. II, Annex 1.
346
Ibid.
347See infra Chap. 3, Section I(B)(1): The Kichwa and Cofán of Sucumbíos.
136 2. Awá Indigenous and Forest Reserve
2.177 The Awá Indigenous and Forest Reserve▯(Reserva Étnico Forestal Awá in
Spanish) is a 120,000 hectare area located immediately adjacent to the Colombian
348
border in the Esmeraldas and Carchi Provinces of Ecuador . The Reserve is
comprised of well-conserved primary forest, including several different
ecosystems ranging from the lowland Chocó rain forest to humid mountain and
cloud forests at the highest elevations 349. The Reserve is home to approximately
3,000 indigenous Awá who depend on its forest resources for their daily
350
survival . It also harbours remarkable biodiversity, including the Brown-
Headed Spider monkey (Ateles fusciceps), the Ecuadorian Sac-Winged Bat
(Balantiopteryx infusca), Jaguar (Panthera onca), Neotropical Otter (Lontra
longicaudis), and the Spectacled Bear (Tremarctos ornatus), among the many the
animal species that inhabit the Reserve 351. As described by Whitten et al., the
region inhabited by the Awá in Ecuador on the border of Colombia “is one of the
352
richest, wettest, high biodiversity rain-forest regions of the worldŽ .
2.178 Colombia’s disregard for the 2 kilometre safety zone that its Ministry of
Environment determined was necessary to protect sensitive areas is evidenced by
348Balslev Report, op. cit., p. 29. ER, Vol. II, Annex 4.
349
Ibid.
350
Whitten et al. Report, op. cit., pp. 45, 47. ER, Vol. II, Annex 5.
351Balslev Report, op. cit., pp. 29-30. ER, Vol. II, Annex 4.
352Whitten et al. Report, op. cit., p. 45. ER, Vol. II, Annex 5.
137the flight data. Colombia sprayed 57 times within 2 kilometres of the Reserve
353
between 2000 and 2008 . Colombia also sprayed 10,913 times within 10
kilometres of the Reserve, the buffer zone which Ecuador has consistently
maintained is necessary to protect its territory, including biologically and
354
culturally sensitive areas such as the Awá Indigenous and Forest Reserve .
2.179 The harm that has been inflicted on the Awá by the spraying programme
355
is described in detail in Chapter 3.
C. C OLOMBIA ’SSPRAYING IN C LOSE PROXIMITY TO E CUADOR S N ON -
INDIGENOUS BORDER C OMMUNITIES
2.180 Ecuador’s indigenous communities are not the only ones subjected to
Colombia’s aerial spraying 2 kilometres or less from their homes, even though in
2001 Colombia enacted regulations prohibiting spraying within 2 kilometres of
human settlements. For example, Colombia sprayed 719 times between 2000 and
2008 within 2 kilometres of the Ecuadorian community of Mataje, located in
Esmeraldas Province 356.
35Hansman & Mena Report, op. cit., p. 14. ER, Vol. II, Annex 1.
35Ibid.
355
See infra Chap. 3, Section I(D)(1): The Awá of Esmeraldas.
356
Hansman & Mena Report, op. cit., p. 14. ER, Vol. II, Annex 1.
1382.181 Other Ecuadorian communities on the Colombian border have
experienced a similar intensity of spraying. For example, Colombia sprayed 174
times within 2 kilometres of the community of Puerto Mestanza, located in
357
Ecuador’s Sucumbíos Province .
2.182 It is not coincidental that the residents of Mataje and Puerto Mestanza,
like the residents of other Ecuadorian border communities in very close proximity
to where Colombia conducts aerial spraying, have experienced serious harm to
their health, crops, animals and livelihoods, as detailed in Chapter 3 358.
Section V. The Spray Mixture Reaches Ecuador in Quantities Sufficient to
Cause Serious Harm
2.183 In this section, Ecuador refutes Colombia’s core contention, which it
argues throughout the Counter-Memorial, that the spray programme cannot cause
harm in Ecuador because the spray mixture does not drift more than a few metres
359
from the target areas where it is released . The evidence overwhelmingly
contradicts Colombia’s argument. It shows that Colombia’s pervasive violations
of its own requirements to prevent spray drift have had a dramatic impact on
357
Hansman & Mena Report, op. cit., p. 14. ER, Vol. II, Annex 1.
358See infra Chap. 3, Sections I and II.
359See, e.g., CCM, Chap. 7, para. 7.3 (“[T]aking into account the scientific evidence on the
limited effect of drift and the strict technical parameters under which the spraying operations are
carried out in Colombia … including the observance of 100m exclusion strips along watercourses …
no damage could have occurred in Ecuadorian territoryŽ.). See also ibid., paras. 7.16-7.30, 7.33,
7.161.
139Ecuador. In particular, it shows the manner in which the spraying is done,
including the height at which the spray mixture is released, the aircraft speed and
application rate, the droplet size, the night time spraying, the disregard of
temperature and wind conditions, combined with the frequency with which areas
near Ecuador’s border are sprayed and the especially toxic nature of the chemical
cocktail, all assure that the spray mixture is deposited far into Ecuador in amounts
which exceed what is needed to cause harm.
A. C OLOMBIA ’SM ISREPRESENTATIONS REGARDING THE EXTENT OF SPRAY
D RIFT
2.184 Colombia has consistently represented to Ecuador and to the international
360
community, and now represents to the Court, that off-target drift from its spray
operations is minimal.
2.185 Starting in the early years of the spray programme, Colombia’s DNE … the
agency responsible for carrying out the spray operations … was warned about the
potential for significant off-target driA report issued by the Colombian
Agriculture Institute (“ICAŽ) in 1999 in response to the DNE’s draft EMP
concluded that the “permissible driftŽ reported by the DNE of less than 2 to 5
36See CCM, Chap. 7, para. 7.3. See also CCM, Chap. 7, paras. 7.93, 7.170.
140metres was “impossibleŽ given the “height and speedŽ of the spray planes 36. The
ICA further explained that the 80 percent rate of recovery reported by the DNE
was “inconsistentŽ given the “considerable speed and height of operationŽ 362.
2.186 This assessment regarding the danger of drift was confirmed by a study
dated July 2004 by Sociedad Las Palmas Ltda., a consulting firm commissioned
by the Colombian government to evaluate the drift associated with different spray
mixtures 363. The Las Palmas study concluded that an astounding 72% of the
spray mixture containing Cosmo-Flux drifted off-target 364. Indeed, the mixture
with Cosmo-Flux had the worst drift of all the mixtures that were assessed,
prompting its authors to recommend that Colombia cease using that adjuvant in
365
the spray, a recommendation that was ignored .
2.187 These warnings were not just disregarded by Colombia. DNE and
Colombia’s Ministry of Foreign Affairs communicated precisely the opposite
message to Ecuador. In a diplomatic note sent to Ecuador on 14 July 2001,
361Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 599, p. 14 (23 Dec. 1999). ER, Vol. V, Annex 132.
362
Ibid., p. 13.
363Las Palmas Report, op. cit., pp. 4-5. ER, Vol. III, Annex 15.
364Ibid., pp. 40-41, 107. ER, Vol. III, Annex 15; see also Giles Report, op. cit., p. 46. Annex 2.
365The Las Palmas study concluded that another available adjuvant, which had the same level of
effectiveness at killing coca plants, resulted in a much lower off-target deposition of
approximately 30%. Las Palmas Report, op. cit., pp. 40-41, 104-107. Nevertheless, Colombia
persisted in using Cosmo-Flux and even claims that it is a “drift control agentŽ. See CCM, Chap.
4, paras. 4.42, 4.51-4.56.
141Colombia’s Ministry of Foreign Affairs represented that strict operational
366
parameters “guarantee a negligible drift, that is, less than 5 metersŽ .
Approximately three years later, a letter from the Director of DNE to the
Ecuadorian Scientific and Technical Commission dated 14 April 2004 repeated
these limited estimates of spray drift, stating that “maximum driftŽ is estimated to
be 12 metres 367. On 18 September 2005, Colombia made the same representation
to the IACHR, stating that the maximum extent of drift was 12 metres 368.
Colombia continued to maintain these gross misrepresentations of spray drift as
recently as 2007, claiming in a Position Statement to the Binational Scientific and
366
Diplomatic Note DM/AL No. 25009, sent from the Ministry of Foreign Affairs of Colombia to
the Ministry of Foreign Affairs of Ecuador, p. 3 (14 July 2001). EM, Vol. II, Annex 42. A
similar estimate is included in Colombia’s Environmental Management Plan issued on 30
September 2003, which states that “Foreseen DriftŽ is less than 5 metres. Resolution Nº 1054 of
30 September 2003 of the Ministry for the Environment of Colombia, p. 173. CCM, Vol. II,
Annex 50. As discussed in Chapter 4, the current version of the EMP was finally adopted by the
Ministry of Environment after a long battle with the DNE. It appears that the DNE ultimately
won.
367
Note No. SARE-142 from the Director of the National Narcotics Directorate of Colombia to
the President of the Scientific and Technical Commission of Ecuador, p. 33 (14 Apr. 2004).
CCM, Vol. II, Annex 13. As discussed in the Giles Report, this estimate was based on an overly
simplistic spray drift calculation, despite the availability of more sophisticated models such as
AGDISP since the 1980s. Colombia’s calculation took into account a mere three variables.
Worse yet, the variables … a maximum spray height of 25 metres, an average droplet size of 650
microns and a wind speed of 4.8 km/hr (1.3 m/s) … do not reflect the actual conditions of
application. Thus, the calculation presented by Colombia to Ecuador’s Scientific and Technical
Commission leads to a gross underestimation of spray drift. Giles Report, op. cit., pp. 45-46. ER,
Vol. II, Annex 2.
368
Diplomatic Note No. DDH. 58003 from the Colombian Foreign Ministry to the Executive
Secretary of the Inter-American Commission on Human Rights, p. 26, Table No. 1, 18 Sept. 2005
(indicating that the “permissible driftŽ is less than 12.3 metres). ER, Vol. V, Annex 154.
142Technical Commission that “under adverse parameters of maximum wind and
369
application altitudeŽ drift would extend 12 to 24.1 metres .
2.188 As explained below, Colombia misrepresented the extent of spray drift …
not by metres but by kilometres. In fact, an internationally-accepted spray drift
model, which is relied upon by Colombia in its Counter-Memorial … and which
has been available to Colombia since the inception of spray operations near
Ecuador’s border in 2000 … demonstrates that the amount of spray deposited as
370
far as 10 kilometres into Ecuador is more than sufficient to cause serious harm .
B. SPRAY D RIFT M ODELING
2.189 The Counter-Memorial relies upon spray drift modeling commissioned by
the Colombian and U.S. governments in 2009 purportedly to show that spray
cannot reach Ecuador. For example, it states at paragraph 7.20, in reliance on this
drift modeling, that the “effects of sprayingŽ are “negligible beyond 120 meters
even for those plants most sensitive to the spray mixtureŽ 37. This statement is
based on the results of a spray drift study conducted by Dr. Andrew Hewitt and
369Republic of Colombia, Position Statement by Colombia to the Binational Scientific and
Technical Commission in Relation to the Destruction in Colombia of Illicit Crops in the Frontier
Zones with Ecuador, p. 23 (8 June 2007). ER, Vol. V, Annex 155.
370See infra Chap. 2, paras. 2.200-2.203; Giles Report, op. cit., pp. 4, 47-48. ER, Vol. II, Annex
2; Weller Report, op. cit., pp. 17-25. ER, Vol. II, Annex 3.
371CCM, Chap. 7, para. 7.20 (citing Hewitt et al., 2009, op. cit., pp. 923, 925, 929. CCM, Vol.
III, Annex 131-B).
143colleagues, attached to Colombia’s Counter-Memorial as Annex 131-B. In his
study, Dr. Hewitt used an internationally accepted spray drift model called
AGDISP, which was “developed and validated by NASA, the U.S. Forest
Service, U.S. Army, Spray Drift Task Force, and others over several decades for
372
aerial forestry and agricultural spray applicationsŽ . He compared the results
obtained by use of that model … an estimated deposition rate in grams per hectare
of glyphosate acid equivalent (a.e.) … to dose-response thresholds for plants.
Based on this comparison, Dr. Hewitt concluded that plants more than 50 to 120
metres from the spray application site would not be exposed to a sufficient
amount of spray to be harmed 373.
2.190 Ecuador has no quarrel with the use of the AGDISP model as a useful
predictive tool. Nor does Ecuador dispute that comparing an estimated deposition
rate with known dose-response values for plants is a sound way to evaluate the
likelihood of harm to those plants. However, as described in greater detail in the
paragraphs that follow, the problem with Colombia’s modeling is that by
definition, models rely on a set of factual assumptions, and the assumptions that
374
were fed into Dr. Hewitt’s model are demonstrably wrong . They are based
entirely on Colombia’s representations, now proven to be false, that it strictly
372Hewitt et al., 2009, op. cit., p. 921 (internal citations omitted). CCM, Vol. III, Annex 131-B.
373
Ibid., p. 921, 928.
374
See infra Chap. 2, paras. 2.191-2.197; Giles Report, op. cit., pp. 9-42, 47. ER, Vol. II, Annex
2.
144complied with all operational requirements for avoiding spray drift contained in
the EMP and boasted of in the Counter-Memorial. When the false assumptions
are corrected to reflect how the aerial spray programme is actually conducted, the
model employed by Colombia actually makes Ecuador’s case: it shows that
significant amounts of the spray mixture drift long distances from the place of
release, and thus reach far into Ecuador, where they are of sufficient toxicity to
375
cause significant harm in Ecuador .
1. Flaws in Colombia’s Modeling
2.191 As indicated, the fatal flaw in Colombia’s modeling is that it assumes that
the aerial spray programme complies with its own operational requirements,
including those relating to aircraft speed and altitude. As demonstrated above,
however, these assumptions are wrong. In reality, Colombia disregards those
requirements and sprays with aircraft travelling much faster and higher than
allowed. For example, Colombia’s modeling assumes that its planes fly at speeds
376
ranging from 226 km/hr to 333 km/hr (140 to 207 mph) , but in reality, they
frequently fly much faster; speeds above 333 km/hr have been recorded for more
than 11,113 flights between 2000 and 2008 within 10 kilometres of the border
375See infra Chap. 2, paras. 2.199-2.202; Giles Report, op. cit., pp. 9-42, 47. ER, Vol. II, Annex
2; Weller Report, op. cit., pp. 17-25. ER, Vol. II, Annex 3.
376
Hewitt et al., 2009, op. cit., p. 923. CCM, Vol. III, Annex 131-B
145with Ecuador 377. As a result, the higher flight speeds disperse spray droplets that
are much smaller, and which drift much longer distances, than would otherwise
be the case 37. Likewise, despite the fact that Colombia’s modeling assumed that
the spray is released at a height of 30.48 metres 379, it is invariably released at far
higher altitudes. The data show more than 81,106 of the 114,525 spray events
(71%) within 10 kilometres of Ecuadorian territory during which the spray
mixture was released at altitudes higher than those assumed in Colombia’s drift
380 381
modeling . This, as shown, also increases the spray’s propensity to drift .
2.192 Nor is the model’s assumption about application rate accurate. Although
the EMP requires an application rate of no greater than 23.65 litres per hectare,
382
Colombia frequently exceeds that limit . Dr. Hewitt, in fact, not only
underestimated the real application rate, and thus the deposition of the spray
mixture, he even misestimated the application rate required by the EMP by a
factor of 2.2. Rather than using 23.65 as the application rate for the total tank
mix, as Colombia’s spray programme dictates, Dr. Hewitt describes his model
377Hansman & Mena Report, op. cit., p. 20. ER, Vol. II, Annex 1.
378See supra Chap. 2, para. 2.97; Hansman & Mena Report, op. cit., p. 20. ER, Vol. II, Annex 1;
Giles Report, op. cit., pp. 6-8, 20-21. ER, Vol. II, Annex 2.
379Hewitt et al., 2009, op. cit., p. 923. CCM, Vol. III, Annex 131-B.
380Hansman & Mena Report, op. cit., Appendix 3, p. 28. ER, Vol. II, Annex 1.
381
Giles Report, op. cit., pp. 16-20. ER, Vol. II, Annex 2.
382
See supra Chap. 2, paras. 2.140-2.141; Hansman & Mena Report, op. cit., p. 23. ER, Vol. II,
Annex 1.
146input for the spray volume rate as “10.4 L/ha (1.11 gallon/acre) total tank mix for
383
coca spraysŽ , i.e., less than half of the maximum allowable rate under the EMP.
2.193 These errors are compounded by another serious flaw: inexplicably,
Colombia’s modeling assumes that the spray planes carry out only one line of
aerial spraying. In fact, the flight data recorded by the spray planes and obtained
from the U.S. Department of State demonstrate that the spray is released in
numerous tightly packed parallel lines 38. Figure 2.11 is a representative sample
of an area sprayed adjacent to Ecuador’s Sucumbíos Province in September 2002;
the two inset boxes show magnified views of 1 square kilometre areas where
Colombia sprayed. As the Court can see, there were over 20 parallel spray lines
in those areas alone. The chemical spray from each individual line drifts and
contributes to deposition in Ecuador 385. This is depicted by Figure 2.12, which
illustrates the cumulative deposition downwind of increasing numbers of spray
lines: two spray lines deposit nearly twice as much herbicide as a single spray
line; three lines deposit almost triple the amount; and ten lines result in close to a
ten-fold increase in deposition. When one considers the large number of parallel
383
Hewitt et al., 2009, op. cit., p. 923. CCM, Vol. III, Annex 131-B; see also Giles Report, op.
cit., pp. 24-25. ER, Vol. II, Annex 2.
384
Hansman & Mena Report, op. cit., p. 29. ER, Vol. II, Annex 1.
385
Giles Report, op. cit., p. 9 (“Many aerial spraying operations, including those conducted in
Colombia, involve multiple parallel flights to cover the target area. Physically and analytically,
each flight is an independent event, i.e., the aircraft wake and the droplets released from one pass
dissipate and do not affect the motion of droplets from subsequent passes. Therefore, the total
spray deposition at any point downwind from multiple passes is an accumulation of the spray
deposition from each individual spray passŽ.) (internal citation omitted). ER, Vol. II, Annex 2.
147 0°30'0"N
km
River Figure 2.11
Putumayo
! Palma Seca
0.5 Km
River
San MiChone 2
76°30'0"W ! Chone 1 76°30'0"W
!
!
0 5 10 15 20 25
! Playeratal
C O L O M B I A 0 1
±
Puertoo
P u t u m a y o
Cor5 de Agosto
Puert! MestaUnion Lojana
Puerto Escondido
! !
!
! !
Monterrey S u c u m b í o s
! E C U A D O R
Dios Peña
! SaMarianitaarapa
0.5 Km ! !
San Miguel
General
!
La Condor
! !
!
0 1 10 de Agosto
!
! San SFarannFcrisacnoci2sco 1
77°0'0"W Salinas 77°0'0"W
Border During a Single Month (September 2002) Virgen del Cisne
Density of Parallel Spray Lines Within 10 Kilometres of Ecuador’s
Peru
River
San Miguel Colombia
Protected areas from international boundary
Ecuador
Legend
N a r i ñ o
Cofán Reserve
0°30'0"N Figure 2.12
WIND
0 km 1 km 2 km 3 km 5 km 10 km
WIND
0 km 1 km 2 km 3 km 5 km 10 km
WIND
2 Spray Lines 3 Spray Lines 10 Spray Lines
0 km 1 km 2 km 3 km 5 km
10 km
Cumulative Impact of Drift From Multiple Spray Lines
WIND
1 Spray Line
(Colombia’s Assumption)
0 km 1 km 2 km 3 km 5 km 10 kmlines depicted on the map in Figure 2.11 (over 20 lines in the inset boxes alone)
the multiplicative effect of Colombia’s spraying practices can be readily
appreciated.
2.194 Nevertheless, Colombia’s modeling only accounts for a single spray line,
ignoring the dozens more that were frequently sprayed. As a result, it
significantly underestimates the amount of spray deposited downwind 386. Indeed,
it is particularly odd that the Hewitt report commissioned by Colombia fails to
take account of the impact of multiple spray lines since the AGDISP model he
used includes a feature designed to calculate the effect of multiple spray lines 387.
Dr. Hewitt apparently chose not to apply this feature or otherwise account for the
multiple spray lines that regularly characterised Colombia’s spray operations.
This is all the more remarkable since previous studies authored by Dr. Hewitt
388
have taken into account the effect of multiple spray lines .
2.195 Further undermining the accuracy of Colombia’s modeling: its exposure
analysis fails to consider the fact that the spray mixture used by Colombia
includes Cosmo-Flux, a powerful chemical that is added to increase the spray’s
386Ibid. (“Using only a single pass to determine drift deposition when multiple passes were made
will significantly underestimate deposition from spray driftŽ.).
387Ibid.
388See, e.g., A. J. Hewitt et al., “Development of the Spray Drift Task Force Database for Aerial
ApplicationsŽ, Environmental Toxicology and Chemistry, Vol. 21, No. 3, pp. 648-658 (2002)
(“four parallel spray swaths (flight line passes) . . . were used in the aerial studiesŽ). ER, Vol. III,
Annex 12.
148lethality389. As detailed in paragraph 5.21 of the Memorial, the manufacturer of
Cosmo-Flux reports that it increases the killing-power of the spray mixture by a
factor of four 390. This four-fold increase in efficacy has also been reported in the
study commissioned by the U.S. government that led to Colombia’s selection of
391
Cosmo-Flux for inclusion in the spray mixture . However, in assessing whether
the amount of spray deposited downwind due to spray drift is sufficient to cause
harm to plants, Colombia’s modeling considers only the toxicity supplied by the
glyphosate element of the spray mixture, and fails to take account of the vastly
392
increased toxicity caused by the addition of Cosmo-Flux .
2.196 These are not the only flaws in Colombia’s modeling. Dr. Hewitt
concludes that the level of damage to plants becomes unacceptable when more
than 43 g/ha of spray is deposited because, in his estimation, 5 percent of plant
389See supra Chap. 2, paras. 2.55-2.56; Weller Report, op. cit., p. 7-8, 15-16. ER, Vol. II, Annex
3.
390Cosmoagro, S.A., Cosmo-Flux 411F. EM, Vol. III, Annex 112.
391Ronald T. Collins & Charles S. Helling, Surfactant-Enhanced Control of Two Erythroxylum
Species by Glyphosate, Weed Technology, Vol. 16, p. 851 (2002). EM, Vol. III, Annex 141
(reporting that two glyphosate-surfactant systems tested “increased glyphosate toxicity fourfoldŽ);
Chemicals Used, op. cit., p. 2 (explaining that Cosmo-Flux was selected because it “most closely
matched the most effective U.S. productsŽ that had been tested by Collins & Helling). EM, Vol.
III, Annex 144; CCM, Chap. 4, para. 4.53 (explaining the “Colombia chose the adjuvant
Cosmoflux 411FŽ as a result of the Collins & Helling study). See also Dobson Report, op. cit., p.
538. CCM, Vol. I, Appendix; Weller Report, op. cit., p. 15-16. ER, Vol. II, Annex 3; Menzie &
Booth Report, op. cit., p. 24-25. ER, Vol. II, Annex 6.
392
Hewitt et al., 2009, op. cit., p. 923. CCM, Vol. III, Annex 131-B; Weller Report, op. cit., pp.
15-16. ER, Vol. II, Annex 3.
149species would be harmed 393. But that 43 g/ha threshold is based on an artificial
selection of plant species. Dr. Hewitt only considers dose-response values for a
selection of hardy Northern Hemisphere crops, including turnips, radishes, beets
and oats, the majority of which are not typically grown in tropical locations like
394
the Ecuador-Colombia border region . Indeed, he considers a very limited
number plants that are representative of tropical agriculture, let alone of the
395
endangered plant species endemic to the Ecuador-Colombia border region .
2.197 The extent to which Dr. Hewitt underestimates the risk of harm in the
Ecuador-Colombia border region is made clear when his 43 g/ha value … the
threshold at which he says harm to plants would become unacceptable … is
compared with the conclusions of other published studies on plant injury caused
by glyphosate. For example, a study sponsored by the Danish Environmental
Protection Agency and Environment Canada by Boutin et al. derived a much
lower value for protecting plant species 396. Instead of 43 g/ha (the level arrived at
393Hewitt et al., 2009, op. cit., pp. 926-928, Table 3. CCM, Vol. III, Annex 131-B; see also
Weller Report, op. cit., pp. 13-14. ER, Vol. II, Annex 3.
394Hewitt et al., 2009, op. cit., p. 927, Table 2. CCM, Vol. III, Annex 131-B; see also Weller
Report, op. cit., pp. 13-14. ER, Vol. II, Annex 3.
395Weller Report, op. cit., pp. 13-14. ER, Vol. II, Annex 3; Balslev Report, op. cit., pp. 5-24, 28.
ER, Vol. II, Annex 4; Whitten et al. Report, op. cit., pp. 10, 19-22, 28-31, 34-37, 40, 47-48, 50.
ER, Vol. II, Annex 5. Leaving aside the methodological flaws highlighted above, the notion that a
5 percent plant injury level is acceptable is a normative judgment which Colombia is not entitled
to make on behalf of Ecuador.
396
Weller Report, op. cit., p. 14. ER, Vol. II, Annex 3 (citing C. N. Boutin, C. Elmegaard and C.
Kjaer, “Toxicity Testing of Fifteen Non-crop Plant Species with Six Herbicides in a Greenhouse
Experiment: Implications for Risk AssessmentŽ, in Ecotoxicology. 13:349…369 (2004)). ER, Vol.
III, Annex 13.
150by Dr. Hewitt) the threshold calculated by the Boutin et al. study was over 90
percent lower: 4.1 g/ha. Moreover, the Boutin et al. study is consistent with other
evaluations of plants injury caused by glyphosate-based products, which conclude
that exposure to significantly less than 43 g/ha can harm plants 397. For example,
corn (maize), which is commonly grown in the Ecuador-Colombia border
region 398, has been shown to be injured by glyphosate at doses as small as 26.25 g
a.e./ha 399, which is 38 percent lower than the threshold Hewitt et al. (2009) stated
was sufficient to guard against harm to “sensitiveŽ plant species 400. Indeed,
tomatoes that are exposed to a mere 3 g/ha of glyphosate exhibit a loss of flowers,
401
which leads to loss of fruits . The discrepancy between Colombia’s view of an
acceptable threshold and those calculated by others is all the more noteworthy
since none of these other studies take into account the effect of Cosmo-Flux,
402
which effectively reduces the threshold for plant injury by a factor of four .
397Weller Report, op. cit., p. 14. ER, Vol. II, Annex 3.
398Whitten et al. Report, op. cit., pp. 3, 10, 21, 28, 31, 36-38, 40, 47, 50. ER, Vol. II, Annex 5
399Weller Report, op. cit., p. 14. ER, Vol. II, Annex 3.
400
Hewitt et al., 2009, op. cit., pp. 926-928, Table 3. CCM, Vol. III, Annex 131-B; Weller
Report, op. cit., p. 14. ER, Vol. II, Annex 3.
401
Ibid.
402
See supra Chap. 2, para. 2.56; Weller Report, op. cit., pp. 15-16. ER, Vol. II, Annex 3.
Moreover, the studies did not take into account the risk to sensitive plant species in Ecuador’s
border region, which is home to thousands of plant species, many of them endemic to the region
or threatened with extinction. Weller Report, op. cit., p. 14. ER, Vol. II, Annex 3; Balslev
Report, op. cit., pp. 5-24. ER, Vol. II, Annex 4.
151 2. Corrected Modeling
2.198 As shown in the preceding paragraphs, the modeling done by Colombia is
flawed by its reliance upon demonstrably false data. When the same AGDISP
model used by Colombia is run using inputs that reflect the actual data, the results
change dramatically and show that the spray mixture is deposited far into Ecuador
… including at least 10 kilometres from the site of application … in amounts
sufficient to cause harm 403. As depicted in Figures 2.13 and 2.14, this
encompasses all of the locations in the Ecuadorian Provinces of Sucumbíos and
404
Esmeraldas that have been harmed .
2.199 For example, at just one kilometre from the site of application, running the
model with median values from the flight path data obtained from the U.S.
th
Department of State, i.e., the 50 percentile value for flight speed (171.20 mph or
th
275.52 km/hr) and the 50 percentile value for altitude (40.61 metres) for the AT-
405
802 aircraft (the spray plane that Colombia admits to using) , yields 4.91 g/ha of
deposition 406. That single line, by itself, deposits at that distance more herbicide
than the Danish and Canadian study (Boutin et al. 2004) concluded can injure
403See infra Chap. 2, paras. 2.198-2.201. The AGDISP drift modeling runs are presented in the
Giles Report (ER, Vol. II, Annex 2); the inputs for each drift modeling run are provided on a data
CD deposited with the Registry.
404
See infra Chap. 3, Section I.
405Hansman & Mena Report, op. cit., p. 25. ER, Vol. II, Annex 1.
406Giles Report, op. cit., p. 13, Table 4. ER, Vol. II, Annex 2.
152 407
plants (4.1 g/ha) . When the increased killing-power caused by Cosmo-Flux is
taken into account, this deposition has the lethality equal to an effective dose of
19.64 g/ha 408. Thus, a plane flying at the median speed and median height that
sprays only one line of herbicide will cause to be deposited a kilometre away an
amount of spray that is nearly five times the amount necessary to kill or injure
plants 409.
2.200 But the situation is actually much worse, because the spray planes do not
just release a single line of the spray mixture, as Colombia’s drift modeling
410
wrongly assumed . Colombia’s failure to account for multiple spray lines
changes the results, significant as they already are, dramatically. For example, if
th th
there are 3 spray lines each at the 50 percentile for speed and the 50 percentile
for height … an extremely conservative approach given the dozens of parallel lines
that characterise Colombia’s spraying practices … the effective deposition, taking
into account the effect of Cosmo-Flux, at 1 kilometre is 57.08 g/ha 411. That is
roughly 25 percent more than the amount that even Colombia concedes causes an
unacceptably high level of damage 412. The effective deposition from 10 spray
407Weller Report, op. cit., p. 18. ER, Vol. II, Annex 3.
408Ibid.
409
Ibid.
410
Hansman & Mena Report, op. cit., p. 29. ER, Vol. II, Annex 1; Giles Report, op. cit., pp. 9,
34-40. ER, Vol. II, Annex 2.
411
Weller Report, op. cit., p. 18. ER, Vol. II, Annex 3.
412
Ibid.
153 0°30'0"N
km
River Figure 2.13
Putumayo
! Palma Seca
River
San Miguel
76°30'0"W ! Chone Chone 1 76°30'0"W
!
!
Playeratal 0 5 10 15 20 25
!
±
Puertoo
Corazón Orense
Puert! Mestanzagosto
Puerto EscondidoLojana
! !
C O L O M BP u t u m a y o !
!! S u c u m b í o s
Monterrey
! E C U A D O R
Dios Peña
! Santa La Charapa
Marianita
! !
San Miguel
Farfán
Province (2000 - 2008) General
!
0-1 k-2 k-3 k-5 k-7 k7-10 km La Condor
! !
!
10 de Agosto
!
Distance from nearest
! San SFarannFcrisacnoci2sco 1
77°0'0"W 77°0'0"W
Salinas Virgen del Cisne
Distance From Spray Events Adjacent To Ecuador’s Sucumbíos
Peru
River
San Miguel Colombia
Spray eventsternational boundary
Protected ar1e0aksm from international boundary
Ecuador
Legend
N a r i ñ o
Cofán Reserve
0°30'0"N 1°30'0"N 1°15'0"N 1°0'0"N
0-1 -2 -3 k-5 -7 k-10 km
Figure 2.14
78°15'0"W km 78°15'0"W
Distasperafnen(re)st
Protected areasfrom international boundary
Legend
C a r c h i
River 0 5 10 15 20 25
78°30'0"W 78°30'0"W
Awá Reserve
±
San !uan
N a r i ñ o
Mataje Alto
C O L O M B I A
Guadualito
River !
Mataje
!
78°45'0"W Mataje 78°45'0"W
Province (2000 - 2008) Awá Reserve
H
San Lorenzo E s m e r a l d a s
E C U A D O R
Cayapas-Mataje
Distance From Spray Events Adjacent To Ecuador’s Esmeraldas
79°0'0"W Peru 79°0'0"W
Colombia
Ecuador
1°30'0"N 1°15'0"N 1°0'0"Nlines … a more accurate but still conservative assumption given how many parallel
lines Colombia normally sprays … is 163.80 g/ha, nearly four times the amount
that Colombia acknowledges is unacceptable 41. It is also nearly 40 times the
threshold calculated by Boutin et al.; 55 times the amount needed to damage
tomato plants; and six times the amount sufficient to injure maize.
2.201 The flight data show that Colombia frequently sprays in a manner than is
much worse than assumed in the preceding paragraphs, which presumed speed
and height were at the median (50 thpercentile) of flights documented in the data.
Consider what happens for a flight by an OV-10 aircraft (responsible for 20,251
spray events within 10 kilometres of Ecuador’s border between 2000 and 2008 414)
th th
operated at the 90 percentile of speed (207.50 mph or 333.94 km/hr) and 90
415
percentile for altitude (42.56 metres) . A single line of spray results in an
416
effective deposition of 54.24 g/ha at 1 kilometre from the application site . This
exceeds the amount that Colombia concedes would injure plants (43 g/ha) and is
over 13 times more spray than the amount that Boutin et al. (2004) determined is
417
enough to cause injury . More concretely, it is twice the amount needed to
413
Ibid.
414Hansman & Mena Report, op. cit., Appendix 3, p. 25. ER, Vol. II, Annex 1.
415Ibid., p. 25.
416Weller Report, op. cit., p. 20. ER, Vol. II, Annex 3.
417
Ibid., p. 18.
154injure maize, and is over 18 times the amount of spray needed to injure tomato
plants 418.
2.202 Thus, Colombia’s spraying sends into Ecuador enough of the spray
mixture to cause serious injury. To be sure, the amount of spray reaching deeper
into Ecuador is less than the amount falling just inside it; but inserting the correct
inputs into Colombia’s model demonstrates that the levels of spray reaching
points 10 kilometres from the application site is still sufficiently toxic to cause
significant harm. For example, four spray lines from an AT-802 aircraft flown at
th th
the 50 percentile for height and 50 percentile for speed will result in an
effective dose of 4.32 g/ha of spray 10 kilometres away, which exceeds the 4.1
419
g/ha threshold for harm derived by Boutin et al. . It would take only two spray
lines from an OV-10 aircraft operating at the 90 thpercentile for height and the
90 percentile for speed to deposit an effective dose at 10 kilometres that exceeds
the same plant injury threshold 420. These are conservative assumptions. Since
dozens of spray lines were routinely documented, the amount of spray deposited
10 kilometres downwind into Ecuador, in actuality, is correspondingly more, and
421
well exceeds the amount necessary to cause significant harm . In that regard,
418Ibid., pp. 14, 20.
419Ibid., pp. 18-19.
420Ibid., pp. 19-20.
421
Hansman & Mena Report, op. cit., p. 29. ER, Vol. II, Annex 1; Giles Report, op. cit., pp. 9,
34-40. ER, Vol. II, Annex 2; Weller Report, op. cit., pp. 17-18. ER, Vol. II, Annex 3.
155using the same example as above (an AT-802 aircraft flying at the 50th percentile
for height and the 50th percentile for speed) but with 10 spray lines (a much more
realistic scenario, according to the flight data or than as hypothesized above) the
deposition of herbicide 10 kilometres from the site of the spraying would be more
than two-and-a-half times the level necessary to harm plants 422.
2.203 As would be expected, the amount of herbicide that is deposited at
distances between one and 10 kilometres is also sufficient to cause serious injury
to plants, including food crops. For example, at 5 kilometres from the site of
release, enough herbicide is deposited (an effective dose of 7.24 g/ha) from only
two lines of spray from an AT-802 flying at the 50 thpercentile for speed and the
th
50 percentile for height, to exceed the threshold for unacceptable damage
423
established by Boutin et al. . Indeed, only a single spray line flown by an OV-
th
10 at the 90 percentiles for speed and height deposits at 5 kilometres downwind
424
an effective dose (5.92 g/ha) that exceeds the Boutin et al. threshold . And it
would take only 8 spray lines from an AT-802 flying at the 50 th percentiles for
speed and height … a conservative scenario given the much higher number of
spray lines routinely flown by Colombia … to deposit an effective dose of 27.44
422Giles Report, p. 36, Table 20. ER, Vol. II, Annex 2. Weller Report, pp. 13-18. ER, Vol. II,
Annex 3.
423Giles Report, p. 36, Table 20. ER, Vol. II, Annex 2; Weller Report, pp. 13-18. ER, Vol. II,
Annex 3.
424Giles Report, p. 39, Table 24. ER, Vol. II, Annex 2; Weller Report, pp. 13-18. ER, Vol. II,
Annex 3.
156g/ha. That is more than the amount that injures maize and 9 times the level
needed to harm tomato plants 425.
Conclusion
2.204 In sum, Colombia’s use of highly toxic herbicides and other dangerous
chemicals (the formulae for which it still keeps hidden), its pervasive violation of
its own operational parameters necessary to control spray drift (on literally tens of
thousands of occasions), and its spraying in close proximity to Ecuador, including
immediately adjacent to human settlements and environmentally sensitive areas,
have ensured that the spray drifts into, and causes damage to people, plants and
425
Giles Report, p. 36, Table 20. ER, Vol. II, Annex 2. Weller Report, pp. 13-18. ER, Vol. II,
Annex 3. Depending on other factors, the deposition could be even higher still. For example, the
paragraphs above assume, as Hewitt et al. did, compliance with the EMP’s wind speed
requirement of 5 knots (2.57 m/s). Hewitt et al., 2009, op. cit., pp. 923. CCM, Vol. III, Annex
131-B; Giles Report, op. cit., pp. 28-29. ER, Vol. II, Annex 2. Stronger wind translates into
greater drift. Giles Report, op. cit., pp. 28-29. ER, Vol. II, Annex 2. Hewitt et al., also assume
favourable atmospheric conditions and the absence of thermal inversions, both of which are not
always the case, particularly since Colombia frequently sprays at night when unfavourable
conditions, including thermal inversions, are common. Giles Report, op. cit., pp. 27-28, 42-44.
ER, Vol. II, Annex 2; Hansman & Mena Report, op. cit., p. 23, n.9. Another important factor that
influences spray drift is the canopy height of the vegetation below the spray planes, which can
intercept spray droplets as they travel through the air. Giles Report, op. cit., p. 23. ER, Vol. II,
Annex 2. The preceding paragraphs assumed, in line with Hewitt et al., a canopy of 25.91 metres.
Hewitt et al., 2009, op. cit., p. 923. CCM, Vol. III, Annex 131-B; see also Giles Report, op. cit.,
p. 23. ER, Vol. II, Annex 2. But large parts on the Colombian side of the border have been
cleared. See e.g., EM, Chaps. 5 and 6, paras. 5.93, 6.40; CCM, Chap. 1, para. 1.38; Solomon et
al., 2005, op. cit., p. 19, Figure 9. CCM, Vol. III, Annex 116. Consequently, there would be
significantly more drift. Giles Report, op. cit., p. 23. ER, Vol. II, Annex 2. Nor does the
modeling described above account for the fact that by the time the droplets reach Ecuador they
have an extremely high concentration of glyphosate and surfactant due to the evaporation of their
water content. Since “[t]he concentration of the spray droplet is a driving force for absorption into
the leafŽ, the “higher glyphosate concentration in the spray droplet increases plant susceptibility to
injuryŽ. Weller Report, op. cit., p. 22. ER, Vol. II, Annex 3. The likelihood of substantial injury
is exacerbated even further due to the high humidity in the region, which makes it easier for the
herbicide to penetrate plants. Weller Report, op. cit., p. 21. ER, Vol. II, Annex 3.
157livestock in, Ecuador. Colombia’s own drift model, when the actual data
generated by the spray planes is inputted, proves this conclusively. In the
following Chapter, Ecuador describes the specific harms that have been caused by
Colombia’s aerial spraying operations along and near the border between the two
States.
158 CHAPTER 3.
THE EVIDENCE OF HARM IN ECUADOR3.1 In the previous Chapter, Ecuador showed that Colombia blatantly
misrepresents: (i) the “harmlessnessŽ of its aerial spray mixture (the full contents
of which still remain undisclosed, but which is proven to harm human skin, eyes,
and respiratory and digestive systems, and to kill all plants it comes in contact
with); and (ii) its “strict complianceŽ with operational requirements that prevent
spray drifting into Ecuador (but which Colombia violated tens of thousands of
times in spray flights along and near enough to the border so that the spray was
bound to reach Ecuador). The flight data automatically recorded by the spray
planes, and furnished to the governments of Colombia and the United States,
along with other official Colombian and U.S. Government reports, unmask
Colombia’s false presentations about the contents of the spray mixture and the
execution of the spraying programme. The data show tens of thousands of spray
flights in close proximity to Ecuador’s border that violated all of Colombia’s
operational requirements for preventing spray drift, which made it inevitable that
the spray would drift into Ecuador in sufficient quantities to harm people and
livestock, kill crops and other lawful plants, and damage the pristine forests,
fields, rivers and streams that make up Ecuador’s unique and diverse natural
environment.
3.2 In its Memorial, Ecuador showed in Chapter 5 the actual harms caused by
spray drift from Colombia’s nearby aerial spraying operations to people, animals,
crops and the environment in Sucumbíos and Esmeraldas Provinces, within 10
161kilometres of the border with Colombia. The evidence is extensive, and comes
from many different sources, all of which corroborate one another. The flight
data that Ecuador subsequently obtained from the U.S. Department of State
provides more corroboration of the harms caused by Colombia’s aerial spraying
programme, because it establishes not only the likelihood but the inevitability of
harm in Ecuador by the reckless and irresponsible manner in which the spray
programme was carried out, which guaranteed that substantial quantities of toxic
spray … sufficient to kill plants deep inside Ecuador … would drift across the
border.
3.3 In this Chapter of the Reply, Ecuador responds to the Counter-Memorial’s
critique of the evidence of harm the sprayings have caused in Ecuador, presented
in the Memorial. As Ecuador will demonstrate, the rebuttal Colombia makes
does nothing to undermine the proof presented in the Memorial. Given the
carelessness with which the sprayings are conducted, spray drift across the border
and resulting harm to people, animals and plants in the border regions abutting
Colombia’s Nariño and Putumayo Provinces is inevitable.
3.4 As shown within, much of Colombia’s effort to refute Ecuador’s showing
of harm is circular. Throughout its analysis of Ecuador’s evidence, Colombia
contends that the sprayings could not have caused the damage Ecuador claims
162because the spray is incapable of causing that damage 42. This argument, of
course, neatly assumes its own conclusion. In truth, as set forth more fully in
Chapter 2, most, if not all, of the harms identified by Ecuador are fully consistent
with the known effects of the admitted components of the spray mixture, and
predictable based on the amount of spray that is deposited in Ecuador, even
without allowing for the toxic effects of the unknown ingredients.
3.5 To the extent the Counter-Memorial does more than assume what it
purports to be showing, Colombia’s main line of attack is to question the
credibility of the eyewitness statements Ecuador presented as one element of its
proof to substantiate the harms caused. According to Colombia, these 38
different witness testimonies (plus nine more from Colombian witnesses) are
entitled to “no weightŽ because they are allegedly uncorroborated by other
sources of evidence 427. Beyond misunderstanding the Court’s jurisprudence on
the subject of witness statements (as discussed in Section II below), Colombia’s
argument is factually incorrect. In the pages to follow, Ecuador will show that
the credibility and reliability of its witness statements are verified by multiple
additional sources of evidence, much of it contemporaneous to the events in
question, and some of it emanating from organs of the Colombian government
426
See, e.g., Counter-Memorial of Colombia, Vol. I, Chap. 7, paras. 7.5, 7.137 (29 Mar. 2010)
(hereinafter “CCMŽ) (“ƒ taking into account the scientific evidence on the limited effect of drift
and the strict technical parameters under which the spraying operations are carried out in
Colombia ƒ no damage could have occurred in Ecuadorian territoryŽ.).
427
CCM, Chap. 7, para. 7.127.
163itself. What emerges is a consistent and coherent pattern of harm that is
consistent in all material respects from year to year and place to place. It is
precisely this coherence and consistency that, in Ecuador’s view, constitutes the
most compelling evidence of the harms Colombia’s sprayings have inflicted on
Ecuador.
3.6 The flight data Ecuador has obtained from the U.S. Government constitute
a key new element in this already sizable bundle of mutually reinforcing and
corroborative evidence. The data provide compelling proof of what Ecuador has
long contended: Colombia has conducted tens of thousands of spray operations
along the border with Ecuador since 2000 without regard for the operational
requirements necessary to prevent spray drift into Ecuador; indeed, in most of
these cases, the safety limits were exceeded by extremely large margins. As will
be discussed and demonstrated graphically in the sections to follow, the flight
data obtained from the U.S. Department of State underscore the connection
between the aerial sprayings and the harms demonstrated in Ecuador’s Memorial.
3.7 In short, when there were sprayings near the border, there was harm. The
consistency between the flight data and Ecuador’s evidence of harm strongly
underscores the reliability of Ecuador’s evidence, precisely because the
information contained in those data is entirely new to Ecuador. Never before had
Ecuador had such detailed information about the dates and locations of spray
164events. As detailed in the Memorial, Colombia never gave Ecuador advance
428
notice of the dates and locations of spray events , and the indigenous people and
campesinos who provided witness statements certainly were not given notice by
Colombia. Even so, the facts show tight congruity between the flight data and
Ecuador’s other evidence. As discussed in the paragraphs to follow, the data
Ecuador received from the United States substantially coincide in temporal and
geographic terms with the evidence of harm presented in the Memorial. In other
words, thanks to this reliable new evidence from a third State, there is now
definitive proof that the harms in Ecuador described in the Memorial occurred in
the wake of nearby aerial sprayings by Colombia. This is compelling new
corroboration of the eyewitness accounts Ecuador has provided.
3.8 This Chapter is presented in the following manner: Section I examines
the mutually corroborating evidence of harm in Ecuador, focusing on the
correspondence between the information the flight data reveal about the dates and
locations of sprayings and the evidence Ecuador has previously introduced … in
the form of witness statements, contemporaneous observation mission reports,
newspaper articles, scientific studies of the chemicals used and their labels, as
well as Colombia’s own reports and flight path data. Where pertinent, Ecuador
supplements the evidentiary record with still other information demonstrating the
428See Memorial of Ecuador, Vol. I, Chap. 3, paras. 3.2-3.3, 3.17, 3.21, 3.25, 3.46 (28 Apr. 2009)
(hereinafter “EMŽ).
165existence of the harms alleged at the times and places spraying occurred. Section
II reviews the evidentiary standards concerning witness statements,
demonstrating Colombia’s misunderstanding of the evidentiary value accorded to
these testimonies. Section III further corroborates the evidence of harm in
Ecuador by looking at the evidence of the same harms found in Colombia.
Notably, much of this evidence comes from the Colombian government itself. As
shown, the same types of harms in Ecuador that Colombia tries so assiduously to
separate from the border sprayings have also systematically been inflicted on the
Colombian areas sprayed by the same planes with the same chemical cocktail.
Put simply, the same causes produce the same effects, in both countries.
Section I. The Mutually Corroborating Evidence of Harm in Ecuador
A. S UCUMBÍOS 2001
3.9 The Court will recall that reports of harm first emanated from Ecuador’s
remote border regions of Sucumbíos Province at the very end of 2000 and early
2001 42. The flight path data confirm this aspect of the timeline. Figure 3.1
depicts the location of spray flights within 10 kilometres of the Ecuador-
Colombia border, along Sucumbíos Province, during December 2000 and
February 2001. Flights in December 2000 are indicated in brown; flights in
January 2001 are indicated in purple; flights in February 2001 are indicated in
42See EM, Chap. 6, paras. 6.4 et seq.
166orange. According to these data, Colombia conducted at least 3,276 spray flights
within 10 kilometres of Ecuador in December 2000 and 8,228 more in January
430
2001 .
3.10 The proximity of the sprayings to the Sucumbíos communities discussed
in Ecuador’s Memorial, including Salinas, Puerto Escondido, Puerto Mestanza,
Corazón Orense and San Francisco 1 and 2 (the locations of which are indicated
on the map) is evident, and needs no further comment here. Given (a) this
proximity, and (b) the substantial drift predicted by Colombia’s own drift model
(corrected only for the true flight parameters, as discussed in the previous
431
Chapter ), it is not surprising that many Ecuadorian residents of the border
region report seeing Colombian aircraft conducting spray operations at this time,
followed in their wakes by the deposit of what witness after witness described as
a white, foul-smelling mist falling on Ecuadorian territory.
3.11 Witness 5, a resident of Salinas, a community less than 1 kilometre from
the spraying, offers a typical description. He states:
“I remember that in 2001 I was working on my farm, which is
located near the San Miguel River, and I saw three white planes
protected by helicopters flying over Salinas. The planes left
behind a white cloud of smoke that had a sour chemical-like odor.
430See R. John Hansman, Ph.D. & Carlos F. Mena, Ph.D., Analysis of Aerial Eradication Spray
Events in the Vicinity of the Border Between Colombia and Ecuador from 2000 to 2008 (Jan.
2011), Appendix 3: Statistics, p. 28. ER, Vol. II, Annex 1.
431See supra Chap. 2, paras. 2.189-2.203.
167 0°30'0"N
km
Figure 3.1
0-1 k1m-2m-3 k3m-m-7 k7m-10 km River
spray line (km) Putumayo
Distance from nearest
Palma Seca
!
River
San Miguel
Chone 2Chone 1
76°30'0"W ! ! 76°30'0"W
!
Oriental 0 5 10 15 20 25
! Playera
±
Puerto
Cora5 de Agosto
Puerto!MestanUnion Lojana
Puerto Escondido
! !
P u t u m a y o !
!!
Monterrey S u c u m b í o s
! E C U A D O R
C O L O M B I A
Dios Peña
! SaMarianitaarapa
! !
San Miguel
Generaln
!
(December 2000 - February 2001) ! ! La Condor
!
December, 20002001 2001
! 10 de Agosto
Spray events (month)
!
San SFarannFcrisacnoci2sco 1
77°0'0"W Salinas 77°0'0"W
Virgen del Cisne
Spray Events Within 10 Kilometres of Ecuador’s Sucumbíos Province
Peru
River
San Miguel Colombia
Protected areasm international boundary
Ecuador
Legend
Cofán Reserve
0°30'0"N This odor could be felt in both the air and in the water we
432
consumedŽ .
He then recounts what happened afterwards:
“About four days after the fumigations, my body ached all over
and my skin itched. I had bumps on my skin that lasted for about a
month and a half . . . . A week after, the coffee began to lose its
flowers, the leaves started to turn yellow and then they turned
black, drying up completely. The cacao also turned yellow and,
when it was cut, one could see inside that the cacao seeds had
rotted. The grass turned a yellowish color that began at the top
433
and moved down to the rootsŽ .
3.12 Similar testimony comes from Witness 4, also a long-time resident of
Salinas, who states that:
“In the year two thousand and one, I remember having seen two
planes followed by helicopters which passed by slowly, several
times a day, above our community dropping something like a mist.
I was working on the farm and I could see them crossing the San
Miguel River and going from one side of the border to the other.
Since I did not know what this was all about and I thought the mist
was not bad, I continued working on the farm while the planes
434
sprayed over me and my children who were playing outsideŽ .
Like Witness 5, she testifies:
“A few days after the spraying, the plants started to turn yellow
and then they turned black and died. I had never experienced
anything like that. I tried to save the crop with fertilizers but it did
not work and we lost everything. Two weeks after the first
spraying, my family and I got bumps all over the body, we had an
432Declaration of Witness 5, 16 Jan. 2009 (hereinafter “Witness 5 DeclarationŽ). EM, Vol. IV,
Annex 193.
433Witness 5 Declaration, op. cit. EM, Vol. IV, Annex 193.
434
Declaration of Witness 4, 22 Dec. 2008 (hereinafter “Witness 4 DeclarationŽ). EM, Vol. IV,
Annex 192.
168 itch that was unbearable. I made home remedies for the children,
like I have done so many times before when they were bitten by
insects435ut this time the medicine I had prepared did not cure
themŽ .
3.13 Spurred by the complaints of area residents, in March 2001, several local
non-governmental organizations (“NGOsŽ) visited the border regions to
investigate and record their observations. The first of the numerous reports
issued by organizations from across civil society on the subject of the aerial
436
sprayings was issued in June 2001 . Echoing the statements of Witnesses 4 and
5 just cited, the report found significant impacts on the health of the local
populations, their crops and their animals. Health symptoms recorded included
eye problems, respiratory distress, dermatological conditions and gastrointestinal
437
difficulties .
3.14 With respect to the damage to crops, the 2001 report stated: “The
campesinos from this entire zone reported significant damages to the crops, to the
extent as to believe they will starve soon. During our trip, we were able to
appreciate all the effects mentioned [by them]Ž 438. Paralleling Witness 5’s
description of what happened to his coffee and cacao plants, the report observes:
435Witness 4 Declaration, op. cit. EM, Vol. IV, Annex 192.
436
Accion Ecologica, Report on the Investigation of the Fumigations’ Impacts on the Ecuadorian
Border (June 2001). EM, Vol. IV, Annex 161.
437
Ibid., pp. 5-6.
438
Ibid., p. 10.
169 “All the campesinos said the coffee has been affected. The plants
turn yellow, dry up and no longer produce seeds. The bean is
empty. The harvest for the year has been lost. We were able to
confirm that the leaves were as if they had been burned and the
fruit is empty.
. . . As with the coffee, the leaves of the cocoa plant were yellow
to the point of drying up and the fruits on the plants appeared dry.Ž
3.15 The report’s conclusions include the following:
“There is a direct temporal relation between the fumigations and
the appearance of the sicknesses.
There is an inverse proportional relationship between the distance
from the site that was sprayed and the symptomatology. As the
distance from the sprayed site increases, the symptomatology in
the population decreases.
. . .
The population that has suffered the impacts of the spraying is
experiencing feelings of anxiety. They have no financial support,
no compensation or appropriate care for their health, which has
been undermined by a spraying program that renders them
invisible.
The negative impacts on the population’s health and its nutritional
state may increase if no adequate measures are taken to offset the
failure of their crops and the death of their livestock and
animalsŽ 439.
3.16 This initial report was followed just a month later, in July 2001, by
another study issued by a delegation composed of representatives from several
NGOs, including the Confederation of Indigenous Nationalities of Ecuador
439
Ibid., p. 11.
170(“CONAIEŽ) 440. As part of the delegation’s mission, it met with affected persons
and conducted site-visits to investigate the allegations of harm first-hand.
Notably, the report contains a section summarizing the testimonies of the local
populations that in all material respects parallels the witness statements Ecuador
submitted with the Memorial. Rather than recapitulate those contemporaneous
testimonies here, Ecuador respectfully refers the Court to the relevant sections of
441
the report, located in Annex 162 of the Memorial .
3.17 The NGO delegation conducted its own investigation of the situation on
the border and, with respect to the health situation, stated:
“Analyzing the reports of the ten leading causes of illness in the
General Farfán Health Subcentre, whose jurisdiction includes San
Francisco 2, it was discovered that the three leading causes of
illnesses in the population continued to be respiratory infections,
which increased by 42% (from 206 to 293 cases) from January to
June 2001, in comparison to the same period of the previous year;
skin infections, which increased by 48% compared to 2000 (from
147 to 218), and malaria, which increased by 33% (from 111 to
148).
The doctor at the Subcentre indicated that the impact of the
fumigations was significantly more noticeable in January, when,
suddenly, people began inhaling the chemical. There were
440Confederation of Indigenous Nationalities of Ecuador (CONAIE) et al., Technical Report of
the International Commission on the Impacts in Ecuadorian Territory of Aerial Fumigations in
Colombia (19-22 July 2001). EM, Vol. IV, Annex 162.
441Ibid., pp. 10-13. See e.g., ibid., p. 10 (“There are numerous affected Kichwa people along the
San Miguel River and the crops are affected by the sprayings. The planes crossed the border and
we have respiratory, eye and skin problems . . . Ž);ibid., p. 11 (“School children, in the
communities close to the border, are having problems. Before they used to play and now they
have headaches and diarrheaŽ.). EM, Vol. IV, Annex 162.
171 numerous cases of rhinitis and eye irritation, which were not
442
recordedŽ .
3.18 The July 2001 report also noted the psychological trauma caused by the
aerial sprayings and their aftermath. The report notes: “The sprayings are causing
situations of insecurity and fear in border populationsŽ 44. As explained by
Salinas resident Witness 5, in addition to the physical symptoms “the sprayings
have also caused psychological problems in our village. It has caused fear,
444
concern, uncertainty and a lot of anxietyŽ . She, like others, was particularly
concerned about the impact on her young child who, years later, “is still scared
that the planes might come backŽ 44. This anxiety has had unfortunate
consequences. Many residents fearing for their health and survival have fled
away from the border. As described by Witness 18, eight of her children have
left and “do not want to return because they are afraid; they hear an airplane and
446
they think that they are going to spray againŽ . Many of those who remain live
with this fear. As echoed by another Salinas resident, “[m]y community lives in
442
Ibid., p. 14.
443
Ibid., p. 22.
444Witness 5 Declaration, op. cit. EM, Vol. IV, Annex 193.
445Ibid.
446Declaration of Witness 18, 15 Jan. 2009 (hereinafter “Witness 18 DeclarationŽ). EM, Vol. IV,
Annex 204.
172constant anguish. We do not know when this nightmare will end. We are afraid
447
that they will spray again and we will not be able to feed our childrenŽ .
3.19 Concerning the effects of the spraying on crops, the NGO report states:
“COFFEE: the crops exhibit an alteration of the green color of
their leaves, with a yellowing of the central vein; followed by total
chlorosis (yellowing) and the presence of brown spots both at the
tip of the leaves and their edge; and the withering of the entire
plant. . . .
YUCCA: yellowing was observed in the leaves and in the root or
edible part. When cut cross-sectionally, one can see a dark
brownish-grey halo near the bark, which appears to be healthy. . . .
PLANTAIN: withering was observed in the bottom leaves of the
mother plant and in the stems of the shoots. The campesinos said
that the growth of the plant has ceased. When cut cross-
sectionally, necrosis was observed in the xylem or conducting
tissue, which prevents the transport of sap. . . .
RICE: there is a yellowing that has markedly reduced the harvest.
A three month-old plot was inspected. At the mere sight of it, one
could detect a discoloration of the entire plant and the onset of
diseases. . . .
PASTURE GRASS: it was observed that there is discoloration or
yellowing that starts at the tips and edges of the leaves, and
subsequently the entire plant dries and diesŽ 448.
3.20 It is notable, particularly in light of Colombia’s criticisms discussed
below, that both of these reports were prepared contemporaneous to the events in
447Declaration of Witness 3, 17 Jan. 2009 (hereinafter “Witness 3 DeclarationŽ). EM, Vol. IV,
Annex 191.
448Confederation of Indigenous Nationalities of Ecuador (CONAIE) et al., Technical Report of
the International Commission on the Impacts in Ecuadorian Territory of Aerial Fumigations in
Colombia, p. 16 (19-22 July 2001). EM, Vol. IV, Annex 162.
173question by organizations that have no other motive than to draw attention to the
problem they found. It is equally telling that both reported on phenomena that are
entirely consistent not only with each other but with the witness statements as
well.
3.21 The findings of these two NGO missions were echoed in
contemporaneous media accounts. In July 2001, for example, a major national
daily, El Universo, published an article detailing the effects of the aerial sprayings
earlier in the year. With respect to the sprayings in northern Sucumbíos, the
newspaper recounted the spray’s impacts on the border community of La
Charapa 44. The President of the Association of Afro-Ecuadorians in Sucumbíos,
a local farmer, is quoted describing how, four or five months earlier, “a strange
rain was swept onto his crops by the wind, and an unbearable smell hung in the
450
airŽ . The article recounts how the residents lived in fear after seeing the
spray’s effects451. Only 20 of the 130 chickens in María Reyna’s farm survived,
the pigs slowly died, the various crops withered and stopped producing 45. In the
neighbouring town of 10 de Agosto, the reporter found the scars left behind on
the residents’ skin from the skin rashes that had followed the border spraying, in
449“The Drama of FumigationsŽ, L UNIVERSO (Guayaquil, 10 July 2001). ER, Vol. IV, Annex
61.
450Ibid.
451
Ibid.
452
Ibid.
174addition to the stomach aches, respiratory problems, dizziness and headaches that
came with it 453.
3.22 In the face of this evidence, Colombia takes the position that none of it is
credible 45. Perhaps to distract the Court from the other sources of evidence, the
Counter-Memorial focuses special attention on the eyewitness statements
presented in the Memorial and attacks them on several fronts. Indeed, it is no
exaggeration to state that the primary thrust of Colombia’s effort to rebut
Ecuador’s showing of harm is to argue that the witness testimonies do not
withstand scrutiny. Yet, as Ecuador will show in the pages to follow, Colombia
is wrong. Taken collectively and together with the other elements of proof … now
supplemented by the spray flight data obtained from the United States … there is
every reason to credit these sworn, eyewitness statements from local residents
who personally observed and experienced the aerial sprayings and the effects on
themselves and their communities.
3.23 The essence of Colombia’s attack on Ecuador’s witness statements is
encapsulated in a single line from the Counter-Memorial: “The allegations of the
campesinos remain wholly unsubstantiatedŽ 45. In Ecuador’s view, this argument
453Ibid.
454
See, e.g., CCM, Chap. 7, para. 7.6.
455
CCM, Chap. 7, para. 7.151.
175is both factually wrong and disturbingly condescending. Surely, the first-hand
accounts of Ecuadorian campesinos are entitled to the same weight as those of
other human beings. The mere fact that they are impoverished citizens of the
developing world does not render them any less able than anyone else to testify
about facts within their personal experience. Indeed, the opposite may be true in
this case. As subsistence farmers, their very survival depends both on their own
health and on the health of their crops. They are thus particularly attuned to
nature’s rhythms, agricultural cycles and even slight disturbances to the delicate
balances of their lives456.
3.24 Equally, and as a matter of fact, Colombia is wrong that Ecuador’s case
rests exclusively on these eyewitness statements. Ecuador has already cited to
NGO and press reports contemporaneous to the earliest aerial sprayings that
confirm the existence of the harms alleged. In addition, Ecuador relies on
multiple other sources to support its claims, including: (a) the reports of several
UN Special Rapporteurs; (b) additional NGO field assessments conducted in later
years; (c) other media accounts; (d) contemporaneous medical inquests; (e)
Colombian eyewitnesses and, not least; (f) reports from organs of the Colombian
456
See Norman E. Whitten, Jr., Ph.D., Dr. William T. Vickers, Ph.D. & Michael Cepek, Ph.D.,
Tropical Forest Cultural Ecology and Social Adaptation in the Ecuadorian Border Region with
Colombia, pp. 3, 13, 31 (Jan. 2011) (hereinafter, “Whitten et al. ReportŽ) (describing how “the
people of the border region are inextricably linked to their natural environment and are thus
extremely vulnerable to environmental perturbations that upset this balanceŽ). ER, Vol. II, Annex
5.
176government itself. Each of these independent sources of evidence is discussed
further in this Chapter. For the moment, the essential point is that Colombia
misrepresents the truth when it suggests that the witness statements stand alone.
In reality, they are just one part of a consistent and mutually reinforcing whole
that together forms a body of consistent and coherent evidence all demonstrating
harm in Ecuador. The spray flight data is only the latest addition to this body of
evidence, although it is a very important one.
3.25 In addition, as was detailed in Ecuador’s Memorial and as is further
explained in Chapter 2 of this Reply, the descriptions of harm that are provided in
the witness testimonies and elsewhere are consistent with the properties of the
known chemicals in the spray mixture. The reports of skin and eye irritation are
not surprising given the fact that surfactants included in glyphosate-based
products, including POEA, are known to cause such effects. In fact, the label of
Roundup SL … a product Colombia admits to using … warns users to “[a]void
457
contact with eyes and skinŽ because the product causes irritation . As discussed
in Chapter 2, another product Colombia has used, Roundup Export, is capable of
causing “irreversible eye damageŽ 458, and in fact its use was discontinued for
457
Colombia Roundup SL Label, p. 1. EM, Vol. III, Annex 115. See also EM, Chap. 5, paras.
5.44…5.45.
458United States Environmental Protection Agency, Office of Prevention, Pesticides and Toxic
Substances, Report on Issues Related to the Aerial Eradication of Illicit Coca in Colombia,
Response from EPA Assistant Administrator Johnson to Secretary of State, p. 8 (19 Aug. 2002)
(hereinafter “EPA 2002 AnalysisŽ). EM, Vol. III, Annex 143; see supra Chap. 2, paras. 2.19-2.23.
177precisely that reason. The health symptoms reported … not just in Ecuador’s
witness statements, but by the UN Special Rapporteurs, NGOs, contemporaneous
medical inquests, the media, and other observers … are also consistent with the
findings of Colombia’s own studies. For example, Colombia’s own assessment
acknowledges that ingestion of glyphosate can cause “erosion of the digestive
tract, which manifests as difficulty in swallowing, sore throat, and gastrointestinal
hemorrhagingŽ 459. In attacking Ecuador’s witness statements, the Counter-
Memorial ignores the remarkable consistency of those statements with its own
conclusions about the risks posed by the spray mixture.
3.26 The harm to plants, animals and the environment is also in line with the
scientific evidence. As explained in the Report of Dr. Stephen C. Weller, the
hallmarks of glyphosate exposure in plants are stunted growth and a yellowing or
blackening of the plant tissue, all symptoms that the affiants describe with great
specificity6. Those very symptoms were universally described in the witness
statements, contemporaneous field reports and press accounts. The product label
for Roundup Export warns “DRIFT MAY CAUSE DAMAGE TO ANY
VEGETATION CONTACTED TO WHICH TREATMENT IS NOT
459Republic of Colombia, Environmental Risk of the Herbicide Glyphosate, Sec. 1.7.1 (date
unknown). EM, Vol. II, Annex 101.
460
See Stephen C. Weller, Ph.D., Glyphosate-Based Herbicides and Potential for Damage to
Non-Target Plants Under Conditions of Application in Colombia, pp. 1, 5-6, 22-23 (Jan. 2011)
(hereinafter “Weller ReportŽ). ER, Vol. II, Annex 3.
178 461
INTENDEDŽ . Thus, the extensive destruction of food crops and other plants
in Ecuador as a result of their exposure to Colombia’s spray mixture is entirely
predictable. Moreover, the witness statements and other accounts invariably
mention similar damage to multiple plant species; the fact that the dead and dying
plants were not specific to a particular species or crop, as would be expected with
a disease or insect infestation, is further evidence that the damage was caused by
an indiscriminate herbicide 462.
3.27 Tellingly, the Counter-Memorial says next to nothing about the early
NGO reports cited above at paragraphs 3.13 to 3.19, which were described in the
463
Memorial . Perhaps Colombia knows they negate its assertion that Ecuador’s
witness statements are uncorroborated by contemporaneous evidence. Perhaps
Colombia has nothing to say consistent with its premise that the witness
statements are latter-day inventions of a few dozen untrustworthy campesinos.
Whatever the reason, rather than respond to these reports or any of the others like
them, Colombia chooses a different tactic. The Counter-Memorial suggests that
in December 2004, Ecuador itself acknowledged that no harm of any kind had
been caused anywhere in the country at any time up to that date and, as a result,
461 United States Roundup Export Label, United States Pesticide Product Label System,
Registration No. 524-308 (9 July 1997). EM, Vol. II, Annex 125.
462
See Weller Report, op. cit., p. 3. ER, Vol. II, Annex 3.
463
See, e.g., EM, Chap. 6, paras. 6.12-6.13, 6.38, 6.43, 6.50, 6.57, 6.83-6.86.
179 464
all of its evidence from before that date can be summarily discarded . And,
what is the basis for this remarkable claim? Statements issued by the Ecuadorian
Foreign Ministry in 2004 to the effect that it had conducted site visits to the
border area and found no traces of glyphosate residue in any of the water or soil
465
samples taken .
3.28 Colombia has gorged itself on Ecuador’s statements that glyphosate was
not detected in these samples. The Counter-Memorial speaks of this for no fewer
than 18 pages. But Colombia’s gluttonous reliance on these statements is both
misplaced and disingenuous.
3.29 It should come as no surprise to Colombia that glyphosate was not
detected in the river waters or the soil that were sampled. Colombia’s Counter-
Memorial and experts recognize that glyphosate “is rapidly removed from water
by absorption to sediments and suspended particulate matterŽ 46. Moreover, in
addition to the chemical’s dissipation in the water, the river’s currents will
quickly carry any materials away from the spray site. Colombia is well aware
this happens. The government agency in Colombia’s Nariño Department (which
abuts Ecuador) that is responsible for environment and natural resources,
464
See, e.g., CCM, Chap. 7, paras. 7.2, 7.112, 7.113.
465
Ibid.
466K.R. Solomon et al. “Environmental and Human Health Assessment of the Aerial Spray
Program for Coca and Poppy Control in ColombiaŽ, OAS, Washington, D.C., 31 March 2005, p.
20 (hereinafter “Solomon, 2005Ž). CCM, Vol. III, Annex 116.
180informed the Colombian Ministry of Environment that testing for glyphosate in a
local river would be fruitless: “Given the strong current of the Chaguí River and
the fact that many months have transpired since the fumigations began, it is not
possible to take samples of water to determine the degree of affectation of the
467 468
riverŽ . Glyphosate also disappears quickly from the soil . After that, it is
undetectable. That glyphosate was not found in the running river waters or soil
samples weeks or months after the aerial sprayings gets Colombia nowhere: it
does not prove glyphosate was not deposited there several weeks earlier by the
spray planes.
3.30 In fact, Ecuador’s searches for glyphosate residues were hopeless
exercises: they were conducted at the wrong times and places. Figure 3.2, shows
why they were in vain. As the Court can see, during the two-month period from
November to December 2004, Colombia conducted only very limited aerial
spraying operations within 10 kilometres of Ecuador; the contrast with other two-
month periods, such as the August to September 2002 period depicted in Figure
3.2, could not be more stark. Thus, to say that testing in December 2004 found
467 Letter from Francisco Santander Delgado, Director General, Corponariño, Republic of
Colombia, to Maria Cecilia Rodriguez, Minister of the Environment, Republic of Colombia (26
Sept. 2002). ER, Vol. V, Annex 144.
468
See Diplomatic Note DM/AL No. 25009, sent from the Ministry of Foreign Affairs of
Colombia to the Ministry of Foreign Affairs of Ecuador, p. 3 (14 July 2001). EM, Vol. II, Annex
43. Weller Report, p. 3. ER, Vol. II, Annex 3; Charles A. Menzie, Ph.D. & Pieter N. Booth, M.S.,
Response to: “Critique of Evaluation of Chemicals Used in Colombia’s Aerial Spraying Program,
and Hazards Presented to People, Plants, Animals and the Environment in Ecuador,” As
Presented in: Counter-Memorial of the Republic of Colombia, Appendix, pp. 25-26 (Jan. 2011)
(hereinafter “Menzie & Booth ReportŽ). ER, Vol. II, Annex 6.
181no glyphosate residue says nothing. If any glyphosate had been present in the
area after the sprayings, it would have been gone, or at least undetectable, by the
time the tests for it were conducted.
3.31 The same applies to Ecuador’s efforts to test for glyphosate earlier in
2004. We know from the flight data that the spraying often occurred more than a
month earlier and more than 10 kilometres away. In the July 2004 testing, for
example, the samples were taken by Ecuador, in Esmeraldas, more than two
months after the most recent sprayings within 10 kilometres, which had been
completed by April 2004 469. There was no spraying along the border in July
2004 47. Similarly, the closest spray lines to the Sucumbíos sites tested in May
471
2004 are all more than 10 kilometres away .
3.32 Thus, the only result that these 2004 tests and statements reach is that, at
that place and time, Colombia’s prior sprayings were sufficiently far removed
geographically and temporally to have left any discernable traces in the local
469See infra Figure 3.6. See also Press Bulletin No. 480 of the Ecuadorian Foreign Ministry, “No
glyphosate residues were found in Esmeraldas, border with ColombiaŽ (26 Aug. 2004). CCM,
Vol. II, Annex 81.
470See infra Figure 3.6.
471See Flight Path Data Received From the U.S. Department of State and Other Technical
Information. See also Press Bulletin No. 388 of the Ecuadorian Foreign Ministry, “No glyphosate
residues exist in the waters of the rivers of the Sucumbíos ProvinceŽ (25 June 2004). CCM, Vol.
II, Annex 80. Even if the sampling at the San Miguel River was at the closest point to the spray
lines, which could make it less than 10 km away, as discussed above, testing in a large, running
river will not detect glyphosate residue several days, weeks, or even months after the most recent
spraying.
182 Spray Events Within 10 Kilometres of Ecuador’s Sucumbíos Province
(August - September 2002)
C O L O M B I A
!
Salinas Puerto Mestanza
! Dios Peña ! ! !
San Francisco 2 ! Puerto Escondido
General ! Santa! ! ! ! !
San Francisco 1 Farfán Marianita Corazón Orense Chone 2
! ! Playera
Virgen del Cisne La Condor 5 de Agosto Oriental
La Charapa Puerto
! ! ! Union Lojana Nuevo ! Chone 1
10 de Agosto Monterrey
San Miguel
Spray events
E C U A D O R
Spray Events Within 10 Kilometres of Ecuador’s Sucumbíos Province
(November - December 2004)
C O L O M B I A
! !
Salinas Puerto Mestanza
!Dios Peña ! Puerto Escondido ! !
San Francisco 2 ! ! ! ! !
San Francisco 1 General Santa ! ! Corazón Orense Chone 2
! Farfán Marianita
Virgen del Cisne ! Playera
La Condor La Charapa 5 de Agosto Oriental
! ! Union Lojana Puerto !
10 de Agosto Nuevo Chone 1
Monterrey
San Miguel
E C U A D O R Spray events
Figure 3.2water or soil that much time after the sprayings were conducted. The statements
are not the silver-bullet Colombia tries to make of them in order to quickly
disregard four years of prior evidence of exposure and harm in Ecuador.
Contrary to Colombia’s misconstrued argument that Ecuadorian authorities
“expressly and publicly stated that, up until December 2004Ž 472 no damage had
occurred in Ecuador, the only conclusions reached were that no damage had been
473
observed at the specific locations visited on the dates of the visits .
3.33 Once Colombia’s attempt to make more out of Ecuador’s inconclusive
2004 water and soil tests than they merit is exposed, Ecuador’s case on harm for
the years between 2001 and 2004 stands effectively unrebutted, except for
Colombia’s dismissive statements about the inherent unreliability of Ecuadorian
campesinos.
3.34 The Counter-Memorial further attacks the witness statements on the
alleged ground that they “are studiously vague as to the timing of the sprayings
and locations where sprayings allegedly took place, making it impossible to check
472
See CCM, Chap. 7, para. 7.2.
473
See, e.g., Memorandum of the Ecuadorian Foreign Ministry, 24 Dec. 2004. CCM, Vol. II,
Annex 83 (stating that it was unaware of “sprayings on the Ecuador-Colombia border in the past
weeksŽ) (emphasis added).
183these assertions against the documented record of spray missionsŽ 474. Colombia
is wrong here, too.
3.35 In the first instance, it is ironic for Colombia to complain about the
difficulty checking the witnesses’ statements against “the documented record of
spray missionsŽ. While Ecuador has no reason to doubt that Colombia has such
records, they have been shared with neither Ecuador nor the Court. In its
475
Counter-Memorial, Colombia admits to having such records , yet does not
bother to produce them as part of its evidentiary showing. Given the discussion
in Chapter 2 about what is revealed by the data Ecuador was able to obtain from
the United States, Colombia’s reticence is understandable.
3.36 In any event, what matters now is the extent to which Ecuador’s witness
statements coincide with the later-acquired spray data. As discussed further
below, the correspondence between the two is impressively close. In Ecuador’s
view, this congruency lends even greater credibility to the witness statements.
Put another way, the fact that the statements match the later-acquired spray data
constitutes an indicium of the statements’ reliability.
474CCM, Chap. 7, para. 7.126.
475See, e.g., CCM, Chaps. 4 & 7, paras. 4.64, 7.17, 7.172.
1843.37 Colombia is also incorrect when it suggests that Ecuador’s witness
statements are anything less than crystal clear on the subject of “the locations
where sprayings allegedly took placeŽ 476. To a person, Ecuador’s witnesses
specify exactly where they were when the sprayings took place. Witness 18, a
resident of San Francisco 1, is typical in this respect. He states:
“I remember the first time the sprayings occurred; I was clearing
ground with my friend in preparation for planting watermelons,
about a kilometre from the San Miguel River. At ten in the
morning, they started spraying. At first I could hear the noise of
the planes and then I began to smell a nasty odor in the airŽ 477.
3.38 As reflected on Figures 2.13 and 2.14, all the places the witnesses
identified are located in very close proximity to the areas where the recently-
acquired flight data confirm that sprayings were being conducted, and certainly
well within the reach of the spray drift as predicted by Colombia’s own model
(corrected only for the actual flight parameters).
3.39 Colombia’s assertion that the witness statements are “studiously vagueŽ as
to when the sprayings took place is similarly off the mark. In fact, a significant
number of them refer specifically to the first sprayings as taking place in “2001Ž.
476CCM, Chap. 7, para. 7.126.
477Witness 18 Declaration, op. cit. EM, Vol. IV, Annex 204.
185Witnesses 4 and 5 quoted already above are just two examples. Others who
478
specifically referred to sprayings beginning in 2001 are cited in a footnote .
3.40 The fact that these witnesses do not refer to the specific month in 2001 is
no reason to discredit their testimonies. It is an attribute of memory that exact
dates become harder to recall as more time passes. And what is true generally is
particularly true for the residents of the border area. As stated in Ecuador’s
Memorial and admitted in the Counter-Memorial, most residents of the border
area are minimally educated subsistence farmers engaged in a daily struggle to
479
put food on the table in a very remote region . Others are indigenous peoples
480
who still live largely in accordance with their ancient ways . They cannot be
viewed through the same lens as professionals in the developed world who clutch
their day-planners wherever they go and then file them away for posterity.
3.41 Ecuador acknowledges that a number of the Sucumbíos witness
statements (sworn in January 2009) refer to the first spray events as taking place
478
See, e.g., Declaration of Witness 2, 16 Jan. 2009 (hereinafter “Witness 2 DeclarationŽ). EM,
Vol. IV, Annex 190; Witness 3 Declaration, op. cit. EM, Vol. IV, Annex 191; Witness 4
Declaration, op. cit. EM, Vol. IV, Annex 192; Witness 5 Declaration, op. cit. EM, Vol. IV,
Annex 193; Declaration of Witness 9, 16 Jan. 2009 (hereinafter “Witness 9 DeclarationŽ). EM,
Vol. IV, Annex 197; Declaration of Witness 11, 16 Jan. 2009 (hereinafter “Witness 11
DeclarationŽ). EM, Vol. IV, Annex 199.
479
See EM, Chap. 2, para. 2.22; CCM, Chap. 2, para. 2.13.
480See EM, Chap. 2, paras. 2.20-2.21.
186 481
“seven or eight years agoŽ . This too should engender no doubt. First, as just
stated, it is unrealistic to expect perfect recall as to dates from any person years
after the fact, much less the people who live along the Ecuador/Colombia border.
Second, and more to the point, saying the sprayings began “seven or eight years
agoŽ is factually correct. Since the statements were sworn in the first month of
2009, seven or eight years before is 2001 or 2002. Third, any apparent lack of
precision can be understood given the frequency with which sprayings took place.
As first described in Ecuador’s Memorial and discussed again below, Colombia’s
spray planes revisited the Sucumbíos border region repeatedly in the years
following 2001, including in the years 2002-2007 inclusive 48. For example, the
area within 10 kilometres of San Francisco 2 village was sprayed during at least 5
different months between 2001 and 2002, with repeated sprayings in the
483
subsequent years . Under the circumstances, with so many intervening spray
events, it is not surprising that a witness did not give the precise month when the
first such event occurred.
481See, e.g., Declaration of Witness 2, 16 Jan. 2009. EM, Vol. IV, Annex 190; Witness 3
Declaration, op. cit. EM, Vol. IV, Annex 191; Witness 9 Declaration, op. cit. EM, Vol. IV,
Annex 197; Witness 11 Declaration, op. cit. EM, Vol. IV, Annex 199.
482Hansman & Mena Report, op. cit., p. 11 et seq. ER, Vol. II, Annex 1. See also Figures 2.6 and
2.13.
483See Figures 3.1 and 3.3. See also Figure 2.56 and 2.13.
1873.42 Colombia also attacks Ecuador’s witness statements on the ostensible
484
ground that they “lack ... any medical evidence or other substantiationŽ . Here
again, Colombia is viewing the matter through an inappropriate lens. The
Ecuador-Colombia border is not The Hague. Hospitals, doctors’ offices and
pharmacies are not well-distributed throughout the region. Indeed, they are
485
extremely rare . Most of the time, and frequently even in the most severe cases,
residents have nowhere to turn for medical assistance other than their families.
As discussed in Ecuador’s Memorial, infrastructure throughout the region is
limited in the extreme. Roads are little more than dirt paths and public
486
transportation, where it exists at all, is infrequent and unreliable . Combined
with the scarcity of medical facilities, it is extraordinarily difficult for ill people to
seek the assistance of medical professionals.
3.43 This aspect of the Counter-Memorial’s argument rings particularly hollow
given Colombia’s repeated recognition of the remoteness, poverty and
underdevelopment of the region. At paragraph 2.13, for instance, the Counter-
Memorial states:
484CCM, Chap. 7, para. 7.126.
485
EM, Chap. 2, para. 2.26. See also Whitten et al. Report, op. cit., p. 26 (“many Cofán people
live far from health centers, which can be more than a day’s travel from their homesŽ.). Ibid. p. 49
(for the Awá “only San Lorenzo [hospital] has expertise in tropical-forest medicine. However,
reaching San Lorenzo from some Awá communities can take up to several days (by foot and then
by boat or by bus). Due to the time and cost of the journey, many illnesses go untreatedŽ.). See
also, pp. 42-43 (describing the limited access to healthcare in Sucumbíos Province). ER, Vol. II,
Annex 5.
486
EM, Chap. 2, para. 2.24; Whitten et al. Report, op. cit., pp. 12-13. ER, Vol. II, Annex 5.
188 “Sucumbíos has traditionally been one of the most neglected and
underdeveloped regions of Ecuador and was until recently
virtually isolated from the rest of the country. As Ecuador itself
acknowledges, the frontier region in general lacks basic
infrastructure and basic sanitation and health servicesŽ 487.
Later, at paragraph 7.37, Colombia similarly states: “The fact that these groups of
Ecuadorians [i.e., those living along the border] live in precarious hygienic
conditions and only have limited access to medical facilities is highly relevant for
488
present purposesŽ .
3.44 Ecuador agrees; it is “highly relevant for present purposesŽ. In particular,
it is relevant precisely because it counters Colombia’s argument that Ecuador’s
witness testimonies should not be credited because they are not supported by
contemporaneous medical documentation. It is disingenuous of Colombia to
insist on documentation that it acknowledges could not exist, through no fault of
the victims.
3.45 It should be added that even the few medical facilities in northern Ecuador
cannot be compared to those facilities with which counsel for Colombia may be
familiar. The hospital at Lago Agrio, for example, which is located more than an
hour’s bus ride (at a cost of US$4-5 per person, which exceeds the daily income
of much of the population) from the nearest frontier communities, is often
487CCM, Chap. 2, para. 2.13.
488CCM, Chap. 7, para. 7.37.
189overcrowded, out of essential medicines, and in ill repair 489. The Counter-
Memorial’s arguments about the lack of contemporaneous medical evidence is
thus a red herring that ignores what Colombia itself admits are the realities of life
in the frontier region.
3.46 Nevertheless, and notwithstanding all of the obstacles, some
contemporaneous medical records do exist. Following the sprayings in August
and September of 2002 a group of doctors and medical students travelled to the
border villages of Sucumbíos that had been affected by the sprayings and
documented the local residents’ resulting symptoms. These symptoms were
490
recorded in individual medical inquests of residents in each village visited .
The consistency in the health impacts reported in the medical inquests is
compelling. The evidence not only shows the repeated occurrence of the same
symptoms across the various towns affected on the border of Ecuador following
the spray; notably, it also reflects the same symptoms reported within Colombia
following local spraying 491, as further discussed in Section III below. In village
after village, the medical inquests reveal a consistent series of ailments following
the appearance of the spray planes … skin irritation and rashes, gastro-intestinal
489
See Whitten et al. Report, op. cit., p. 26. ER, Vol. II, Annex 5.
490Ecuadorian Medical Inquests (Sept. … Nov. 2002). ER, Vol. III, Annex 31.
491See Colombian Medical Inquests (Sept. 2002). ER, Vol. III, Annex 30.
190problems, respiratory problems, headaches, fever, and eye irritation 492… the same
symptoms predicted in the spray chemicals’ warning labels and expected from
493
improper exposure to the spray .
B. S UCUMBÍOS 2002
3.47 The flight data Ecuador secured from the U.S. Government show that
after the end of heavy sprayings along the Sucumbíos-Putumayo border in
January 2001, Colombia conducted some limited sprayings near Sucumbíos in
February, March and April that year.
3.48 Aerial spraying along the border resumed in earnest in November 2001
and lasted through a significant portion of 2002. Figure 3.3 depicts the heavy
spraying that took place in November 2001 to January 2002, and then again
between August and October 2002. As depicted, Colombia conducted massive
spray operations across huge swaths of Putumayo that directly abut Ecuador’s
Sucumbíos Province, including the Cofán-Bermejo Ecological Reserve.
3.49 The data also show that in the two-month period between December 2001
and January 2002, Colombia conducted 10,487 spray flights within 10 kilometres
492See Ecuadorian Medical Inquests (Sept. … Nov. 2002). ER, Vol. III, Annex 31.
493See supra Chap. 3, para. 3.25; Chap. 2, paras. 2.27-2.41; EM, pp. 132-152.
191 0°30'0"N
km
Figure 3.3
River
Putumayo
! Palma Seca
River
San Miguel
76°30'0"W 0-1 k1m-2m-3 k3m-m-7 k7m-10 km ! Chone Chone 1 76°30'0"W
!
spray line (km) !
Distance from nearest Playeratal 0 5 10 15 20 25
!
±
Puertoo
Corazón Orense
Puert! Mestanzagosto
Puerto EscondidoLojana
! !
P u t u m a y o
!
!! S u c u m b í o s
Monterrey
C O L O M B I A ! E C U A D O R
Dios Peña
! Santa La Charapa
Marianita
! !
San Miguel
Farfán
General
!
La Condor
November, 200120010022002r, 200202 ! !
!
(November 2001 - October 2002) 10 de Agosto
!
Spray events (month)
! San SFarannFcrisacnoci2sco 1
77°0'0"W 77°0'0"W
Salinas Virgen del Cisne
Spray Events Within 10 Kilometres of Ecuador’s Sucumbíos Province Peru
River
San Miguel Colombia
International boundary
Protected areas international boundary
Ecuador
Legend
Cofán Reserve
0°30'0"Nof the Ecuador border 49. Between August and October 2002, Colombia
conducted an additional 28,121 flights in the same area, during more than 15,000
of which it similarly disregarded the speed requirement that Colombia deemed
necessary to prevent spray drift 495.
3.50 Tellingly, what the flight data show is described in the witness statements,
which were submitted to the Court long before the flight data became available to
Ecuador. Witness 20 of Puerto Escondido, which is located on the river border,
for example, describes the first spraying during this time:
“The first spraying that I remember was in the year two thousand
and two. It was late in the morning. I was with the pigs by the
plantain fields when I saw the planes. There were also helicopters.
The planes were flying like vultures fighting for food, going up
and down repeatedly. They were dropping white liquid that
extended throughout the air. In some areas it fell directly, in
others it drifted with the wind. It smelled bad, I could barely stand
it. I felt the mist go into my eyes. My eyes became sticky. I
started to feel sick and I immediately returned homeŽ 49.
3.51 The consequences began almost immediately and, over time, affected his
health, the health of children, and his farm:
“When I got home, I shouted to my children to go into the house
because they were outside playing, watching the planes. Still, a
few days later my seven children had stomach aches and diarrhea.
Before, they were healthy. They had never been sick like this
494Hansman & Mena Report, Appendix 3, op. cit., p. 28. ER, Vol. II, Annex 1.
495Ibid.
496
Declaration of Witness 20, 16 Jan. 2009 (hereinafter “Witness 20 DeclarationŽ). EM, Vol. IV,
Annex 206.
192 before; also, the other children in the community became sick with
the same thing. I did not know what to give them, except for
chamomile tea. The plants died a week or two after the sprayings.
The maize started to bend. I had three hectares of yucca and I was
not able to harvest any; it all dried up. I also had ten hectares of
coffee and cocoa, all of which turned yellowŽ 497.
Other witness statements that specifically refer to sprayings in 2002 are cited in
498
the footnote .
3.52 The Counter-Memorial attacks Witness 20’s testimony directly but does
nothing more than argue that “all of this has nothing to do with the characteristics
499
of glyphosateŽ . Like so much of the Counter-Memorial, this argument
assumes its own conclusion. The spray could not have caused the harm
described, Colombia says, because the spray does not cause that kind of harm!
Besides being circular, it is inaccurate.
497
Ibid. EM, Vol. IV, Annex 206.
498Salinas: Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 190; Witness 3 Declaration, op.
cit. EM, Vol. IV, Annex 191; Witness 4 Declaration, op. cit. EM, Vol. IV, Annex 192; Corazon
Orense: Witness 9 Declaration, op. cit. EM, Vol. IV, Annex 197; San Francisco II: Declaration of
Witness 12, 16 Jan. 2009 (hereinafter “Witness 12 DeclarationŽ). EM, Vol. IV, Annex 200;
Declaration of Witness 13, 15 Jan. 2009 (hereinafter “Witness 13 DeclarationŽ). EM, Vol. IV,
Annex 201; Declaration of Witness 17, 16 Jan. 2009 (hereinafter “Witness 17 DeclarationŽ). EM,
Vol. IV, Annex 203; Puerto Escondido: Witness 20 Declaration, op. cit. EM, Vol. IV, Annex
206; Declaration of Witness 22, 16 Jan. 2009 (hereinafter “Witness 22 DeclarationŽ). EM, Vol.
IV, Annex 208; Declaration of Witness 23 (hereinafter Witness 23 DeclarationŽ), 16 Jan. 2009.
EM, Vol. IV, Annex 209; Awá: Declaration Witness of 40, 20 Feb. 2009 (hereinafter “Witness 40
DeclarationŽ). Annex 223; Declaration of Witness 41, 20 Feb. 2009 (hereinafter “Witness 41
DeclarationŽ). EM, Vol. IV, Annex 224.
499CCM, Chap. 7, para. 7.137.
1933.53 Before proceeding further, it bears mention here that, as in other parts of
the Counter-Memorial, Colombia plays a shell-game with the facts. It says that
the effects described have nothing to do with “the characteristics of glyphosateŽ.
Whether or not that is strictly true (it is not), Colombia’s statement does not fairly
meet the evidence. A significant part of the problem is that: (a) the spray mixture
is not composed exclusively of glyphosate; and (b) Colombia has never fully
disclosed exactly what else is in it. Thus, by trying to focus the Court exclusively
on the ostensible effects of glyphosate as such, Colombia clearly hopes to elide
the larger question of the composition and toxicity of the spray mixture as a
whole, and of the other elements of the mixture that have never been fully
disclosed.
3.54 That said, the fact of the matter is that the effects Witness 20 (and others)
describe are exactly the attributes of chemical herbicide exposure. Eye irritation,
headaches, dizziness and gastro-intestinal irritation, along with the yellowing and
wilting of plants, are all classic and expected consequences of exposure to
glyphosate-based herbicides 500. Colombia’s own Counter-Memorial
acknowledges that the mixture “may cause temporary symptoms, such as eye or
skin irritationŽ50. In addition, Colombia’s own National Health Institute has
linked POEA (an ingredient Colombia has admitted to including in the spray
500See supra para. 3.25; EM, pp. 132-152.
501CCM, Chap. 7, para. 7.179.
194mixture) to gastrointestinal damage, breathing difficulties, and other symptoms
502
experienced by the border residents . And the witnesses’ description of the
yellowing and death of multiple crops … maize, yucca, coffee and cocoa … 1 to 2
weeks after the spraying, is precisely what one would expect from exposure to a
503
glyphosate-based herbicide .
3.55 The testimonies of Witness 20 and other witnesses who specifically refer
to sprayings in 2002 are amply supported by contemporaneous accounts. El
Universo reported on 7 September 2002 that Ecuadorian residents in the border
towns of Sucumbíos, including Puerto Nuevo, confirmed having witnessed spray
504
planes operating across the river the week before . The reporters personally
505
observed the same skin rashes on children exposed to the spray in Colombia as
506
those described in the Ecuadorian witness statements .
502
Government of Colombia National Health Institute, Evaluation of Effects of Glyphosate on
Human Health in Illicit Crop Eradication Program Influence Zones, p. 5 (2003) (hereinafter
“Evaluation of Effects of Glyphosate on Human HealthŽ). EM, Vol. II, Annex 96; see also supra,
Chap. 2, para. 2.45.
503Weller Report, op. cit., pp. 3-6. ER, Vol. II, Annex 3.
504“Hunger and Misery from FumigationsŽ, E L U NIVERSO (Guayaquil, 7 Sept. 2002). ER, Vol.
IV, Annex 68.
505Ibid.
506See, e.g., Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 190; Witness 3 Declaration, op.
cit. EM, Vol. IV, Annex 191; Witness 4 Declaration, op. cit. EM, Vol. IV, Annex 192; Witness 9
Declaration, op. cit. EM, Vol. IV, Annex 197; Witness 12 Declaration, op. cit. EM, Vol. IV,
Annex 200; Witness 13 Declaration, op. cit. EM, Vol. IV, Annex 201; Witness 17 Declaration,
op. cit. EM, Vol. IV, Annex 203; Witness 22 Declaration, op. cit. EM, Vol. IV, Annex 208;
Witness 23 Declaration, op. cit. EM, Vol. IV, Annex 209; Declaration of Witness 28, 17 Feb.
1953.56 Twelve days later, on 19 September 2002, El Universo published another
story about more spraying near the border “by six planes and eight helicoptersŽ
that caused harm in Ecuador, including the village of Chone 2, located less than 1
507
kilometre from the border . (The location of Chone 2 is depicted on Figure 3.3
above). The reporter interviewed a Mr. Ángel Encarnación, who watched the
aircraft spraying across the river in Colombia 508. The story also quotes a Mr. José
Aldaz as saying that after the spraying, “the Ecuadorian plantations of sugarcane,
corn, plantains, coffee, fruits, yucca and other crops have lost their natural color
509
and as a result have been damagedŽ .
3.57 The human health effects of the sprayings recounted in the witness
statements annexed to the Memorial are also reflected in contemporaneous
medical inquests of residents in Chone 2 taken on 12 September 2002. For
example, Ms. Isabel Campoverde and her husband both fell ill following the
spraying, suffering from throat irritation, headache, eye irritation and a skin
rash510. Similarly, Ms. Obdulia Pineda, also of Chone 2, suffered from eye
irritation and stress, and her husband had both eye problems and respiratory
2009 (hereinafter “Witness 28 DeclarationŽ). EM, Vol. IV, Annex 212; Witness 40 Declaration,
op. cit. EM, Vol. IV, Annex 223; Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224.
507“Farmers Against FumigationsŽ, E L UNIVERSO (Guayaquil, 19 Sept. 2002). ER, Vol. IV,
Annex 69.
508Ibid.
509Ibid. ER, Vol. IV, Annex 69.
510Medical Inquest of Ms. Isabel Campoverde (12 Sept. 2002) in Ecuadorian Medical Inquests
(Sept.…Nov. 2002). ER, Vol. III, Annex 31.
196distress. Their children were affected too: their three year-old child experienced
stomach problems and their 13 year-old developed a skin rash 511.
3.58 Contemporaneous press accounts show that families in nearby villages
were experiencing the same harms. On 26 September 2002, the Ecuadorian daily,
La Hora, reported that residents of the border villages, including Santa Marianita,
Monterrey, Puerto Mestanza and Puerto Nuevo, had watched as Colombian spray
planes crossed over the border one week earlier 512. Residents affirmed “the
damage caused to crops, animals, and the human beings that inhabit this region as
a result of the Colombian government fumigating in Ecuadorian territoryŽ 513.
The consequences included the following harm to people: “many children have
suffered skin rashes and a type of uncontrollable allergyŽ; harm to plants: “the
type of fumigating agent employed has caused damages to all the vegetation,
especially to pastures, rice, cacao, plantain, coffee, and sugar cane, causing a true
511Medical Inquest of Obdulia Pineda (12 Sept. 2002) in Ecuadorian Medical Inquests (Sept.…
Nov. 2002). ER, Vol. III, Annex 31. See also Medical Inquest of Rosa Margarita Jimenez (12
Sept. 2002) (symptoms included stomach problems, throat and eye irritation, insomnia, and
chills); Medical Inquest of Ma Encarnación (12 Sept. 2002) (symptoms included stomach
problems, dizziness, eye and skin irritation, and a cough); Medical Inquest of Isabel Campoverde
(12 Sept. 2002) (symptoms included throat irritation, eye irritation, a skin rash, and headaches);
Medical Inquest of Alicia Calero (12 Sept. 2002) (symptoms included eye irritation and
headaches); Medical Inquest of Maria Ilbay (12 Sept. 2002) (symptoms included stomach
problems, eye irritation, skin irritation, headaches, and dizziness) in Ecuadorian Medical Inquests
(Sept.…Nov. 2002). ER, Vol. III, Annex 31.
512
“Ecuadorians Demand CompensationŽ, L A H ORA (Quito, 26 Sept. 2002). ER, Vol. IV, Annex
70.
513
Ibid.
197desert in the midst of the jungleŽ; and harm to animals: “the animals that have
514
been most affected are barnyard fowlŽ .
3.59 Shortly thereafter, on 8 October 2002, another article was published in El
Comercio, in which the reporter visited the areas hit by the recent sprayings on
both the Colombian and Ecuadorian sides of the border. On the Colombian side,
the reporter saw “[d]ozens of guarumos, native trees in the areas fumigated, are
barely able to support their enormous leaves. Others, with completely barren
branches, look like enormous candlesticks. The banana trees are almost falling
and their stalks barely support them. The impact of fumigations is so strong that
even the grass has been completely destroyedŽ 515. Crossing the border into
Ecuador he found a similar, though less extreme, sight in villages including
General Farfán: “Plantain plants starting to show dry leaves can be seen from the
road. Walking along the parcels, you can see how the yucca leaves are wilted,
516
how the corn leaves look yellowed and wrinkledŽ . The reporter details how,
following sprayings witnessed by local residents 15 days earlier, whole families
514Ibid.
515“Glyphosate Affects Crops in SucumbíosŽ, L COMERCIO (Quito, 8 Oct. 2002). ER, Vol. IV,
Annex 71.
516Ibid.
198had became ill with headaches, dizziness, stomach aches and skin rashes; plants
dried out; and animals became sick or slowly starved to death 517.
3.60 Just over a week after this article was published, an official from the
Sucumbíos provincial government wrote a letter to the Ecuadorian Ministry of the
Environment in which he reported that the day before (15 October 2002) he
journeyed to the villages affected to personally verify the damage. He confirmed
that:
“as a result of the fumigations the orito, plantain, banana, corn,
and yucca crops, and, in general, all the different agricultural crops
that are basis of sustenance for peasants in the area were burnt.
Similarly, the fish in pools, which comprise part of the sector’s
industry, have died. Barnyard fowl has also been affected. As a
result, negative effects exist which harm the inhabitants of the
Border with Colombia. I was also able to confirm the effects
caused to individuals’ health, such as itchiness, boils on their
bodies, skin affectations and respiratory problemsŽ 518.
3.61 The medical inquest records confirm that the health effects of the 2002
sprayings were widespread throughout the border region. For example, Mr. Italo
Ramón Bene Cosa, a resident of General Farfán, located less than a kilometre
from the border, reported fever-like symptoms, head-ache, eye irritation,
519
coughing and intense skin irritation . He recounted that in late September he
517Ibid.
518Letter from Victor Velasco Tapia, Government of Sucumbíos, to Lourdes Luque, Minister of
Health (16 Oct. 2002). ER, Vol. III, Annex 33.
519Medical Inquest of Ramon Bene Cosa Italo (13 Nov. 2002) in Ecuadorian Medical Inquests
(Sept.…Nov. 2002). ER, Vol. III, Annex 31.
199 520
had seen six spray planes escorted by four helicopters . As he worked outside
on his farm, just 600 metres from the border, his face began to break out in a rash,
and then became hot and swollen 521.
3.62 In the neighbouring town of Puerto Nuevo, itself just abutting the river
border, several residents reported very similar ailments following the September
2002 sprayings: stomach aches, headaches, dizziness, eye irritation, throat
522
irritation and skin rashes . As one resident put it, all this occurred after they
523
“received ‘rain water’ from the planesŽ .
3.63 Other towns in which the occurrence of harm is documented in the
medical inquest forms include Palma Seca and Playera Oriental, 524both of which
are less than 2 kilometres from the border and close to the 2002 sprayings as
525
disclosed in the recently-obtained spray flight data .
520Ibid. ER, Vol. III, Annex 31.
521Ibid.
522Medical Inquest of Zacarias Garcia Chavez (12 Sept. 2002); Medical Inquest of Jose Hilberto
Reyes Ramirez (12 Sept. 2002); Medical Inquest of Jose Felix Guerra Rodriguez (12 Sept. 2002);
Medical Inquest of Lucelia Torres Garcia (12 Sept. 2002); Medical Inquest of Edith Garcia (12
Sept. 2002) in Ecuadorian Medical Inquests (Sept.…Nov. 2002). ER, Vol. III, Annex 31.
523
Medical Inquest of Lucelia Torres Garcia (12 Sept. 2002) in Ecuadorian Medical Inquests
(Sept.…Nov. 2002). ER, Vol. III, Annex 31.
524
See Ecuadorian Medical Inquests (Sept.…Nov. 2002). ER, Vol. III, Annex 31.
525
See Figure 3.3.
2003.64 The accounts of the harm resulting from the sprayings are still further
corroborated in the report issued by representatives of several Ecuadorian NGOs
who visited various communities in the border area in late September 2002, to
investigate reports of the damage for themselves. In the first two Ecuadorian
border towns visited, Playera Oriental and Chone 2, “evident damage from the
fumigation of 6 September 2002 on the Colombian side of the banks of the San
526
Miguel River was observedŽ . Similarly, in the Ecuadorian village facing the
Colombian town of La Pedregosa across the river, “damage was found in the
banana plantations and other crops, which showed signs of chlorosis (yellowing),
as a result of the proximity of fumigations in Colombia. The accounts pointed
out that these impacts stemmed from the fumigations done on the Colombian side
527
between 30 August and 6 September 2002Ž . Ecuador first cited to this report
528
in the Memorial . In response, Colombia says only that it is “wholly
unsubstantiatedŽ 529. Given the corroborative evidence Ecuador has proffered, one
might be justified in wondering at what point exactly Colombia would accept any
evidence as “substantiatedŽ.
526
Association of American Jurists et al., Report on Verification Mission: Impacts in Ecuador of
Fumigations in Putumayo as Part of Plan Colombia, pp. 7-8 (Oct. 2002). ER, Vol. III, Annex 32.
527
Ibid., p. 8.
528
EM, Chap. 6, para. 6.14, n. 417.
529CCM, Chap. 7, para. 7.151.
2013.65 In any event, the truth of the events recounted in the witness statements
and the NGO and press reports is compellingly affirmed by the then-UN Special
Rapporteur on the Right to Health, Mr. Paul Hunt, who issued a press statement
and a report on the subject of Colombia’s aerial sprayings. This unquestionably
impartial observer visited the border area, interviewed the inhabitants, reviewed
relevant materials, including those of a scientific and technical nature, and
concluded that the residents’ descriptions of harm were credible and reliable.
Perhaps for that reason, the Counter-Memorial entirely fails to come to grips with
the Special Rapporteur’s findings.
3.66 The Special Rapporteur travelled to the region in May 2007 in order to
investigate the reports of harm for himself. In his own words, the “focus of my
mission [to Ecuador] was aerial spraying of glyphosate, combined with additional
530
components, along the Colombia-Ecuador borderŽ . His assessment involved
reviewing all pertinent sources of evidence: “The Mission . . . reviewed the
existing scientific evidence, took personal testimonies, consulted with experts,
collected additional information … and examined all of this material through the
lens of the human right to healthŽ 531. In other words, the Special Rapporteur took
all the steps necessary to make reliable findings of fact, including visiting
530Paul Hunt, UN Special Rapporteur on the Right to the Highest Attainable Standard of Health,
Closing Remarks to the Press, Quito, Ecuador (18 May 2007). ER, Vol. IV, Annex 107.
531Ibid.
202 532
communities in the northern border zone . His conclusions thus deserve special
attention, particularly insofar as they validate the campesinos’ descriptions of the
harm that they experienced and observed. In that regard, Mr. Hunt specifically
determined that: “There is credible, reliable evidence that the aerial spraying of
glyphosate along the Colombia-Ecuador border damages the physical health of
people living in EcuadorŽ 533. Based on this finding of “credibleŽ and “reliable
evidenceŽ, Mr. Hunt concluded “there is an overwhelming case that the aerial
spraying of glyphosate along the Colombia-Ecuador border should not re-
commenceŽ 534. Indeed, the Special Rapporteur found the evidence so compelling
that he concluded “there is no doubt in my mind that Colombia should not
recommence aerial spraying of glyphosate on its border with Ecuador . . .
535
Colombia should respect a ten-kilometre no-spray zone along the borderŽ .
532Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest
Attainable Standard of Physical and Mental Health, Paul Hunt: Preliminary Note on Mission to
Ecuador and Colombia, Addendum, A/HRC/7/11/Add.3, para. 6 (4 Mar. 2007). EM, Vol. II,
Annex 31.
533Ibid. (emphasis added).
534
Paul Hunt, UN Special Rapporteur on the Right to the Highest Attainable Standard of Health,
Closing Remarks to the Press, Quito, Ecuador (18 May 2007) (emphasis added). ER, Vol. IV,
Annex 107.
535Ibid. See also Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of
the Highest Attainable Standard of Physical and Mental Health, Paul Hunt: Preliminary Note on
Mission to Ecuador and Colombia, Addendum, U.N. Doc. A/HRC/7/11/Add.3, para. 17 (4 Mar.
2007). (“While in Ecuador, the Special Rapporteur’s preliminary view was that there was
credible and reliable evidence that the aerial spraying of glyphosate along the border damages the
physical and mental health of people living in Ecuador. The Special Rapporteur’s preliminary
conclusion was that the evidence provided during the mission was sufficient to call for the
application of the precautionary principle and that, accordingly, Colombia should not recommence
aerial spraying in the 10-km border zone with Ecuador, thus ensuring conformity with its
international human rights responsibilitiesŽ.). EM, Vol. II, Annex 31.
2033.67 The Special Rapporteur’s findings are all the more notable because they
were made after affording the Colombian government an opportunity to present
its side of the story. As part of his investigation, the Special Rapporteur visited
Colombia and conducted discussions about the aerial spraying programme with
senior Colombian officials, including the Vice-President, the Deputy Minister of
Health and the Director of the Anti-Narcotics Police, among others 536. Nothing
he heard in Colombia altered his determination that “credible and reliable
evidenceŽ demonstrated that “aerial spraying of glyphosate along the Colombia-
537
Ecuador border damages the physical health of people living in EcuadorŽ .
3.68 The Counter-Memorial seems wary of reminding the Court about the
Special Rapporteur’s report. Thus, Colombia adopts what might charitably be
characterized as a hit-and-run approach, devoting just one paragraph to it. And
even then, it does not dispute any of the Special Rapporteur’s findings. It argues
only that they are irrelevant because they do not purport to be “a scientific
assessment of the effects of the fumigationsŽ 53.
3.69 Here, as in so many places, Colombia twists the truth. True, the report
does state that the Special Rapporteur’s visit “was not a scientific missionŽ, but it
536Ibid., paras. 6-7. EM, Vol. II, Annex 31.
537
Ibid.
538
CCM, Chap. 7, para. 7.118.
204also notes that he “reviewed the existing scientific evidenceŽ and “consulted with
539
expertsŽ . More to the point, the importance of the Special Rapporteur’s
findings does not lie in whether it can be labelled “scientificŽ. What matters is
that the individual specifically charged by the United Nations with responsibility
for these issues explicitly determined that all the evidence taken as a whole …
including the scientific evidence that he considered … credibly and reliably
supported the conclusion that the sprayings were causing harm in Ecuador. And
he specifically credited the testimonies of the Ecuadorian campesinos whom he
540
interviewed, declaring them “credibleŽ and “reliableŽ .
3.70 Viewed in light of the spray flight data, especially insofar as they show
Colombia’s wanton disregard for its own operational parameters, the Special
Rapporteur’s findings are more than corroborated. As discussed above, the data
confirm that Colombia was conducting massive spray operations along the border
with Sucumbíos in late 2000/early 2001 and in 2002 at exactly the time that the
other evidence, including NGO reports and eyewitness statements, indicate that
541
harm materialized there . The Special Rapporteur’s findings thus constitute one
539
Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest
Attainable Standard of Physical and Mental Health, Paul Hunt: Preliminary Note on Mission to
Ecuador and Colombia, Addendum, U.N. Doc. A/HRC/7/11/Add.3, para. 10 (4 Mar. 2007). EM,
Vol. II, Annex 67.
540
See supra Chap. 3, para. 3.66.
541
See supra Figures 3.1 and 3.3.
205more highly probative element underscoring the validity and veracity of
Ecuador’s case.
3.71 A second UN Special Rapporteur, the Special Rapporteur on the Situation
of Human Rights and Fundamental Freedoms of Indigenous People, made similar
findings during his mission to Ecuador in 2006, especially in regard to the special
harms suffered by indigenous peoples as a result of Colombia’s aerial sprayings
542
along the Ecuadorian border .
1. The Kichwa and Cofán of Sucumbíos
3.72 With respect to the special harms caused to indigenous peoples, Colombia
is hard-pressed to dispute that their unique communal structures, modes of living,
spiritual traditions, and inter-connection with the land, make them particularly
susceptible to the health and environmental impacts that result from exposure to
the chemical spray mixture. In response to Ecuador’s evidence, Colombia simply
relies on the same refrain used throughout its Counter-Memorial: “the time is
vague, the living conditions are precarious, and we want more evidenceŽ. The
hollowness of this robotic response has already been addressed at paragraphs 3.23
to 3.46 above. Colombia accepts that the problems faced by the indigenous
542Report of the Special Rapporteur on the Situation of Human Rights and Fundamental
Freedoms of Indigenous People, Rodolfo Stavenhagen: Mission to Ecuador (25 April-4 May
2006), U.N. Doc. A/HRC/4/32/Add.2, paras. 28-34 (28 Dec. 2006). EM, Vol. II, Annex 30.
206communities along the border are “real and urgentŽ but prefers to side-step the
543
evidence and blame them on something else . This approach simply doesn’t
work. The same confluence of evidence that demonstrates the harm caused by
Colombia’s aerial spraying elsewhere in the border regions of Ecuador confirms
the particular harms suffered by the indigenous communities during the same
time periods.
3.73 For the Kichwa people residing along the border of Sucumbíos, the harm
544
began with the initial sprayings in late 2000/early 2001 . A July 2001 report, in
which delegates from CONAIE had travelled to the Kichwa village of San
Francisco 2, located approximately 3 kilometres from the border, to document the
effects of the sprayings, already describes “the departure of the shamansŽ from
the village as “a clear indication of the cultural impact on bordering
545
communitiesŽ . The witness statement by Ms. Blanca Chancosa, a Kichwa
leader and member of the observation mission, explains the critical role that
543
CCM, Chap. 9, para. 9.155.
544See Figure 3.1; Hansman & Mena Report, op. cit., p. 14. ER, Vol. II, Annex 1.
545Confederation of Indigenous Nationalities of Ecuador (CONAIE) et al., Technical Report of
the International Commission on the Impacts in Ecuadorian Territory of Aerial Fumigations in
Colombia, p. 22 (19-22 July 2001). EM, Vol. IV, Annex 162.
207shamans, or yachaks, play as the spiritual leaders and medicinal healers in the
Kichwa community 546.
3.74 The 2001 CONAIE report explains that the shamans had left “because
their healing plants have been contaminated and they can no longer use them to
cure peopleŽ 54. Underscoring the gravity of this event, the report concludes,
“[t]he cultural impact of the fumigations on this community has been
548
enormousŽ .
3.75 The CONAIE report observes that the harms caused by Colombia’s aerial
sprayings to the Cofán, Kichwa, and Shuar nationalities of Ecuador were in
common with those experienced by numerous indigenous communities that had
“seen their collective rights violated by indiscriminate sprayings over their
549
territories in ColombiaŽ . The threat was so great and so common, that in April
2002 the indigenous coastal tribes from both Ecuador and Colombia held their
546
Declaration of María Blanca Chancosa Sánchez, 14 Jan. 2009, para. 3. EM, Vol. IV, Annex
187. The report, written by experts on the indigenous communities of northern Ecuador, found at
Annex 5, further describes the shamans’ role in maintaining the community’s traditions and
protecting the health of its people. Whitten et al. Report, op. cit., p. 27. ER, Vol. II, Annex 5.
547Confederation of Indigenous Nationalities of Ecuador (CONAIE) et al., Technical Report of
the International Commission on the Impacts in Ecuadorian Territory of Aerial Fumigations in
Colombia, p. 13 (19-22 July 2001). EM, Vol. IV, Annex 162. See also ibid., p. 22.
548Ibid.
549Ibid., p. 22.
208first binational meeting in history to discuss their shared experience of the
sprayings 550.
3.76 The extent of harm that year was notable … so notable that, belying
Colombia’s attempt to depict the indigenous peoples as time-challenged, the
indigenous Kichwa residents cite to 2002 as a particularly intense year of aerial
spraying and adverse consequences therefrom 551. A 2003 report recording the
impacts in the Kichwa community of Yana Amarum, explains that the community
was “just recovering from the effect of the sprayings in July, August and
September 2002Ž 55. The spray flight data for 2002, shown in Figure 3.3 above,
confirms that the sprayings in that year and in that location were quite intense and
close to the Kichwa communities. The available evidence on the spray mixture
during that time shows that in 2002 the spray was particularly toxic … so toxic that
553
the formulation had to be changed . The evidence is not only consistent on the
locations and dates of the sprayings near the Kichwa communities, but also on the
special harms suffered as reflected in the witness testimonies.
550
“Binational Meeting of Indigenous Communities … Plan Colombia terrorizes the communitiesŽ,
LA H ORA (7 Apr. 2002). ER, Vol. IV, Annex 65.
551
See, e.g., Witness 12 Declaration, op. cit. EM, Vol. IV, Annex 200.
552Ecuadorian Ministry of Environment et al., Impacts in Ecuador by the Fumigations Carried
Out in the Putumayo Province under Plan Colombia, pp. 17-18 (July 2003) (hereinafter “Impacts
in Ecuador by the Fumigations Carried Out in the Putumayo Province, 2003Ž). EM, Vol. IV,
Annex 166.
553
See supra Chap. 2, para. 2.22.
2093.77 The Kichwa people are particularly reliant on plants for their sustenance,
554
their medicine, and their cultural and spiritual traditions . The 2002 sprayings
and destruction of plants in and around the Kichwa villages thereby caused
especially significant harm. Most immediately, the destruction of plants left the
Kichwa without medicine to treat the ailments caused by the spray mixture. The
report provided by experts on the indigenous communities of northern Ecuador,
explains the critical function that local plants serve in Kichwa medicinal
555
practice . As illustrated by a Kichwa mother in her witness statement, when she
and her children became ill following the 2002 sprayings … with eye irritation,
vomiting, diarrhea and headaches (the typical symptoms of exposure to
glyphosate-based herbicides and their adjuvants) … she could not use or treat her
children with the traditional medicine “passed from parents to childrenŽ as the
556
cure for diarrhea . She explains that “with the sprayings, the plants have dried
up and we can no longer prepare natural remediesŽ 557.
3.78 The same Kichwa mother explains that her crops of maize, coffee,
plantain and cacao all “dried upŽ after the 2002 spraying, leaving “no food for
554
See Whitten et al. Report,op. cit.,pp. 29-30. (“The Amazonian Kichwa are renowned for
their extensive knowledge of hundreds of useful plant species, many of which are medicinalŽ.).
ER, Vol. II, Annex 5.
555
Ibid.
556
Witness 12 Declaration, op. cit. EM, Vol. IV, Annex 200.
557Ibid.
210 558
[her] childrenŽ . As further detailed by experts in the anthropology of the
Kichwa people, the Kichwa maintain many of their traditional agricultural
practices, “making them heavily reliant on environmental resources for their well-
beingŽ and particularly vulnerable to environmental changes 559. With the 2002
sprayings, the Kichwa in the community of Yana Amarum saw their primary food
crops, including their staple food source, yucca or manioc, turn yellow and die.
Kichwa resident Witness 28, explains that in Yana Amarum “the situation was
very serious, because these plants are the basis of our diet. After the sprayings,
560
we had nothing to eatŽ .
3.79 With the integral aspects of the Kichwa life damaged by Colombia’s
aerial sprayings and “nothing to harvestŽ, Witness 28, like other Kichwa before
and after him, left to live elsewhere … leaving the indigenous community and its
561
traditions behind .
3.80 Another indigenous group, the Cofán, who also reside on the Sucumbíos
border, fared no better than the Kichwa. Approximately 1,200 Cofán people live
in 13 communities in Ecuador’s Sucumbíos province, including within the Cofán-
Bermejo Ecological Reserve. Experts on the Cofán people explain that “the
558Ibid.
559Whitten et al. Report, p. 26. ER, Vol. II, Annex 5.
560
Witness 28 Declaration, op. cit. EM, Vol. IV, Annex 212.
561
Ibid.
211Cofán rely heavily on forest resources for their culture, spiritual traditions, and
livelihoods. Therefore, the survival of this important indigenous group depends
upon continuous access to the healthy and intact environments . . . Ž 562. As
explained in Cofán Witness Statement 31, “[f]or the Cofán, nature is very
important, she provides us with everything that we need to live . . . if nature gets
563
sick, we also get sick; our life depends on natureŽ .
3.81 The spray flight data from 2002, shown in Figure 3.3, demonstrate the
extent and proximity of Colombia’s sprayings to the Cofán-Bermejo Ecological
Reserve, where Cofán have resided since pre-colonial times 564. In 2002, there
were more than 8,950 spray flights within 10 kilometres of the Reserve 565. As
indicated in Chapter 2, between 2000 and 2008 there were more than 12,400
spray flights within 10 kilometres of the Reserve, and more than one thousand
566
within just 2 kilometres .
3.82 The Cofán people produce the great majority of their food directly from
the forest environment 567. Following the sprayings, the staple crops on which
562Whitten et al. Report, op. cit., p. 17. ER, Vol. II, Annex 5.
563Declaration of Witness 31, 27 Feb. 2009 (hereinafter “Witness 31 DeclarationŽ). EM, Vol. IV,
Annex 215.
564Whitten et al. Report, op. cit., p. 17. ER, Vol. II, Annex 5.
565Hansman & Mena Report, op. cit., Appendix 3, p. 28. ER, Vol. II, Annex 1.
566Ibid., p. 14.
567
See Whitten et al. Report, op. cit., pp. 20-21. ER, Vol. II, Annex 5.
212they relied were dead or dried up 568. The animals on which they base their daily
diet were also dead or gone … as one witness recounts, following the spraying
they found the birds dead on the ground, a guanta whose hair had fallen off, and
almost no fish in the water 569. The harm goes beyond damage to the Cofán’s food
sources; it has also impacted their medicinal and spiritual practices. As explained
in the anthropological expert report: “Cofán medicinal practices also depend upon
570
residence in a relatively intact environmentŽ . They use as many as 250
571
different plant species for medicinal purposes . But, as described in the Cofán
witness statements, many of the medicinal plants were destroyed upon exposure
572
to the spray . The debilitation of the medicinal plants left no remedies for the
Cofán who were sickened (including skin irritation, vomiting, throat irritation,
and diarrhea) by exposure to the chemicals deposited by Colombia’s spray
planes 573.
3.83 Many of the Cofán have not been able to withstand the damage to their
way of life caused by Colombia’s aerial spraying, and have moved away from the
568
See Declaration of Witness 26, 17 Feb. 2009 (hereinafter “Witness 26 DeclarationŽ). EM, Vol.
IV, Annex 210.
569
Witness 31 Declaration, op. cit. EM, Vol. IV, Annex 215.
570
Whitten et al. Report, op. cit., p. 22. ER, Vol. II, Annex 5.
571
See Ibid.
572Declaration of Witness 27, 17 Feb. 2009 (hereinafter “Witness 27 DeclarationŽ). EM, Vol. IV,
Annex 211; Declaration of Witness 29, 16 Jan. 2009 (hereinafter “Witness 29 DeclarationŽ). EM,
Vol. IV, Annex 213; Witness 31 Declaration, op. cit. EM, Vol. IV, Annex 215.
573
See Witness 29 Declaration, op. cit. EM, Vol. IV, Annex 213; Witness 31 Declaration, op. cit.
EM, Vol. IV, Annex 215.
213sprayings and their ancestral lands. Cofán Witness 26 and her family used to live
by the border but “moved to live in the interior of Bermejo River, thinking that
574
this way we were going to be safe, but we were still affectedŽ . She continues,
“this displacement has affected our traditions, [as] it is very important for the
Cofán people to keep their rootsŽ 575.
C. E SMERALDAS 2000
3.84 The evidence from Mataje, a village located in the westernmost part of the
Ecuadorian province of Esmeraldas, demonstrates that the aerial sprayings began
576
to exert their effects on the village also in late 2000 . The village was originally
situated along the banks of the Mataje River bordering Putumayo, Colombia, but
following repeated aerial sprayings in the border area has since been re-
established farther inland 577.
3.85 The witnesses describe exactly where they were and what they saw when
the spray planes, previously unfamiliar to them, arrived in 2000. Witness 34
relates that she:
574
Witness 26 Declaration, op. cit. EM, Vol. IV, Annex 210.
575
Ibid.
576See infra Figure 3.4.
577See Declaration of Witness 37, 19 Feb. 2009 (hereinafter “Witness 37 DeclarationŽ). EM, Vol.
IV, Annex 220.
214 “was in Mataje when the first spraying occurred in the year 2000 .
. . When the first spraying occurred, I was working in the field,
clearing the land with a machete and removing the weeds in order
to plant. I saw several planes above. . . . They came and went
several times. They made a noise and dropped a liquid. In the air
it looked like white dust. With the wind, it moved through the air
and descended to the ground. When it fell on the plants, I noticed
that it looked like oil on top of themŽ 57.
3.86 Similarly, Witness 36 describes what she saw from her home on the banks
of the Mataje River at that time:
“The first time that I saw the sprayings was in the year 2000. I
was clearing the land in my farm, accompanied by my younger
son. I saw the planes and helicopters flying over the river. From
the planes, a white rain was coming out. That rain fell on top of 579
me and also on top of my son; it looked like grease on the skinŽ .
The statements from Mataje residents … Witnesses 30, 32, 33, 37, 38 and 39 …
further describe the witnesses’ experience of the first aerial spraying in 2000 and
its effects580. These descriptions belie Colombia’s criticism of the “vagueŽ dates
described in the statements of the Mataje residents.
578
Declaration of Witness 34, 19 Feb. 2009 (hereinafter “Witness 34 DeclarationŽ). EM, Vol. IV,
Annex. 218.
579
Declaration of Witness 36, 19 Feb. 2009 (hereinafter “Witness 36 DeclarationŽ). EM, Vol. IV,
Annex 219
580
Mataje residents: Declaration of Witness 30, 19 Feb. 2009 (hereinafter “Witness 30
DeclarationŽ). EM, Vol. IV, Annex 214; Declaration of Witness 32, 19 Feb. 2009 (hereinafter
“Witness 32 DeclarationŽ). EM, Vol. IV, Annex 216; Declaration of Witness 33, 19 Feb. 2009
(hereinafter “Witness 33 DeclarationŽ). EM, Vol. IV, Annex 217; Witness 37 Declaration, op. cit.
EM, Vol. IV, Annex 220; Declaration of Witness 38, 19 Feb. 2009 (hereinafter “Witness 38
DeclarationŽ). EM, Vol. IV, Annex 221; Declaration of Witness 39, 19 Feb. 2009 (hereinafter
“Witness 39 DeclarationŽ). EM, Vol. IV, Annex 222.
2153.87 Colombia’s own flight data confirm what the witnesses describe. The
data, depicted in Figure 3.4, show that Colombia began its sprayings along the
border with Esmeraldas in August 2000, and continued through September 2000.
3.88 Contemporaneous news articles further corroborate what is evident from
the witness statements and flight data, confirming the timing of the initial
sprayings and the resultant harms. A newspaper report from La Hora on 18
September 2000 states that the Mataje residents were suffering from health
581
impacts following sprayings during that time . Due to the extent of harm
reported, the article explains that the local district of San Lorenzo formed a
commission to travel to Mataje to further investigate. A member of the
commission confirmed that “at this momentŽ Mataje residents were suffering
from “skin infections, ongoing diarrhea, and eye irritations . . . seemingly as a
consequence of [Colombia’s] fumigationsŽ 582. The article provides Mataje
residents’ reports of frequent flights by Colombian spray planes and
583
helicopters .
581
“In Mataje the Implementation of Plan Colombia Causes First RavagesA,HORA (Quito, 18
Sept. 2000). ER, Vol. IV, Annex 57.
582
Ibid..
583Ibid.
216 1°30'0"N 1°15'0"N 1°0'0"N
0-1m-2 k2mm-5 k5m-m-10 km Figure 3.4
August, 2000r, 2000
78°15'0"W spray line (km) km 78°15'0"W
Distance from nearest Spray events (month)
Protected areas international boundary
Legend
River C a r c h i
San Juan 0 5 10 15 20 25
78°30'0"W 78°30'0"W
Awá Reserve ±
!
N a r i ñ o
Mataje Alto
C O L O M B I A
Gua!ualito
River
Mataje
!
Mataje Awá Reserve
78°45'0"W 78°45'0"W
(August - September 2000)
H
San Lorenzo E s m e r a l d a s
E C U A D O R
Cayapas-Mataje
79°0'0"W Peru 79°0'0"W
Spray Events Within 10 Kilometres of Ecuador’s Esmeraldas Province Colombia
Ecuador
1°30'0"N 1°15'0"N 1°0'0"N3.89 The harms reported in the contemporaneous news article … “skin
584
infections, ongoing diarrhea, and eye irritationsŽ … are the same harms
described by the witness statements (and the same ones associated with exposure
585
to glyphosate-based herbicides and adjuvants like POEA) . Witness 36, for
example, explains how after feeling the spray on her skin and drinking water from
a bucket exposed to the white mist, she became sick “with a stomach ache,
vomiting, diarrhea and itchiness on [her] bodyŽ 58. Likewise, Witness 34
describes her immediate reactions upon being exposed to the spray in 2000, “[t]he
liquid also fell on me, on my head, arms, and all over my body. Immediately, I
587
felt my skin itch intensely. . . . Above all, my face became very swollenŽ . She
describes how her face was so “disfiguredŽ that she was ashamed to speak to the
visiting newspaper reporters 588.
3.90 Despite the Counter-Memorial’s rote denials, these harms are the classic,
and acknowledged, health effects of exposure to the known elements of
Colombia’s spray mixture. As described in Ecuador’s Memorial, Colombian and
U.S. government studies expressly acknowledge that the spray mixture, and its
584Ibid.
585
See supra Chap. 3, para. 3.25; EM pp. 132-152.
586
Witness 36 Declaration, op. cit. EM, Vol. IV, Annex 219.
587Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218.
588Ibid.
217recognized chemical components, cause eye irritation 589. In fact, the U.S.
Environmental and Protection Agency confirms that the product being used at
590
this time could cause “irreversible eye damageŽ . As previously discussed, the
product labels for the known chemical elements used in the spray mixture warn
that improper exposure to the products cause “skin irritationŽ, “gastrointestinal
tract irritationŽ, and “eye irritationŽ 591 … all the symptoms felt by the Mataje
residents.
3.91 Colombia cites a lack of contemporaneous medical evidence to
corroborate these testimonies, and points to statements in a 2001 observation
mission report to the effect that, beyond the impacts seen and reported, there were
no studies showing that the sprayings and the immediate appearance of
glyphosate-related illnesses were connected. As noted earlier, medical records of
the kind typically found in European health clinics are not maintained in Mataje
where there was but one nurse and no electricity in the makeshift “health centreŽ
that was established on 15 September 2000, a few days before it was
overwhelmed by an unprecedented wave of sick patients with similar but
592
unfamiliar symptoms . The observation mission reports and witness statements
589
EM, Chap. 5, para. 5.37.
590See supra Chap. 2, para. 2.21
591See supra Chap. 2, paras. 2.19-2.20, 2.24, 2.27, 2.29, 2.32-2.34, 2.37-2.41.
592“44 Affected by the FumigationsŽ,LECOMERCIO (Quito, 22 Oct. 2000). ER, Vol. IV, Annex
58.
218provided in the Memorial explain that after the spray drift had been felt by the
local residents, the children became ill first, soon after followed by the adults …
593
ultimately totalling nearly 40 people . As described by Witness 36, when she
went to see the nurse “there were so many people sick with vomiting and
594
headaches that there was nowhere to sitŽ . With or without contemporaneous
medical records or scientific studies, the uncontradicted statements of numerous
witnesses in regard to what they themselves experienced and observed constitutes
reliable evidence of the health impacts of Colombia’s aerial sprayings near
Mataje.
3.92 Although she made no written records, the nurse at Mataje had no
difficulty connecting the illnesses she treated in September 2000 to the recently
conducted aerial spraying adjacent to the village 595. In a contemporaneous news
593
Ibid. ER, Vol. IV, Annex 58; EM, Chap. 6, paras. 6.38-6.43; Confederation of Indigenous
Nationalities of Ecuador (CONAIE) et al., Technical Report of the International Commission on
the Impacts in Ecuadorian Territory of Aerial Fumigations in Colombia, p. 17 (19-22 July 2001).
EM, Vol. IV, Annex 162; Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222; Witness 36
Declaration, op. cit. EM, Vol. IV, Annex 219; Witness 32 Declaration, op. cit. EM, Vol. IV,
Annex 216. See also Witness 30 Declaration, op. cit. EM, Vol. IV, Annex 214; Witness 33
Declaration, op. cit. EM, Vol. IV, Annex 217; Witness 34 Declaration, op. cit. EM, Vol. IV,
Annex 218; Witness 37 Declaration, op. cit. EM, Vol. IV, Annex 220; Witness 38 Declaration,
op. cit. EM, Vol. IV, Annex 221.
594Witness 36 Declaration, op. cit. EM, Vol. IV, Annex 219.
595
Hospital staff in southern Colombia similarly had no trouble seeing the causal link between the
sprayings and illnesses seen there: “These symptoms also coincide with observations made by the
medical staff at the hospitals in southern Colombia, who said that from the moment that the
sprayings began, they observed a marked increase of these illnessesŽ. Confederation of
Indigenous Nationalities of Ecuador (CONAIE) et al., Technical Report of the International
Commission on the Impacts in Ecuadorian Territory of Aerial Fumigations in Colombia, p. 20
(19-22 July 2001). EM, Vol. IV, Annex 162.
219article from 22 October 2000, she is reported as confirming that since the
596
September sprayings had ended, the symptoms had not reappeared . That is,
until Colombia resumed spraying along the border with Mataje the following
year, and in each year after that.
3.93 Colombia cites a report including statements by area doctors … none of
whom treated the victims of aerial spraying in Mataje … suggesting that their
symptoms may have resulted from exposure to glyphosate-based herbicides used
at palm plantations in Ecuador 59. But the Counter-Memorial fails to point out
that, on the same page of the cited report, a plantation worker explains that the
palm plantation spraying is “done with a [hand] pump, and not with planes; and,
that the stream into which the water for these crops drain is downstream, below
598
MatajeŽ . The UN Special Rapporteur on the Right of Everyone to the
Enjoyment of the Highest Attainable Standard of Physical and Mental Health
further refutes Colombia’s “misleadingŽ argument: “The Special Rapporteur
notes that the use of glyphosate in Ecuador (direct and manual) is different from
the method used on the border by Colombia (aerial spraying). Furthermore, as
596“44 Affected by the FumigationsŽ,LECOMERCIO (Quito, 22 Oct. 2000). ER, Vol. IV, Annex
58.
597CCM, Chap. 7, paras. 7.145-7.146 (citing Confederation of Indigenous Nationalities of
Ecuador (CONAIE) et al., Technical Report of the International Commission on the Impacts in
Ecuadorian Territory of Aerial Fumigations in Colombia (19-22 July 2001). EM, Vol. IV, Annex
162).
598
Confederation of Indigenous Nationalities of Ecuador (CONAIE) et al., Technical Report of
the International Commission on the Impacts in Ecuadorian Territory of Aerial Fumigations in
Colombia, p. 18 (19-22 July 2001). EM, Vol. IV, Annex 162.
220the composition and concentration of the spraying appear to differ between
Ecuador and Colombia, the suggested equivalence between Ecuadorian and
Colombian practice is misleadingŽ 599.
3.94 The harms caused in Mataje extended beyond human health. The Mataje
residents also witnessed a wave of fish deaths in the border river immediately
following the sprayings. As confirmed by Witness 34, “after the spraying, there
were a lot of dead fish and shrimp. Usually, the fish and shrimp are below the
water level. But, after the sprayings, they were floating on the surface of the river
and going downstream with the current. I observed this immediately after the
sprayingsŽ 600. Mataje Witnesses 33, 37, 38, and 39 similarly recount their
601
sighting of the fish-kill . The witness statements are corroborated by
contemporaneous news articles reporting that on 22 September 2000, in addition
to the ill effects on humans following the spraying, “fish and other speciesŽ had
602
also died in the border Mataje River .
599Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest
Attainable Standard of Physical and Mental Health, Paul Hunt: Preliminary Note on Mission to
Ecuador and Colombia, Addendum, U.N. Doc. A/HRC/7/11/Add.3, para. 18 (4 Mar. 2007). EM,
Vol. II, Annex 31.
600
Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218.
601Witness 33 Declaration, op. cit. EM, Vol. IV, Annex 217; Witness 37 Declaration, op. cit.
EM, Vol. IV, Annex 220; Witness 38 Declaration, op. cit. EM, Vol. IV, Annex 221; Witness 39
Declaration, op. cit. EM, Vol. IV, Annex 222.
602
“44 Affected by the FumigationsŽ, EL COMERCIO (Quito, 22 Oct. 2000). ER, Vol. IV, Annex
58.
2213.95 Colombia tries to ignore the confluence of evidence, including its own, to
claim that the fish deaths are “unsubstantiatedŽ 603. The Counter-Memorial
provides two principal reasons for this unsupported assertion … both of which fail.
First, Colombia resorts to its usual tactic of referring only to the base chemical
glyphosate, and not the actual spray mixture, arguing that “glyphosateŽ has
exhibited “little chronic toxicity to fishŽ 604. This is not the understanding of the
manufacturers of glyphosate-based herbicides, however. The label for GLY-41,
one of the herbicide formulations that Colombia admits to using, for example,
605
provides this warning in regard to the product’s toxicity to fish :
Figure 3.5 Warning Symbols From GLY-41 Label … Fish and Other Aquatic
Organisms
3.96 Moreover, Colombia’s own experts recognize the difference in toxicity
between glyphosate itself and the actual spray mixture used in the aerial spraying
programme. The Dobson Report, which is annexed to the Counter-Memorial, for
example, admits that “fish exposed to the spray formulation as used in Colombia
603CCM, Chap. 7, para. 7.82.
604
Ibid. (internal quotations omitted).
605
See supra Chap. 2, para. 2.41.
222(including the Cosmo-flux adjuvant) show greater toxicity than to the formulation
aloneŽ 606. The Solomon study further warns that “moderate risks could occur in
aquatic organisms in shallow surface waters that are over-sprayed during the
eradication programŽ 607. Accordingly, “[i]f shallow waters are routinely found
close to fields, it is recommended that other formulates be tested for the purposes
608
of selecting products that present a lower risk to aquatic organismsŽ . There is
no evidence that Colombia changed the formula in response to this study. Thus,
it should come as no surprise that fish-kills similar to the one at Mataje were also
produced by the aerial sprayings in Ecuador’s Sucumbíos Province and in many
609
Colombian villages exposed to the sprayings .
3.97 Colombia’s second attempt at avoiding responsibility for fish-kills
resulting from its aerial sprayings is based on its alleged observance of “no-spray
610
buffer zones along watercoursesŽ . As with all of the studies Colombia relies
on, its expert presumes Colombia’s strict compliance with the buffer zone
606
CCM, Appendix, p. 25, para. 99.
607
Solomon 2005, op. cit., p. 11. CCM, Vol. III, Annex 116. The report continues: “However,
the frequency of occurrence and extent to which this happens are unknown as data on the
proximity of surface waters to coca fields were not available.Ž
608
Ibid., op. cit., p. 12.
609
See, e.g., Association of American Jurists et al., Report on Verification Mission: Impacts in
Ecuador of Fumigations in Putumayo as Part of Plan Colombia, p. 3 (Oct. 2002). EM, Vol. IV,
Annex 165; Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 190; Witness 5 Declaration, op.
cit. EM, Vol. IV, Annex 193; Declaration of Witness 8, 16 Jan. 2009 (hereinafter “Witness 8
DeclarationŽ). EM, Vol. IV, Annex 196; Declaration of Witness 10, 16 Jan. 2009 (hereinafter
“Witness 10 DeclarationŽ). EM, Vol. IV, Annex 198; Witness 28 Declaration, op. cit. EM, Vol.
IV, Annex 212.
610
CCM, Chap. 7, para. 7.82.
223restrictions and operational requirements in the Environmental Management Plan
to reach the conclusion that the fish-kills could not have been caused by the
spray. Significantly, he acknowledges that the spray mixture could cause fish-
kills either “through the direct toxicity of the surfactants in the spray mix or from
indirect effects due to oxygen depletion caused by biodegradation of dead plant
611
materialŽ . However, based on Colombia’s supposed strict adherence to
operational requirements and buffer zones, he concludes that it would be “highly
improbableŽ for the spray to have reached Ecuador and caused these effects 61.
As shown in Chapter 2, the presumptions on which the opinion of Colombia’s
expert are based are thoroughly negated by the flight data recorded by the spray
planes which Ecuador obtained from the U.S. Department of State. The opinion
remains interesting, however, for this reason: its acknowledgement that
Colombia’s spray mixture, if it reached the water bodies in or near Mataje where
the dead fish were observed, could have been responsible for killing them. The
flight data tell the rest of the story: spray drift into Mataje was inevitable given
the pervasive violations of all of the operational requirements by the spray pilots,
including deposition of huge volumes of spray near the river that runs
immediately next to Mataje.
611CCM, Appendix, p. 26, para. 100.
612Ibid.
224 D. E SMERALDAS 2007
3.98 After the impacts felt for the first time in September 2000, the residents of
Mataje were repeatedly revisited by Colombia’s spray planes depositing the spray
mixture along the border river, and by the matrix of ills that would immediately
follow. The greater Mataje area, including the Cayapas-Mataje Ecological
Reserve to the west and the Awá Indigenous Reserve to the east, was particularly
hard hit in 2004 and 2005, as shown by the flight data illustrated on Figure 3.6.
613
By the time aerial spraying in this area finally ended in 2007 , Colombia had
sprayed along the southernmost 10 kilometres of Nariño Province at least 28,638
times 61.
3.99 The last of these spraying campaigns, in early 2007, was particularly
intense and especially close to the Esmeraldas border … as can be appreciated
from the flight data in Figure 3.7. This spraying campaign was especially
troubling because Colombia had earlier promised that it would notify Ecuador in
613
Contrary to Colombia’s claims, the evidence shows that its fumigations along the Ecuadorian
border ended on or after 9 February 2007, not January 2007, as repeatedly presented in
Colombia’s Counter-Memorial, and not in January 2006, as recently claimed by the Colombian
government in its press release of 11 November 2010.Colombia chose not to provide any
evidence to support its assertions in the Counter-Memorial regarding the dates on which it
sprayed or suspended spraying. What the flight data show, and what the witness statements and
contemporaneous news reports further corroborate down to the day, is that Colombia continued to
spray in this area through at least 9 February 200See “More Refugees As A Result of
Fumigations Along the BorderŽ, E NIVERSO (Guayaquil, 8 Feb. 2007). ER, Vol. IV, Annex 81;
“Colombia Sprayed Within 1 km of the BorderŽ, EL UNIVERSO (Guayaquil, 10 Feb. 2007).
Annex 83.
614Hansman & Mena Report, Appendix 3, p. 28. ER, Vol. II, Annex 1.
225 1°30'0"N 1°15'0"N 1°0'0"N
Figure 3.6
0-1 k1mm-3 k3mm-7 k7m-10 km March, 2004J2a0n0u4ary, 20055, 2005
78°15'0"W spray line (km) km 78°15'0"W
Distance from nearest
Spray events (month)
Protected areasinternational boundary
Legend
River
C a r c h i
San Juan
0 5 10 15 20 25
78°30'0"W 78°30'0"W
±
! Awá Reserve
N a r i ñ o
Mataje Alto
C O L O M B I A
Guadualito
River
Mataje
!
Mataje Awá Reserve
78°45'0"W 78°45'0"W
(March 2004 - December 2005)
H
San Lorenzo E s m e r a l d a s
E C U A D O R
Cayapas-Mataje
Spray Events Within 10 Kilometres of Ecuador’s Esmeraldas Province
79°0'0"W Colombia Peru 79°0'0"W
Ecuador
1°30'0"N 1°15'0"N 1°0'0"N 1°30'0"N 1°15'0"N 1°0'0"N
Figure 3.7
0-1 k1m-m-3 k3m-5 k5m-7 k7m-1February, 2007
78°15'0"W spray line (km) km 78°15'0"W
Distance from nearest Spray events (month)
Protected areasinternational boundary
Legend
River
C a r c h i
San Juan 0 5 10 15 20 25
78°30'0"W 78°30'0"W
±
! Awá Reserve
N a r i ñ o
Mataje Alto
C O L O M B I A
River !
Guadualito
(February 2007) Mataje
!
Mataje
78°45'0"W Awá Reserve 78°45'0"W
H
San Lorenzo E s m e r a l d a s
E C U A D O R
Cayapas-Mataje
Spray Events Within 10 Kilometres of Ecuador’s Esmeraldas Province
79°0'0"W 79°0'0"W
Colombia Peru
Ecuador
1°30'0"N 1°15'0"N 1°0'0"Ncase of future sprayings, thereby allowing both States to put in place investigators
615
to determine whether the spray entered or harmed Ecuador . Regrettably, no
such advance notice was provided by Colombia.
3.100 Yet, on 8 February 2007, as reported in contemporaneous news articles in
El Universo, residents and Ecuadorian military officers watched five Colombian
spray planes accompanied by helicopters spraying within 1 kilometre of the
616
Mataje River . On 9 February 2007, after Colombia’s announcement that
spraying had been suspended, Ecuadorian residents watched again as four spray
planes and helicopters returned at 10:00 a.m., and continued to deposit the aerial
spray mixture, this time less than 1 kilometre from the border. The El Universo
article reports that local residents had witnessed this spraying campaign since the
previous week 617.
3.101 As happened many times before, immediately following exposure to the
spraying, area residents fell ill with the now-familiar symptoms. By 10 February
615
“Colombia Announces Ceasing of Fumigations to Ease Relations with QuitoŽ, E L
U NIVERSAL.COM (Caracas, 9 Feb. 2007). ER, Vol. IV, Annex 82.
616
“Colombia Sprayed Within 1 km of the BorderŽ, EL U NIVERSO (Guayaquil, 10 Feb. 2007).
ER, Vol. IV, Annex 83.
617
Ibid.
2262007, local children were sick with skin rashes covering their bodies, and eye
618
irritation .
3.102 The children, being the most vulnerable, were often the most hard-struck
by the spray mixture’s effects. Mataje Witness 33 describes how “many in the
community were affected. A lot of the children had diarrhea and vomiting,
including the children in my family. The adults were also sick but the children
were affected moreŽ 619.
3.103 The harms to human health were not the only injuries caused by the
spraying campaign. As in years past, in February 2007 Colombia’s aerial
sprayings posed serious risks to the environment on Ecuador’s side of the border.
Mataje is located near the Cayapas-Mataje Ecological Reserve, which was
620
established to protect the area’s mangrove forests . As described in Professor
Balslev’s expert report, the mangroves on Ecuador’s western coast “are the
largest in Ecuador and the only ones where the mangrove tree Pelliceria
rhizophorae can be foundŽ 62. These mangroves are particularly important for the
variety of ecosystem services they provide:
618
Ibid.
619
Witness 33 Declaration, op. cit. EM, Vol. IV, Annex 217.
620See Henrik Balslev, Ph.D., The Vulnerability of the Ecuador-Colombia Border Region to
Ecological Harm, p. 28 (Jan. 2011) (hereinafter “Balslev ReportŽ). EM, Vol. II, Annex 4.
621Ibid., p. 22.
227 “Their position in the tidal zone makes them important for many
species of marine life. The tidal zone placement produces a
salinity gradient and each zone has its own fish and invertebrate
fauna, so in a very limited space fishermen can find a wide range
of different species. Mangrove ecosystems are also important as
hatchment areas for larvae of a variety of marine organisms,
including shrimp and lobster. Many of these species, including
oysters, crab, lobster, shrimp and many types of fish are important
to local human diets. Mangroves also provide an important habitat
for a variety of bird species, many of which are residents of the
Cayapas▯Mataje mangrove protected area along the Colombian
622
borderŽ .
3.104 The local Esmeraldas fisherman are thus heavily reliant on the health of
this ecosystem. It was they who, soon after the February 2007 sprayings, raised
claims against Colombia for the harms caused to the mangroves and their sole
source of income. According to a contemporaneous press report, the President of
the Esmeraldas Fishermen’s Union expressed this concern about the effects of the
aerial sprayings on the mangroves: “The only natural laboratory we have in
Esmeraldas, which is the northern mangroves, and which is key to breeding and
maintaining the ecosystem, is being seriously affected and consequently fishing
623
will decrease in a very short timeŽ . He explained that the timing of the
sprayings was particularly troubling because they occurred during red snapper
622Ibid.
623
“Fishermen in Esmeraldas Fear Spraying with Glyphosate Affects MangrovesLU NIVERSO
(Guayaquil, 20 Feb. 2007). ER, Vol. IV, Annex 84.
228fishing season, and “the poison just alienates all coastal species and kills the
624
larvae, as it is in this area that females lay their eggs Ž .
3.105 The largely Afro-Ecuadorian communities in Mataje and other parts of
Esmeraldas Province are not the only ones impacted adversely by Colombia’s
aerial sprayings along and near the border. Also affected are the Awá indigenous
people. The UN Special Rapporteur on the Right of Everyone to the Enjoyment
of the Highest Attainable Standard of Physical and Mental Health observed that
the sprayings affected different groups: “the aerial spraying of glyphosate along
the northern border had to be seen in the context of the conditions of the people …
refugees, indigenous peoples, Afro-Ecuadorians . . . living on the northern
625
zoneŽ .
1. The Awá of Esmeraldas
3.106 The Awá are an indigenous group numbering only about 3,000 individuals
626
in Ecuador . As described in the report written by experts on the Awá
community of northern Ecuador, “[t]he Awá live mostly in very remote areas in
moist pristine forests on the western slopes of the Andes in the provinces of
624
Ibid.
625Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest
Attainable Standard of Physical and Mental Health, Paul Hunt: Preliminary Note on Mission to
Ecuador and Colombia, Addendum, U.N.Doc. A/HRC/7/11/Add.3, para. 21 (4 Mar. 2007). EM,
Vol. II, Annex 31.
626
Whitten et al. Report, op. cit., p. 45. ER, Vol. II, Annex 5.
229Carchi, Imbabura and EsmeraldasŽ 627. Most are concentrated in the Awá
Indigenous Forest Reserve located east of Mataje along the Esmeraldas/Carchi
border adjoining Colombia’s Nariño province, shown in Figures 2.14, 3.6 and 3.7
above. Anthropologists familiar with this group explain that “[t]his is not a
628
population of ‘several generations,’ but of millenniaŽ . “Contemporary Awá
use a system of horticulture (agriculture) practices that includes maize cultivation
629
that dates to at least four thousand years agoŽ . Their mode of living and
relationship to the environment around them is thus deeply ingrained. It is not
just that they are intimately connected to the land; it is that they are intimately
connected with this land.
3.107 The environment they inhabit, and where they have lived for millennia, is
one of the most biodiverse places in the world. It is a rare “biodiversity hotspotŽ
because of the exceptional concentrations of unique species that exist nowhere
else on earth 630. In fact: “The best preserved parts of this forest are those along
the Ecuador▯Colombia borderŽ 63. This area is not only notable for the large
627
Ibid.
628
Ibid., p. 47.
629
Ibid.
63Balslev Report, pp. 19-22. ER, Vol. II, Annex 4. The endemism is particularly high in plant
species, where 25 percent or 2,750 plant species found in the hotspot occur nowhere else in the
world. There are close to 900 species of birds in the hotspot, 110 of which are endemic.
Amphibian diversity is also very high, with 200 different species including 30 endemics, such as
the famous poison dart frog.
631Balslev Report, op. cit., p. 19. ER, Vol. II, Annex 4.
230number of endemic species found there but also for the number of them that are
632
“highly threatenedŽ .
3.108 Given the depth of the connection of the Awá to this area, any significant
change to its unique environment has wide-spread consequences for them. Such
has been the result of Colombia’s aerial sprayings adjacent to this habitat. As
expressed by Witness 40, an Awá from the Mataje Alto village situated in the
Reserve: “The sprayings came and broke our connections with our earth and our
633
way of livingŽ .
3.109 As indicated in Chapter 2, the flight data obtained by Ecuador from the
U.S. State Department show that Colombia sprayed within 10 kilometres of the
Awá Reserve more than 10,900 times between 2000 and 2008, and within a mere
634
2 kilometres at least 57 times during this period . Colombia began spraying in
635
the area of Nariño Province bordering the Awá Reserve in late 2000 .
Thereafter, the sprayings increased in intensity and also in proximity to the Awá
Reserve, reaching their peak in 2005. As shown in Figure 3.6 above, in that year
Colombia blanketed the border area with its chemical spray mixture. The data
show a series of spray lines skimming the Mataje River marking the border
632Ibid.
633
Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223.
634
Hansman & Mena Report, op. cit., Appendix 3, p. 14. ER, Vol. II, Annex 1.
635See supra Figure 3.4.
231between Colombia and the Awá Reserve, and coming particularly close to the
village of Mataje Alto, the home of Witness 40, quoted above.
3.110 The witness statements, NGO reports, flight data and report of the UN
Special Rapporteur on Indigenous Peoples all corroborate the harm caused to
indigenous people, plants and animals exposed to the spray in the north-western
area of the protected reserve. As stated in an observation mission report from
November 2005, the Awá complained that as a result of the spraying “the animals
have decreased, the leaves have dried up. The produce turns hard, the maize dries
636
up leaving only the cob. There are no fish anymoreŽ .
3.111 Within days of the spraying, Awá children arrived at the health post “sick
637
with diarrhea, vomiting, high fever, and stomach acheŽ . Soon after, adults
638
followed with the same set of symptoms, as well as skin rashes . These were
unlike any other symptoms previously experienced by the local Awá … they
639
occurred for the first time following the first spraying in the area . Since it has
636Interamerican Association for Environmental Defense et al., Ecolex and AIDA Environmental
Report on the Impacts of the Fumigations under Plan Colombia, p. 5 (Nov. 2005). EM, Vol. IV,
Annex 170.
637Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223.
638Ibid.
639Ibid.
232nothing better to say, Colombia again complains about lack of contemporaneous
medical records 640.
3.112 Ecuador agrees it would be nice if such records existed … and much nicer
still if there were physicians or nurses accessible to the Awá who might have
prepared them. But the reality is that from Mataje Alto it takes five hours on foot
and another 1.5 hours by car to reach the town with the nearest hospital 64. Thus,
when ill Awá usually self-medicate using medicinal-plants or go to the traditional
642
healer, the shaman . The shaman does not keep a doctor’s notebook while
performing his healing rituals. If the patient remains uncured, on the best of days
643
the health promoter can provide the scant treatment that is available . At other
times, as recounted by Witness 40, he is over-run and over-whelmed with
patients, unable to attend to all their needs, let alone keep a medical log he does
not have 64.
3.113 The death of plants traditionally gathered by the Awá as a principal food
source also followed closely upon Colombia’s aerial sprayings. Witness 40
640CCM, Chap. 7, paras. 7.141-7.142.
641Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223. See also Whitten et al. Report, pp.
45, 49. ER, Vol. II, Annex 5.
642Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223; Whitten et al. Report, pp. 48-49.
ER, Vol. II, Annex 5.
643See, e.g., Witness 41 Declaration. EM, Vol. IV, Annex 224.
644Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223.
233explains that “from the first time they sprayed, our food supply was affectedŽ 645.
The statement by Awá Witness 41 describes in more detail how “[a]fter the
spraying, all the crops began to dry up. They turned yellow two or three days
after the sprayings. The leaves of the sugarcane became withered and they fell
off. The yucca leaves also withered, and the maize completely dried within a
week after the sprayingŽ 64. As in Mataje and elsewhere, immediately following
the spraying, the fish in the river were found to be affected. Witness 40 describes
the “bumpsŽ on the fish’s skin, like “blistersŽ, their unusually pale eyes and the
647
skin’s strange change to a “reddishŽ color . The witness statements and the
report of the UN Special Rapporteur on the Rights of Indigenous Peoples also
document the death and disappearance of wild animals normally found in the
Reserve 648.
3.114 Tragically, despite their ancient cultural connection to the land of their
ancestors, the Awá have been forced to choose between remaining in their homes
and enduring the consequences of future sprayings by Colombia, or abandoning
their traditional lands for greater safety. The evidence shows that the sprayings
645Ibid.
646Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224.
647
Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223.
648
Report of the Special Rapporteur on the Situation of Human Rights and Fundamental
Freedoms of Indigenous People, Rodolfo Stavenhagen: Mission to Ecuador (25 April-4 May
2006), U.N. Doc. A/HRC/4/32/Add.2, para. 30 (28 Dec. 2006). EM, Vol. II, Annex 30; Witness
41 Declaration, op. cit. EM, Vol. IV, Annex 224.
234have caused some Awá … like the Cofán and the Kichwa, as described previously
… to choose the latter, leaving a culture and millennia of history behind. The UN
Special Rapporteur on the Rights of Indigenous Peoples reports that following
Colombia’s sprayings the entire Sumac Pamba Awá community abandoned their
649
ancestral village . As described by Witness 40:
“as a result of the damages to health, nature and our sources of
food and spirituality, some people had to move to other Awá
communities within the reserve, which were farther from the
border and not affected by the sprayings. They made this decision
in order to avoid the health problems caused by the fumigations
and the death of their crops, because they no longer had the means
to surviveŽ650.
Section II. The Evidentiary Value of Witness Statements
3.115 A main feature of the Counter-Memorial’s effort to undermine Ecuador’s
witness statements is its argument that, under the Court’s jurisprudence, they
should be disregarded. In particular, after citing the Court’s Judgment in
Territorial and Maritime Dispute between Nicaragua and Honduras in the
Caribbean Sea (Nicaragua v. Honduras), Colombia asserts: “Unless
independently corroborated, [witness statements] are entitled to no weight;
651
notably insofar as they purport to express any opinion as to causationŽ . This
649Report of the Special Rapporteur on the Situation of Human Rights and Fundamental
Freedoms of Indigenous People, Rodolfo Stavenhagen: Mission to Ecuador (25 April-4 May
2006), U.N. Doc. A/HRC/4/32/Add.2, para. 30 (28 Dec. 2006). EM, Vol. II, Annex 30.
650Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223.
651CCM, Chap. 7, para. 7.127.
235aspect of the Counter-Memorial’s argument both defeats itself and is wrong as a
matter of law.
3.116 Colombia’s argument defeats itself because, as demonstrated in Section I
above, Ecuador’s eyewitness accounts are, in fact, “independently corroboratedŽ,
not least by the spray flight data that has recently come into Ecuador’s
possession, as well as by contemporaneous observation mission reports,
newspaper articles, the reports of various UN Special Rapporteurs,
contemporaneous medical inquests, the scientific literature on the known effects
of glyphosate-based herbicides, health warnings on product labels, and official
reports of governmental agencies in third States, inter alia.
3.117 And Colombia is wrong as a matter of law because the Court has never
652
said that witness statements should be accorded “no weightŽ . In making this
claim Colombia has disregarded the historic practice of the Court. Since as early
as the Corfu Channel case, the Court has admitted sworn statements as
653
evidence . Indeed, the Court noted that it “gave much attention to this
652Ibid.
653Corfu Channel (United Kingdom v. Albania), Judgment, I.C.J. Reports 1949, p. 19.
236evidenceƒŽ in the Corfu Channel decision 654. Since then, witness statements
have regularly been accepted as sources of factual evidence 655.
3.118 Consistent with this approach, what the Court actually said in Nicaragua
v. Honduras was that “the Court will not find it inappropriate as such to receive
affidavits produced for purposes of litigation if they attest to personal knowledge
of facts by a particular individualŽ 656. The Court explained that such affidavits
may be treated with a degree of “cautionŽ, depending on a number of specific
factors, including: (i) the affidavit attests to facts or only offers an opinion; (ii)
the witness’ “capacity to attest to certain factsŽ; (iii) “the utility of what is saidŽ;
(iv) when the affidavits were made; and (v) whether the affiant has an interest in
657
the outcome of the proceedings . Balancing these factors in the circumstances
of this case dictates that substantial weight be given to the witness statements
attached to Ecuador’s Memorial.
654
Ibid., p. 16. (referring to written and verbal witness statements presented by the United
Kingdom).
655
See e.g., Military and Paramilitary Activities (Nicaragua v. United States), Judgment, I.C.J.
Reports 1986, p. 42, para. 72; Armed Activities on the Territory of the Congo (Democratic
Republic of the Congo v. Uganda), Judgment, I.C.J. Reports 2005, p. 219, para. 129;
Guyana/Suriname, Arbitral Award, pp. 141-144, paras. 432-439 (2007).
656Territorial and Maritime Dispute between Nicaragua and Honduras in the Caribbean Sea
(Nicaragua v. Honduras), Judgment, I.C.J. Reports 2007, p. 731, para. 244. See also Military and
Paramilitary Activities (Nicaragua v. United States), Judgment, I.C.J. Reports 1986, p. 42, para.
68.
657
Territorial and Maritime Dispute between Nicaragua and Honduras in the Caribbean Sea
(Nicaragua v. Honduras), Judgment, I.C.J. Reports 2007, p. 731, para. 244.
2373.119 Ecuador offers the statements for the truth of the facts stated, not any
incidental opinions they might contain. To a person, the witnesses’ statements
are devoted to recounting historical facts within their personal knowledge and
experience; something they plainly have the capacity to do. In fact, there are
literally no other people in the world who are in a better position to speak to the
matters in dispute in this case; they are the ones who directly observed and
experienced the impacts of the spray that drifted across the border into Ecuador.
3.120 In this respect, the context of this case should not be forgotten. Colombia
was conducting massive aerial spraying operations in one of the more remote
corners of the planet, sparsely inhabited only by isolated communities of
impoverished peasants and indigenous peoples largely cut off from centres of
communication and commerce. But for their complaints calling attention to the
harms inflicted on them, the truth of what happened might never have come out.
Certainly Colombia … which still hides the full contents and formula of the spray
mixture, and still keeps confidential the dates and precise location of its spraying
events … would not have been forthcoming.
3.121 The testimonies are therefore highly useful. Most offer detailed accounts
of what transpired when the sprayings began, and include very specific statements
as to where they were, what they were doing and the consequences that followed.
They are not summary assertions of ultimate conclusions.
2383.122 Nonetheless, Ecuador does agree with Colombia in one respect. If,
instead of offering 37 witness statements (plus 10 more from eyewitnesses in
Colombia), Ecuador had offered only one, or even just a handful, the weight to be
accorded them would have to be assessed differently. They then might plausibly
be treated as a few isolated allegations notwithstanding the specificity with which
each is made. But that is not the case; it is not a question of each statement
standing alone. Rather, they stand together both with each other and with all the
other elements of proof that Ecuador has presented (now including the spray
flight data obtained from the United States) to form a consistent, coherent and
mutually reinforcing whole that is entitled to substantial weight.
3.123 In Ecuador’s view, it is this consistency that makes the statements so
remarkable. They describe phenomena that are similar in all material respects.
Their descriptions of the sprayings themselves, for example, are remarkably
consistent though, of course, never precisely the same. Witness 37 from Mataje
near the Pacific Coast, describes seeing the spray planes for the first time as
follows: “The first time was in the year two thousand. I was working on my
farm, at the edge of the river. I saw several planes and some helicopters coming
from the Colombian side, dropping a liquid. The liquid looked like smoke and it
658
fell on the ground and on my body, it looked shinyŽ . More than 250 kilometres
away in Sucumbíos, Witness 2 from Salinas describes the spraying that he
658Witness 37 Declaration, op. cit. EM, Vol. IV, Annex 220.
239observed thusly: “I could observe maybe four planes and some helicopters
passing by the San Miguel River and, when turning around, they would fly over
Ecuadorian territory. The planes were flying, dropping a white liquid that with
the wind came quickly toward usŽ 659.
3.124 The witnesses were equally consistent in their portrayal of the spray mist
itself. Invariably, it was described as appearing “whiteŽ, “like smokeŽ or “a
cloudŽ 660. Many specifically reported seeing it drift across the border and
watching it land in Ecuadorian territory, including directly on them. It was
uniformly described as “foul-smellingŽ, and looking “greasyŽ or “like a light
oilŽ661. Some even gave minute descriptions of the spray droplets that can only
have come from direct experience. As Witness 33 from Mataje described the
659
Declaration of Witness 2, 16 Jan. 2009 (hereinafter “Witness 2 DeclarationŽ). EM, Vol. IV,
Annex 190.
660
Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 3 Declaration, op. cit. EM,
Vol. IV, Annex 191; Witness 5 Declaration, op. cit. EM, Vol. IV, Annex 193; Witness 8
Declaration, op. cit. EM, Vol. IV, Annex 196; Witness 9 Declaration, op. cit. EM, Vol. IV,
Annex 197; Witness 12 Declaration, op. cit. EM, Vol. IV, Annex 200; Witness 13 Declaration,
op. cit. EM, Vol. IV, Annex 201; Declaration of Witness 14, 17 Jan. 2009 (hereinafter “Witness
14 DeclarationŽ). EM, Vol. IV, Annex 202; Witness 17 Declaration, op. cit. EM, Vol. IV, Annex
203; Witness 20 Declaration, op. cit. EM, Vol. IV, Annex 206; Witness 22 Declaration, op. cit.
EM, Vol. IV, Annex 208; Witness 23 Declaration, op. cit. EM, Vol. IV, Annex 209; Witness 26
Declaration, op. cit. EM, Vol. IV, Annex 210; Witness 27 Declaration, op. cit. EM, Vol. IV,
Annex 211; Witness 28 Declaration, op. cit. EM, Vol. IV, Annex 212; Witness 29 Declaration,
op. cit. EM, Vol. IV, Annex 213; Witness 31 Declaration, op. cit. EM, Vol. IV, Annex 215;
Witness 32 Declaration, op. cit. EM, Vol. IV, Annex 216; Witness 33 Declaration, op. cit. EM,
Vol. IV, Annex 217; Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218; Witness 36
Declaration, op. cit. EM, Vol. IV, Annex 219; Witness 38 Declaration, op. cit. EM, Vol. IV,
Annex 221; Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222.
661Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 23 Declaration, op. cit. EM,
Vol. IV, Annex 209; Witness 32 Declaration, op. cit. EM, Vol. IV, Annex 216; Witness 3
Declaration, op. cit. EM, Vol. IV, Annex 191.
240spraying, “I saw them releasing something that looked like a cloud, but when it
662
fell on the grass it was shiny, oily and it stayed on the plantsŽ .
3.125 The witnesses’ descriptions of what happened after the sprayings are also
mutually corroborating. The description of the specific symptoms of skin
irritation … the burning itch, the bumps, the pus upon scratching … are remarkably
consistent despite the witnesses’ distance from each other and relative isolation.
Witness 31, a Cofán from a remote village in the Cofán-Bermejo Reserve stated,
“[i]t was there when the smoke also fell on the clothes and that continued to
affect our skin. That lasted for about two weeks, first we had small bumps and
663
then a week later they burst. The bumps itched a lotŽ . Across the country in
Mataje, Witness 34 testified that “the liquid also fell on me, on my head, arms,
and all over my body. Immediately, I felt my skin itch intensely. My whole body
was itching. Above all, my face became very swollen . . . I also got bumps all
664
over my skin. I had a rash that burned and my skin peeled quite a bitŽ . The
witness statements are also consistent with the known reactions to the chemicals
665
in the spray mixture . The flight path data further affirm that the villages in
which these witnesses resided were the same villages exposed to Colombia’s
662Witness 33 Declaration, op. cit. EM, Vol. IV, Annex 217.
663
Witness 31 Declaration, op. cit. EM, Vol. IV, Annex 215.
664
Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218.
665See supra Chap. 3, para. 3.25; EM pp. 132-152.
241spraying, and at the same time that the witnesses say they observed the spray
planes in operation.
3.126 The witness testimonies are equally consistent in their descriptions of the
effects the sprayings had on plants. They invariably described how each different
variety of crops they had planted showed the same signs of damage following the
sighting of the spray planes and the deposition of the spray mixture. The
witnesses provide similar details of the plants becoming yellow, often starting
666
with the leaves, until completely wilted . Upon opening the crop’s fruits, the
667 668
insides were found to be rotten . There were also reductions in crop yields .
666
Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 2 Declaration, op. cit. EM,
Vol. IV, Annex 190; Witness 5 Declaration, op. cit. EM, Vol. IV, Annex 193; Declaration of
Witness 6, 16 Jan. 2009 (hereinafter “Witness 6 DeclarationŽ). EM, Vol. IV, Annex 194;
Declaration of Witness 7, 16 Jan. 2009. EM, Vol. IV, Annex 195; Witness 8 Declaration, op. cit.
EM, Vol. IV, Annex 196; Witness 9 Declaration, op. cit. EM, Vol. IV, Annex 197; Witness 12
Declaration, op. cit. EM, Vol. IV, Annex 200; Witness 17 Declaration, op. cit. EM, Vol. IV,
Annex 203; Witness 22 Declaration, op. cit. EM, Vol. IV, Annex 208; Witness 23 Declaration,
op. cit. EM, Vol. IV, Annex 209; Witness 30 Declaration, op. cit. EM, Vol. IV, Annex 214;
Witness 32 Declaration, op. cit. EM, Vol. IV, Annex 216; Witness 37 Declaration, op. cit. EM,
Vol. IV, Annex 220.
667
Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 4 Declaration, op. cit. EM,
Vol. IV, Annex 192; Witness 8 Declaration, op. cit. EM, Vol. IV, Annex 196; Witness 13
Declaration, op. cit. EM, Vol. IV, Annex 201; Witness 36 Declaration. EM, Vol. IV, Annex 219.
668
Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 4 Declaration, op. cit. EM,
Vol. IV, Annex 192; Witness 6 Declaration, op. cit. EM, Vol. IV, Annex 194; Witness 8
Declaration, op. cit. EM, Vol. IV, Annex 196; Witness 12 Declaration, op. cit. EM, Vol. IV,
Annex 200; Witness 17 Declaration, op. cit. EM, Vol. IV, Annex 203; Declaration of Witness 19,
17 Jan. 2009 (hereinafter “Witness 19 DeclarationŽ). EM, Vol. IV, Annex 205.
2423.127 In addition to the mutually corroborating details of the signs of damage
shown by the plants, what is notable is that all plants were indiscriminately
affected. As described by Witness 30 of Mataje, Esmeraldas:
“After the sprayings, my crops were affected. Eight days after the
sprayings, the leaves on the cacao tree started falling off, until not
a single leaf was left; the tree dried up completely and it died. The
same thing happened with the yucca. Within three days, the leaves
fell off the yucca and even the root rotted; the root smelled. The
guineo also dried up, the leaves were drying up and withering.
The plants turned yellow. After a few weeks, everything was
669
dead. All the plants were dead on the groundŽ .
3.128 In the remote reaches of the protected Awá Indigenous and Forest
Reserve, the symptoms were the same following the appearance of the spray
planes. Witness 40 testifies:
“It was probably five days later that some hectares of the natural
forest, near the Mataje River, died. Three days later the plants
began to dry up and fall off, as if they were burned. The leaves
fell off the plants and all the branches died. All the plants, big and
small, were destroyed. Several species of wild plants that were in
that hectare died. I estimate that at least some thirty species of
plants that died were used by us in the Awá traditional medical
670
treatmentsŽ .
3.129 The damage suffered by numerous species is consistent with exposure to a
671
broad spectrum herbicide, not to plant disease or insect infestation .
669Witness 30 Declaration, op. cit. EM, Vol. IV, Annex 214.
670Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223.
671
See Weller Report, op. cit., p. 3. ER, Vol. II, Annex 3.
2433.130 Ecuador observes further that many of the witnesses offer the same or
similar esoteric details that further highlight the credibility of their accounts. A
number, for instance, recount that the yellowing of the plants in many cases
began at the top of the plant and worked its way down to the ground. After
describing the effect on his crops, Witness 18 from San Francisco 1, states:
“I had never seen this type of disease before. When bugs attack
the plants, they do it from the root towards the top. In this case,
the plants were dying from top to bottom. Besides, I had seen on
some occasions in the past that when the plants get sick, only one
species is attacked, without affecting other plants. But, during
those 672s, all the plants were affected, from pasture to fruit
treesŽ .
3.131 In a similar way, Witness 1 of Salinas recounts that:
“the tallest fruit trees . . . were the first to dry up at the top. They
did not die completely although they did dry up, and no longer
produced fruit. The plantain trees were also destroyed quickly.
The plantain, planted next to my house, which is a few meters
from the river, died first. The plant was undernourished, falling to
one side and the fruit started to dieŽ673.
3.132 Still others are frank in admitting that as bad as the damage was in
Ecuador, it was even worse on the Colombian side of the border, precisely as one
would expect because of its closer proximity to the spray target. Witness 10, a
Colombian resident who had earlier moved to Sucumbíos testifies, for instance,
that “[i]n San Miguel and Dios Peña, one can see the same effects from the
672Witness 18 Declaration, op. cit. EM, Vol. IV, Annex 204.
673Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189.
244fumigations; after the liquid is left in the air, plants, animals and people get sick.
674
In Colombia the effect is the same as in Ecuador but a little more excessiveŽ .
Further to the west in Salinas, Sucumbíos, Witness 1 echoes this testimony: “from
my house, one can see the river and Colombia. On the other side, I noticed that
the trees were yellow, dry, and dead. It was very similar to what had happened to
my crops, it looked like a trail of destruction; although, the Colombian side was
slightly more severeŽ 675.
3.133 Unable to rely on their crops to feed themselves, the residents had to buy
their food. By afflicting their domestic animals too, however, the sprayings left
them with less money to do so. As explained by anthropologists familiar with the
region, the border residents often use animals as a form of “bank accountŽ, in
676
which they invest their earnings and then sell when money is needed . The
witness statements provide accounts with mutually enforcing details of the
illnesses that befell their animals and the consequences to their families. In
Salinas, Witness 2 testified: “In the following years, they sprayed again and we
lost what little we had over again. History repeated itself: children became sick
again and the animals lost their hair and died. Fifty percent of my chickens died,
674
Witness 10 Declaration, op. cit. EM, Vol. IV, Annex 198.
675Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189. See also Witness 19 Declaration, op.
cit. EM, Vol. IV, Annex 205. “It was incredible. On the Colombian side, the land was a desert,
where it used be to full of forest. Everything, but everything, was dead . . .Ž
676
See Whitten et al. Report, op. cit., p. 12. ER, Vol. II, Annex 5.
245 677
the same with the fishŽ . Notably, many of the witnesses, including residents
from the villages of Puerto Mestanza, Mataje, Mataje Alto of the Awá Reserve,
and Salinas share their recollections of the spray’s particularly deadly effect on
fish678.
3.134 The indigenous witness statements corroborate these observations, with a
particular emphasis on the spray’s effects on wild animal species. As recounted
by a Cofán resident of the Cofán-Bermejo Reserve: “The chickens that I had
would vomit everything they ate, shake and then die, now I do not have many
chickens. We also saw many of the jungle birds become stiff and fall dead to the
679
ground, we saw this about four days after the sprayingŽ .
3.135 Ecuador submits that the coherence and consistency of the witness
statements it has offered are particularly probative in the circumstances of this
case. As stated, both Parties agree that Ecuador’s border regions are among the
least developed areas of the country. Their chronic lack of basic infrastructure,
including transportation and communication, has made them, in Colombia’s
677Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 190.
678See, e.g., Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 190; Witness 10 Declaration, op.
cit. EM, Vol. IV, Annex 198; Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222; Witness
40 Declaration, op. cit. EM, Vol. IV, Annex 223.
679Witness 31 Declaration, op. cit. EM, Vol. IV, Annex 215.
246words, “virtually isolated from the rest of the countryŽ 680. Exactly right. As
Ecuador previously observed in its Memorial: “Roads are usually no more than
hardened dirt paths and, where it exists at all, public transportation (via an
occasional bus) is scarce and infrequent. Communication with the outside world,
and even other villages in the region, is generally limited to periodic radio
681
contactŽ . It is precisely for this reason that the parallels among the witness
accounts … from diverse individuals of indigenous, Afro-Ecuadorian and mestizo
origin, spread out along the border and physically cut-off from one another … are
so remarkable.
3.136 Whatever their origin or wherever they live, be it in Esmeraldas, the
scattered outposts along the Sucumbíos-Putumayo border or the indigenous
reserves in the region, the Ecuadorian witnesses offer consistent accounts of the
effects Colombia’s sprayings have had on their health, their crops, their animals
and the wild flora and fauna. Colombia would like the Court to believe that these
similarities represent a collective delusion, or even a mass conspiracy. But the
truth is that achieving the coordination necessary to produce such compelling
commonality is quite literally impossible in the remote, impoverished frontier
regions. In the end, the only plausible explanation is the simplest: the witnesses
680CCM, Chap. 2, para. 2.13; see also CCM, Chap. 2, para. 2.15 (stating that “their present
difficulties are a continuation of long-term isolation . . .Ž).
681
EM, Chap. 2, para. 2.24.
247are accurately recounting that the spray drifted into Ecuador, and that it impacted
them and their surroundings in the ways they have described.
3.137 Throughout the Counter-Memorial, Colombia intimates that the region’s
remoteness and poverty make it impossible to tease out the harms the sprayings
have caused. Things were already so bad, Colombia suggests, that it is more
plausible to believe that the harms the witnesses identify represent a natural
outcome in these already poor conditions than that they are the effects of its
682
sprayings . Colombia’s argument in this respect ties into its broader argument
that, to the extent they express an opinion as to causation, the witness statements
683
presented with the Memorial are entitled to “no weightŽ .
3.138 As a matter of law, Ecuador certainly agrees that the witnesses are not
qualified as experts to offer a scientific opinion on the question of causation. But
Colombia’s argument misses a key point. While the witnesses’ opinions on
causation, as such, may not constitute proof in and of themselves 684, their
statements of fact constitute evidence from which conclusions about causation
682
See CCM, Chap. 7, para. 7.37 (“[I]t is impossible to tell whether the ailments complained of …
in particular gastrointestinal disorders such as vomiting and diarrhea … resulted from the
sprayings, or whether they were due to other causes which are common among poorly nourished
populations living in precarious hygienic conditionsŽ.).
683CCM, Chap. 7, para. 7.127.
684
Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of
America), Judgment, I.C.J. Reports 1986, para. 68 (“An opinion expressed by a witness is a mere
personal and subjective evaluation of a possibility . . . it may, in conjunction with other material,
assist the Court in determining a question of fact, but is not proof in itselfŽ.).
248may be drawn, by experts and most importantly by the Court. In Ecuador’s view,
the witness statements offer convincing factual evidence relevant to the issue of
causation. To a person, the witnesses testify that before the sprayings one state of
affairs existed and after the sprayings, in close proximity to them, and after
observing the spray fall inside Ecuador, a different state of affairs came into
being. The change they all speak of coincided precisely with the advent of
Colombia’s aerial sprayings, and precisely with the known effects of glyphosate
and POEA. This is compelling evidence from which conclusions as to cause and
effect may be drawn by the Court.
3.139 Just two examples will suffice for present purposes. Witness 4 from
Salinas, Sucumbíos states:
“In a short period of time, they sprayed for several days, on our
community and neighbouring communities. Usually, they
fumigated during the day and on clear days, and not when it was
rainy. On my farm I had planted about twelve hectares of pasture
land, plantain, yucca, coffee, and cacao. The spraying completely
ruined all of it. A few days after the spraying, the plants started to
turn yellow and then they turned black and died. I had never
experienced anything like that. I tried to save the crop with
fertilizers but it did not work, and we lost everything . . . . Before
the fumigations, a hectare of coffee would yield sixty quintals, and
a hectare of maize would yield forty quintals. Now, the coffee
yields about five quintals per hectare, and the maize about two
quintals. Never before, not even in the cas685f a drought or in the
rainy season, had the land yielded so littleŽ .
685
Witness 4 Declaration, op. cit. EM, Vol. IV, Annex 192.
249In a similar vein, Witness 41, an Awá resident of the Reserve in Esmeraldas
declares:
“Before the spraying, we were healthy. But after the spraying,
many people in my community became sick. Some of the people
in the community had bumps all over their bodies. It was strange,
I had not seen that before . . . . Not only were the people affected
but the animals were too. After the spraying, many of the chickens
that we raised became sick, they would not walk but rather
remained seated, and within a few days some of them died. The
chicks either remained small or did not survive . . . . After the
sprayings, all the crops began to dry up. They turned yellow two
or three days after the sprayings. The leaves of the sugarcane
became withered and they fell off. The yucca leaves also
withered, and the maize completely died within a week after the
sprayings. The community used to live off the crops that grew in
our land, but after the sprayings we lost several crops such as
maizeŽ 686.
3.140 Ecuador considers this particularly probative factual evidence of causation
in the circumstances of this case. As stated, the subsistence farmers and
indigenous peoples alike are tied to the land and the rhythms of nature. Many
testify that they have lived on the land in the same location for their entire lives.
Indeed, for indigenous populations, the tie with “mother earthŽ is a central
687
component of their culture . As such, they are finely attuned to even minor
686
Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224. See also Witness 3 Declaration, op.
cit. EM, Vol. IV, Annex 191.
687
See Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223; Declaration of María Blanca
Chancosa Sánchez, 14 Jan. 2009. EM, Vol. IV, Annex 187; Whitten et al. Report. ER, Vol. II,
Annex 5. (e.g., p. 20, “ƒCofán people do not believe that they would be able to maintain their
culture and identity without residing in their traditional territory’s mountain and lowland
ecosystems. In their native language of A’ingae, Cofán call themselves tsampini can’jensundeccu
(dwellers of the forest). In their political discourse, Cofán leaders proclaim, ‘Without our Forest,
which has been the one constant throughout our history, we are no longer Cofán.’ Cofán language,
250disruptions in the environment. They are thus uniquely well-qualified to speak to
the health of the environment over time and how the changes they observed
coincide with the introduction of noxious external elements. Notably,
international tribunals have not shied away from assigning witness testimony
from indigenous populations significant probative value based on their intimate
688
relationship with the subject-matter of their testimony . As the distinguished
arbitral tribunal noted in the Abyei arbitration:
“One other potential source of evidence is witness testimony. For
its part, the [Government of Sudan] has criticized the reliability of
witness evidence. This Tribunal agrees that where the witnesses
rely on knowledge passed down through one or two generations,
the precise dating of the evidence which they supply may
sometimes be difficult. Nevertheless, depriving witness evidence
per se of all probative value would be unjustifiable. When
defining the historic area of a tribe, an inherently difficult exercise,
it is reasonable, and indeed quite logical, to seek information from
the tribe members themselvesŽ 689.
3.141 Colombia appears to take issue with demonstrating causation, at least in
part, in this manner. Instead, Colombia contends that if it has not been measured
and observed in a laboratory, it does not count. This is not only a new rule of
evidence, invented by Colombia for this case; it is an entirely inappropriate one,
cosmology, social life, healing practices, and subsistence patterns interweave profoundly with the
Amazonian environmentŽ.).
688
The Government of Sudan/The Sudan People's Liberation Movement/Army (Abyei Arbitration),
Arbitral Award, pp. 247, 256, paras. 717, 742 (22 July 2009); Case of the Indigenous Community
Yakye Axa v. Paraguay, Judgment, IACHR, Series C No. 125, para. 201 (17 June 2005).
689
The Government of Sudan/The Sudan People's Liberation Movement/Army (Abyei Arbitration),
Arbitral Award, pp. 247, para. 717 (22 July 2009) (emphasis added).
251especially in this case. Indeed, it is ironic in the extreme for Colombia to argue
that only scientific data can be considered: to this day, Colombia has never
truthfully acknowledged what is … or was … in the spray mixture, particularly in
the period when the evidence shows the chemicals were even more toxic than in
690
later years . How can it be demonstrated scientifically which unidentified
substances caused what harms? No litigating State can be expected to hit an
invisible (indeed, hidden) target. Moreover, Colombia never gave Ecuador
advance … or even after-the-fact … notice of the dates when and locations where
sprayings were carried out. Ecuador was thus never in a position to have
scientific personnel on-site ready to collect spray samples as they wafted over the
border and settled on people, plants, animals, water bodies and the ground.
3.142 Colombia’s argument is also legally incorrect. Notably, the Counter-
Memorial cites no authority for the proposition that only scientific evidence
counts … because there is no such authority. In this case, as in all cases, it is for
the Court to “[e]xamine all the facts relevant to each of the component elements
of the claims advanced by the Parties. In so doing, it will identify the documents
relied on and make its own clear assessment of their weight, reliability and
valueŽ 69. Thus, as the Court recently reaffirmed in the Pulp Mills case, “in
690
See supra Chap. 2, paras. 2.17-2.63.
691Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda),
Judgment, I.C.J. Reports 2005, pp. 200-201, para. 59. See also Military and Paramilitary
252keeping with its practice, the Court will make its own determination of the facts,
on the basis of the evidence presented to it, and then it will apply the relevant
692
rules of international law to those facts which it has found to have existedŽ .
The extent of scientific evidence must be weighed in connection with the record
as a whole. This is particularly true in the circumstances of the present case,
where obtaining corroborative physical evidence in the field is unusually difficult
because of the remoteness of and lack of resources in the areas involved, and the
rapidity with which glyphosate dissipates into soil or water.
3.143 In any event, as described above, there is abundant scientific evidence
linking Colombia’s aerial spraying to the specific harms to people, animals,
plants and the environment reported in the witness testimonies, and in
contemporaneous reports by the news media and NGOs who visited the affected
693
areas , not least of which is the fact that the internationally accepted drift model
predicts deposition of herbicide far into Ecuador in amounts that can cause
serious harm 694.
Activities (Nicaragua v. United States), Judgment, I.C.J. Reports 1986, p. 40, para. 60; Case
Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, p. 52, para. 168.
692
Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, p. 52,
para. 168.
693
See supra Chap. 3, paras. 3.25, 3.44. 3.46, 3.54, 3.89.
694See supra Chap. 2, paras. 2.198-2.203/
2533.144 Interestingly, Colombia’s otherwise strident Counter-Memorial is notably
timid on the subject of what else, other than its sprayings, may have caused the
harms universally described by the witnesses and corroborated by multiple
contemporaneous accounts and impartial investigations. Colombia identifies just
two other possible culprits: first, the deleterious effects of coca cultivation; and
second, environmental contamination caused by the exploration for and
exploitation of petroleum. Neither alternative Colombia identifies is a plausible
cause of the harms suffered by the witnesses.
3.145 Colombia seems particularly enamoured of its claim that the harms
identified may be the result of coca cultivation. It offers this as a possible
explanation repeatedly in both Chapters 1 and 7 of the Counter-Memorial. In
Chapter 1, it states: “Health problems in the border area may . . . have something
to do with the very reasons for the spray program, since the unlawful cultivation
of coca plants carries a serious risk of personal injury poisoning by much more
toxic chemicals and harm to the environmentŽ 695. And in Chapter 7 it repeats:
“Alternatively, [the ailments complained of] may be the result of the much more
toxic chemicals used in the cultivation and processing of coca in those areasŽ 696.
695CCM, Chap. 1, para. 1.38; see also ibid., para. 1.39.
696CCM, Chap. 7, para. 7.37.
2543.146 The trouble for Colombia is that, unlike Colombia, Ecuador does not have
a coca cultivation problem. According to the reports of the United Nations Office
on Drugs and Crime (“UNODCŽ), there is no significant cultivation of coca in the
697
northern border regions of Ecuador (or anywhere else in Ecuador) . This
absence of coca cultivation in Ecuador is reflected in the following graphic,
Figure 3.8, from the UNODC report entitled Coca Cultivation in the Andean
Region, a Survey of Bolivia, Colombia, Ecuador and Peru, published in June
2007 (shortly after Colombia stopped aerial spraying within 10 kilometres of the
border with Ecuador); the graphic itself is captioned “Coca cultivation density in
698
the Andean Region, 2006Ž . As the Court can see, unlike Colombia, there are
no concentrations of coca cultivation in Ecuador.
3.147 In 2010, the UNODC stated in its World Drug Report that “surveys
implemented by UNODC in cooperation with the Government of Ecuador in
2006 and 2008Ž covering the “provinces in the north of Ecuador bordering
ColombiaŽ, had “confirmed that the level of coca cultivation was
insignificantŽ699. Colombia offers no evidence to the contrary; it makes no
attempt at showing that there is any coca cultivation on Ecuador’s side of the
697United Nations Office on Drugs and Crime, World Drug Report 2010, p. 161 & n.1 (2010).
ER, Vol. IV, Annex 110.
698
United Nations Office on Drugs and Crime, Coca Cultivation in the Andean Region, A Survey
of Bolivia, Colombia, Ecuador and Peru, p. 2 (June 2007). ER, Vol. IV, Annex 108.
699
United Nations Office on Drugs and Crime, World Drug Report 2010, p. 161 & n.1 (2010).
ER, Vol. IV, Annex 110.
255 Coca Cultivation Density in the Andean Region, 2006
Regional Overview
Map 1: Coca cultivation density in the Andean Region, 2006
80°W 70°W Colombia
Caribbean Sea Peru
10°N Bolivia 10°N
Catatumbo
N A M A
PA
South America
Norte de VENEZUELA
Antioquia
Atrato
GUYANA
COLOMBIA
^ Meta
Bogota Vichada
Magdalena Guaviare
Meta -
Guaviare
Nariño
Putumayo -
0° 0°
Caquetata
ECUADOR ay o
Amazonas
BRAZIL
Marañon
Alto Ucayali
Huallaga
Huallaga
10°S Purus 10°S
PERU
Ene
UrubLa Convencion
Lima ^
Apurimac and Lares Madre de Dios Mamore
Ene Guapare
Pacific Beni
Apurimac
Ocean
Titicaca Yungas
Lake
Chapare
^
La Paz
BOLIVIA
Cultivation density
(ha/km )2
0.1 - 1.0
20°S 20°S
1.1 - 4.0 Pilcomayo
> 4.0 PARAGUAY
International 0 250 500
km
boundaries Geographic coordinates WGS 84
Department CHILE
boundaries ARGENTINA
80°W 70°W 60°W
Sources: National monitoring systems supported by UNODC - Governments of Bolivia, Colombia and Perú
The boundaries and names shown and the designations used on this map do not imply official endorsement or acceptance by the United Nations
Figure 3.8border. Colombia’s would-be alternative explanation for the harms experienced
in Ecuador is thus not viable.
3.148 Equally without basis is the Counter-Memorial’s suggestion that
hydrocarbon exploration and exploitation activities are to blame for the harms
described in the Memorial 70. There is no petroleum activity in Esmeraldas
Province, where harms following Colombia’s spraying campaigns have
repeatedly resulted. And there is certainly no oil exploration in the Cofán-
Bermejo Ecological Reserve, the Awá Indigenous Reserve or the Cayapas-Mataje
Ecological Reserve … all of which are documented by the evidence as having
suffered the same effects following Colombia’s aerial spraying in close proximity
to those locations. Insofar as other parts of Ecuador have suffered environmental
degradation associated with petroleum production in Sucumbíos, they are
generally remote from the areas of Sucumbíos affected by Colombia’s spraying.
3.149 Colombia provides not a shred of evidence to support its argument that the
harms suffered in Ecuador are attributable to some cause other than its aerial
spraying of toxic chemicals in close proximity to the border under operating
conditions guaranteed to produce spray drift into Ecuador. There is not a single
fact to show that the injuries resulted from the (non-existent) cultivation of coca
on the Ecuadorian side of the border; or from petroleum production far removed
700CCM, Chap. 2, para. 2.31.
256from the places where injuries occurred; or from malnutrition, poverty, lack of
infrastructure, presence of illegal armed bands, or any of the other potpourri of
hypothetical possibilities thrown out by Colombia to avoid responsibility for the
harms the evidence shows it has caused. To the extent that Colombia asserts that
other causes may have contributed to these harms, it bears the onus of providing
the necessary proof, and it has manifestly failed to do so.
3.150 In the end, despite Colombia’s efforts to rid itself of the evidence of harm
by any means available, the outcome remains the same. The witness statements
and numerous other sources of evidence of harm are more than admissible, they
are undeniable. The consistent corroboration of the harm inflicted by Colombia’s
spraying seen across the spectrum of witness statements, NGO and UN reports,
newspaper articles, and more, only strengthens their probative weight and furthers
the unavoidable conclusion that Colombia’s repeated spraying of chemical
herbicides along the border caused harm to the people, plants, and animals of
Ecuador.
Section III. The Evidence of Harm Caused in Colombia Corroborates the
Evidence of Harm Caused in Ecuador
3.151 The evidence of the harm that Colombia’s aerial sprayings have caused in
Ecuador is further corroborated by the evidence of the damage they have inflicted
257 701
in Colombia itself. As initially described in the Memorial and demonstrated
further below, the sprayings have caused precisely the same sorts of harms in
Colombia as they have in Ecuador, although they have been more severe on the
Colombian side of the border because there have been more of them and they
have been carried out directly over Colombian territory. In Ecuador’s view, this
fact is relevant for at least two reasons. First, the evidence of harm from
Colombia validates the evidence from Ecuador. Taken together with the other
elements of proof, the Colombian evidence adds to the corpus of consistent,
mutually reinforcing evidence that proves Ecuador’s case. Second, and relatedly,
it underscores the cause-and-effect relationship between the sprayings and the
harms in Ecuador. Put simply, the fact that the sprayings caused nearly identical
harms in Colombia shows that they are, in fact, the injury-causing agent in
Ecuador as well.
3.152 Before proceeding further, a threshold point must be dispensed with. The
Counter-Memorial professes confusion about whether or not, by invoking harms
in Colombia, Ecuador purports to be bringing claims on behalf of Colombian
702
nationals in addition to its own citizens . Indeed, the Counter-Memorial spends
703
no less than nine pages of Chapter 1 voicing its confusion . But Ecuador
701EM, Chap. 5. paras. 5.100-5.115.
702
CCM, Chap. 1, paras. 1.14-1.25.
703
CCM, Chap. 1, paras. 1.14-1.25.
258specifically addressed this issue in the Memorial, stating: “Ecuador, of course, is
not before the Court to press claims on behalf of the people of Colombia. The
harms inflicted in Colombia nonetheless merit the Court’s attention because they
constitute proof of the impacts of the spray mixture Colombia employsŽ 70.
Under the circumstances, Ecuador considers that there is no serious basis for the
confusion Colombia claims to be afflicted by. Although it is wholly unnecessary,
Ecuador here reiterates that it brings claims only on behalf of its own citizens.
The evidence of harm in Colombia is invoked for corroborative purposes only.
3.153 Curiously, even as it professes confusion about why Ecuador is offering
evidence of events in Colombia, the Counter-Memorial makes the converse
argument; that is, it argues that Ecuador’s case is not credible because the
sprayings have not caused any appreciable harm in Colombia. At paragraph 1.34
of the Counter-Memorial, for example, Colombia states:
“If mere drift of the spraying mixture across the border into
Ecuador had caused the catalogue of harms recited by Ecuador,
what would be the position in Colombia itself, the actual target of
many thousands of spray missions over 10 years? The result would
be carnage, hundreds if not thousands of deaths of humans and
705
large animals, environmental devastation, economic collapseŽ .
3.154 As detailed further in the paragraphs to follow, this description is, sadly,
not far from the truth, although more exaggerated. For present purposes the point
704EM, Chap. 5, para. 5.101.
705CCM, Chap. 1, para. 1.34(1).
259is simply that with this argument, Colombia itself recognizes that what happened
in Colombia is relevant to understanding what happened in Ecuador. Elsewhere
in the Counter-Memorial, Colombia helpfully articulates just why that is so. At
paragraph 7.43, it states: “It bears emphasizing again that in the aerial sprayings
carried out up until 2007 over Colombian territory situated close to the border
with Ecuador, exactly the same modalities, mix and procedures as were applied in
706
the rest of the Colombian territory were usedŽ . In Ecuador’s view, it is
precisely for this reason that the evidence of harm in Colombia sheds important
light on the question of harm in Ecuador. Since the same spray mixtures and
operational procedures were employed, the fact that the same injuries were
caused in Colombia underscores the cause of the harm in Ecuador.
3.155 Aside from professing confusion about why Ecuador presents evidence of
the harms the sprayings have caused in Colombia, the Counter-Memorial has
little to say by way of rebutting the facts Ecuador introduced. As demonstrated in
the Memorial, the damage the sprayings have caused in Colombia is borne out by
reports of international observers and civil society organizations,
contemporaneous news reports and even the findings of organs of the Colombian
government not engaged in the execution of the aerial spraying programme 707.
Rather than respond, the Counter-Memorial elects to disregard this important
706CCM, Chap. 7, para. 7.43.
707EM, Chap. 5. paras. 5.100-5.115.
260evidence. In Ecuador’s view, Colombia’s failure to engage on the point speaks as
powerfully as anything it does say in its otherwise ample Counter-Memorial.
3.156 Many of the most alarming reports of damage within Colombian territory
come from the Colombian authorities themselves. Because they constitute
official statements against interest, these reports are entitled to great weight. As
the Court stated in the Case Concerning Military and Paramilitary Activities in
and against Nicaragua (Nicaragua v. United States of America):
“The Court takes the view that statements of this kind, emanating
from high-ranking official political figures, sometimes of the
highest rank, are of particular probative value when they
acknowledge facts or conduct unfavorable to the State represented
by the person who made them. They may be construed as a form
708
of admissionŽ .
3.157 Ecuador brought many such examples to the Court’s attention in its
Memorial. Rather than revisit them here, it respectfully refers the Court to the
709
relevant sections of the Memorial cited in the footnote below . For purposes of
this Reply, a few additional examples will suffice.
708
Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of
America), Judgment, I.C.J. Reports 1986, p. 41, para. 64.
709See EM, Ch. 5, paras. 5.102-5.108; Comptroller General of the Republic of Colombia,
Comptroller for the Environment, Special Audit Regarding Illicit Crop Eradication Policies, p. 34
(July 2001) (“This drift effect is the result of the combination of different technical and
meteorological variables that make this strategy highly susceptible to errFactors like the
height of spraying, the velocity and direction of the wind and the relative humidity are difficult to
control, which affects the precision of the sprayingsŽ.). EM, Vol. II, Annex 93; Republic of
Colombia, Administrative Department of Health (DASALUD) Putumayo Province, Office of
2613.158 In 2000, the Colombian Ombudsman Delegated for Collective Rights and
Environment wrote to Colombia’s Minister of Environment to express his
concern about the severe impacts caused by the aerial spraying program:
“In the last two years, the number of complaints arising primarily
from the departments of Guaviare, Putumayo and Cauca have
increased. ƒ In many of these cases, a critical situation in the
regions after conducting fumigation programs can be observed.
The affectation of illicit crops as well as the destruction of legal
crops leave whole groups of humans without any sustenance,
waves of insecurity and violence are unleashed, valuable
Planning, Epidemiology Section, Effects of Aerial Spraying with Glyphosate Valle del Guamuez –
San Miguel – Orito, Putumayo, p.2 (Feb. 2001) (hereinafter “DASALUD Putumayo Health
StudyŽ) (“According to information from the Administrator of the La Dorada Health Center in the
municipality of San Miguel, in the town of Agua Clara the poisoning of people exposed to the
fumigations was apparent, with symptomology related to skin and eye irritation, nausea, and acute
respiratory infection, as well as bronchitis, the flu, colds, and abdominal pain, among others,
which corresponds to the findings of epidemiological studies carried out in other placesŽ.). EM,
Vol. II, Annex 90; Republic of Colombia, Office of the Ombudsman, Ombudsman Report No. 1,
Fumigations and Alternative Development Projects in Putumayo, pp. 9-10 (9 Feb. 2001)
(hereinafter “Colombia Ombudsman Report No. 1Ž). EM, Vol. II, Annex 91; see also ibid., p. 11
(“Indiscriminate destruction of the little remaining forest, of subsistence crops and medicinal
plants, as well as of pastures and fish-farming ponds, among othersŽ.); Republic of Colombia,
Office of the Ombudsman, Ombudsman Resolution No. 4, On the Impact of Fumigations on 11
Alternative Development Projects in Putumayo, p. 5 (12 Feb. 2001) (“the fumigations condemned
by this resolution destroyed not only the illicit crops … the target of manual eradication … but also
other species necessary for the household subsistence of the beneficiaries of the pacts. Now, these
people and communities are facing both the ruin of their household finances as well as a severe
food security problem. Given the precarious conditions of this group of people, the action by the
State gives rise to a violation of their right to subsistence, which translates into a serious harm to
the physical integrity and dignity of the family and its membersŽ.). EM, Vol. II, Annex 92;
Republic of Colombia, Office of the Ombudsman, Ombudsman Resolution No. 28, The Coffee
Crisis and the Possible Fumigations in the Province of Caldas, pp. 23-25 (21 May 2003)
(footnotes removed). EM, Vol. II, Annex 97. Of course, there is abundant additional evidence
from other sources in Colombia regarding the harm to health and the environment caused by the
aerial spraying. See, e.g., Marcella Ceballos & Carlos Duarte, Report of the Observation Mission
on the Human Rights Situation in Lower Putumayo, pp. 15-16 (June 2008). EM, Vol. IV, Annex
171; Declaration of Colombia Witness 3, 20 Feb. 2009. EM, Vol. IV, Annex 227; Declaration of
Colombia Witness 9, 5 Mar. 2009. EM, Vol. IV, Annex 232; Declaration of Colombia Witness 8,
4 Mar. 2009. EM, Vol. IV, Annex 231; Declaration of Colombia Witness 1, 20 Feb. 2009. EM,
Vol. IV, Annex 225; Declaration of Colombia Witness 2, 20 Feb. 2009. EM, Vol. IV, Annex 226;
Declaration of Colombia Witness 5, 20 Feb. 2009. EM, Vol. IV, Annex 229; Declaration of
Colombia Witness 6, 20 Feb. 2009. EM, Vol. IV, Annex 230; Declaration of Colombia Witness
10, 5 Mar. 2009. EM, Vol. IV, Annex 233.
262 ecosystems to the country are destroyed and the health of the
population is affectedŽ 71.
3.159 Two years later, in 2002, Colombia’s national Office of the Ombudsman
(Defensoría del Pueblo) visited Putumayo Department and other areas affected by
the aerial spraying programme. It subsequently issued a position statement in
which it reported that:
“a commission comprised of officials from the Ombudsman and
other entities recently, in mid November 2001, visited the
municipalities of Valle del Guamuez, San Miguel and Puerto Asís
and verified the impacts on food crops. The crops that were
primarily affected were yucca, plantain, pineapple, corn, cane and
rice; similarly, the deaths of 2500 young fish were reportedŽ 711.
On the basis of this and similar findings, the Ombudsman’s Office concluded that
“[t]he implementation of the eradication program besides posing a threat to the
environment and public health in the fumigation zones, has considerably affected
the vulnerable segments of the population such as small scale farmers and
712
childrenŽ .
710
Letter from Medardo Galindo Hernandez, Ombudsman, Republic of Colombia, to Juan Mayr
Maldonado, Minister of the Environment, Republic of Colombia, p. 1 (24 July 2000). ER, Vol. V,
Annex 134.
711
Republic of Colombia, Office of Ombudsman, The Implementation of the Strategy of Aerial
Eradication of Illicit Crops With Chemicals, From a Constitutional Perspective, p. 44, n. 19 (Apr.
2003). ER, Vol. V, Annex 146.
712Ibid., pp. 55-56. The report also noted that “despite 22 years having elapsed since fumigation
operations started in the country, no relevant scientific studies were conducted to determine this
substance’s effects on healthŽ. ER, Vol. V, Annex 146.
2633.160 In a later statement, the Ombudsman’s Office emphasized that the “[t]he
damage to the food crops of farmers and natives has put their food safety at riskŽ
because, without such crops, “an adequate food supply that covers their
713
nutritional needs cannot be guaranteedŽ . These observations echoed similar
statements in which the Ombudsman’s Office explained that as a result of the
damage caused to crops and animals used for subsistence, people affected by the
sprayings faced “both the ruin of their household finances as well as a severe food
714
security problemŽ .
3.161 The Colombian Comptroller General’s Office has likewise recognized the
injuries caused by Colombia’s sprayings. In 2001, it reported on the occurrence
of “symptoms of nausea, vomiting, diarrhea, and burning of the eyes, skin and
throat after the spraying; reports that coincide with information in the literature
715
and are consistent with the position of the Ministry of HealthŽ .
713
Republic of Colombia, Office of the Ombudsman, National Ombudsman Resolution No. 26,
Human Rights and International Humanitarian Law in the Context of Armed Conflict and
Fumigation of the Coca Crops in the Province of Putumayo, p. 25 (9 Oct. 2002). ER, Vol. V,
Annex 145.
714Republic of Colombia, Office of the Ombudsman, Ombudsman Resolution No. 4, p. 4 (12 Feb.
2001). EM, Vol. II, Annex 92; see also EM, Chap. 5, paras. 5.106-5.108.
715Comptroller General of the Republic of Colombia, Plan Colombia: Second Evaluation Report,
p. 44 (10 Dec. 2001). EM, Vol. II, Annex 94.
2643.162 As discussed in the Memorial, these same symptoms were documented by
Colombia’s health authorities 716. For example, the Putumayo Department of
Health recorded a dramatic increase in symptoms, including acute respiratory
infections, diarrhea, dermatitis, and skin infections following sprayings just 20
kilometres north of the Ecuadorian border in early 2001 71. Ecuador notes that
these harms are precisely the types of harms that have been documented in
Ecuador.
3.163 As has been true in Ecuador, the Colombian Office of the Ombudsman
reported that the aerial sprayings took a particularly heavy toll on children:
“In this regard, the increase in medical visits related to skin
problems, gastrointestinal, respiratory infections and
conjunctivitis, after the fumigations, in the area sprayed, is cause
for concern. Even more alarming is that, in most cases it is the
children in these regions who are presenting such symptoms. Due
to718eir fragile state, their symptoms tend to become more acute . .
.Ž .
3.164 The Ombudsman’s Office also reported that, on a number of occasions,
the exposure to the spray mixture appeared to have contributed to the death of
716EM, Chap. 5, paras. 5.102-5.105.
717
EM, Chap. 5, para. 5.103.
718
Republic of Colombia, Office of Ombudsman, The Implementation of the Strategy of Aerial
Eradication of Illicit Crops With Chemicals, From a Constitutional Perspective, p. 6 (Apr. 2003).
ER, Vol. V, Annex 146.
265small children: “the Ombudsman has also recognized some cases in which the
death of five-year old minors is related to chemical-based aerial sprayingŽ 71.
These findings bring into rather stark relief the Counter-Memorial’s bald
assertion that “[n]o substantiated complaint of death or serious harm to human
720
health has been presented in Colombia since the inception of the programŽ .
3.165 Colombia’s Comptroller General has recognized that the extent of the
harm in Colombia is a function, as in Ecuador, not just of the toxicity of the spray
mixture and Colombia’s recklessness in applying it, but also the unique
vulnerability of the population:
“One of the most troubling aspects [of the aerial spraying
programme] are the continual complaints from communities
located in the zones targeted by the program, composed primarily
of campesinos and colonists with precarious incomes, low levels
of nutrition, far from medical treatment centers, and with limited
access to health services. Both the physical and mental health of
the population has been affected, and family finances have been
impacted as well; first, as a direct and indirect result of the
fumigations, and second, due to the damage to their financial well-
719Republic of Colombia, Office of Ombudsman, The Implementation of the Strategy of Aerial
Eradication of Illicit Crops With Chemicals, From a Constitutional Perspective, p. 51 (Apr. 2003)
(“according to complaints filed by the children’s parents, they showed symptoms of poisoning
such as vomiting, diarrhea and respiratory and skin problems after the spraying. The poisoning,
caused by direct exposure to the herbicide and by the consumption of contaminated water, is
considered to be a possible cause of their deaths, hours and days laterŽ. ER, Vol. V, Annex 146.
720
CCM, Chap. 1, para. 1.34(1).
266 being, whic721s represented primarily by legal crops and
livestockŽ .
3.166 A number of Colombian departmental and local representatives, who do
not have the same vested interest in extolling the virtues of the aerial spraying
programme as elements of the national government in Bogotá, have also
complained of the severe damage caused by the spraying carried out in their
regions. In 2007, for instance, the Governor of Putumayo Department denounced
the sprayings, stating that the damage caused to legal crops by the aerial spraying
722
program was “causing an economic crisis and displacement of the populationŽ .
Other examples include the mayor of Puerto Guzmán, Putumayo, who in 2000
reported that at least seven people had died as a consequence of aerial sprayings
723
conducted in the area . Later, in 2001, the Governors of six departments,
including Putumayo and Nariño along the Ecuadorian border, denounced the
sprayings due to their impacts on human health, legal crops, and the environment,
and appealed to the national government to instead pursue a strategy of manual
eradication 724.
721Comptroller General of the Republic of Colombia, Plan Colombia: Third Evaluation Report,
p. 61 (Aug. 2002). ER, Vol. V, Annex 143.
722
“Putumayo: Governor Denounces FumigationsŽ, H OY (Quito, 29 July 2007). ER, Vol. IV,
Annex 85.
723
“Mayor Denounces FumigationsŽ, E LU NIVERSO (Guayaquil, 22 Aug. 2000). ER, Vol. IV,
Annex 56.
724
“No To Fumigation: GovernorsŽ, L TIEMPO (Bogotá, 15 Jan. 2001). ER, Vol. IV, Annex 59.
2673.167 The extensive harm in Colombia has also been substantiated by
international observers. Following a March 2004 visit to Colombia, the UN
Situation of Human Rights and Fundamental Freedoms of Indigenous People, Mr.
Rodolfo Stavenhagen, found “adverse effects of indiscriminate spraying,
including environmental damage to the topsoil, fauna, flora and water, the
destruction of subsistence crops and direct damage to human healthŽ 725. The
Special Rapporteur was sufficiently concerned about the damage caused by
Colombia’s aerial spraying program that he concluded his report with the
following recommendation: “[e]xcept where expressly requested by an
indigenous community which has been fully apprised of the implications, no
aerial spraying of illicit crops should take place near indigenous settlements or
sources of provisionsŽ 726. He arrived at this conclusion after meeting with high-
ranking governmental officials in Bogotá, including then-President Álvaro Uribe,
727
as well as personally visiting the departments of Cauca, César and Putumayo .
During these local visits, he met with departmental and local officials, members
725
Report of the Special Rapporteur on the Situation of Human Rights and Fundamental
Freedoms of Indigenous People, Mr. Rodolfo Stavenhagen, Mission to Colombia U.N. Doc.
E/CN.4/2005/88/Add.2, para. 50 (10 Nov. 2004). ER, Vol. IV, Annex 102; see also ibid., para.
82. Mr. Stavenhagen reported that the Awá had been particularly affected: “The Awá community
in Nariño has informed the Special Rapporteur of various kinds of damage caused over the last
three years to large tracts of rainforest in several areas of the municipalities of Tumaco and
Barbacoas, as a result of spraying with glyphosate. The greatest damage was done, they say, to
sources of fresh water, killing native fish and affecting human health, causing aching bones,
vomiting, dizziness, fever and other ailments, particularly among children.Ž Ibid., para. 51.
726Report of the Special Rapporteur on the Situation of Human Rights and Fundamental
Freedoms of Indigenous People, Mr. Rodolfo Stavenhagen, Mission to Colombia U.N. Doc.
E/CN.4/2005/88/Add.2, para. 106 (10 Nov. 2004). ER, Vol. IV, Annex 102.
727Ibid., para. 8.
268of civil-society and grass-roots associations, and representatives of more than 30
728
indigenous communities .
3.168 The Special Rapporteur on the Right to Food, Mr. Jean Ziegler, reached
the same conclusions as Mr. Stavenhagen, also finding cause for concern about
the impacts of aerial spraying in Colombia. Mr. Ziegler highlighted the lack of
clarity and information regarding the chemicals used in the spray mixture and
their concentrations, noting that the “proportion of glyphosate being employed
and the actual composition of the final product being used are unknownŽ 729. As
to food, Mr. Ziegler explained: “the concern of the Special Rapporteurs is not just
limited to food security risk but also to the right to food free from harmful
730
substancesŽ . Mr. Ziegler cited, in particular, to evidence from the Colombian
police documenting extensive damage to crops, resident health, and animals
731
following a 2001 spraying campaign . He also commented on how the harms
experienced in Colombia have direct ramifications for Ecuador: “Even the
Ecuadorian Red Cross acknowledges that the second largest case for
728Ibid., para. 9.
729 Report of the Special Rapporteur on the Right to Food, Jean Ziegler, Addendum:
Communications Sent to Governments and Other Actors and Replies Received, U.N. Doc.
A/HRC/4/30/Add.1, para. 17 (18 May 2007). EM, Vol. II, Annex 33.
730Ibid., para. 17 (18 May 2007). EM, Vol. II, Annex 33.
731Ibid.
269displacement of the Colombian population to Ecuador, 54%, is due to the
732
sprayings that affect their productsŽ .
3.169 There is also a flood of contemporaneous news reports corroborating the
extensive damage caused by the aerial spraying programme inside Colombia.
Ecuador has collected a number of these reports in the Annexes to this Reply. It
will not burden the Court by examining their contents here. Instead, it
733
respectfully refers the Court to the relevant annexes . The essential point is that
these media reports contain contemporaneous accounts of harms in Colombia that
further corroborate the accounts of harm emanating from Ecuador.
3.170 To the extent that the Counter-Memorial makes any effort whatsoever to
controvert this evidence, it focuses on the limited number of successful claims
made under the compensation program it established for farmers who have
suffered harm to their lawful crops. According to the Counter-Memorial,
732Ibid.
733See e.g., Larry Rohter, “To Colombians, Drug War is Toxic EnHEyNEWT YORK TIMES
(New York, 1 May 2000). ER, Vol. IV, Annex 54; “The Void of the FumigatioLsTIEMPO
(Bogotá, 28 May 2000). ER, Vol. IV, Annex 55; Juan Forero, “No Crops Spared in Colombia’s
Coca WarŽ, T HE NEW YORK TIMES (New York, 31 Jan. 2001). ER, Vol. IV, Annex 60;
“Fumigation DisputeŽ, ETIEMPO (Bogotá, 22 July 2001). ER, Vol. IV, Annex 62; “Colombia
Denounces Indiscriminate Spraying in PutumayoŽ, EERCIO (Quito, 10 Jan. 2002). ER, Vol.
IV, Annex 64; “Another Controversy Over FumigationŽ, EERCIO (Quito, 9 July 2002). ER,
Vol. IV, Annex 67; “Requesting an End to FumigationsŽ, EO (Bogotá, 10 Oct. 2002). ER,
Vol. IV, Annex 72; “Fumigations Cause Concern in PutumayoŽ, EMERCIO (Quito, 10 Nov.
2002). ER, Vol. IV, Annex 73; “Glyphosate RainŽ, EPO (Bogotá, 25 Feb. 2003). ER, Vol.
IV, Annex 74; “Between Faith and FumigationsŽ, EEMPO(Bogotá, 10 May 2002). ER, Vol.
IV, Annex 66; “Spray Program on Indigenous Territories Is StrugglingŽ, EBogotá, 28
Apr. 2003). ER, Vol. IV, Annex 75.
270between 2002 and 2008, there were only 117 cases in which compensation was
actually awarded 73. Presumably, this is meant to show that the off-target impacts
are more limited than the evidence cited above might suggest. If that is indeed
Colombia’s point, it is manifestly not credible in light of the spray flight data
discussed in Chapter 2. The data Ecuador secured shows a wholesale disregard …
by a staggering margin … of what Colombia itself refers to as “mandatoryŽ flight
parameters. Given this, combined with the fact that Colombia has conducted
literally hundreds of thousands of spray flights since Plan Colombia began, the
fact that there have only been 117 successful claims since the programme began
says more about the unfairness of the compensation program than the
harmlessness of the sprayings. Moreover, 117 compensated claims is certainly
not evidence of the lack of harm in Colombia.
3.171 Colombia also does not bother to mention that during the same time
period (2002-2008), many thousands of complaints were submitted. According to
the Colombian Comptroller General, in 2002 alone, the Ministry of Justice
received 4,500 complaints 73. Still other complaints were directed to the
734CCM, Chap. 4, para. 4.22; see also CCM, Chaps. 1 and 7, paras. 1.34(1), 7.174.
735
Comptroller General of the Republic of Colombia, Plan Colombia: Fifth Evaluation Report, p.
36 (Dec. 2004). ER, Vol. V, Annex 152.
271Colombian Ombudsman’s Office, which reported receiving 6,553 complaints by
736
the end of 2001 .
3.172 The one thing that is clear from these numbers is that the vast majority of
claims have gone uncompensated. Given that the cards are stacked so heavily
against the claimants, this is not surprising. DIRAN’s compensation program
was created in 2001 and involves a lengthy and mind-numbingly complex
process, requiring, among other things, the presentation of a deed (a legal
instrument which many landowners in the remote locations affected by the
spraying programme simply do not have); two field visits (which are not
conducted if there is “public unrestŽ, a near ubiquitous condition in Colombia’s
southern Departments); and additional confirmatory evidence (including spray
records and satellite images) 737. To say that these requirements exceed the means
of the vast majority of campesinos is to belabour the obvious. For this reason,
736Republic of Colombia, Office of the Ombudsman, National Ombudsman Resolution No. 26,
Human Rights and International Humanitarian Law in the Context of Armed Conflict and
Fumigation of the Coca Crops in the Province of Putumayo, p. 24 (9 Oct. 2002). ER, Vol. V,
Annex 145.
737Resolution No. 017 of 4 October 2001 of the National Narcotics Council of Colombia, Art. 4,
5, 8, 13. CCM, Vol. II, Annex 43. In 2007, the National Narcotics Council found it “necessary to
amend [the compensation program] in order to make its implementation faster and efficient and
this way determine the alleged liability of the StateŽ, however the program enacted in 2007
continues to be structured in a way that make a compensation award nearly impossible for the
claimant. See Resolution No. 008 of 2 March 2007 of the National Narcotics Council of
Colombia. CCM, Vol. II, Annex 61.
272among others, the Colombian Ombudsman’s Office has called the programme
“inadequate and ineffectiveŽ 73.
3.173 The evident inadequacy of the Colombian compensation programme
stems at least in part from the fact that it is administered by the very entities that
are responsible for the spray programme, i.e., the DNE and the Antinarcotics
Police. Noting this conflict of interest, the Colombian Comptroller General stated
that:
“This procedure has serious flaws, among others, and most
obvious is that the same agencies responsible for carrying out the
sprayings, the DNE [National Narcotics Directorate] and the Anti-
Narcotics Division of the Police, are charged with evaluating the
claims for damages, thus becoming judge and party in the
conflictŽ739.
Under the circumstances, perhaps the most remarkable feature of the
compensation program is that any of the claims managed to succeed.
3.174 With the totality of mutually corroborating evidence decisively against it,
it is difficult to understand how Colombia can continue to maintain that the
massive aerial spraying of chemical herbicide across swathes of territory in
reckless disregard of all operational and safety requirements, without ever having
738
Republic of Colombia, Office of the Ombudsman, National Ombudsman Resolution No. 26,
Human Rights and International Humanitarian Law in the Context of Armed Conflict and
Fumigation of the Coca Crops in the Province of Putumayo, p. 22-23 (9 Oct. 2002). ER, Vol. V,
Annex 145.
739
Comptroller General of the Republic of Colombia, Plan Colombia: Fourth Evaluation Report,
p. 37 (July 2003). EM, Vol. II, Annex 98.
273conducted an environmental impact assessment to assess the harms, has caused
no harms. Its own evidence attests to the contrary in Colombia, as it does in
Ecuador.
Conclusion
3.175 In summary, this Chapter provides incontrovertible evidence that aerial
spraying drifted across the boundary and caused measurable and significant harm
to people, to livestock, to animals, to crops and to the environment, in areas
within 10 kilometres of the border with Colombia, in Sucumbíos and Esmeraldas
Provinces. The harm extends to indigenous peoples and their communities. The
evidence is extensive and conclusive, coming as it does from a range of different
sources, all of which confirm and are consistent with each other. In particular,
this Chapter allows the following conclusions to be drawn as to the evidence of
harm:
(1) Newly available flight data information proves the dates and locations of
sprayings and the reckless conditions under which they were carried out,
and provides clear evidence in support of the facts set forth in the
Memorial; of particular probative value are individual witness statements,
contemporaneous UN and other observation mission reports, press reports,
scientific studies and reports from Colombia.
274(2) Colombia has misunderstood and misrepresented the evidentiary value of
the material on which Ecuador relies, including witness testimonies, that
provide conclusive evidence as to the effects of the spraying on people,
crops, property and the environment.
(3) The evidence of harms occurring in Ecuador are confirmed by the
consistent evidence of the same harms arising in Colombia.
275 CHAPTER 4.
COLOMBIA’S FAILURE TO CONDUCT AN ENVIRONMENTAL
IMPACT ASSESSMENT, AND OTHER FAILURES TO PREVENT, OR
EVEN ASSESS, HARM TO ECUADOR4.1 The Parties agree that Colombia’s aerial spraying programme required a
transboundary environmental impact assessment (“EIAŽ). The Counter-
Memorial concedes, “Colombia was, no doubt, obliged to consider potential
impacts on its neighbours, and on the environment, in formulating and
implementing the aerial spraying programŽ 74. Colombia also does not dispute
that it was required to carry out a transboundary EIA prior to beginning the
spraying programme along or near the border with Ecuador. This is self-evident.
An EIA is the means by which regulatory authorities determine whether a project
should be permitted, and if so, for deciding what restrictions must be enforced to
prevent or minimize risk to the surrounding environment, including human
settlements. Carrying out an EIA before a project begins is therefore a
fundamental necessity. This is especially so when the contemplated undertaking
is as potentially damaging as the aerial spraying of entire areas with a potent
herbicide mixture whose known elements are understood to harm human health
and indiscriminately kill the plants it comes in contact with. The Counter-
Memorial does not dare to disagree.
4.2 Yet, as shown in this Chapter, Colombia did not carry out a transboundary
EIA before spraying along the border with Ecuador. Nor did it conduct an EIA at
any time thereafter. The Counter-Memorial does not deny that a transboundary
740Counter-Memorial of Colombia, Vol. I, Chap. 1, para. 1.40 (29 Mar. 2010) (hereinafter
“CCMŽ).
279EIA was never carried out. Instead, it tries to defend that failure by asserting that
an EIA is not required under Colombian municipal law. How does Colombia
justify such a remarkable claim? By stating that it enacted legislation which
exempted the programme from requiring an EIA. Of course, Colombia’s self-
serving domestic legislation cannot trump its international legal obligations in
regard to conducting a transboundary EIA. But the legislative manoeuvring is
nonetheless relevant because it reveals the lengths to which Colombia has gone to
avoid carrying out an EIA in regard to the aerial spraying programme … even
eliminating the domestic legal requirement to assess environmental impacts in
Colombia … because it knows very well that if it ever carried out an EIA in regard
to the programme the results would make aerial spraying impossible. Simply put,
the spraying programme could not survive an EIA that meets even the most
minimal standards, and Colombia knows it. Hence it dispensed with the EIA
altogether, and conveniently changed its own laws to give its omission a veneer
of legality.
4.3 In place of an EIA, the Counter-Memorial says that Colombia substituted
an “Environmental Management PlanŽ (“EMPŽ) 74. This document bears scant
742
relation to an EIA, and does not even purport to be one . The Court has already
741See CCM, Chap. 4, paras. 4.10-4.11.
742See CCM, Chap. 6, paras. 6.23-6.24 (“Colombia had only developed an Environmental
Management Plan . . . and not an Environmental Impact AssessmentŽ.) (internal quotation
omitted). See also Claudia Rojas Quiñonez, Esq., The Aerial Spray Program and Violations of
280been introduced to the EMP in Chapter 2 of this Reply: this is the document that,
inter alia, establishes the spray programme’s operational requirements for the
purpose of preventing spray drift onto non-target locations, including the
maximum altitude for spray dispersal, spray application rate and droplet size.
Ecuador has already shown that Colombia’s spray pilots routinely ignored these
requirements on tens of thousands of spray flights along or near the border,
making spray drift into Ecuador inevitable. Although these massive and pervasive
violations of the EMP are of critical importance in this case, they are not
Colombia’s only violations of the EMP. In this Chapter, Ecuador shows that
Colombia also failed to comply with its obligations under the EMP to assess the
environmental impacts of its aerial spraying programme, despite the fact that it
was ordered in 2004 by the nation’s highest administrative tribunal … the Council
of State … to do so. Chiefly, the evidence of these violations of the EMP is
supplied by the Government of Colombia itself. On repeated occasions
Colombia’s own Ministry of Environment, Comptroller General and National
Ombudsman criticised the National Narcotics Directorate (Dirección Nacional de
Estupefacientes, “DNEŽ) for failing to fulfil its responsibilities under the EMP to
carry out the required environmental impact studies. In fact, the DNE’s refusal to
conduct any environmental impact assessments in regard to the aerial spraying
Colombia’s Domestic Laws Regarding the Environment and the Rights of Indigenous Peoples,
paras. 36-41 (describing the difference between an EIA and an EMP under Colombian law)
(hereinafter “Rojas Report) (Jan. 2011). ER, Vol. II, Annex 8.
281programme led the Environment Ministry to impose sanctions on it 743. But it still
refused to comply.
4.4 In light of these facts, it is not surprising that no actual environmental
impact studies were ever carried out … neither prior to (as required by
international law) nor at any time after the aerial spraying programme was put
into operation. The Counter-Memorial is left to cite only three studies of any
kind: a minor scientific study on glyphosate that occurred 13 years before aerial
spraying began near Ecuador; and two others that took place five and nine years,
744
respectively, after spraying commenced along the border . None of these
studies included even the most rudimentary elements of a transboundary EIA. In
particular, they failed to evaluate the specific herbicide formulations that
Colombia uses (or has used) in the programme; and they failed to assess
transboundary impacts in Ecuador, including via spray drift.
4.5 What is apparent from Colombia’s consistent conduct is that
environmental protection and human rights are set aside in favour of drug
eradication. This has been the case from the time the aerial spraying programme
began until the present day, and at all times in between. To be sure, the
743See infra, Chap. 4, para. 4.65; Republic of Colombia, Ministry of Environment Resolution No.
0670, Whereby a sanction is imposed and other decisions are made (19 June 2003). EM, Vol. II,
Annex 19.
744CCM, Chap. 4, paras. 4.8-4.19.
282Environment Ministry, the Comptroller General and the National Ombudsman
have spoken out against the programme … but no one in authority pays any
attention to them: certainly not the DNE or the Anti-Narcotics Direction of the
Colombian National Police (“DIRANŽ), the drug enforcement and anti-narcotics
agencies which are charged with carrying out the aerial spraying programme.
That they have been free to ignore these other government agencies … as well as
Colombia’s highest administrative tribunal … attests to the priority given at the
highest levels of the Government of Colombia to drug eradication, even at the
expense of environmental protection and the human rights of mostly
impoverished farmers, villagers, and indigenous peoples regularly doused with
the spray mixture. It is not for Ecuador to comment on the wisdom or morality of
this policy. Colombia is free to act in this manner if it so chooses … at least in
regard to its own territory and nationals. But, as important as the fight against
illicit drugs may be to Colombia (and Ecuador, as well, for that matter), Colombia
is not free to conduct aerial spraying operations in a manner that harms the
environment in Ecuador, or harms or violates the human rights of Ecuadorian
nationals.
4.6 This Chapter is organized as follows. Section I establishes that Colombia
failed to assess environmental impacts before it started to spray near Ecuador.
This point is hardly in dispute. The Counter-Memorial makes only the feeblest of
attempts to claim any impact assessments were performed before January 2000
283when the spraying began along the border with Ecuador. The scant evidence that
Colombia cites … a small-scale test on glyphosate conducted over a decade before
spraying commenced … is facially inadequate. So inadequate, in fact, that the
745
Counter-Memorial relegates it to a footnote .
4.7 In Section II, Ecuador demonstrates that Colombia also did not assess
environmental impacts after it started spraying. Colombia cites only two studies
after the commencement of spraying in 2000, neither of which comes close to an
EIA, let alone a transboundary assessment of impacts on Ecuador. The first study
did not take place until 2005, over five years after Colombia began to spray near
Ecuador, and the second was not until 2009, four years later still. Critically,
neither study assessed environmental risks to Ecuador, its people, animals or
plant-life. Nor did they evaluate the impacts of the particular chemical mixture
the Counter-Memorial claims Colombia used for the first five years of the aerial
spraying programme, much less the elements in the mixture which Colombia has
used, but not disclosed. And, worst of all, the 2005 and 2009 studies invoked by
the Counter-Memorial were based on Colombia’s false representations that it
operates the spray programme in strict compliance with the operational
requirements that prevent spray drift, including aircraft speed, height of spray
release, application rate, droplet size, time of day, etc. Therefore, although the
model used for measuring drift is an acceptable one, the results are meaningless
745CCM, Chap. 4, para. 4.8, n. 253.
284because the inputs are completely wrong. As the saying goes: garbage in,
garbage out.
4.8 Section III demonstrates that the aerial spraying programme also violates
other important operational and safety requirements, both within and external to
the EMP. These include compliance with label instructions for safe usage of each
of the products that Colombia has sprayed near Ecuador, including the label
instructions for Roundup Export, Roundup Ultra, Roundup SL and GLY-41.
These violations are highly probative of a likelihood of harm, and evidence a
breach of Colombia’s duty of due diligence, especially because compliance with
product labelling instructions for pesticides is mandatory in Colombia. The
violations thus transgress Colombia’s own standards for what is required for the
safe application of the products used in the spray mixture, including those
required to avoid spray drift.
4.9 Finally, Section IV demonstrates that Colombia’s aerial spraying
programme could not be conducted in other jurisdictions throughout the world,
including the European Union (“EUŽ), which has banned aerial spraying, subject
only to a limited ability to derogate in narrow and highly controlled
circumstances. Colombia’s aerial spraying programme would also be prohibited
by the environmental laws of numerous other States. These are further
285indications that the irresponsible manner in which Colombia sprays toxic
herbicides near Ecuador is dangerous, and prone to cause transboundary harm.
Section I. Colombia’s Failure to Assess Environmental Impacts Before the
Aerial Spraying Programme Was Commenced
4.10 It will doubtless not have escaped the Court’s attention that the Counter-
Memorial nowhere presents the results of an EIA. The reason for that omission is
not in dispute: no EIA was carried out.
4.11 Colombia’s defence of its failure to carry out an EIA is risible: it claims
none was required under its domestic law because when new legislation was
enacted in 1993 it created a “transitional regimeŽ that exempted pre-existing un-
licensed projects from having to carry out EIAs. As Colombia states at
paragraphs 4.10-4.11 of the Counter-Memorial:
“In December 1993, the Law on the Environment was passed. In
1994, a regulatory decree adopted under that Law provided that
activities … including the aerial spraying of illicit crops … that had
been authorized and begun prior to that date could continue . . .
The new legal regime expressly set out the activities requiring
prior environmental authorizati746 the application of pesticides
was not included among themŽ .
4.12 This won’t fly. Colombia cannot avoid its obligation under general
international law to carry out a transboundary EIA before starting a project
746CCM, Chap. 4, paras. 4.10-4.11.
286capable of damaging a neighbouring State by declaring that, as a matter of
municipal law, no EIA is required 747.
4.13 Absent an EIA, it is hardly surprising that the evidence that Colombia
conducted any sort of assessment of environmental impacts prior to spraying near
the border with Ecuador is presented in a single footnote … Footnote 253.
According to this footnote, at some point “in the 1980sŽ, a private company, SGS
Colombia, “analyzed the results of several soil and water samples following the
application of glyphosateŽ in a national park in Sierra Nevada de Santa Marta,
Colombia. That is all. The Counter-Memorial, in effect, concedes that there are
no other scientific assessments showing that Colombia carried out anything
resembling an EIA prior to spraying near Ecuador.
4.14 Ecuador need not detain the Court by dwelling on why this limited study,
conducted well over a decade before spraying commenced near its territory, was
inadequate to assess the risks of harm to people, plants, animals or the
environment in Ecuador. The study was limited to measuring glyphosate residue
747
As discussed in greater detail in Chapter 6, the Court recently confirmed in the Case
Concerning Pulp Mills on the River Uruguay that States have an obligation under general
international law to “undertake an environmental impact assessment where there is a risk that the
proposed activity may have a significant adverse impact in a transboundary context, in particular,
on a shared resource. Moreover, due diligence, and the duty of vigilance and prevention which it
implies, would not be considered to have been exercised, if a party planning works liable to affect
the [environmental quality of a shared or transboundary resource] did not undertake an
environmental impact assessment on the potential effects of such worksŽ. Case Concerning Pulp
Mills on the River Uruguay (Argentina v. Uruguay), Judgment, p. 60, para. 204 (2010). The
Court made clear that “an environmental impact assessment must be conducted prior to the
implementation of a projectŽ. Ibid., p. 60, para. 205 (emphasis added).
287in a small number of soil and water samples taken from areas that had been
748
sprayed between two and five months earlier , despite the fact that, according to
749
Colombia, glyphosate biodegrades in tropical soils in less than 7 days . Nor did
it assess spray drift, including spraying at various heights, speeds and application
rates750. It also did not assess impacts to human health, off-target plant-life or
animals 75. No analysis was done on either of the two spray formulations that the
Counter-Memorial admits Colombia sprayed in areas adjacent to Ecuador,
namely Roundup SL and GLY-41, or on Roundup Export (the product the U.S.
government reported Colombia was using), or on Roundup Ultra (the product
752
Colombia claims it sprayed in diplomatic exchanges with Ecuador) .
4.15 Indeed, the study is so inadequate that the Counter-Memorial
acknowledges Colombia could not comply with a request from Ecuador, made in
October 2003, that it provide any “Environmental Impact AssessmentŽ that had
748
SGS (Societé Génerale de Surveillance, S.A.) Colombia S.A., “Report of Contamination
Control for glyphosate application at the Sierra of Santa MartaŽ, p. 5 (1987). CCM, Vol. III,
Annex 123.
749Diplomatic Note DM/AL No. 25009, sent from the Ministry of Foreign Affairs of Colombia to
the Ministry of Foreign Affairs of Ecuador, p. 3 (14 July 2001). EM, Vol. II, Annex 42.
750SGS (Société Générale de Surveillance, S.A.) Colombia S.A., “Report of Contamination
Control for glyphosate application at the Sierra of Santa MartaŽ (1987). CCM, Vol. III, Annex
123.
751Ibid.
752Ibid. The study refers only to “RoundupŽ without stating which, if any, of the formulations
Colombia has used near Ecuador was studied.
288been “conducted prior to sprayings of GlyphosateŽ 753. The reason this simple
request could not be granted is patent: there was nothing to provide.
754
4.16 Colombia argues that in place of an EIA it prepared an EMP . Without
elaborating, the Counter-Memorial asserts that an EMP “is equivalent to an
755
environmental impact assessmentŽ . Colombia knows better. Its EMP comes
nowhere close to being an EIA. An environmental impact assessment is just that …
it is a comprehensive assessment of the potential environmental impacts of a
project. Its fundamental objective is to identify the harms a project might cause,
before they occur, in order to determine whether a project should proceed and to
determine the measures required to prevent or limit these harms 75. In order to
attain an environmental license in Colombia, one must first execute an EIA to
757
determine the harms . An EMP is just one of the documents that accompany
758
the EIA in this process . It is only after the extent and type of harms have been
identified in the EIA that the EMP can be created to manage and prevent the
project’s identified harms once it is underway. Thus, Colombia carried out its
753CCM, Chap. 6, paras. 6.24.
754
Ibid., Chap. 4, paras. 4.10-4.11.
755
Ibid., Chap. 4, para. 4.10
756
See Reinhard Joas, Ph.D., The Development of the 2009 European Union Pesticides Directive
With Particular Focus on Aerial Spraying, p. 15 (Jan. 2011) (hereinafter “Joas ReportŽ). ER, Vol.
II, Annex 7; Rojas Report, op. cit., paras. 36-38, 96. ER, Vol. II, Annex 8.
757
See Rojas Report, op. cit., paras. 33-38, 96 (discussing the requirements of Law 99 of 1993).
ER, Vol. II, Annex 8.
758
See ibid., paras. 39-41, 96. ER, Vol. II, Annex 8.
289mass-scale aerial spraying programme without ever having first assessed and
understood the harms it would cause. The ostensible plan to address these harms,
the EMP, was designed in a void, uninformed of the damage it was created to
prevent.
4.17 The Colombian Ministry for the Environment recognized that the EMP
was not equivalent to, and could not substitute for, an EIA. And it should know: it
was the agency responsible for approving the EMP. As shown below, it
repeatedly demanded that EIAs be carried out as part of the EMP. Each of these
demands was rebuffed by the organ of the Colombian government that was
assigned responsibility for conducting the impact assessments of the aerial
spraying programme: the DNE. This, self-evidently, was a major flaw in the
EMP. By putting environmental protection in the hands of the anti-narcotics
police … whose main mission is drug eradication … the EMP effectively placed the
fox in charge of the henhouse. Thus, whenever it was called on … or ordered … by
the Environment Ministry to assess the environmental impacts of the aerial
spraying programme, the DNE just said “NoŽ. And no one was able to make the
DNE comply. The DNE was fined and sanctioned; Colombian courts and
agencies even ordered the aerial spraying programme to be suspended until the
DNE complied with the EMP. Yet the DNE continued to carry out the spraying
programme in defiance of these orders.
2904.18 The Counter-Memorial makes only a brief reference to the development
of the EMP. It says: “The Ministry for the Environment and the National
Narcotics Directorate jointly worked in developing the EMP. In November 2001,
following several adjustments arising from initial field experience, the EMP was
759
formally adopted by Resolution 1065 of 2001Ž . The Counter-Memorial is
silent in regard to the contents of the EMP, its requirement that environmental
impacts be assessed, and the Environment Ministry’s unsuccessful efforts to get
the DNE to make these assessments. It thus falls to Ecuador to tell the story.
4.19 What happened is this. When the Colombian Government first began
aerial spraying on an experimental basis in northern Colombia in 1984,
INDIRENA (as the Ministry of Environment was then known) insisted that the
Colombian National Police (the agency then responsible for the spraying) carry
out the legally required impact studies. Specifically, it presented the Ministry’s
“demand that an ecological and environmental study be submitted prior to the
glyphosate-based aerial spraying actions in any area of the countryŽ 760. To that
end, INDIRENA provided the National Police with terms of reference for the
759
CCM, Chap. 4, para. 4.13 (emphasis added).
760
Republic of Colombia, Ministry of Agriculture, Study Proposal for the Environmental
Management of the National Park Areas of Sierra Nevada de Santa Marta and Buffer Zones
Affected by Marijuana Crops and their Destruction by Aerial Spraying with Glyphosate, pp. 3-4
(31 July 1986). ER, Vol. V, Annex 120. See also Rojas Report, op. cit., paras. 25 & 64
(explaining how the “EES, according to article 28 of the CNRNR, was required prior to obtaining
[an Environmental] License and consisted of a study that had to be carried out before the
performance of activities that could produce serious environmental deteriorationŽ.). ER, Vol. II,
Annex 8.
291studies, and registered an official protest for having been excluded from the
761
decision to spray glyphosate and for the manner in which it was being applied .
762
The National Police ignored INDIRENA’s directive and continued spraying .
4.20 Two years later, INDIRENA again insisted that environmental studies be
carried out. It stated that “[t]hese studiesŽ had to “consist of basic research on
Ecology, Socioeconomics, Culture and the identification of adverse effects on
human populations, plants and animals caused by the spraying of GlyphosateŽ,
and were necessary to “enable the proposal of environmental management
measures of a socio-economic and ecological natureŽ 763. In light of its serious
concerns with the spray programme, INDIRENA stated that “legal and moral
obligationsŽ compelled it to “oppose[]Ž the “spraying with glyphosate in areas
belonging to Colombia’s National Park SystemŽ and, with respect to other areas
being considered for aerial spraying, to conclude that “the preparation,
presentation, evaluation and approval of the environmental studies required by
LawŽ was “essentialŽ 764.
761
Republic of Colombia, Ministry of Agriculture, Study Proposal for the Environmental
Management of the National Park Areas of Sierra Nevada de Santa Marta and Buffer Zones
Affected by Marijuana Crops and their Destruction by Aerial Spraying with Glyphosate, p. 4 (31
July 1986). ER, Vol. V, Annex 120.
762Ibid., p. 5.
763
Ibid., pp. 3-4.
764
Republic of Colombia, Ministry of Agriculture, Study Proposal for the Environmental
Management of the National Park Areas of Sierra Nevada de Santa Marta and Buffer Zones
2924.21 When three scientists from INDIRENA later observed first-hand spraying
operations and their environmental impacts, their report was brutally frank in its
criticism, finding that the sprayings were “ecologically . . . destroyingŽ the
765
area . With respect to the missing environmental impact studies, INDIRENA
stated:
“Glyphosate is a harmful chemical agent, it is not recommended
for aerial application, its danger is even greater if the manner and
intensity of application during almost the entire year is taken into
account. Its effects have not been studied for the type of tropical
ecosystems such as those exists in the Sierra Nevada de Santa
MartaŽ 766.
And it further noted that “the National Police did not respond to prior orders and
767
that no type of research regarding the effect of fumigation has been initiatedŽ .
INDIRENA therefore reiterated the urgent need to carry out the required
768
studies .
Affected by Marijuana Crops and their Destruction by Aerial Spraying with Glyphosate, p. 12 (31
July 1986). ER, Vol. V, Annex 120.
765Republic of Colombia, Ministry of Agriculture, Technical Commission, Specification of the
Terms of Reference for Environmental Research in the Sierra Nevada of Santa Marta Affected by
Marijuana Crops and Spraying with Glyphosate, p. 2 (1986). ER, Vol. V, Annex 121.
766
Ibid., p. 1 (1986).
767Ibid., p. 27 (1986).
768Ibid., p. 2 (1986) (“[T]he Terms of Reference proposed by INDERENA in the document
submitted to the National Narcotics Council on July 1986 are necessary in order to, in an
immediate fashion, further studies as well as the necessary measures that should be taken to
overcome the problem that is ecologically . . . destroying historic areas of the Sierra Nevada de
Santa MartaŽ.).
2934.22 These studies were never done. Eight years later, in 1994, the Ministry of
Environment was still calling for environmental impact studies to be carried out,
and making it clear that such studies remained obligatory notwithstanding the
recent passage of the environmental legislation that the Counter-Memorial cites
to claim an exemption from EIA obligations. In a December 1994 letter to the
Minister of Justice, the Minister of Environment stated: “the Ministry of
Environment will solicit an environmental impact assessment for the fumigations
with glyphosate from the National Narcotics Council, in order to supplement the
environmental measures adopted so that they respond to the new requirements of
769
the new law 99 of 1993Ž .
4.23 Soon thereafter, in February 1995, representatives from the relevant
Colombian governmental agencies, including the Ministry of Environment, the
Agriculture Institute, the Ministry of Health, and the Ombudsman for Health and
Social Security, met to assess whether the Colombian government was in
compliance with the regulations in force concerning the aerial spraying
programme 770. They concluded, among other things, that the aerial spraying was
being carried out “without the environmental licenses that both the Ministry of
Health and the Ministry of the Environment must provideŽ, that the spraying was
769
Letter from Cecilia Lopez Montano, Minister of Environment, Republic of Colombia, to
Nestor Humberto Martinez Neira, Minister of Justice and Law, Republic of Colombia, p. 1 (20
Dec. 1994). ER, Vol. V, Annex 123.
770
Republic of Colombia, Office of the Ombudsman, Meeting Minutes, p. 1 (17 Feb. 1995). ER,
Vol. V, Annex 124.
294“causing impacts to substitution crops, food crops, bodies of water and tropical
forests, including conservation areas and parks which have been declared natural
reservesŽ, and that the “fumigation techniques ha[d] yet to be assessedŽ 771.
4.24 In October 1995, the Environment Ministry ordered the DNE within two
months to design a field study to evaluate the environmental impact of aerial
spraying of a glyphosate-based herbicide, with guidance from the ICA and the
772
Ministry of Environment . Rather than comply with this order, the DNE
773 774
appealed it . When the appeal was rejected in December 1995 , the DNE
775
appealed again. That appeal was denied as well .
4.25 In August 1996, the Environment Ministry issued a categorical rejection
of the DNE’s submission, declaring that it was plagued by “deficiencies and
771Ibid., p. 2 (17 Feb. 1995). The official’s reached other equally negative conclusions, including
that “[t]he dosage and frequency of the glyphosate application has not been prior approved by the
ICAŽ; that “[t]he Anti-narcotics Police, in accordance with orders received from the National
Narcotics Council, decides the areas that are fumigated which it determines at the same moment
that the operation is conducted, in other words, without prior evaluation . . . which makes it
impossible to determine the affected areas with any precisionŽ; and that “[t]he techniques for
direct fumigation have yet to be evaluatedŽ.
772Republic of Colombia, Ministry of Environment, Legal Department, Order No. 558A, p. 2 (13
Aug. 1996). ER, Vol. V, Annex 126.
773
Ibid.
774Republic of Colombia, Ministry of Environment, Legal Department, Order No. 558A, p. 2 (13
Aug. 1996). ER, Vol. V, Annex 126.
775Republic of Colombia, Ministry of Environment, Legal Department, Order No. 557A, p. 2 (13
Aug. 1996). ER, Vol. V, Annex 127.
295technical and logistical flawsŽ 77. Not least of these deficiencies were the
technical assessments that underpinned the use of Turbo Thrush airplanes and the
high concentration of glyphosate, which the Ministry said were “unreliableŽ 777.
As a result, the Ministry of Environment ordered the DNE to carry out “studies
and evaluations [] that permit re-evaluating and determining the parameters and
technical specifications for the fumigationsŽ. This was necessary, the Ministry
said, to “prevent harm or danger to renewable natural resources and to human
778
healthŽ . Given the DNE’s continuing inaction, the Ministry promulgated Terms
of Reference that included the required field study within a more complete set of
779
specific environmental assessment obligations .
4.26 The Terms of Reference included, inter alia, the need to “[i]dentify the
negative impacts that the Program could cause in the different environmental,
780
physical-biological and socio-economic componentsŽ . They also included the
obligation to “[i]ndicate the information deficiencies that generate uncertainty in
776Republic of Colombia, Ministry of Environment, Legal Department, Order No. 558A, p. 1 (13
Aug. 1996) (emphasis added). ER, Vol. V, Annex 126.
777Ibid. (emphasis added).
778Ibid.
779
Ibid, p. 2. Two inter-institutional meetings were held, in May and June 1996, to develop the
terms of reference for the EMP. The DNE … an invited and critical participant in this Plan …
notably did not participate in the second meeting finalizing the EMP’s terms. See Republic of
Colombia, Ministry of Environment, Division of Environmental Licenses, Order No. 599, p. 2 (23
Dec. 1999). ER, Vol. V, Annex 132.
780
Republic of Colombia, Ministry of Environment, Legal Department, Order No. 558A, p. 5 (13
Aug. 1996). ER, Vol. V, Annex 126.
296the estimation, scale and evaluation of the impactsŽ 78. In addition, the Ministry
of Environment required the DNE to include in the EMP a chapter devoted to the
“Identification and Evaluation of ImpactsŽ, which had to provide an “evaluation
and ranking of temporary or permanent impacts produced by the Program . . .
782
Ž . Further, the Ministry ordered the DNE to compare an unsprayed area with
areas that were sprayed. Among the subjects that had to be evaluated were the
risks to human health, animal species, the ecosystems, and air, soil and water
quality783. In assessing these risks the study had to evaluate:
x “Potential toxic or adverse effects that may result from
herbicides, metabolites, or the products resulting from their
transformation
x Physical and chemical properties of the herbicides
associated with their distribution, mobility, and persistence
in various environmental compartments
x Environmental characteristics (physical-biotic and socio-
economic aspects) of the exposed areas
x Conditions of exposure, type of exposure (direct or
indirect), concentration, duration, and routes of exposure
x In all cases, there will be a description of the methodology
used for evaluating impacts, as well as for establishing the
parameters and criteria for scoring and ranking them,
including the following considerations:
o The nature and magnitude of the negative effect
o The probability of occurrence
o Resources affected and ecological significance of
the effect
o Reversibility
781
Ibid.
782Ibid., p. 15 (emphasis in original).
783Ibid.
297 o The impact’s area of influence
o Lack of information causing any uncertainty or
limiting the scope of the evaluationŽ 784.
The Ministry also ordered that the evaluation analyze both “unavoidableŽ and
785
“cumulativeŽ impacts . The Ministry expressly ordered the DNE to consider
risks to human health and the territory of indigenous peoples, as well as
786
environmental harm generally .
4.27 Nearly two years elapsed after the Environment Ministry issued these
Terms of Reference, but the DNE did not comply. As a result, the Ministry was
compelled to repeat its order for the DNE to submit the required
787
documentation . The Ministry could not have been clearer that the DNE’s
failure to present these materials breached its legal obligations; it declared that
“the DNE could not continue evading what the Constitution and the law required
788
of itŽ . The Ministry therefore warned that unless the situation was remedied, it
789
would be required to take “other actionŽ .
784Republic of Colombia, Ministry of Environment, Legal Department, Order No. 558A, pp. 15-
16 (13 Aug. 1996). ER, Vol. V, Annex 126.
785Ibid., pp. 18-19.
786Ibid., p. 17.
787Republic of Colombia, Ministry of Environment, Division of Licenses, Technical Report No.
419.99, p. 4 (21 Dec. 1999). ER, Vol. V, Annex 131.
788Ibid. (emphasis added).
789
Ibid.
2984.28 When the DNE finally submitted a response, the Environment Ministry
promptly rejected it. And for good reason. The DNE’s submission was missing
the entire Chapter that was supposed to present the “Identification and Evaluation
790
of Environmental ImpactsŽ .
4.29 In light of the DNE’s failure to carry out the required impact assessments,
791
the Environment Ministry repeatedly reiterated its demand for them . When the
792
full report finally arrived, the Ministry again rejected it .
4.30 The Counter-Memorial’s reticence about the DNE’s foot-dragging in
regard to EIA is not surprising, in light of the criticisms that Colombia’s own
government agencies levelled against it. These criticisms, which are set forth in a
document dated 23 December 1999 … just days before Colombia began to spray
along the border with Ecuador in January 2000 … further confirm that the
environmental impacts of the aerial spraying had not been evaluated prior to the
790
Letter from Guillermo Acevedo Mantilla, Subdirector of Environmental Licenses, Ministry of
Environment, Republic of Colombia, to Ruben Olarte Reyes, Director, National Drug Directorate,
Republic of Colombia (13 Nov. 1998). ER, Vol. V, Annex 130.
791
Letter from Guillermo Acevedo Mantilla, Subdirector of Environmental Licenses, Ministry of
Environment, Republic of Colombia, to Ivon Alcala Arevalo, Director, National Drug
Directorate, Republic of Colombia (8 Oct. 1998). ER, Vol. V, Annex 129; Letter from Guillermo
Acevedo Mantilla, Subdirector of Environmental Licenses, Ministry of Environment, Republic of
Colombia, to Ruben Olarte Reyes, Director, National Drug Directorate, Republic of Colombia (13
Nov. 1998). ER, Vol. V, Annex 130.
792Republic of Colombia, Ministry of Environment, Division of Licenses, Technical Report No.
419.99, p. 4 (21 Dec. 1999). ER, Vol. V, Annex 131; Republic of Colombia, Ministry of
Environment, Division of Environmental Licenses, Order No. 599, pp. 31-33 (23 Dec. 1999). ER,
Vol. V, Annex 132. See also Colombian Ministry of Environment, Resolution No. 341 of 2001,
p. 2. EM, Vol. II, Annex 14.
299 793
spraying of areas immediately adjacent to Ecuador . Indeed, it is not just that no
prior EIA was carried out; the evidence is that the DNE refused to do it.
4.31 Among the government agencies critical of the DNE for its refusal to
conduct impact studies was the Colombian Agriculture Institute (“ICAŽ) … an
agency under the jurisdiction of the Ministry of Agriculture. Its criticisms
included the following:
x The ICA rejected the DNE’s assertion that the equipment and spray
mixture used in the programme “guarantee[d] that the ecosystem will not
794
suffer any damageŽ . In particular, the ICA stated that the DNE had
never evaluated spraying with fixed-wing aircraft, but instead had only
assessed spraying with helicopters, which do not fly nearly as high or fast,
795
“despite the fact that applications [we]re being carried out via airplaneŽ .
The ICA further stated that the relevant “technical conceptsŽ had “not
796
been taken into accountŽ .
x The ICA disputed the DNE’s assertion that it had decided which
territories to spray in coordination with the ICA. Specifically, it stated
793
Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 599 (23 Dec. 1999). ER, Vol. V, Annex 132.
794
Ibid., p. 11.
795
Ibid.
796Ibid.
300 that after the initiation of the spraying programme the ICA had “never
797
againŽ been involved . Nor, the ICA said, was it aware of any of the
recent inter-institutional “assessment meetingsŽ that the DNE claimed had
798
taken place during which the programme was reassessed . To the
contrary, the ICA stated that the last meetings of which it was aware had
taken placed 15 years earlier, in 1984 799. The experts convened at those
earlier meetings, which Ecuador described in the Memorial, had decided
against aerial spraying, stating: “[T]he Committee reiterates its opposition
of not recommending the use of glyphosate or any other herbicide by
aerial application for the destruction of marijuana crops . . . the
implementation of the program is advised against because it would be
accepting experimentation on humansŽ 800.
x The ICA found serious flaws in the DNE’s discussion of nearly every
operational parameter relevant to spray drift, including height, speed and
application rate 801. It therefore rejected the DNE’s assertion that spray
drift would be limited to 2 to 5 metres. In particular, the ICA declared
797
Ibid.
798
Ibid, p. 12.
799Ibid.
800See Memorial of Ecuador, Chap. 2, para. 2.34 (28 Apr. 2009) (hereinafter “EMŽ).
801Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 599, pp. 12-14 (23 Dec. 1999). ER, Vol. V, Annex 132.
301 that “the height and speed of operationŽ made that level of drift
“impossible to achieve with airplanesŽ 80.
x The ICA determined that much more information was needed to identify
the impacts caused by spray drift and to understand its implications for the
environment and human health. The ICA explained that, given the toxicity
of the spray mixture and the lack of information on its potential impact,
the Ministry of Environment should require assessments to evaluate the
spray’s impact in Colombia’s “environmentŽ 803.
4.32 The Ministry of Environment was particularly critical of the DNE’s
failure to take impact assessments seriously. Its Ecosystems Department wrote
that the insufficient level of detail provided in the DNE’s environmental
characterization of the areas to be assessed made it “impossible to carry out
evaluations or rankings of the temporary and permanent impacts caused by the
eradication efforts, and therefore, to define the control, prevention, compensation
804
and recuperation measures for the EMPŽ . The lack of specificity in the plan
meant that there was:
802The ICA observed that the DNE would be more likely to achieve that level of drift if it used
helicopters. Ibid., p. 14 (emphasis added).
803
Ibid., p. 16.
804
Ibid., 19.
302 “no accuracy regarding the critical and risk areas corresponding to
the different degrees of erosion and geotechnical stability, the
inventory of the primary uses of water related to municipal
aqueduct supply, fields and production activities, and specific
characterizations regarding qualitative and quantitative vegetative
aspects, the presence of endemic flora and fauna species in areas
of direct influence within eradication areas, basic parameters of
impact identification and evaluationŽ 80.
In other words, there was no way to determine the type and extent of harm that
could be caused and would need to be assessed in the areas to be sprayed.
4.33 The Subdivision for the Planning and Management of National Parks of
the Special Administrative Unit for National Parks was equally critical. It stated
that the DNE had not evaluated the vulnerabilities of the ecosystems exposed to
the spray, had failed to evaluate the impacts caused by the sprayings, had failed to
secure the required approval from the Ministry of Environment for areas to be
studied, and had failed to address the legal requirement to establish buffer zones
around sensitive areas, such as national parks 806.
4.34 As was to be expected, the DNE challenged this order in an administrative
807
appeal .
805Ibid., pp. 19-20.
806
Ibid., pp. 21-24.
807
Colombian Ministry of Environment, Resolution No. 341 of 2001, p. 2. EM, Vol. II, Annex
14.
3034.35 Thus, the state-of-affairs with respect to environmental impact
assessment as of 23 December 1999 … that is, just before aerial spraying along the
border with Ecuador began in January 2000 … was as follows: The only study
that Colombia felt was worth mentioning in the Counter-Memorial’s section on
“Studies Leading to the Environmental Management PlanŽ had occurred over a
decade earlier, in 1987, and was totally inadequate; and the Colombian
Environment Ministry had repeatedly demanded impact studies for many years,
but was refused at every turn by the DNE.
Section II. Colombia’s Failure to Assess Environmental Impacts After the
Aerial Spraying Programme Commenced
A. T HES TUDIES INVOKED IN THE C OUNTER -M EMORIAL
4.36 The first study mentioned in the Counter-Memorial that allegedly
assessed the impacts of the aerial spraying programme is the Solomon study,
which was not completed until March 2005 808. Colombia thus concedes that
between January 2000 (when spraying near Ecuador began) and March 2005
(when the Solomon study was completed), there were no assessments that qualify
as an EIA 80. As a result, the Counter-Memorial does not contest that, for over
five years, Colombia sprayed vast quantities of chemical herbicides in areas
immediately adjacent to Ecuador without having assessed environmental impacts,
80CCM, Chap. 4, paras. 4.15-4.17.
80See CCM, Chap. 4, 4.8-4.17.
304let alone transboundary impacts. Indeed, by the time the Solomon study was
completed in March 2005, Colombia had sprayed within 10 kilometres of
810
Ecuador no fewer than 64,285 times .
4.37 The 2005 Solomon study did not purport to be a transboundary EIA. It
was a report conducted at the request of the Governments of the United States,
811
the United Kingdom and Colombia . By its own terms, it was confined to
812
assessing impacts in Colombia exclusively . Moreover, the value of this or any
other scientific study necessarily depends on the quality of the data or
assumptions upon which it is based. In this case, the Solomon study was based
on factual representations given by Colombia … that the spray planes strictly
complied with all operational requirements regarding flight speed, altitude,
application rate, droplet size, and time of day … representations which the
evidence described in Chapter 2 of this Reply now shows were false.
Accordingly, contrary to the Counter-Memorial, the 2005 Solomon study is not
810
R. John Hansman, Ph.D. & Carlos F. Mena, Ph.D., Analysis of Aerial Eradication Spray
Events in the Vicinity of the Border Between Colombia and Ecuador from 2000 to 2008,
Appendix 3, p. 29 (Jan. 2011) (hereinafter “Hansman & Mena ReportŽ). ER, Vol. II, Annex 1.
811
Keith R. Solomon et al., Environmental and Human Health Assessment of the Aerial Spray
program for Coca and Poppy Control in Colombia, p. 2 (31 Mar. 2005) (“This report was
prepared . . . in response to requests from the Governments of Colombia, the United Kingdom,
and the United States of AmericaŽ.) (hereinafter “Solomon et al., 2005Ž). CCM, Vol. III, Annex
116. See also 2004 Memorandum of Understanding between the Government of Colombia and
the Organization of American States (OAS) for the execution of the study on the effects of the
Program for the Eradication of Illicit Crops by aerial spraying with Glyphosate Herbicide
(PECIG) and of illicit crops, on human health and environment, pp. 8-9 (4 Feb. 2004). CCM,
Vol. III, Annex 113.
812Ibid, p. 9.
305only not an EIA, but it is also devoid of any sustainable findings about the
environmental impacts of the aerial spraying programme in either Colombia or
Ecuador. It certainly has nothing useful to say about spray drift … into Ecuador or
813
anywhere else .
4.38 Further undermining the reliability of the 2005 Solomon study, it did not
purport to assess the impacts of spray mixtures that Colombia actually used. As
described in Chapter 2, at least four different formulations have been reported:
814
Roundup Export, Roundup Ultra, Roundup SL and GLY-41 . The Solomon
study does not state that any of these products were evaluated. All the study says
is that its authors received a toxicity assessment that had been conducted three
years earlier by a Colombian laboratory of a “sample product GLYPHOSATE
44% + COSMOFLUX 1% + WATER 55%Ž 81. If the report of the Colombian
813For example, the Solomon study assumed that the spray droplets were 300-1,500 m in
diameter. Solomon et al., 2005, op. cit., p. 28. CCM, Vol. III, Annex 116. Nevertheless, later
research by the same research team revealed that at least 50 percent of the droplets were less than
128 m in size. A.J. Hewitt et al., “Spray Droplet Size, Drift Potential, and Risks to Nontarget
Organisms from Aerially Applied Glyphosate for Coca Control in ColombiaŽ in Journal of
Toxicology and Environmental Health, Part A, 72:921, p. 921 (2009) (hereinafter “Hewitt et al.,
2009Ž). CCM, Vol. III, Annex 131-B. As discussed in Chapter 2, droplet size is one of the most
critical parameters influencing spray drift, and this drastic overestimation of the size of the
droplets would materially affect any conclusions regarding impacts caused by spray drift. In
addition, the Solomon study assumed that the spray was released from a height of 30 meters.
Solomon et al., 2005, op. cit., p. 30. CCM, Vol. III, Annex 116. Again, Solomon’s assumptions
were far off the mark, invalidating any of its conclusions regarding the health and environmental
impacts of spray drift.
814
See supra Chap. 2, paras. 2.18-2.42.
815O. Saavedra, Laboratorio Immunopharmos Ltda., Toxicity Study on Laboratory Animals for
two concentrations of Glyphosate 44% + Cosmoflux 1% + Water 55%, Bogotá, p. 281 (15 Feb.
2002). CCM, Vol. III, Annex 128; Solomon et al., 2005, op. cit., pp. 104-105. CCM, Vol. III,
Annex 116.
306laboratory is taken at face value, no formulated version of glyphosate was
assessed, since it refers only to “GlyphosateŽ, thus indicating that glyphosate in
its pure form was tested, not a formulated product such as Roundup Export or
GLY-41 that contains surfactants like POEA (which the Counter-Memorial
admits is present in all formulations that have been sprayed) 816. Regardless,
since only one mixture was assessed, at least three of the four products that
Colombia has sprayed were not studied.
4.39 The 2005 Solomon study identified numerous topics relating to
environmental risk and protection of human health that it did not address, even
though it stressed their importance. Colombia’s own expert acknowledges that
the “Solomon et al. report identified data gaps and areas with insufficient
information to conclude a risk assessment satisfactorily and outlined a further
programme of research to fill these gapsŽ 817. Chief among these was the absence
of any proper study of spray drift, since it had “not been measured under
818
conditions of use in ColombiaŽ . Since spray drift is the principal means by
816
CCM, para. 4.50, n. 312.
817Dr. Stuart Dobson, Critique of “Evaluation of Chemicals Used in Colombia’s Aerial Spraying
Program, and Hazards Presented to People, Plants, Animals and the Environment in EcuadorŽ
Menzie et al., p. 528 (2009) (hereinafter “Dobson ReportŽ). CCM, Vol. I, Appendix.
818 2006 Memorandum of Understanding between the General Secretariat of the Organization of
American States (SG/OAS) and the Government of Colombia for the execution of the study on the
effects of the Program for the Eradication of Illicit Crops by aerial spraying with Glyphosate
Herbicide (PECIG) on human health and the environment, p. 228 (23 May 2006). CCM, Vol. III,
Annex 119. Inter-American Drug Abuse Control Commission (CICAD), Second Phase
Environmental and Human Health Assessment of the Aerial Spray Program for Coca and Poppy
Control in Colombia, p. 1 (date unknown). EM, Vol. III, Annex 159.
307which Ecuador has been affected by the aerial spraying programme, the study’s
failure to assess drift confirms that Colombia failed to evaluate the programme’s
transboundary impacts, even as late as five years into it. Even the Counter-
Memorial concedes that the 2005 study left numerous “remaining uncertaintiesŽ
819
that were “identified as subjects for further studyŽ . These included, according
to Colombia, “the issues of spray drift, the effects on sensitive wildlife such as
820
amphibians and the effects on humansŽ . For example, although spraying had
been ongoing for over five years near Ecuador, the impact on amphibians (which
are particularly sensitive to herbicides) had not been assessed as of 2005. In that
regard, even as late as 2009, the authors of the study could state that “there was
no data on the susceptibility of amphibiansŽ to the spray used in the eradication
programme 821. The Solomon Study also observed that the spray’s impact on
human health in important respects had not been adequately studied, including
with respect to reproductive health 82.
819CCM, Chap. 7, para. 7.12.
820
Ibid.
821
K.R. Solomon et al., “Human Health and Environmental Risks from the Use of Glyphosate
Formulations to Control the Production of Coca in Colombia: Overview and ConclusionsŽ, in
Journal of Toxicology and Environmental Health, Part A 72:914-920, p. 915 (2009) (hereinafter
“Solomon et al., 2009Ž). CCM, Vol. III, Annex 131-A. See also Memorandum of Understanding
between the General Secretariat of the Organization of American States (SG/OAS) and the
Government of Colombia for the execution of the study on the effects of the Program for the
Eradication of Illicit Crops by aerial spraying with Glyphosate Herbicide (PECIG) on human
health and the environment, pp. 228-229 (23 May 2006). CCM, Vol. III, Annex 119.
822Ibid, pp. 914, 917-919. See also Memorandum of Understanding between the General
Secretariat of the Organization of American States (SG/OAS) and the Government of Colombia
for the execution of the study on the effects of the Program for the Eradication of Illicit Crops by
3084.40 Ecuador submits that a document that fails to assess these matters cannot
be deemed an adequate impact assessment, especially in a transboundary context.
4.41 The authors of the 2005 Solomon study emphasized the need for studies
of spray drift. They proposed a specific study, to be completed by 2007, which
was an aerial application “field experimentŽ to be conducted “under conditions in
Colombia to determine spray driftŽ 82. The contemplated study was described as
follows:
“Three non-overlapping spray swaths will be sprayed and spray
targets (water-sensitive paper and filter papers for dye catching)
set out in transects so that a spatial representation of deposition
can be mapped. Test plants of pot-grown maize will also be
placed with targets to evaluate the biological impact of drift
outside the garget area. Applications can be repeated for four
wind speeds (1, 2, 4, and 8 knots). A comparison of predictions
with EPA drift models will evaluate the accuracy of risk
predictions from drift events. Different spray heights will be used
824
as well (30, 50, and 70 m)Ž .
4.42 The Counter-Memorial made no mention of this study or its results. The
Court is thus left with two possibilities: either the study was never conducted,
aerial spraying with Glyphosate Herbicide (PECIG) on human health and the environment, p. 230
(23 May 2006). CCM, Vol. III, Annex 119.
823Memorandum of Understanding between the General Secretariat of the Organization of
American States (SG/OAS) and the Government of Colombia for the execution of the study on the
effects of the Program for the Eradication of Illicit Crops by aerial spraying with Glyphosate
Herbicide (PECIG) on human health and the environment, p. 230 (23 May 2006). CCM, Vol. III,
Annex 119.
824
Inter-American Drug Abuse Control Commission (CICAD), Second Phase Environmental and
Human Health Assessment of the Aerial Spray Program for Coca and Poppy Control in Colombia,
p. 4 (date unknown). EM, Vol. III, Annex 159.
309despite the recommendation by the authors of the 2005 Solomon study, or
Colombia has chosen not to present the results.
4.43 The authors of the 2005 study also proposed a “wind tunnelŽ study “to
characterize the spray droplet spectrum for the nozzles and speed of spraying in
825
ColombiaŽ . This study was carried out by Dr. Andrew Hewitt as part of a
group of studies conducted in collaboration with Dr. Solomon, but not until 2009,
that is, nine years after spraying began near the border and two years after
Ecuador submitted its Application to the Court 826. It was paid for by the
Governments of Colombia and the United States 82.
4.44 The main deficiency in the 2009 Hewitt et al. study … indeed, its fatal flaw
… was already identified in paragraphs 2.191 and 4.7: its predictions of spray drift
and toxicity levels at various distances from the target depended on accurate data
regarding spray plane flight speed and altitude of dispersion, among other
825
Memorandum of Understanding between the General Secretariat of the Organization of
American States (SG/OAS) and the Government of Colombia for the execution of the study on the
effects of the Program for the Eradication of Illicit Crops by aerial spraying with Glyphosate
Herbicide (PECIG) on human health and the environment, p. 230 (23 May 2006). CCM, Vol. III,
Annex 119.
826Hewitt et al., 2009. CCM, Vol. III, Annex 131-B. The Hewitt et. al. (2009) study was part of
a group of studies conducted by the Solomon et al. team, and published in 2009. See CCM, Vol.
III, Annexes 131-A to 131-I.
827See page one of CCM Annexes 131-A through 131-I (“This paper was prepared as part of a
Study entitled ‘Production of Illicit Drugs, the Environment and Human Health,’ financed with
contributions from the Governments of Colombia and the United States of America. The
conclusions and opinions expressed herein are those of the authors and not necessarily those of the
Organization of American States and its General Secretariat, which as of the date of this
copyright, have not formulated any opinion with respect to themŽ.).
310important parameters; but the “dataŽ provided by Colombia were far from
accurate. In fact, Colombia apparently provided no data to the experts it
contracted to perform the study; there is no indication in the reports that Hewitt et
al. received or reviewed any actual spray flight data, even though it had been
automatically generated by each of thousands of spray flights between 2000 and
2009. Instead, the experts made the assumption (or were instructed by Colombia)
that the spray flights rigorously observed the operational requirements mandated
by the EMP in regard to these parameters, and these are the numbers they fed into
their model … even though, as shown in Chapter 2, supra, they bear no relation to
reality. Thus, the 2009 Hewitt et al. study was predicated upon assumptions that
do not reflect the actual spray programme and the conclusions it reached based on
these assumptions have no validity.
4.45 The set of Solomon et al. studies from 2009 also lacks merit because, like
Solomon 2005, they purported to measure the properties, including toxicity levels
at various distances of drift, of a substance other than the one Colombia actually
sprays (or sprayed). The Counter-Memorial fails to mention that the only product
Solomon et al. evaluated in 2009 was a chemical called “GlyphosŽ 82. This name
828
Solomon et al., 2009, op. cit., pp. 915, 916. CCM, Vol. III, Annex 131-A; Hewitt et al., 2009,
op. cit., pp. 921, 922. CCM, Vol. III, Annex 131-B; E.J.P. Marshall et al., “Coca (Erythroxylum
coca) Control is Affected by Glyphosate Formulations and AdjuvantsŽ, in Journal of Toxicology
and Environmental Health, Part A 72:930-936, p. 930 (2009). CCM Annex 131-C; R.A. Brain et
al., “Comparison of the Hazards Posed to Amphibians by the Glyphosate Spray Control Program
versus the Chemical and Physical Activities of Coca Production in ColombiaŽ, in Journal of
Toxicology and Environmental Health, Part A 72:937-948, p. 937 (2009). CCM, Vol. III, Annex
311does not match any of the products that the Counter-Memorial says have been
sprayed (Roundup SL and GLY-41); nor does it correspond to the product
mentioned in the U.S. government records (Roundup Export), or to the
formulation that Colombia referenced in diplomatic exchanges with Ecuador
(Roundup Ultra) 829. Thus, it does not appear that the study addressed any of the
relevant spray mixtures used along the border with Ecuador.
4.46 The Counter-Memorial informs the Court that Solomon et al. 2009
830
evaluated impacts on amphibians . However, even the experts’ unrealistically
low estimate of the spray’s “normal applicationŽ rate resulted in 30-35%
831
mortality for three species . That is a disturbing result for an herbicide that is
131-D; M.H. Bernal et al., “Toxicity of Formulated Glyphosate (Glyphos) and Cosmo-Flux to
Larval and Juvenile Colombian Frogs 1. Field and Laboratory Microcosm Acute ToxicityŽ, in
Journal of Toxicology and Environmental Health, Part A, 72:961-964, p. 961 (2009). CCM, Vol.
III, Annex 131-F; M.H. Bernal et al., “Toxicity of Formulated Glyphosate (Glyphos) and Cosmo-
Flux to Larval and Juvenile Colombian Frogs 2. Field and Laboratory Microcosm Acute
ToxicityŽ, in Journal of Toxicology and Environmental Health, Part A, 72:966-973, p. 966
(2009). CCM Annex 131-G; C. Bolognesi et al., “Biomonitoring of Genotoxic Risk in
Agricultural Workers from Five Colombian Regions: Association to Occupational Exposure to
GlyphosateŽ, in Journal of Toxicology and Environmental Health, Part A 72:986-997, p. 988
(2009). CCM, Vol. III, Annex 131-I. Dr. Dobson asserted that GLY-41 is Glyphos (“Glyphos as
Gly41Ž), but provided no basis for that assertion. Dobson Report, op. cit., p. 521. CCM, Vol. I,
Appendix.
829See supra Chap. 2, paras. 2.18-2.42.
830CCM, Chaps. 3 and 7, paras. 3.56, 7.12, 7.21, 7.76.
831M.H. Bernal et al., “Toxicity of Formulated Glyphosate (Glyphos) and Cosmo-Flux to Larval
and Juvenile Colombian Frogs 2. Field and Laboratory Microcosm Acute ToxicityŽ, in Journal of
Toxicology and Environmental Health, Part A, 72:966-973, p. 971, Table 3 (2009). CCM, Vol.
III, Annex 131-G. See also Charles A. Menzie, Ph.D. & Pieter N. Booth, M.S., Response to:
“Critique of Evaluation of Chemicals Used in Colombia’s Aerial Spraying Program, and Hazards
Presented to People, Plants, Animals and the Environment in Ecuador,” As Presented in:
Counter-Memorial of the Republic of Colombia, Appendix, pp. 19-23 (Jan. 2011) (describing the
deficiencies of the Solomon et al. and Dobson studies on amphibians). ER, Vol. II, Annex 6.
312indiscriminately applied to the home of some of the earth’s most unique,
832
concentrated, and vulnerable amphibian populations . Also troubling is the fact
833
that other spray mixtures were considered less toxic to aquatic life . In other
words, Colombia sprayed for almost a decade near the border … discharging
hundreds of thousands of gallons of a formulated herbicide mixture within 10
kilometres of Ecuador in 114,525 different aerial spraying events 834 … before
considering which substances would be the least harmful to animal-life, and when
a study finally evaluated that issue, Colombia disregarded the answer.
4.47 In sum, the evidence presented in the Counter-Memorial to support
Colombia’s claim to have discharged its EIA obligations, or to have assured itself
that its aerial spraying programme did not affect Ecuador, falls far short of the
mark. Neither of the two Solomon studies cited in the Counter-Memorial … both
of which occurred long after the spray programme commenced near Ecuador …
can be characterized as an EIA. Neither addressed risks to Ecuador, and both
relied upon grossly inaccurate data in regard to evaluation of spray drift.
832
See EM, Chap. 2, paras. 2.13-2.14. Henrik Balslev, Ph.D., The Vulnerability of the Ecuador-
Colombia Border Region to Ecological Harm, pp. 15, 19, 20, 21, 34, 37, 44 (Jan. 2011) (“Ecuador
is ranked third in amphibian diversity worldwide with 415 described species . Only Brazil and
Colombia have more species than Ecuador. . . . Amphibians may serve as indicators of more
extensive environmental change because they are sensitive to environmental contamination and
live in both aquatic and terrestrial environmentsŽ.). ER, Vol. II, Annex 4.
833
E.J.P. Marshall et al., “Coca (Erythroxylum coca) Control is Affected by Glyphosate
Formulations and AdjuvantsŽ, in Journal of Toxicology and Environmental Health, Part A
72:930-936, p. 930 (2009). CCM Annex 131-C.
834
R. John Hansman, Ph.D. & Carlos F. Mena, Ph.D., Analysis of Aerial Eradication Spray
Events in the Vicinity of the Border Between Colombia and Ecuador from 2000 to 2008, pp. 1, 2,
11 (Jan. 2011). ER, Vol. II, Annex 1.
313 B. T HEC OLOMBIAN ENVIRONMENT M INISTRY’SO NGOING FAILURE TO
SECURE AN EIA
4.48 As shown above, Colombia’s Environment Ministry tried, but failed, to
get the executing agency, the DNE, to carry out environmental impact studies for
the aerial spraying programme prior to the first known spray flights along the
border with Ecuador in 2000. The evidence shows that the Environment Ministry
continued its effort to have the environmental impacts of the programme assessed
through 2003, with the same effect: total failure. The Counter-Memorial
mentions none of this, or that not only the Environment Ministry, but also the
Comptroller General and the National Ombudsman repeatedly demanded impact
studies from DNE during this period, but were completely rebuffed. In fact, the
Environment Ministry even fined the DNE for failing to present the required
835
impact assessments .
4.49 In March 2000, two months into the aerial spraying programme in the
border region, the Environment Ministry issued an order that:
x Required the DNE to carry out an analysis of the aerial spraying’s
environmental impacts over time. The Ministry reiterated that in order to
assess potential impacts, it was “necessary to correctly identify the
negative impacts on different environmental, physical-biotic and socio-
83See Republic of Colombia, Ministry of Environment, Resolution 0670, Whereby a sanction is
imposed and other decisions are made, p. 13 (19 Jun. 2003). ER, Vol. V, Annex 148. See also
Rojas Report, op. cit., para. 78. ER, Vol. II, Annex 8.
314 economic componentsŽ 836. The Ministry further explained that the DNE
could not fulfil this obligation by relying, as it had, on secondary
837
international literature .
x Ordered the DNE to conduct studies regarding the spray’s impacts on soil
and animal species, including aquatic organisms. The Ministry specified
that the DNE’s responses to date had merely “open[ed] a range of
information gaps regarding the effects of the glyphosate used in the doses
838
and spray conditions of the Illegal Crop Eradication Program.Ž .
x Required the DNE to explain whether the technical parameters
established for the spray operations had taken into consideration the use
839
of Cosmo Flux 411F .
x Required the DNE to consider measures to address potential harm to
territories bordering national parks, rather than merely stating the special
840
measures taken within national parks .
836
Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 143, pp. 26, 31 (29 Mar. 2000). ER, Vol. V, Annex 133.
837
Ibid., pp. 27.
838Ibid., pp. 29, 31.
839Ibid., pp. 26, 31.
3154.50 The Environment Ministry ordered the DNE to produce these impact
841
studies within three months . The DNE appealed and requested a one-year
delay. The appeal was denied 842. But, still nothing changed. In August 2000, the
Colombian National Narcotics Council ordered changes to the aerial spraying
programme because of the need to address complaints of “damagesŽ to “the
people, the environment, and agricultural and livestock activitiesŽ by “spraying
with glyphosateŽ 84. It acknowledged, in that regard, the need “to strengthen
effective control, follow-up, and monitoring mechanisms that make it possible to
evaluate environmental, agronomic, and health impacts generated by the illicit
844
crops eradication programŽ . Specifically, the Council required the Colombian
National Police and the DNE to assess risks to human health, the environment,
845
agriculture and livestock in the targeted areas . It directed these agencies to
obtain and analyze local information on the “environmental contextŽ of areas
considered for aerial spraying to “determin[e] potential risk for human health, the
840Republic of Colombia, Ministry of Environment, Division of Environmental Licenses, Order
No. 143, pp. 25, 30 (29 Mar. 2000). ER, Vol. V, Annex 133.
841Ibid., p. 31 (29 Mar. 2000).
842See Colombian Ministry of Environment, Resolution No. 341 of 2001, p. 2. EM, Vol. II,
Annex 14. See also, Rojas Report, op. cit., para. 78. ER, Vol. II, Annex 8.
843Resolution No. 005 of 11 August 2000 of the National Narcotics Council of Colombia, p. 118.
CCM, Vol. II, Annex 41.
844
Ibid. See also Rojas Report, op. cit., paras. 54-56. ER, Vol. II, Annex 8.
845Resolution No. 005 of 11 August 2000 of the National Narcotics Council of Colombia, p. 119.
CCM, Vol. II, Annex 41. The resolution also stated, in the only section that Colombia did not
translate in its annex, that the spray operations had been required to comply with several
environmental and health protection measures since as early as 1986. It also reiterated the
requirement, dating back to 1991, that the Health Ministry carry out epidemiological monitoring
plans and implement sanitary controls for use of the herbicide. Republic of Colombia, National
Narcotics Council, Resolution No. 005, pp. 26-27 (11 Aug. 2000). ER, Vol. V, Annex 135.
316environment, and agriculture and livestockŽ for those areas selected “for aerial
spraying with glyphosateŽ 846. The Council also ordered the DNE to hire an
external technical auditor to carry out an “assessment on impacts on the
847
environment, human health, and on agricultural activitiesŽ . Although the
Counter-Memorial briefly refers to the retention of an auditor to ensure that the
spray mixture “conforms with the relevant regulations as to permitted
848
composition and dosage under Colombian lawŽ , it neglects to say that the DNE
was also required to contract for an audit of whether operations are carried out as
required by law and to analyze impacts on the environment, human health and
agricultural activities 849.
4.51 Regardless, the DNE again failed to comply. The Environment Ministry
concluded in September 2000 that “DNE has not undertaken in any of the
submitted documents, a technical or methodological assessment of impact
850
identificationŽ or of environmental risks. It further found that the DNE’s
846Resolution No. 005 of 11 August 2000 of the National Narcotics Council of Colombia, p. 119.
CCM, Vol. II, Annex 41. See also Rojas Report, op. cit., para. 55. ER, Vol. II, Annex 8.
847Resolution No. 005 of 11 August 2000 of the National Narcotics Council of Colombia, p. 122.
CCM, Vol. II, Annex 41.
848CCM, Chap. 6, para. 6.20.
849The Ministry of Environment also pointed out the DNE’s attempt to substitute the terms and
function of the external audit in Resolution 670 of 2003, sanctioning the DNE for its continuing
EMP violations. Republic of Colombia, Ministry of Environment, Resolution 0670, Whereby a
sanction is imposed and other decisions are made, p. 6 (19 June 2003). ER, Vol. V, Annex 148.
850 Republic of Colombia, Ministry of Environment, Division of Environmental Licenses,
Technical Report, No. 589,Evaluation of the Additional Information Provided by the National
Narcotics Directorate, p. 9 (20 Dec. 2000). ER, Vol. V, Annex 136. See also, Rojas Report, op.
cit., para. 78. ER, Vol. II, Annex 8.
317response was “not environmentally viable . . . due to the uncertainty regarding
[the programme’s] potential environmental impactsŽ 851.
4.52 In December 2000, the Environment Ministry, having had all of its prior
requests for impact studies frustrated, demanded that representatives of the DNE
appear before it. At the ensuing meeting, the Ministry again informed the DNE
that the information it had provided, and its level of cooperation more generally,
852
“did not satisfy the [Ministry’s] requirementsŽ . The Counter-Memorial fails to
mention this meeting or its aftermath, but it is clear from statements by the
853
Colombian National Ombudsman that the DNE continued to stonewall on EIA .
4.53 When the DNE submitted a new response, the Ministry of Environment
854
was again compelled to reject it in May 2001 . The reason: none of the 10
topics it addressed evaluated environmental risks 855. Specifically, the Ministry
declared:
“The analysis to determine which areas are environmentally
affected by the eradication program is not supported from a
technical or scientific point of view. It contains neither qualitative
851
Ibid., pp. 13-14.
852
Colombian Ministry of Environment, Resolution No. 341 of 2001, p. 3. EM, Vol. II, Annex
14.
853
Republic of Colombia, Office of Ombudsman, The Implementation of the Strategy of Aerial
Eradication of Illicit Crops With Chemicals, From a Constitutional Perspective, p. 34 (Apr.
2003). ER, Vol. V, Annex 146.
854
Colombian Ministry of Environment, Resolution No. 341 of 2001, p. 9. EM, Vol. II, Annex
14. See also Rojas Report, op. cit., para. 78. ER, Vol. II, Annex 8.
855
Colombian Ministry of Environment, Resolution No. 341 of 2001, p. 3 et seq. EM, Vol. II,
Annex 14.
318 nor quantitative supporting information, nor any supporting
research on programs related to the purpose of the evaluation that
could allow a determination and evaluation of the impacts caused
to natural resources by the application of Glyphosate. Therefore,
the risk assessment is in a very general form, which does not allow
for a detailing or specification of actions to determine risks and to
plan for their management in advanceŽ 856.
4.54 In light of this critique, the Ministry “concludedŽ that the “documents that
have so far been submittedŽ by the DNE did not satisfy “the scope and objectives
defined in the terms of reference, nor the informational requirements that this
Ministry has repeatedly requested from the National Narcotics Directorate
857
(DNE)Ž . In making this determination, the Environment Ministry made the
following specific criticisms:
x There was “no presentation of an evaluation of the supply and
vulnerability of the individual ecosystems and natural resources
contained within the core areas of illicit crops which are the object of
the program, based on which, from a technical point of view,
environmentally critical, sensitive and important ecosystems could be
determined and differentiated, along with the areas that should be
excluded, treated or specially managed in the development and
execution of the program. Nevertheless, the program has resources
and technological tools available . . . to comply with the requirements
of both the environmental plan and the stipulations of the National
858
Narcotics Council in article 2 of resolution 0005 of 2000Ž ;
856
Ibid., p. 4.
857Ibid.
858Ibid., pp. 4-5.
319 x There was “no formulation of evaluation parameters for the
environmental impacts and effects caused by the Glyphosate
eradication program, which could be used to establish in a clear and
acceptable manner a level of certainty to support decision-making on
environmental management measures, which must be considered in
859
the planning and implementation processes of the spraying projectŽ ;
x “In the documents which reference the national Environmental
Management Plan and the Environmental Management Plan for the
Province of Putumayo, no set of programs, actions and concrete
management measures for prevention, control, mitigation,
compensation and correction have been put forward for the possible
impacts and effects caused by the programŽ 860;
x “No concrete, systematic actions are proposed for the tracking and
monitoring of the environmental management measures and their
results, supported by a technical design of a set of environmental
861
quality indicators that the program must comply withŽ ; and
x The DNE failed to hire an external technical auditor to assess the
862
environmental and other impacts of the fumigation program .
4.55 Confronted with the DNE’s intractable opposition to preparing even a
rudimentary impact assessment, the Environment Ministry resorted to stronger
measures. In an Order referred to as Resolution 341 of 2001, it decreed that:
859
Ibid., p. 5.
860Ibid.
861Ibid.
862
Ibid., p. 8, Art. 8.
320 “it is necessary to immediately apply preventative measures as
established in the resolution portion of this decree, with the aim
that these measures, in their development and results, will serve as
a means for this Ministry to definitively impose an Environmental
Management Plan for the activities of the transitio863 regime,
according to article 38 of Decree 1753 of 1994Ž .
4.56 Article 2 of the Decree ordered the DNE, within six months, to conduct
“environmental impact assessments in order to establish the nature and
characteristics of possible environmental impacts generated by said activity in the
4 months before [this decree], to investigate potential environmental effects
according to the findings and to impose the necessary measures to mitigate and/or
compensate for themŽ 864. In Article 7, the Ministry further ordered the DNE to
prepare two specific impact assessments, namely, a regeneration and dynamic
ecological study of sprayed areas; and a glyphosate residue study, including an
analysis of its effects on soil properties 865. In addition to these impact
assessments, the Environment Ministry ordered the DNE to conduct other
activities relevant to its EIA obligation:
x that the DNE comply with the existing prohibition on spraying over
National Parks (Article 3);
x that the DNE identify and map environmentally and socio-
economically sensitive areas, and that it propose and implement buffer
zones, with the approval of the Ministry of Environment (Article 5);
863
Ibid, p. 5.
864Ibid., pp. 5-6.
865Ibid., p. 6.
321 x that the DNE establish “in an immediate and efficient mannerŽ a
contingency plan for “undesired events,Ž as well as an inspection,
verification, and control plan for verifying the adequate
implementation of environmental management measures (Article 6);
and
x that the DNE retain an external, independent auditor, as it was already
required, but had failed to do, under Resolution 005 of 2000 (Article
866
8) .
4.57 Repeating the pattern, the DNE again brushed off the Environment
Ministry’s order. Two months later, in July 2001, the Comptroller General of
Colombia released an audit of the DNE’s operations, drawing attention to the
DNE’s failure to assess environmental impacts. The Comptroller General noted
that despite “sufficient evidence regarding the existence of negative
environmental impactsŽ, the DNE had failed to carry out the necessary
environmental studies:
“Despite sufficient evidence regarding the existence of negative
environmental impacts by forced eradication by aerial spraying on
illicit use crops, which has been reported by various local and
regional authorities as well as the Ombudsman, and which have
caused widespread protests of rejection and questioning at national
and international levels, there is no true monitoring or control of
the eradication policy so long as the Environmental Management
Plan is not approvedŽ 867.
866
Ibid., pp. 5-8.
867Comptroller General of the Republic of Colombia, Appointed Comptroller for the
Environment, Special Audit of the Policy for Eradication of Illicit Crops, p. 2-3 (July 2001). ER,
Vol. V, Annex 137.
3224.58 The Comptroller General urged the Environment Ministry to apply “the
principle of precautionŽ by ordering “the suspension of the chemical eradication
via spraying until the program’s environmental, social and economic implications
868
are determinedŽ . In defiance of the Comptroller’s call to suspend operations,
869
as well as a temporary Colombian court order to do so , Colombia’s Anti-
870
Narcotics Chief ordered the aerial sprayings to continue .
4.59 The continued spraying prompted the Environment Ministry to commence
a formal investigation 871. This included inquiries into:
x the failure of the DNE to conduct the environmental impacts
assessments necessary to identify the scope and character of potential
environmental impacts generated by the spray, and resulting
868Ibid., p. 29.
869
See Rojas Report, op. cit., para. 150, n. 77. ER, Vol. II, Annex 8. The Colombian court
ordered the temporary suspension of the aerial fumigations upon admitting a case presented by the
Organization of Indigenous Populations of the Colombian Amazon (“OPIACŽ) for the spray
program’s violation of indigenous and human rights. After the Government’s open rejection of
the court order, the suspension was terminated on 6 August 2001.
870
“Colombia Drug Czar to Keep SprayingŽ, THE NEW YORK TIMES (New York, 31 July 2001).
ER, Vol. IV, Annex 63.
871
Republic of Colombia, Ministry of Environment, Resolution No. 1066 (26 Nov. 2001). ER,
Vol. V, Annex 139. See also Republic of Colombia, Ministry of Environment, Resolution No.
108, p. 4 (31 Jan. 2002). ER, Vol. V, Annex 141. (“[A]n administrative proceedings has been
initiated aimed at opening an investigation into the failure to comply with some of the
requirements contained in Resolution 341 of 2001 . . . Ž); Republic of Colombia, Ministry of
Environment, Resolution 0670, Whereby a sanction is imposed and other decisions are made (19
Jun. 2003). EM, Vol. II, Annex 19; Republic of Colombia, Office of the Ombudsman, National
Ombudsman Resolution No. 26, Human Rights and International Humanitarian Law in the
Context of Armed Conflict and Fumigation of the Coca Crops in the Province of Putumayo, p. 27
(9 Oct. 2002). ER, Vol. V, Annex 145; Organization of Indigenous Nations of Colombia (ONIC),
Evaluation of the Fumigations in Colombia: Destruction of Rural Areas from Plan Colombia, pp.
3-42…3-43 (Aug. 2002). ER, Vol. III, Annex 29.
323 872
mitigation measures . Specifically, the Environment Ministry
criticized the DNE’s position that carrying out an impact evaluation
was “not possibleŽ 87. The Ministry maintained that the DNE’s
opinion did not obviate the need to produce the required studies,
particularly since they were indeed possible to perform. It thus
ordered the DNE to conduct the assessments “immediately,Ž and to
874
provide quarterly reports on their status .
x the failure to implement the two environmental impact studies within
three months of Resolution 341, as mandated under Article 7, that is,
an ecological regeneration study and a study on the persistence in, and
875
effect of, glyphosate on soil ; and
x the failure to immediately establish an inspection, verification, and
control program to monitor the effectiveness of environmental
protection measures during the spray operations, as required under
Article 6 of Resolution 341 876. In that connection, the Ministry
emphasized that “these activities should have been carried out
immediately, therefore efforts should proceed in the manner
877
already indicated, that is, in an immediate mannerŽ .
4.60 The urgency the Ministry of Environment placed on carrying out each of
these impact assessments could not have been clearer. It reiterated seven times
872Republic of Colombia, Ministry of Environment, Resolution No. 1066, pp. 1-2, 5 (26 Nov.
2001). ER, Vol. V, Annex 139.
873Ibid., p. 2.
874Ibid., pp. 2-3, 5.
875
Ibid.
876
Ibid., pp. 3, 5.
877Ibid., p. 3.
324that the DNE’s compliance must be “immediate Ž, highlighting its importance in
878
bold . None of these facts is mentioned in the Counter-Memorial.
4.61 Concerned about harms already being caused by the aerial spraying and
frustrated by the DNE’s failure to implement an adequate EMP after so many
years, on the same day, the Ministry of Environment finally imposed an EMP,
879
under Resolution 1065 .
4.62 Yet, the DNE still kept refusing to carry out environmental impact
assessments. Thus, in January 2002, the Environment Ministry again felt
880
compelled to order the production of the long-overdue impact studies . It
reiterated that the DNE had failed to present, among other things, environmental
impacts assessments as required by Article 2 of Resolution 341. Nor had it
produced the two studies on glyphosate (as required by Article 7), on ecological
regeneration and glyphosate’s effects on soil 881.
878
Ibid. (emphasis in original).
879Republic of Colombia, Ministry of Environment, Resolution No. 1065 (26 Nov. 2001). EM,
Vol. II, Annex 15. See also Rojas Report, op. cit., paras. 78-79. ER, Vol. II, Annex 8.
880Republic of Colombia, Ministry of Environment, Resolution No. 108 (31 Jan. 2002). ER, Vol.
V, Annex 141.
881Ibid. The Ministry also informed the DNE that, as the authority responsible for the aerial
fumigation program, it was “responsible for ensuring that in the development of the PECIG, the
management measures, prevention, mitigation, environmental control and compensation are
observed and taken into account by the actors under its coordination, as appropriateŽ. Ibid., pp. 3,
12.
3254.63 At a February 2002 meeting with Ecuadorian officials, representatives of
the DNE and the Anti-Narcotics Police agreed “that there is a lack of objective
and impartial scientific research to study the short- and long-term impacts to the
environment and to health, as well as the chemical formulations used to eliminate
882
illicit cropsŽ . The Colombian delegation also admitted that since the spray
mixture’s “application is not completely uniform in the field . . . it cannot be said
883
that there are no risks to the ecosystemŽ .
4.64 Faced with the persistent refusal by the DNE to assess risks, and
confronted by mounting evidence of actual harm, Colombia’s National
Ombudsman, in October 2002, called for suspending aerial sprayings in
Putumayo department (bordering Ecuador), and declared that the suspension
should remain in place until the spraying programme was brought into
compliance with the existing legal obligations 88. The need for the suspension
882Republic of Ecuador, Ministry of Environment, Joint Report from the Workshop: Eradication
of Illicit Crops, Bogotá, Colombia, p. 2 (13-15 Feb. 2002). EM, Vol. IV, Annex 163.
883Ibid., p. 10.
884These included “particularly those obligations relating to the prohibition from fumigating over
indigenous territories and bodies of waterŽ. Republic of Colombia, Office of the Ombudsman,
National Ombudsman Resolution No. 26, Human Rights and International Humanitarian Law in
the Context of Armed Conflict and Fumigation of the Coca Crops in the Province of Putumayo, p.
37 (9 Oct. 2002). ER, Vol. V, Annex 145. The Ombudsman also explained how, despite the fact
that the DNE had the geographic information necessary to avoid these vulnerable and excluded
areas, the DNE had sprayed over these areas nonetheless. See Republic of Colombia, Office of
the Ombudsman, National Ombudsman Resolution No. 26, Human Rights and International
Humanitarian Law in the Context of Armed Conflict and Fumigation of the Coca Crops in the
Province of Putumayo, p. 27, para. 3.10.4 (9 Oct. 2002) (explaining that Art. 5(d) of the 2001
EMP “prohibits the forced eradication in still and running bodies of water, in areas of human
settlement, in indigenous reservations and areas of productive and agreed upon projects. In order
to preserve these areas, the Portfolio defines minimum safety strips, ranging from 200 to 2,000
326was underscored, the Ombudsman said, by the DNE’s long-standing failure to
address fundamental obligations, including those concerning monitoring and
mitigating the spray’s health and environmental impacts 88. The Ombudsman
also cited the DNE’s failure to perform an Environmental Audit and the further
886
failure to adopt or implement an Epidemiological Monitoring Plan .
4.65 By June 2003, that is, three-and-a-half years after the spraying began
along the border with Ecuador, there still was no EIA or even a rudimentary
impact study regarding the aerial spraying programme. As a result, the Ministry
of Environment took the unprecedented step of formally sanctioning and fining
the DNE for failing to assess environmental impacts, on 19 June 2003 88. Rather
than a proper assessment of the programme’s environmental impacts, the DNE
meters. . . . Despite the fact that, on April and July of this year, the PNDA submitted a compact
disc to the DNE with information regarding the alternative development projects and indigenous
territories, these were sprayed, as shown in the complaints filed before the Ombudsman and the
DNEŽ.). ER, Vol. V, Annex 145.
885Republic of Colombia, Office of the Ombudsman, National Ombudsman Resolution No. 26,
Human Rights and International Humanitarian Law in the Context of Armed Conflict and
Fumigation of the Coca Crops in the Province of Putumayo, p. 33 (9 Oct. 2002) (“The health of
inhabitants of sprayed areas and the environment are equally threatened, since the spraying is
carried out without fulfilling the essential preventive measures designed to prevent, control and
mitigate potential damage to the population’s health and the environment. Such measures include
the engaging of an environmental auditor, the implementation of an Epidemiological Monitoring
Plan and the strict compliance with the Environmental Management Plan. However, as is clear
from this resolution, the aerial spraying has been conducted ignoring the Environmental
Management Plan, especially since such spraying was effectuated on indigenous communities and
contaminated certain water suppliesŽ.). ER, Vol. V, Annex 145.
886Ibid., pp. 27, 33, 37. In addition, the Ombudsman urged the CNE to revise the complaint
mechanisms created under Resolution No. 17, due to ongoing and fundamental flaws in the
procedure. Ibid., p. 37.
887Republic of Colombia, Ministry of Environment, Resolution 0670, Whereby a sanction is
imposed and other decisions are made, p. 13 (19 Jun. 2003). ER, Vol. V, Annex 148. See also,
Rojas Report, op. cit., para. 78. ER, Vol. II, Annex 8.
327had merely presented a literature review of impacts caused by coca crops (not the
888
spray mixture that was used to eradicate them) . Brazenly, the DNE defended
its inaction based on what it called “the impossibility of carrying out
environmental impact evaluations during the execution of the programŽ 88. In
other words, after blocking all efforts to carry out environmental impact
assessments before the aerial spraying programme became operational, the DNE
argued that it was impossible to carry out such assessments afterwards, while it
was being executed. The DNE thus created its own version of Catch 22. The
Environment Ministry was not amused 890.
4.66 By June 2003, upon fining the DNE for its refusal to comply with its legal
obligations to assess environmental impacts, the Ministry of Environment had
confirmed that:
x there was still no impact study designed “to establish the nature and
characteristics of possible environmental impactsŽ, as required by
Article 2 of Resolution 341;
888Republic of Colombia, Ministry of Environment, Resolution 0670, Whereby a sanction is
imposed and other decisions are made, p. 8 (19 Jun. 2003). ER, Vol. V, Annex 148. See also
ibid., p. 6 (“the argument presented does not meet the requirements under letter b), Article 6 of
Resolution 341 of 2001, for verification and assessment of environmental management measures
during PECIG operations, because it identifies the impacts generated by the establishment of
illegal crops and not the impacts of aerial sprayingŽ).
889
Ibid., p. 6.
890
Ibid. (“[T]his Ministry does not share this opinion since it is a very common practice to
conduct ex post facto evaluations to determine the effects of an activity on the environment and,
based on the analysis of this evaluations, ascertain if the foreseen impacts and environmental
management measures give optimum results and/or suggest measures to mitigate and/or offset
said impactsŽ.).
328 x there was still no study assessing either the ecological regeneration in
sprayed areas or the degree to which glyphosate remains in soil as
residue and its effects on the physio-chemical and biological
properties of the soil, as required by Article 7; and
x there was still no inspection, verification and control programme to
evaluate the effectiveness of the spray programme’s environmental
891
management measures, as required by Article 6 .
In other words, the DNE had complied with none of the environmental impact
obligations that had been imposed upon it by the Ministry of Environment. The
Counter-Memorial is silent about this. Instead, it tells the Court that the
892
Environment Ministry and the DNE “jointly workedŽ to develop the EMP .
4.67 The Environment Ministry’s decision to fine the DNE produced a
backlash that ended its efforts to require an assessment of the aerial spraying
programme’s environmental impacts. Thereafter, the Minster of Environment
893
was compelled to resign . The new Minister had no environmental experience,
891
Ibid., pp. 4-9,12-13.
892CCM, Chap. 4, para. 4.13 (emphasis added).
893“Suarez new Environment Minister as Rodriguez QuitsŽ,USINESS N EWS A MERICAS (14 Nov.
2003). ER, Vol. IV, Annex 79. The National Narcotics Council also retaliated against the
Environment Ministry’s sanctions by eviscerating the environmental protections that had been
imposed to protect national parks. In that regard, the Council renounced any intention to abide by
the exclusion areas and buffer zones that had protected these areas since the inception of the spray
programme. Resolution No. 013 of 27 June 2003 of the National Narcotics Council of Colombia.
CCM, Vol. II, Annex 49. As a result, the DNE was given plenary authority to spray directly over
protected areas, including national parks. The Resolution was signed by the President of the
Council, Mr. Fernando Londoño Hoyos, who also held the position of Minister of Interior and
Justice. This was all done in open disregard of Colombia’s existing environmental protections.
See Rojas Report, op. cit., paras. 80-82. ER, Vol. II, Annex 8.
329but had served as the Presidential Advisor on Plan Colombia, that is, the person
894
responsible for promoting and coordinating the aerial spraying programme . In
a classic understatement, the Colombian newspaper El Tiempo commented, “It is
895
not clear whether the new minister has the capacity and experienceŽ.
4.68 Unsurprisingly, the Ministry of Environment never again challenged the
DNE’s failure to assess the environmental impacts of the aerial spraying
programme. Nor, apparently, did any other agency of the Government of
Colombia.
C. T HE FAILED E FFORTS OF COLOMBIA ’S COURTS TO SECURE AN EIA
4.69 Attempts to compel the DNE to conduct an EIA in regard to the aerial
spraying programme were also made by Colombia’s courts. They, too, met with
no success. And they, too, are ignored in the Counter-Memorial.
4.70 In June 2003, the Administrative Court of Cundinamarca issued a
judgment in a class action lawsuit brought by Colombian citizens who had been
harmed by the aerial spraying programme 89. The court ruled that the DNE had
89CCM, Chap. 5, para. 5.18; “Suarez new Environment Minister as RodrigueUSINESSŽ, B
NEWS A MERICAS(14 Nov. 2003). ER, Vol. IV, Annex 79.
89“Billiard Shot on Three GangLTIEMPO (Bogotá, 13 Nov. 2003). ER, Vol. IV, Annex 78.
896Republic of Colombia, Administrative Tribunal of Cundinamarca, Claudia Sampedro and
Others, Judgment (13 June 2003). ER, Vol. V, Annex 147. See also Rojas Report, op. cit., paras.
87-88. ER, Vol. II, Annex 8.
330failed to meet its obligations under the EMP, as established by Resolution 1065,
and that the sprayings violated the citizens’ right to a healthy environment. It
thus ordered the sprayings to cease until the DNE fully complied with its
obligations under Articles 2, 6, 7, and 8 of Environment Ministry Resolution
341 897. These included the DNE’s obligations to carry out environmental impact
studies on territories sprayed (article 2); to establish an inspection, verification
and control plan for monitoring the implementation of the environmental
management measures (article 6); to conduct two specific soil impact studies to
establish the chemicals’ permanence in Colombian soil (article 7); and to hire an
external, independent auditor to evaluate impacts on the environment, human
898
health, and agriculture and livestock (article 8) . The court also ordered studies
on whether the aerial spraying was harmful to the environment and human health,
including a cohort study comparing the health of Colombian citizens exposed to
899
the sprayings with a control group . The court further ordered the DNE to
identify the harm caused by the sprayings “with glyphosate plus POEA plus
897Republic of Colombia, Administrative Tribunal of Cundinamarca, Claudia Sampedro and
Others, Judgment, p. 113 (13 June 2003). ER, Vol. V, Annex 147. See also Rojas Report, op.
cit., para. 88. ER, Vol. II, Annex 8.
898
Republic of Colombia, Administrative Tribunal of Cundinamarca, Claudia Sampedro and
Others, Judgment, p. 113 (13 June 2003). ER, Vol. V, Annex 147. See also Rojas Report, op.
cit., para. 88. ER, Vol. II, Annex 8.
899
Republic of Colombia, Administrative Tribunal of Cundinamarca, Claudia Sampedro and
Others, Judgment, p. 113-114 (13 June 2003). ER, Vol. V, Annex 147. See also Rojas Report,
op. cit., para. 88. ER, Vol. II, Annex 8.
331Cosmo FluxŽ and to implement the correction, mitigation and compensation
900
measures required by law under CNE Resolution 17 of 2001 and the EMP .
4.71 The Colombian government appealed the decision. In the meantime, it
refused to comply with the court order to suspend the aerial sprayings, and it
refused to conduct the impact studies or the external audit ordered by the court.
Colombia President Álvaro Uribe announced, “while I am president, we will not
suspend the fumigationŽ 90. This made it indisputably clear that, as between drug
eradication and environmental protection (or protection of Colombian nationals
against the health consequences of exposure to the spray mixture), the priority
was given to the former, at the expense of the latter, even in the face of a court
order.
4.72 In October 2004, the Colombian Council of State ruled on the appeal.
Although it recognized that the aerial sprayings do cause harm (“the evidence
clearly demonstrates ƒ that certain problems and complaints do ariseŽ) 902, the
Council of State determined that the harms are outweighed by Colombia’s
900
Republic of Colombia, Administrative Tribunal of Cundinamarca, Claudia Sampedro and
Others, Judgment, p. 114 (13 June 2003). ER, Vol. V, Annex 147. See also Rojas Report, op.
cit., para. 88. ER, Vol. II, Annex 8.
901“We Will Continue To Fumigate While I Am PresidentŽ, EL TIEMPO (Bogotá, 30 Jun. 2003).
ER, Vol. IV, Annex 76.
902State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p. 10 (19 Oct. 2004). ER, Vol. V, Annex 151. See
also Rojas Report, op. cit., paras. 89-90. ER, Vol. II, Annex 8.
332pressing need to eradicate illicit drugs, and the harms could be limited if strict
safeguards are applied to the programme: “permanent and strict controls of
903
fumigation activities are requiredŽ .
4.73 Thus, the Council of State ordered the aerial spray program to adhere with
“strict complianceŽ to “the Environmental Management Plan, imposed by
resolution No. 1065, as well as the obligations noted in articles 2, 6, 7, and 8 of
resolution No. 341 of 2001Ž 904:
“Clearly, the guidelines stated by the environmental authorities
should be followed when illicit crops are being sprayed, and not
even the slightest deviation from these should be permitted, which
means that it is therefore necessary for permanent controls to be
undertaken, with continuous evaluations, of any effects which
905
might begin to appearŽ .
4.74 The Council of State further ordered the DNE to assess the environmental
effects of the complete spray mixture, that is, “glyphosate plus POEA plus
CosmofluxŽ, on areas that had been sprayed … adopting as its own the Ministry of
903
State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p. 10 (19 Oct. 2004). ER, Vol. V, Annex 151. See
also Rojas Report, op. cit., paras. 89-90. ER, Vol. II, Annex 8.
904
State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca (19 Oct. 2004), p. 10. ER, Vol. V, Annex 151. See
also Rojas Report, op. cit., paras. 89-90. ER, Vol. II, Annex 8.
905State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p. 10 (19 Oct. 2004). ER, Vol. V, Annex 151. See
also Rojas Report, op. cit., paras. 89-90. ER, Vol. II, Annex 8.
333 906
Environment’s long-standing demand . The Council also ordered the Ministry
of Social Protection (Colombia’s Health Ministry) to evaluate Colombians who
had been exposed to “glyphosate plus POEA plus Cosmo FluxŽ, for the purpose
of determining the spray’s impact on their health and lives 907. The Council
further ordered that both agencies “receive the necessary supervision for ensuring
908
that follow up is carried out of the effects of fumigationŽ .
4.75 The Counter-Memorial discusses the decision of the Council of State, but
fails to mention any of these elements of it. The Counter-Memorial also fails to
report that none of the orders issued by the Council of State were complied with
by the DNE or the Ministry of Social Protection. Specifically, the DNE failed to
comply with the Council’s orders to: (i) follow “the guidelines stated by the
environmental authorities . . . when illicit crops are being sprayed, and not even
the slightest deviation from these should be permittedŽ; (ii) undertake “permanent
controls . . . with continuous evaluations, of any effects which might begin to
appearŽ; or (iii) assess the environmental effects of the complete spray mixtureŽ
on areas subject to aerial spraying. And the Ministry of Social Protection failed
906
State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p. 11 (19 Oct. 2004). ER, Vol. V, Annex 151. See
also Rojas Report, paras. 89-90. ER, Vol. II, Annex 8.
907
State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p. 11 (19 Oct. 2004). ER, Vol. V, Annex 151. See
also Rojas Report, op. cit., paras. 89-90. ER, Vol. II, Annex 8.
908State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p. 11 (19 Oct. 2004). ER, Vol. V, Annex 151. See
also Rojas Report, op. cit., paras. 89-90. ER, Vol. II, Annex 8.
334to obey the Council’s order to protect Colombia’s citizens by “determining the
spray’s impact on their health and livesŽ. At least, the Counter-Memorial
presents no evidence of compliance by either agency with these orders.
*
4.76 In sum, Colombia has no case on EIA. Contrary to the assertions made in
the Counter-Memorial, it conducted no such assessment, either before
commencing aerial spraying along the border with Ecuador in January 2000, or at
any time thereafter. To the contrary, the Colombian government agency
responsible for carrying out the programme … the DNE … stubbornly (and
successfully) resisted all attempts by other governmental authorities to cajole,
persuade or even order it to carry out any form of assessment of the programme’s
environmental impacts. It was able to get away with this because it had the
support of Colombia’s highest authorities, including the President of the
Republic. Colombia could not allow an EIA to be conducted without running the
risk that its findings would be so devastating that it would be impossible to
continue the spraying. In the President of Colombia’s own words: “[W]hile I am
president, we will not suspend the fumigationŽ 909.
909“We Will Continue To Fumigate While I Am PresidentŽ, EL TIEMPO (Bogotá, 30 Jun. 2003).
ER, Vol. IV, Annex 76.
335 Section III. The Spray Programme Violates Other Critical Operational
and Safety Requirements
4.77 In this Section, Ecuador further demonstrates the falsity of the Counter-
Memorial’s assertion that the aerial spray programme is conducted “in
910
accordance with the environmental provisions in forceŽ . It has already been
established that no EIA has ever been conducted, and that the programme makes
a mockery of the operational requirements in the EMP that are intended to
prevent spray drift. In this Section, Ecuador shows how the programme also
violates other important safety requirements, including obligations imposed by
Colombian law that are also intended to avoid spray drift and its associated harms
to people, animals, plants and the environment. Colombia can hardly claim to
satisfy its obligation of due diligence when it routinely and blatantly violates its
own safety laws and regulations regarding the handling and use of toxic
pesticides like those contained in the spray mixture deposited along and near the
border with Ecuador.
4.78 Since pesticides, if not used properly, can cause serious harm to human
health and the environment, they are required to have warning labels that give
legally binding instructions for how and under what circumstances they may be
used. A label thus represents a State’s considered view on what is required for a
particular pesticide to be applied safely.
910CCM, Chap. 4, para. 4.23.
3364.79 In most jurisdictions … including Colombia … using a pesticide in a manner
inconsistent with its label is illegal, and violations carry civil or criminal
penalties 91. Indeed, as the UN Food and Agriculture Organization’s (“FAOŽ)
guidelines on aerial spraying explain, “[i]n most countries, adhering to the label
912
recommendations is a legal obligationŽ .
4.80 In aerial spraying of herbicides, compliance with a label has an especially
close link to environmental protection and human health. For example, the
Australian Pesticides and Veterinary Medicines Authority (“APVMAŽ) explains
that enforcing label instructions is critical to preventing aerial spraying from
causing harm to people and the environment:
“When the APVMA considers registering an agricultural chemical
product, it must satisfy itself, according to scientific principles,
that the product can be used to achieve its intended purpose and at
the same time not be likely to harm human health, the environment
or Australia’s international trade. To achieve this end, the
APVMA determines instructions for use and limitations on use for
each product and places them on the product’s label. User
911
EM, Chaps. 2 and 5, paras. 2.8, 5.39-5.40; see also European Union, Council Directive
Concerning the Placing of Plant Protection Products on the Market, 91/414/EEC, Art. 3(3) (15
July 1991) (“Member States shall prescribe that plant protection products must be used properly.
Proper use shall include compliance with the conditions . . . specified on the labellingŽ); infra
Chap. 4, n. 181.
912Food and Agriculture Organization of the United Nations, Guidelines on Good Practice for
Aerial Application of Pesticides, p. 8 (2001). ER, Vol. IV, Annex 98. The FAO further explains
that “[t]he product label carries statutory instructions for the user, and must cover the crops for
which it is registered, the recommended dose rate, the number of treatments permitted during the
growing season and how many days before harvest the last treatment may be applied.
Additionally, the label will ƒ advise on environmental protection measures to be carried out.
Such measures may refer to a ‘non-spray’ barrier (buffer zone). . . . The product label should
provide application details, which should include nozzle selection, volume applied, and
application timingŽ. Ibid., pp. 8-9.
337 compliance with these instructions and limitations falls under the
913
enforcement powers of the states and territoriesŽ .
4.81 The U.S. Environmental Protection Agency (“U.S. EPAŽ) makes the same
point, emphasizing that label compliance is important for preventing damage to
human health and the environment from spray drift:
“In the U.S., the Agency can assure significant controls on use and
potential health and environmental impacts through the pesticide
label, and through a state infrastructure which governs label
compliance to address issues such as drift and worker and
bystander exposureŽ 914.
4.82 Like other States, Colombia only allows application of a pesticide if its
Agriculture Institute, which falls under the jurisdiction of the Ministry of
915
Agriculture, has approved a label for it . Thus, the approval of a label indicates
what the Colombian regulatory authorities themselves believe is necessary to
protect human health and prevent environmental harm. Since a label is specific to
913
Australian Pesticides and Veterinary Medicines Authority (APVMA), AVPM Operating
Principles in Relation to Spray Drift Risk, p. 2 (15 July 2008) (hereinafter “APVMA Operating
PrinciplesŽ). ER, Vol. III, Annex 22.
914
United States Environmental Protection Agency, Office of Prevention, Pesticides and Toxic
Substances, Report on Issues Related to the Aerial Eradication of Illicit Coca in Colombia,
Response from EPA Assistant Administrator Johnson to Secretary of State, p. 2 (19 Aug. 2002).
EM, Vol. III, Annex 143. The importance of following label instructions is also recognized by
pesticide manufacturers. The Spray Drift Task Force, a group of 38 agricultural chemical
companies which collaborated on spray drift studies in the United States, notes that “[u]se of
pesticide products is strictly governed by label instructions.Always read and follow label
directionsŽ. Spray Drift Task Force, A Summary of Aerial Application Studies (1997). ER, Vol.
III, Annex 10.
915Rojas Report, op. cit., paras. 104-105. ER, Vol. II, Annex 8.
338a single herbicide, it is a reflection of the authorities’ views on the precautions
916
that are necessary to prevent harm from use of that herbicide .
4.83 Unable to justify or even explain Colombia’s failure to comply with the
labels required by its regulatory authorities, the Counter-Memorial seeks to
minimize their importance. For example, paragraph 7.15 attempts to dismiss a
pesticide label as “similar to that commonly found … and generally required … to
appear on over-the-counter and prescription medicines, household products, and
917
even processed food productsŽ . But Colombia’s position in the Counter-
Memorial contradicts its own laws, which make it unlawful to use a pesticide in
918
ways that are at variance with the label . Regulations governing the use of
pesticides, promulgated by the Colombian Ministry of Health, provide that:
“All persons involved in the management and use of pesticides
must comply with the norms related to the respective activity, as
laid out in this decree . . . . Use products according to the
instructions on the labels or with the technical assistance of the
919
companyŽ .
916
Ibid.
917CCM, Chap. 7, para. 7.15; see also CCM, Chap. 7, paras. 7.150, 7.165.
918Rojas Report, op. cit., para. 104. ER, Vol. II, Annex 8.
919Republic of Colombia, Ministry of Health, Decree No. 1843 (22 July 1991), as amended by
Decree No. 695 (26 April 1995) and Decree No. 4368, Arts. 180, 181(h) (4 Dec. 2006). EM, Vol.
II, Annex 11; see also Rojas Report, op. cit., para. 104. ER, Vol. II, Annex 8.
3394.84 In Colombia, as elsewhere, failure to use a pesticide in accordance with its
920
label instructions subjects the user to civil and criminal penalties
4.85 As detailed below, Colombia’s aerial spraying programme consistently
and flagrantly violates the label instructions for the pesticides used in its spray
mixtures. The pervasive violations of standards that Colombia itself has adopted
for the specific chemical products in question is strong evidence of a likelihood of
harm and of Colombia’s failure to exercise due diligence.
A. R OUNDUP SL
4.86 Consider, for example, the label for Roundup SL, the herbicide that the
Counter-Memorial says was used from 2000 until 2005, when it was replaced
because of its propensity to damage human eyes. The label for Roundup SL,
approved by the Colombian Agriculture Institute, sets mandatory limits on,
among other things, droplet size, height of spray release and wind speed, all of
which reflect the regulatory agency’s views on what is necessary to prevent spray
drift and avoid unsafe application of the herbicide 92. The Counter-Memorial
acknowledges that this label corresponds to one of the herbicide products used in
920Rojas Report, op. cit., para. 104, n. 43. ER, Vol. II, Annex 8.
921Colombia Roundup SL Label. EM, Vol. III, Annex 115.
340the spray program 92, but avoids mentioning that the aerial sprayings violate each
of the label’s requirements.
4.87 In particular, Colombia’s Roundup SL label requires droplets to be no
923
smaller than 250 microns . The actual size of the droplets, according to a study
paid for by the Colombian and U.S. governments, is much smaller: they have a
924
median diameter of 128 microns . That is approximately half the size of the
smallest droplets permitted by the label. In fact, according to the same study, half
of the droplets are smaller still 925.
4.88 Droplet size smaller than the allowable minimum is not the only way in
which the spray programme violates Colombia’s Roundup SL label. In addition,
Colombia allows spraying at much higher temperatures than the label permits.
Although the label instructs against spraying when the temperature is higher than
29qC, the programme’s EMP permits spraying at temperatures as high as 35qC.
As explained in Chapter 2, warmer air increases evaporation of the spray droplets,
926
thereby reducing their size and making them more prone to drift .
4.89 The aerial spraying programme further violates Colombia’s Roundup SL
label because the spraying takes place at much greater heights than is permitted
922CCM, Chap. 4, para. 4.50, n. 310 (referencing Roundup SL Label, EM, Vol. III, Annex 115).
923
Colombia Roundup SL Label. EM, Vol. III, Annex 115.
924
Hewitt et al, 2009, op. cit., p. 921. CCM, Vol. III, Annex 131-B.
925See supra Chap. 2, para. 2.136.
926See supra Chap. 2, para. 2.151.
341by the label. Although the label forbids spraying at heights more than 2 metres
above the crop, the programme permits spraying as high as 50 metres (and higher
if there are obstacles) 927. That is approximately twenty-five times higher than
allowed by Colombia’s label. And, in reality, as shown in Chapter 2, the planes
frequently spray from even higher altitudes: no fewer than 16,143 flights between
2000 and 2008 and within 10 kilometres of Ecuador’s border have dispersed the
928
spray mixture above 50 metres . The excessive height of Colombia’s spray
operations significantly increases spray drift, as Ecuador showed in Chapter 2.
4.90 The programme also violates the Roundup SL label’s restrictions
929
regarding wind speed. This too has significant implications for spray drift .
Although the label does not allow spraying when the wind is blowing faster than
7 kilometres per hour, the programme’s EMP permits it to occur in conditions as
930
windy as 9.26 km/h .
927Resolution No. 1054 of 30 September 2003 of the Ministry of Environment of Colombia, p. 173
(hereinafter (“2003 Environmental Management PlanŽ). CCM, Vol. II, Annex 50.
928See supra Chap. 2, para. 2.103.
929See supra Chap. 2, para. 2.152
9302003 Environmental Management Plan, p. 174. CCM, Vol. II, Annex 50. Colombia also
violates the Roundup SL label by using the product for an unauthorized purpose. In that regard,
the label states that Roundup SL may only be applied for “AGRICULTURAL USEŽ. Colombia
Roundup SL Label. EM, Vol. III, Annex 115. The aerial spraying programme, however, cannot
by any stretch of the imagination, be construed as an agricultural use. Indeed, Colombia does not
even try to claim that it is. To the contrary, Colombia has insisted that “coca eradicationŽ is a
“non-agricultural use of glyphosateŽ. CCM, Chap. 7, para. 7.14 & n. 551. Pesticides in Colombia
may only be employed for authorized uses. Rojas Report, op. cit., paras. 110-119. ER, Vol. II,
Annex 8.
3424.91 Accordingly, contrary to the Counter-Memorial, Colombia’s aerial
spraying of Roundup SL violates the restrictions that Colombia itself has imposed
for that very herbicide, precisely in order to prevent spray drift and its associated
931
harms to off-target areas, including human settlements.
B. GLY-41
4.92 Like Roundup SL, the Counter-Memorial refers to the label for GLY-41,
the other herbicide that Colombia admits to spraying 93. But, as with Roundup
SL, the Counter-Memorial fails to mention that the spray programme regularly
violates the restrictions set forth in that label, too.
4.93 The label for GLY-41, approved by Colombia’s Agriculture Institute,
establishes what the agency considers to be the necessary restrictions for its safe
application, including those meant to prevent spray drift.
4.94 For example, Colombia’s GLY-41 label states that because “coarse sprays
are less likely to driftŽ, users are forbidden from employing “nozzles or nozzle
931
Failure to comply with the product labels has also been noted by the Colombian Agricultural
Institute (ICA). In a report published in 1999, the ICA informed the DNE that the addition of
Cosmo-Flux 411F violates the label requirement of “not adding adjuvantsŽ. Republic of
Colombia, Ministry of Environment, Division of Environmental Licenses, Order No. 599, p. 17
(23 Dec. 1999). ER, Vol. V, Annex 132. The ICA also noted that the application rate of 10.4
litres per hectare is 42.4% higher than the application rate of 6 litres per hectare recommended by
the manufacturer. Ibid., p. 15.
932
CCM, Chap. 4, para. 4.50, n. 310 (referencing Label and Safety Data Sheet for GLY-41, CCM,
Vol. III, Annex 134).
343configurations which dispense spray as fine spray dropletsŽ 933. However,
Colombia’s own study determined that the droplets dispersed by its spray planes
934
are “very fine to fineŽ . In other words, Colombia sprays in precisely the
manner proscribed by the Agriculture Institute’s label, which specifically
prohibits “fine spray dropletsŽ. As described in the Giles Report, the size of the
spray droplets is one of the principal drivers of spray drift and long-distance
deposition. The “fineŽ spray droplets created under the conditions of application
in Colombia are especially prone to spray drift 935.
4.95 Colombia also violates the GLY-41 label in another way that has
important implications for human health and environmental protection: it sprays
the herbicide at an impermissibly high concentration. In that regard, the label
requires an applicator to “[u]se the recommended dose of herbicide in 20 to 140
936
litres of water volume per hectare unless otherwise specified on the labelŽ .
937
Contrary to this instruction, Colombia adds only 13 litres of water per hectare .
933
Label and Safety Data Sheet for GLY-41, p. 433. CCM, Vol. III, Annex 134.
934
Hewitt et al, 2009, op. cit., p. 921. CCM, Vol. III, Annex 131-B.
935See supra Chap. 2, para. 2.138.
936Label and Safety Data Sheet for GLY-41, p. 433. CCM, Vol. III, Annex 134. The label, as
provided by Colombia in Annex 143 to its Counter-Memorial, does not provide any alternative
specifications. Ibid.
937Report by the Anti-Narcotics Direction of the Colombian National Police (DIRAN), p. 306 (8
Feb. 2010). CCM, Vol. II, Annex 67.
344The excessive concentration is important because a more concentrated spray
938
mixture enhances its toxicity and the risk of injury to off-target plants .
4.96 By implementing the spray programme in the manner it has … in disregard
of its own legally-mandated product label requirements … Colombia has violated
its own legal protections against harm to human health, animals, non-target plants
and the environment. Despite the high risks inherent in aerial spraying of toxic
herbicides, Colombia ignored its legal obligations to carry out an EIA 939, to
comply with the operational requirements of the EMP (which has the status of
law in Colombia) 940, and to adhere to the legal mandate to use these herbicides in
941
strict conformity with label instructions . As discussed in Chapter 7,
Colombia’s disregard of laws and binding regulations in all these respects
demonstrates, at the least, its failure to fulfil its duty of due diligence in the
conduct of the aerial spraying programme.
938Stephen C. Weller, Ph.D., Glyphosate-Based Herbicides and Potential for Damage to Non-
Target Plants Under Conditions of Application in Colombia, pp. 21-22 (Jan. 2011). ER, Vol. II,
Annex 3.
939
See supra Chap. 4, Sections I and II.
940See supra Chap. 2, paras. 2.65-2.73
941See supra Chap. 4, paras. 4.78-4.95.
345 Section IV. Colombia’s Spray Programme Would Not Be Allowed
Elsewhere
4.97 In this Section, Ecuador shows that the conduct of Colombia’s aerial
spraying programme is not just unlawful under Colombian law; it is also
irreconcilable with environmental laws the world over.
4.98 The Counter-Memorial is exaggerating when it asserts that the
programme is “heavily regulatedŽ 94. In reality, the rules that govern Colombia’s
aerial spraying … even if they were enforced (as shown in this and Chapter 2, they
are not) … are among the most lenient in the world. Indeed, many jurisdictions,
including the European Union, have banned aerial spraying outright, except in
very limited circumstances. Those jurisdictions that do allow aerial spraying
subject it to restrictions that are far stronger than those in Colombia, in order to
minimize the risks of spray drift. This further demonstrates that Colombia sprays
in a manner likely to cause significant harm, and violates its duty of due
diligence.
A. E UROPE
4.99 The Counter-Memorial falsely claims that Colombia’s aerial spraying
programme is endorsed by the European Union. The only “supportŽ for this
improbable claim is a reference to a book published in Bogotá by one of
94CCM, Chap. 7, para. 7.31.
346Colombia’s Foreign Ministers during the execution of the programme, Mr.
Guillermo Fernández de Soto. The cited parts of his book are not annexed to the
Counter-Memorial 943.
4.100 Colombia should have checked with the original source, the European
Union itself. In fact, the European Parliament rejected participation in Plan
944
Colombia . Among the reasons cited was Colombia’s “aerial crop-sprayingŽ,
which the Parliament said was causing “the forced displacement of families and
945
communitiesŽ and was “seriously affecting Colombia’s rich biodiversityŽ . The
Parliament therefore resolved that the European Union “must take the necessary
steps to secure an end to the large-scale use of chemical herbicidesŽ given “the
946
dangers of their use to human health and the environment alikeŽ . The
943CCM, Chap. 3, paras. 3.45-3.46 & n. 206-211. Ecuador provides the relevant extracts at ER,
Vol. IV, Annex 111. Guillermo Fernández-Soto, The Possible Illusion: Testimony on Colombian
Foreign Policy (Grupo Editorial Norma, 2004). ER, Vol. IV, Annex 111. As is readily apparent,
the cited parts do not, in fact, evidence support for aerial spraying, but rather for structural reform
to reduce inequality and instability, support for local human rights organizations, the
establishment of a peace promoting institution, and aid programs for people displaced by the
aerial fumigations and conflict. Ibid., p. 109.
944European Parliament, Resolution on Plan Colombia and Support for the Peace Process in
Colombia, EUR. PARL. DOC. B5-0087 (1 Feb. 2001). ER, Vol. IV, Annex 99.
945Ibid., para. D.
946Ibid., para. 15 (emphasis added). Although the EU does not support aerial spraying in
Colombia, it does support other strategies for reducing coca production. For example, although
the EU Drugs Action Plan for 2009-2012 contains a detailed policy for promoting alternative
development programmes to reduce the supply of illicit drugs, the Action Plan does not
recommend aerial eradication. European Union, “EU Drugs Action Plan for 2009-2012Ž, Official
Journal of the European Union, 2008/C 326/09 (20 Dec. 2008). Nor is there anything in the
earlier EU Drugs Action Plan for 2005-2008 regarding support for aerial eradication. European
Union, “EU Drugs Action Plan for 2005-2008Ž, Official Journal of the European Union, 2005/C
168/01 (8 July 2005). In that regard, the EU set as an “ObjectiveŽ for “International CooperationŽ
the goal to “[p]romote and implement the EU approach to alternative development . . . in
347European Parliament’s condemnation of aerial spraying in Colombia was later
repeated by the EU’s Commissioner for External Relations, Mr. Chris Patton,
who stated that the spraying programme is “not effective,Ž “affects other cropsŽ,
947
and “harms health and the environmentŽ . None of this is mentioned in the
Counter-Memorial, presumably because it defeats Colombia’s attempt to portray
the European Union as supportive of its spray programme.
4.101 In fact, the EU’s stance against aerial spraying in Colombia is consistent
with its approach to the aerial application of pesticides generally: that it is
dangerous for human health and the environment and should not be done except
in rare and tightly controlled circumstances. The policy against aerial spraying is
unambiguous. EU Directive 2009/128/EC requires that, subject to certain limited
and narrow exceptions, “Member States shall ensure that aerial spraying is
cooperation with third countries, taking into account human rights, human security and specific
framework conditionsŽ. European Union, “EU Drugs Action Plan for 2009-2012Ž, Official
Journal of the European Union, 2008/C 326/09, para. 17 (20 Dec. 2008).
947
“EU criticises Colombia on rightsŽ, BBC (London, 22 Jan. 2004). ER, Vol. IV, Annex 80.
348prohibitedŽ 948. The prohibition on aerial spraying applies equally to the spraying
of glyphosate-based herbicides and all other pesticides 949.
4.102 As explained by Dr. Reinhard Joas, the international expert on chemicals
regulation who served as technical advisor to the European Commission in
developing the Directive that banned aerial spraying, the prohibition represents
the consensus view on the minimum standard that all 27 EU Member States agree
is necessary to protect human health and to prevent environmental harm 950. The
EU consensus is the culmination of a lengthy consultation process, beginning in
2002, which involved extensive discussions among Member States, EU
regulatory bodies, scientific and technical experts, and representatives of affected
951
industries and other stakeholders . It was the subject of considerable
deliberation by expert working groups and studies that reviewed, among other
things, the human health and environmental impacts of different policy
948
This ban on aerial spraying is part of a broader EU policy “establishing a framework for
Community action to achieve the sustainable use of pesticidesŽ. European Parliament and the
Council of the European Union, Directive 2009/128/EC: Establishing a Framework for
Community Action to Achieve the Sustainable Use of Pesticides, Art. 9(1) (21 Oct. 2009). ER,
Vol. IV, Annex 109. Joas Report, op. cit., p. 3, 11-12. ER, Vol. II, Annex 8.
949
European Parliament and the Council of the European Union, Directive 2009/128/EC:
Establishing a Framework for Community Action to Achieve the Sustainable Use of Pesticides,
Art. 2(1), 3(10)(a) (21 Oct. 2009). ER, Vol. IV, Annex 109.
950
Joas Report, op. cit., pp. 3, 5, 16. ER, Vol. II, Annex 8.
951
Ibid, pp. 7-11. European Commission, Sustainable Use of Pesticides: Historical Background,
available athttp://ec.europa.eu/environment/ppps/history.htmast visited 16 Jan. 2011).
349 952
alternatives . This deliberative process resulted in the following determination
by the European Union:
“Aerial spraying of pesticides has the potential to cause significant
adverse impacts on human health and the environment, in
particular from spray drift. Therefore, aerial spraying should
generally be prohibited with derogations possible where it
represents clear advantages in terms of reduced impacts on human
health and the environment in comparison with other spraying
methods, or where there are no viable alternatives, provided that
953
the best available technology to reduce drift is usedŽ .
4.103 As explained in a report prepared by Dr. Joas, the Directive is based upon
the recognition of the following risks:
“Pesticides have an adverse impact on human health when the
degree of exposure exceeds the level considered to be safe. Both
direct exposure (workers and operators) and indirect exposure
(consumers, residents, and bystanders) are of concern in this
respect. Indirect risks, via spray drift or otherwise, can be
amplified for vulnerable population groups such as children, the
elderly, immunologically compromised people, and agricultural
workers who receive more intensive exposure.
952
Joas Report, op. cit., pp. 7-11. ER, Vol. II, Annex 8; see also Commission of European
Communities, The Impact Assessment of the Thematic Strategy on the Sustainable Use of
Pesticides, SEC(2006) 894 (12 July 2006) (hereinafter “Impact AssessmentŽ), p. 97-102. ER,
Vol. IV, Annex 103; BiPro, Assessing Economic Impacts of the Specific Measures to be Part of
the Thematic Strategy on the Sustainable Use of Pesticides, ENV.C.4/ETU/2003/0094R (Oct.
2004), available at ec.europa.eu/environment/ppps/pdf/bipro_ppp_final_report.pdf (last visited 23
Jan. 2011); Commission of European Communities, A Thematic Strategy on the Sustainable Use
of Pesticides, Technical Annex, SEC(2006) 895 Final (12 July 2006), pp. 4-5, 7. ER, Vol. IV,
Annex 104. The European Union notes that the Directive is “accompanied by a detailed impact
assessment and a legislative proposal to create an overall coherent and consistent policy
framework for pesticide useŽ. European Commission, EU Policy for a Sustainable Use of
Pesticides: The Story Behind the Strategy, pp. 7, 13 (2007). ER, Vol. IV, Annex 106.
953 European Parliament and the Council of the European Union, Directive 2009/128/EC:
Establishing a Framework for Community Action to Achieve the Sustainable Use of Pesticides,
Preamble, para. 14 (21 Oct. 2009). ER, Vol. IV, Annex 109. Joas Report, op. cit., p. 5. ER, Vol.
II, Annex 8.
350 . . .
Pesticides, which may enter the environment through direct
application, leaching, runͲoff or spray drift, have adverse impacts
by contaminating water, air and soil, damaging plants and wildlife,
954
and causing a loss of biodiversityŽ .
B. INDIVIDUAL S TATES
4.104 Many individual States have adopted regulations in regard to aerial
spraying of pesticides that are as fully protective of human health and the
954Joas Report, op. cit., pp. 5-6. ER, Vol. II, Annex 8. Derogation from this general rule is
allowed only in limited and narrow circumstances, when the following conditions are met: there
are no viable alternatives, or clear advantages in terms of health and environmental impacts; the
pesticides used are explicitly approved for aerial spraying following a specific assessment
addressing risks from aerial spraying; and the operator and responsible enterprise are properly
certified to conduct aerial spraying. Moreover, if the area to be sprayed is in close proximity to
areas open to the public, specific risk management measures are required to ensure that there are
no adverse effects on the health of bystanders. The area to be sprayed must not be in close
proximity to residential areas. Further, aircraft must be equipped with best available technology
to reduce spray drift. Parliament and the Council of the European Union, Directive 2009/128/EC:
Establishing a Framework for Community Action to Achieve the Sustainable Use of Pesticides,
Art. 9(2)(a)-(f) (21 Oct. 2009). ER, Vol. IV, Annex 109; see also Joas Report, op. cit., p. 11. ER,
Vol. II, Annex 8. The burden is on the applicator to demonstrate compliance with the above
conditions, in order to receive approval to conduct aerial spraying from the relevant regulatory
authority. Parliament and the Council of the European Union, Directive 2009/128/EC:
Establishing a Framework for Community Action to Achieve the Sustainable Use of Pesticides,
Art. 9(4) (21 Oct. 2009). ER, Vol. IV, Annex 109; see also Joas Report, op. cit., p. 12. ER, Vol.
II, Annex 8. In the rare cases where approvals to conduct aerial spraying are granted, important
notification and monitoring requirements must be adhered to: the approval to conduct aerial
spraying must contain measures necessary for warning residents and bystanders in due time and to
protect the environment in the vicinity of the area sprayed; national authorities must keep records
of requests and approvals including relevant information such as the area to be sprayed, the
provisional day and time of spraying, and the type of pesticide used; and monitoring must be
conducted to ensure compliance with the above conditions (e.g. no adverse effects on bystanders).
Parliament and the Council of the European Union, Directive 2009/128/EC: Establishing a
Framework for Community Action to Achieve the Sustainable Use of Pesticides, Art. 9(3), (4), (5)
(21 Oct. 2009). ER, Vol. IV, Annex 109; see also Joas Report, op. cit., pp. 11-12. ER, Vol. II,
Annex 8. Dr. Joas concludes that “[b]ased on the information provided by Colombia, its aerial
spraying program does not fulfil several of the conditions required to obtain an exceptional permit
under the EU Directive. In light of the risk-prevention rationale of Directive 2009/128/EC and its
general ban on aerial spraying, the aerial spraying program to eradicate coca crops in Colombia
would not be authorized in the EUŽ. Joas Report, op. cit., p. 16. ER, Vol. II, Annex 8
351environment as the European Union’s. Even prior to EU Directive 2009/128/EC,
some European States, including Estonia and Slovenia, instituted a total ban on
955
aerial spraying with no possibility for exceptions . Others permit spraying only
in extremely limited circumstances. For example, as reported by the
Organization for Economic Co-operation and Development, the regulations in
Denmark and Switzerland are so restrictive that aerial spraying “seldomŽ
956
occurs . In France and Italy, aerial spraying is banned as a general rule, with
957
permits issued only in exceptional cases . In Finland and Sweden, aerial
958
spraying is allowed only in what the OECD describes as “exceptional casesŽ .
In fact, Sweden has authorized only two aerial spraying operations in the last 20
years 959.
955Republic of Estonia, Plant Protection Products Act, entered into force 1 May 2004, amended
July 1, 2008, Art. 78(4). ER, Vol. III, Annex 20; Republic of Slovenia, Act on Plant Protection
Products, Art. 8 (9 Sept. 2004). ER, Vol. III, Annex 16; see also Impact Assessment, op. cit., p.
99. ER, Vol. IV, Annex 103. Aerial spraying is also banned in parts of Austria. Austrian
Federated State of Vorarlberg, Ordinance on Plant Protection Products, LGB1.Nr. 18/2008, § 1(1)
(2008). ER, Vol. III, Annex 21.
956
Organization for Economic Co-operation and Development (OECD), Activities to Reduce
Pesticide Risks in OECD and Selected FAO Countries, Part I: Summary Report,
OCDE/GD(96)121, p. 44 (1996). ER, Vol. IV, Annex 96.
957
Italian Republic, Legislative Decree No. 194, Art. 5(22)(b) (17 Mar. 1995). ER, Vol. III,
Annex 9; French Republic, Code rural et de la pêche maritime, Article L253-3, p. 32 (2010). ER,
Vol. III, Annex 24.
958Organization for Economic Co-Operation and Development (OECD), Activities to Reduce
Pesticide Risks in OECD and Selected FAO Countries, Part I: Summary Report,
OCDE/GD(96)121, p. 44 (1996). ER, Vol. IV, Annex 96; Sweden, Environmental Code 808,
Chap. 14, § 18 (last amended 2009) (1998). ER, Vol. III, Annex 11.
959
Organization for Economic Co-Operation and Development (OECD), Joint Meeting of the
Chemicals Committee and The Working Party on Chemicals, Pesticides and Biotechnology,
Report of the OECD Pesticide Risk Reduction Steering Group: The Second Risk Reduction
Survey, ENV/JM/MONO (2006) 14, p. 103 (19 July 2006). ER, Vol. IV, Annex 105.
3524.105 Numerous States that allow aerial spraying require advance warning to
nearby residents and resource managers prior to the spraying. For example, in the
United Kingdom, where regulations require that “all reasonable precautionsŽ be
taken to prevent spray drift, notice must be given “well beforeŽ the pesticide is
applied, and “certainly not after the minimum consultation period set by lawŽ 960.
In Nova Scotia, Canada, aerial applicators must notify local residents and
businesses at least 30 days before spraying, and post signs on access roads 30
961
days prior to spraying . Unlike Colombia, these States regulate aerial spraying
in a manner that is consistent with the UN FAO Guidelines, which explain that:
“Members of the public, not directly involved with the spray
operation, may also be affected by an aerial pesticide application
so the contractor/farmer may have a mandatory obligation to issue
‘prior warnings’ to any person or organisation that might be
affected or concerned. Warnings must be given in ample time to
beekeepers, owners of adjacent crops, livestock owners and those
962
responsible for nearby environmentally sensitive sitesŽ .
960
United Kingdom, Department for Environment, Food and Rural Affairs, Code of Practice For
Using Plant Protection Products, p. 31, 144 (2006). ER, Vol. III, Annex 17. UK regulations
require that the following information be furnished when preparing for aerial spraying: name,
address, and phone number of the person applying the pesticide; name of the pesticides intended
for use and their active ingredients; date and time of intended spraying; and confirmation that the
same details were provided to the Chief Environmental Health Officer for the district. Ibid., p.
144. In France, aerial spraying operations conducted in locations frequently used by people must
be noticed well in advance. French Republic, Decree On the Use of Products Mentioned in
Article L.253-1 of Rural Code, p. 33 (5 Mar. 2004). ER, Vol. III, Annex 14.
961Nova Scotia Environment and Labour, Media Backgrounder: Herbicide Management, pp. 1-2
(July 2007). ER, Vol. III, Annex 19.
962Food and Agriculture Organization of the United Nations, Guidelines on Good Practice for
Aerial Application of Pesticides, p. 21 (2001). ER, Vol. IV, Annex 98.
3534.106 No such warnings have ever been given to Ecuadorian nationals in areas
affected by Colombia’s spray programme.
4.107 States that allow aerial spraying impose strict regulations in regard to
operational requirements. Regarding droplet size, the United Kingdom requires
that the “coarsest appropriate spray qualityŽ be used 963. Colombia’s programme
could not meet this standard because, as discussed above, its own hired experts
964
classify the droplet size as “fine to very fineŽ . Colombia’s droplets would not
be permitted in Costa Rica either, where aerial spraying regulations require the
average droplet size of the spray mixture to be “between 200 … 300 microns in
order to minimize drift due to drops with a slower terminal velocity and greater
965
potential for evaporationŽ . As indicated above, the median droplet size in
Colombia is much smaller: 128 microns 966.
4.108 Height of spray release is also subject to strict limitations. In Costa Rica,
spray planes may not release chemicals more than 5 metres above the crop
canopy 967; in the Netherlands, spraying is prohibited more than 4 metres above
963United Kingdom, Department for Environment, Food and Rural Affairs, Code of Practice For
Using Plant Protection Products, § 4.7.4 (2006). ER, Vol. III, Annex 17.
964Hewitt et al, 2009, op. cit., p. 921. CCM, Vol. III, Annex 131-B.
965Costa Rica, Executive Decree No. 34202-MAG-S-MINAE-MOPT-G-MSP, Art. 1(c) (21 May
2007). ER, Vol. III, Annex 18.
966Hewitt et al., 2009, op. cit., p. 921. CCM, Vol. III, Annex 131-B.
967
Costa Rica, Executive Decree No. 34202-MAG-S-MINAE-MOPT-G-MSP, Art. 1(b) (21 May
2007). ER, Vol. III, Annex 18.
354 968 969
crops . In contrast, Colombia’s EMP allows spraying at 50 metres . As
shown in Chapter 2, even this dangerously high limit has been violated by
Colombia’s spray planes on over 16,143 spray flights along or near the border
with Ecuador 970.
4.109 Nor does Colombia’s programme meet the meteorological requirements
imposed by other States in regard to aerial spraying. Colombia permits spraying
971
in weather as warm as 35°C . In contrast, the Netherlands forbids spraying
when the temperature rises above 25°C … a 10°C difference … to avoid the greater
risks of spray drift at elevated temperatures 97. The Colombian aerial spraying
programme is also incompatible with the law in the United Kingdom, which bars
spraying when the temperature is higher than 30°C because, according to the
Department for Environment, Food and Rural Affairs, “rising air currents may
carry spray droplets and vapour in an unexpected wayŽ 973. In Costa Rica, a
968Organization for Economic Co-Operation and Development (OECD), Joint Meeting of the
Chemicals Committee and The Working Party on Chemicals, Pesticides and Biotechnology,
Report of the OECD Pesticide Risk Reduction Steering Group: The Second Risk Reduction
Survey, ENV/JM/MONO (2006)14, p. 81 (19 July 2006). ER, Vol. IV, Annex 105.
969
Resolution No. 1054 of 30 September 2003 of the Ministry for the Environment of Colombia,
p. 173. CCM, Vol. II, Annex 50.
970See supra Chap. 2, para. 2.103.
971
Resolution No. 1054 of 30 September 2003 of the Ministry for the Environment of Colombia,
op. cit., p. 173. CCM, Vol. II, Annex 50.
972
Organization for Economic Co-Operation and Development (OECD), Joint Meeting of the
Chemicals Committee and The Working Party on Chemicals, Pesticides and Biotechnology,
Report of the OECD Pesticide Risk Reduction Steering Group: The Second Risk Reduction
Survey, ENV/JM/MONO (2006)14, p. 81 (19 July 2006). ER, Vol. IV, Annex 105.
973United Kingdom Department for Environment, Food and Rural Affairs, Code of Practice For
Using Plant Protection Products, p. 90 (2006). ER, Vol. III, Annex 17.
355tropical country where conditions are similar to Colombia, aerial spraying must
974
be suspended if the temperature exceeds 29ºC .
4.110 Differences between Colombia and the rest of the world exist with respect
to wind conditions as well. For example, in Australia, a minimum wind speed of
3 kilometres per hour is required 97. The Australian regulatory agency explains
that:
“A minimum speed of 3 km/hr is required because times of no
wind (essentially below 3 km/hr) often precede or accompany
periods of highly stable air and surface temperature inversion
conditions both of which can greatly increase spray drift risk.
Moreover, when wind resumes after periods of calm, its direction
is not predictable. Spraying only when there is at least some wind
ensures that wind direction is known (so that drift onto sensitive
areas can be avoided) and greatly reduces the likelihood of surface
temperature inversions forming during or shortly after
976
applicationŽ .
4.111 Similarly, aerial application guidance from the United Kingdom states that
“[t]he safest conditions in which to spray are when it is cool and humid with a
steady wind of 2 to 4 miles an hour or 3.2 to 6.5 kilometres an hour (light breeze)
977
blowing away from any sensitive areas or neighbours’ landŽ .
974Costa Rica, Executive Decree No. 34202-MAG-S-MINAE-MOPT-G-MSP, Art. 1(g)(2)-(3) (21
May 2007). ER, Vol. III, Annex 18.
975APVMA Operating Principles, op. cit., p. 21. ER, Vol. III, Annex 22.
976Ibid., p. 21.
977
United Kingdom Department for Environment, Food and Rural Affairs, Code of Practice For
Using Plant Protection Products, p. 89 (2006). ER, Vol. III, Annex 17. In Saskatchewan,
Canada, the regulators direct: “[d]o not spray under dead calm conditions in early morning, night,
3564.112 Unlike these and other jurisdictions, Colombia’s spray program has no
minimum threshold for wind speed. This is particularly dangerous because calm
wind conditions favour temperature inversions that produce greater spray drift
and off-target deposition 978.
4.113 The dangers inherent in aerial spraying … and thus the need for strict
regulation … are also recognized by courts across many jurisdictions. For
example, a 2009 decision by the U.S. Court of Appeals of Arizona found that
despite the use of modern technology, aerial spraying of pesticides remains an
inherently dangerous activity because “the risk of harm cannot be eliminated
through the exercise of reasonable careŽ 97. In holding the aerial applicator liable
for damage caused to an adjacent landowner’s property, the Arizona court found
that the risk of harm to nearby land, property and people could not be eliminated
through the use of advanced technologies such as “improved spray nozzles, new
computer-controlled release systems, use of GPS navigation systems and
inclusion of ‘thickening agents’ in spray solutionsŽ 980. The court also considered
the state’s strict regulation of aerial spraying as evidence of the inherent danger of
or late evening. These are often associated with temperature inversions and the combination of
these factors can result in long-distance spray drift (2 km or more)Ž. Government of
Saskatchewan, Ministry of Agriculture, 2010 Guide to Crop Protection, p. 12 (2010). ER, Vol.
III, Annex 25.
978Durham K. Giles, Ph.D., Spray Drift Modeling of Conditions of Application for Coca Crops in
Colombia, pp. 27-28, 42-44 (Jan. 2011). ER, Vol. II, Annex 2; Hansman & Mena Report, op. cit.,
p. 23, n.9. ER, Vol. II, Annex 1; supra Chap. 2, para. 2.152.
979
Pride of San Juan v. Pratt, 221 Ariz. 337, 338, 340 (Ct. App. 2009).
980Ibid., p. 340.
357the activity, explaining that “[t]he legislature regulates this activity in part to
avoid the serious potential harm that can be caused by pesticides and chemical
driftŽ, and that violation of the regulations can result in criminal penalties81.
4.114 Courts in other jurisdictions have reached the same conclusion. For
instance, in 2009 a court in Argentina granted a writ of amparo filed by the
citizens of Santa Fé, enjoining for six months the aerial spraying of pesticides,
including Roundup. The court found that aerial application caused adverse
982
consequences for public health and the environment . The decision was upheld
on appeal by the Second Chamber of the Court of Civil and Commercial Appeals
of Santa Fé 983. The appellate court affirmed that the use of agrochemicals such as
glyphosate could cause severe harm to the environment, to animals, and to the
health and quality of life of the population of Santa Fé, in violation of the law984.
981Ibid., p. 342, n. 8.
982“A Constitution Appeal Is Ordered In San Jorge: A Judge Recognizes the Risk of Glyphosate
FumigationsŽ, ETRE R IOSENTRE TODOS (Entre Rios, 13 Apr. 2009). ER, Vol. IV, Annex 86.
983Shane Romig, “Argentina Court Blocks Glyphosate Spraying Near Rural TownŽ,OW ONES
N EWSWIRES (21 Mar. 2010). ER, Vol. IV, Annex 91.
984
“Santa Fe: A Ruling In Favor of LifeŽENACE (4 Jan. 2010). ER, Vol. IV, Annex 88. In
May 2009, an environmentalist group petitioned Argentina’s Supreme Court, seeking a temporary
ban on the use of Roundup after reported high incidence of birth defects and carcinogenesis in
people living near rural areas having been sprayed with herbicides, and scientific evidence linking
genetic malformations in amphibians in those areas to glyphosate. “Weed Killer Kills Human
Cells: Study Intensifies Debate over ‘Inert’ IngredientsŽ, EMENTAL HEALTH NEWS (22
June 2009). ER, Vol. IV, Annex 87. Furthermore, the Municipality of Paraná has prohibited
fumigations with Roundup close to urban areas, after complaints of damaged crops and the death
of animals by local residents. The only application of herbicides allowed is manual fumigation
outside urban areas, after the processing of the corresponding permi“It Is Warned that
Fumigations Are Being Carried Out in Paraná Despite Them Being ProhibitedŽ, LOZ (9 Jan.
2010). ER, Vol. IV, Annex 89. And in San Pedro Peninsula, the local authorities prohibited the
358 Conclusion
4.115 Colombia’s aerial spraying programme is not only conducted in a manner
that is impermissible in countries around the world, it is executed in ways that are
impermissible even in Colombia. In sum, Colombia sprays huge swaths of
territory immediately adjacent to Ecuador with a toxic chemical herbicide without
ever having carried out an EIA to determine the spray’s impacts on human health
or the environment, in defiance of the demands of its Environment Ministry,
National Ombudsman, Comptroller General, and courts; it executes the aerial
spraying programme in a manner that flagrantly disregards the operational
requirements of its own EMP, which has the status of law and is intended to
prevent or minimize harm to human health and the environment; and it ignores
legally mandated requirements intended to avoid spray drift and associated harms
to health and the environment. These failures to exercise even a minimum
amount of diligence in carrying out what is an inherently dangerous activity all
but assure spray drift into Ecuador at toxic levels sufficient to harm the local
population, kill legitimate crops, and wreak havoc on the delicate natural
environment characteristic of the border region. In short, these facts destroy
Colombia’s claim that it satisfied its obligation of due diligence.
use of Roundup, invoking scientific uncertainty over the consequences of its application.
“Fumigations with Glyphosate Are Not Permitted on the PeninsulaŽ (10 Feb. 2010). ER, Vol. IV,
Annex 90.
359 CHAPTER 5.
VIOLATION OF TERRITORIAL SOVEREIGNTY5.1 In its Memorial, Ecuador showed how the adverse effects of the use of
toxic herbicides by Colombia in the border area with Ecuador has violated
Ecuador’s territorial sovereignty. As noted by Ecuador, and not contradicted by
Colombia, respect for a State’s territorial sovereignty is a fundamental obligation
under general international law as well as the treaties applicable to the Parties in
985
this dispute, and it gives rise to a distinct cause of action in international law .
In this case, by failing to take steps to prevent the drift of toxic herbicides onto
the territory of Ecuador, Colombia has violated the duty to respect Ecuador’s
territorial sovereignty. As described in the Memorial and in more detail in
Chapters 2 through 4 of this Reply 986, Colombia has allowed this to occur by
failing to require a proper environmental assessment to be carried out in
accordance with national and international legal requirements (including but not
limited to the obligations to carry out (1) the transboundary environmental impact
assessment required by general international law 98, and (2) the assessment
required by Article 7(3) of the 1989 Indigenous and Tribal Peoples’ Convention
(ILO Convention No. 169 988), and by failing to exercise proper diligence in
authorising the spraying activities, namely by:
985Memorial of Ecuador, Vol. I, Chap. 7, paras. 7.3-7.8 (28 Apr. 2009) (hereinafter “EMŽ).
986EM, Chap. 8, Section C, “Colombia Failed to Take Adequate Precautionary Measures;Ž see
supra Chaps. 2-4.
987See infra Chap. 6; Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay),
Judgment, 2010, pp. 60-61, paras. 203-206.
988See infra Chap. 7; ILO Convention No. 169 Concerning Indigenous and Tribal Peoples, Art.
7(3) (hereinafter “ILO Convention 169Ž).
363 989
▯ allowing the use of inappropriate chemicals ;
▯ failing to prevent planes from operating at a speed and height that
will prevent spray drift 990;
991
▯ permitting inappropriate aircraft to be utilised ;
▯ failing to prevent small droplet sizes from being sprayed 992;
993
▯ allowing night spraying ;
▯ paying insufficient attention to climatic temperatures and wind
conditions 994; and
▯ permitting the herbicide spray to be applied at an excessive
application rate 995.
Colombia does not claim any right to allow overflights of the territory of
Ecuador; nevertheless, there is evidence that at least some such flights have
occurred in a manner not authorised by Ecuador, in violation of Ecuadorian
996
sovereignty .
5.2 In its Counter-Memorial, Colombia has chosen to avoid engaging with
Ecuador’s arguments. It has invoked a well-trodden path frequently adopted by a
Respondent, namely to rewrite the arguments made by the Applicant and respond
989
See supra Chap. 2, paras. 2.17-2.64.
990See supra Chap. 2, paras. 2.88-2.107.
991See supra Chap. 2, paras. 2.115-2.122.
992See supra Chap. 2, paras. 2.135-2.138.
993
See supra Chap. 2, paras. 2.145-2.149.
994
See supra Chap. 2, paras. 2.151-2.154.
995
See supra Chap. 2, paras. 2.140-2.143.
996See supra Chap. 2, para. 2.163, n320 . See also R. John Hansman, Ph.D. & Carlos F. Mena,
Ph.D., Analysis of Aerial Eradication Spray Events in the Vicinity of the Border Between
Colombia and Ecuador from 2000 to 2008, p. 13 (Jan. 2011). ER, Vol. II, Annex 1.
364to arguments that have not actually been made. Thus, Colombia has recast
Ecuador’s claim as being that “no detectable or measurable trace of spray mix, no
997
matter how small, should be allowed to enter Ecuador’s . . . territoryŽ . There is
no citation to any such assertion, which is not Ecuador’s position. As shown in
Chapters 2 and 3 of the Reply, the evidence of substantial harm is well-
established.
5.3 As noted in Chapter 3, Colombia has ignored the harm caused to its own
territory998, despite the reports of extensive harm there. Colombia’s Office of the
Ombudsman has concluded that the programme had affected “thousands of
ColombiansŽ and had “high socio-economic and environmental costŽ 999,
identifying damage to crops, animals and people, and raising concerns about “the
1000
ruin of their household financesŽ and the “severe food security problemŽ .
Departmental and local authorities have also complained of the severe damage
caused in Colombia: in 2002 the Governor of Putumayo Department, adjacent to
Ecuador, denounced the aerial spraying, stating that the damage caused to legal
997Counter-Memorial of Colombia, Vol. I, Chap. 8, para. 8.35 (29 Mar. 2009) (hereinafter
“CCMŽ).
998CCM, Chap. 1, para. 1.34(1).
999
Republic of Colombia, Office of Ombudsman, The Implementation of the Strategy of Aerial
Eradication of Illicit Crops With Chemicals, From a Constitutional Perspective, p. 1 (Apr. 2003).
ER, Vol. V, Annex 146.
1000Republic of Colombia, Office of the Ombudsman, Ombudsman Resolution No. 4, p. 4 (12
Feb. 2001). EM, Vol. II, Annex 92. See also EM, Chap. 5, paras. 5.106-5.108.
365crops by the program was “causing an economic crisis and displacement of the
1001
populationŽ .
5.4 This directly contradicts Colombia’s claim of limited effects 1002.
Colombia’s Counter-Memorial confirms that between 2002 and 2008 some 117
persons were compensated, thereby admitting that some harm has occurred 1003.
Yet Colombia has ignored the thousands of complaints of property damage which
are awaiting compensation. Indeed, Colombia’s Comptroller General reports that
4,500 complaints were received by the Ministry of Justice in 2002 alone 1004.
Similarly, in 2001 the Colombian Office of the Ombudsman reported receiving
1005
6,553 complaints .
5.5 As described in Chapter 3, the harm in Colombia has been extensive. To
cite but a few examples: in 2000, the mayor of Puerto Guzmán reported that at
1006
least seven people had died as a consequence of aerial spraying in that area ; in
1001
“Putumayo: Governor Denounces FumigationsŽ, H OY (Quito, 29 July 2007). ER, Vol. IV,
Annex 85.
1002
CCM, Chap. 1, para. 1.34(1).
1003
CCM, Chap. 4, para. 4.22. See also ibid., Chap. 1, para.1.34(1).
1004
Comptroller General of the Republic of Colombia, Plan Colombia: Fifth Evaluation Report
(Dec. 2004), p. 36. ER, Vol. V, Annex 152.
1005
Republic of Colombia, Office of the Ombudsman, National Ombudsman Resolution No. 26,
Human Rights and International Humanitarian Law in the Context of Armed Conflict and
Fumigation of the Coca Crops in the Province of Putumayo (9 Oct. 2002), p. 24. ER, Vol. V,
Annex 145. See also supra Chap. 3, paras. 3.170-3.173.
1006“Mayor Denounces FumigationsŽ, E L U NIVERSO (Guayaquil, 22 Aug. 2000). ER, Vol. IV,
Annex 56.
3662001, the Colombian Comptroller-General reported that spraying had caused
“nausea, vomiting, diarrhea, and burning of the eyes, skin and throatŽ, all
symptoms that “coincide with information in the literature and are consistent with
the position of the Ministry of HealthŽ 1007; also in 2001, the Putumayo
Department of Health recorded a sharp increase in acute respiratory infections,
diarrhea, dermatitis, and skin infections following sprayings just 20 kilometres
from Ecuador 100; and in 2003, the Colombian Office of the Ombudsman reported
an “increase in medical visits related to skin problems, gastrointestinal,
respiratory infections and conjunctivitis after the fumigationsŽ, particularly with
“childrenŽ who, due to their “fragile stateŽ, have had their “fundamental right to
life and health . . . affectedŽ00. Thus, Colombia’s claim that “[n]o substantiated
complaint of death or serious harm to human health has been presented in
1010
Colombia since the inception of the programŽ disregards the findings of its
1011
own public officials . That harm, as noted in Chapter 3, has been substantiated
1007
Comptroller General of the Republic of Colombia, Plan Colombia: Second Evaluation Report,
p. 43…44 (10 Dec. 2001). EM, Vol. II, Annex 94.
1008
EM, Chap. V, para. 5.103.
1009
Republic of Colombia, Office of Ombudsman, The Implementation of the Strategy of Aerial
Eradication of Illicit Crops With Chemicals, From a Constitutional Perspective, p. 6 (Apr. 2003).
ER, Vol. V, Annex 146.
101CCM, Chap. 1, para. 1.34(1).
1011It also disregards the numerous independent accounts that confirm the extensive damage
caused by the aerial spraying. See, e.g., “The Void of the FumigationsŽ, EO (Bogotá, 28
May 2000) (“[i]n the countryside, reporters from this newspaper confirmed the destruction of
plantain plants and corn, as well as expanses of virgin forestŽ). ER, Vol. IV, Annex 55. Larry
Rohter, “To Colombians, Drug War is Toxic EnemyŽ, TE NEW YORK TIMES (New York, 1 May
2000) (describing damage to a remote Yanacona Indian village and noting that “dozensŽ of
residents became ill and numerous farms were damaged). ER, Vol. IV, Annex 54; Juan Forero,
367by international observers, including the UN Special Rapporteur on the situation
of human rights and fundamental freedoms of indigenous people, who, in 2004,
noted the “adverse effects of indiscriminate spraying, including environmental
damage to the topsoil, fauna, flora and water, the destruction of subsistence crops
1012
and direct damage to human healthŽ .
5.6 Colombia seeks to elide Ecuador’s claims of violation of territorial
sovereignty into a renewed discussion of the legal issues relating to
transboundary environmental harm, a matter that Ecuador has addressed
separately, in Chapter 8 of its Memorial and in Chapter 6 of this Reply. The
evidence shows that the amount of herbicide that is transported into Ecuador’s
territory as a result of Colombia’s activity is significant and gives rise to adverse
“No Crops Spared in Colombia's Coca WarŽ,HEN EW Y ORK TIMES (New York, 31 Jan. 2001).
ER, Vol. IV, Annex 60; “Fumigation DisputeŽ, EEMPO (Bogotá, 22 July 2001). ER, Vol. IV,
Annex 62; “Colombia Denounces Indiscriminate Spraying in PutumayoŽ, EMERCIO (Quito,
10 Jan. 2002). ER, Vol. IV, Annex 64; “Another Controversy Over FumigationŽ, ERCIO
(Quito, 9 July 2002). ER, Vol. IV, Annex 67; “Requesting an End to FumigatiLTIEMPO
(Bogotá, 10 Oct. 2002). ER, Vol. IV, Annex 72; “Fumigations Cause Concern in PutumLyoŽ, E
COMERCIO (Quito, 10 Nov. 2002). ER, Vol. IV, Annex 73; “Glyphosate RainŽ,LETIEMPO
(Bogotá, 25 Feb. 2003). ER, Vol. IV, Annex 74; “Between Faith and FumigationsŽ, EO
(Bogotá, 10 May 2002). ER, Vol. IV, Annex 66; “Spray Program on Indigenous Territories Is
StrugglingŽ, ETIEMPO (Bogotá, 28 Apr. 2003). ER, Vol. IV, Annex 75.
1012
Report of the Special Rapporteur on the Situation of Human Rights and Fundamental
Freedoms of Indigenous People, Mr. Rodolfo Stavenhagen, Mission to Colombia U.N. Doc.
E/CN.4/2005/88/Add.2, para. 50 (10 Nov. 2004). ER, Vol. IV, Annex 102; see also ibid., para.
82. ER, Vol. IV, Annex 102. Mr. Stavenhagen reported that the Awa had been particularly
affected: “The Awá community in Nariño has informed the Special Rapporteur of various kinds of
damage caused over the last three years to large tracts of rainforest in several areas of the
municipalities of Tumaco and Barbacoas, as a result of spraying with glyphosate. The greatest
damage was done, they say, to sources of fresh water, killing native fish and affecting human
health, causing aching bones, vomiting, dizziness, fever and other ailments, particularly among
children.Ž Ibid., para. 51. ER, Vol. IV, Annex 102.
368 1013
effects . Such adverse effects include the inducing of fear and apprehension,
and on occasion even panic, in sectors of the Ecuadorian population, just as fear
and apprehension and panic are induced in elements of the Colombian
population 1014(and also in the generation of a very large number of compensation
1015
claims in Colombia) . It is these consequences that contribute to the violation
by Colombia of Ecuador’s sovereignty: in accordance with general international
law, and the 1988 Narcotics Convention, Colombia has an obligation to respect
the territorial integrity and sovereignty of Ecuador, and it has failed to do so. The
deposit of toxic substances is not insignificant, giving rise to a distinct violation
of Colombia’s international legal obligations, consistent with the approach taken
in numerous judgments of the Court. Colombia has made no effort to provide
any response to those authorities. It has not sought to distinguish them or argue
that they were wrongly decided. It simply chooses to ignore them.
5.7 There is no need to recall in great detail the factual and legal arguments
made by Ecuador in its Memorial: these are largely uncontested by Colombia.
This chapter focuses on the two issues raised by Colombia. The first concerns
Colombia’s response to the very notion that an obligation to respect sovereignty
1013See supra Chap. 2, paras. 2.199-2.202; Durham K. Giles, Ph.D., Spray Drift Modeling of
Conditions of Application for Coca Crops in Colombia, pp. 47-48 (Jan. 2011). ER, Vol. II,
Annex 2; Stephen C. Weller, Ph.D., Glyphosate-Based Herbicides and Potential for Damage to
Non-Target Plants Under Conditions of Application in Colombia, pp. 17-21, 25 (Jan. 2011). ER,
Vol. II, Annex 3.
1014
See, e.g., supra Chap. 3, paras. 3.18, 3.21, 3.158-3.166.
1015
See supra Chap. 3, paras. 3.170-3.173.
369adds in any way to the obligation to prevent transboundary harm; the second
concerns Colombia’s arguments on unauthorised overflights by Colombian
aircraft of Ecuadorian territory.
5.8 As recognised by this Court in the Corfu Channel case, “[b]etween
independent States, respect for territorial sovereignty is an essential foundation of
international relationsŽ 1016. This obligation, and its corresponding duty to not
intervene in the internal and external affairs of other States, is part and parcel of
customary international law 1017. It has been recognized in numerous international
instruments, such as General Assembly Resolution 2625 (XXV) on the
Declaration of Principles of International Law concerning Friendly Relations and
Co-operation among States in accordance with the Charter of the United
1018
Nations , and the Montevideo Convention on the Rights and Duties of
1019
States .
1016Corfu Channel (United Kingdom v. Albania), Judgment, I.C.J. Reports 1949, p. 35.
1017
Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of
America), Judgment, I.C.J. Reports 1986, p. 14, paras. 202, 292 (5).
1018
United Nations General Assembly Resolution 2625 (XXV) Declaration of Principles of
International Law concerning Friendly Relations and Cooperation among States in accordance
with the Charter of the United Nations (24 Oct. 1970).
1019Montevideo Convention on the Rights and Duties of States, Art. 8 (26 Dec. 1933), entered
into force 26 Dec. 1934, OAS Treaty Series No. 37, available at
http://www.oas.org/juridico/english/sigs/a-40.html. Both Ecuador and Colombia are parties to the
Convention.
3705.9 Of particular relevance to this case is Article 2 of the 1988 Narcotics
Convention, which provides:
“2. The Parties shall carry out their obligations under this
Convention in a manner consistent with the principles of sovereign
equality and territorial integrity of States and that of non-
1020
intervention in the domestic affairs of other StatesŽ .
Colombia argues in its Counter-Memorial that the principles stated in Article 2 of
this Convention “do not have the effect of imposing any additional substantive
obligations on the States Parties to the 1988 ConventionŽ 102. Implicitly,
therefore, Colombia recognizes that the obligation to respect Ecuador’s territorial
integrity arises under general international law, and that Article 2(2) merely
reflects and underscores that legal obligation. In short, general international law
and Article 2 of the 1988 Narcotics Convention impose upon Colombia a clear
obligation to respect the principles of territorial integrity and non-intervention in
respect of Ecuador. By causing and allowing toxic sprays to cross into the
territory of Ecuador, Colombia is violating these obligations, which exist under
conventional and customary law. Colombia accepts that exposure to glyphosate
has a “toxicity to humans and animalsŽ, even if it is asserted to be “minimalŽ, and
1022
that it causes “minor irritationŽ (for the avoidance of doubt, Ecuador does not
accept as a matter of international law that even the most minor of “irritationsŽ is
1020United Nations Convention Against Illicit Traffic in Narcotic Drugs and Psychotropic
Substances, Art. 2, U.N. Doc. E/CONF.82/15 (20 Dec. 1988), reprinted in 28 I.L.M. 493 (1989)
(hereinafter “1988 Narcotics ConventionŽ). EM, Vol. II, Annex 3.
1021CCM, Chap. 8, para. 8.33.
1022
CCM, Chap. 8, para. 8.39.
371acceptable, and actions giving rise to such “irritationsŽ constitute harm that is to
be prohibited). Ecuador’s right to territorial sovereignty encompasses a right that
its people … as well as animals found on its territory … should not be subject to
exposure to such “toxicityŽ or “irritationŽ, the consequences are neither tolerable
nor insignificant, and it cannot be the position of Colombia that it would tolerate
such consequences in a reverse scenario. This is all the more so having regard to
the consequences of the herbicide on plant life and on biodiversity, a matter on
which Colombia retains a conspicuous silence.
5.10 The obligation to respect a State’s territorial sovereignty involves not only
the duty for other States to not intervene in the internal affairs of other States, but
the exclusive right to display the activities of a State. This was recognised in the
1023
arbitral decision in the Island of Palmas case , and by this Court in the case
1024
concerning Military and Paramilitary Activities in and against Nicaragua .
The Court, when addressing the content of the principle of non-intervention,
noted:
“in view of the generally accepted formulations, the principle
forbids all States or groups of States to intervene directly or
indirectly in internal or external affairs of other States. A
prohibited intervention must accordingly be one bearing on
matters in which each State is permitted, by the principle of State
sovereignty, to decide freely. One of these is the choice of a
1023Island of Palmas Case (Netherlands v. United States of America), Reports of International
Arbitral Awards, Vol. II, p. 839 (1928).
1024
Military and Paramilitary Activities (Nicaragua v. United States), Judgment, I.C.J. Reports
1986, p. 108, para. 205.
372 political, economic, social and cultural system, and the
1025
formulation of foreign policyŽ .
To this formulation may be added the violation of environmental and health
standards that Ecuador has adopted, as well as international environmental norms
and those relating to the protection of fundamental human rights and the rights of
indigenous peoples, including those set forth in ILO Convention No. 169. In
addition, Ecuador invokes the right of all its citizens and all persons living within
its territory not to be subject to exposure to toxic chemicals that cause any degree
of harm.
5.11 The exercise of sovereignty by Ecuador inevitably includes the right to
exercise permanent sovereignty over the natural resources that are to be found
within its territory. As Ecuador made clear in its Memorial, the principle of
permanent sovereignty over natural resources is recognised in numerous
1026
instruments of international law and has been confirmed by the Court to be “a
1027
principle of customary international lawŽ .
5.12 The principle of permanent sovereignty over natural resources has
received special recognition in the context of environmental policy through its
formulation provided for the first time in Principle 21 of the 1972 Stockholm
102Ibid.
102EM, Chap. 7, para. 7.14.
1027
Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v.
Uganda), Judgment, I.C.J. Reports 2005, p. 251, para. 244.
373Declaration on the Human Environment, which asserts that States have “the
sovereign right to exploit their own resources pursuant to their own
1028
environmental policiesŽ . The principle is also reflected in Article 7(4) of ILO
Convention No. 169, which provides that “Governments shall take measures, in
co-operation with the peoples concerned, to protect and preserve the environment
1029
of the territories they inhabitŽ . Principle 21 establishes as the sole limitation
of this sovereign right the duty to respect the principles of international law and to
not cause damage to the environment of other States or of areas beyond the limits
of national jurisdiction 1030. The 1992 Convention on Biological Diversity,
binding upon Ecuador and Colombia, incorporates Principle 21 of the Stockholm
1031
Declaration as a legal obligation of States in its Article 3 . Colombia does not
dispute the content or legal status of these obligations.
5.13 In the same manner that States have the right to freely determine their
“choice of a political, economic, social and cultural system, and the formulation
1032
of foreign policyŽ , they have the right to freely determine their own standards
for the protection of the environment and the well-being of their population.
1028Declaration of the United Nations Conference on the Human Environment, U.N. Doc.
A/CONF/48/14/REV.1 (1972) (hereinafter “Stockholm DeclarationŽ).
1029ILO Convention 169, Art. 7(4).
1030Stockholm Declaration, Principle 21.
1031
Convention on Biological Diversity, Art. 3 (22 May 1992), entered into force on 29 Dec.
1993. Colombia and Ecuador are parties to this Convention.
1032
Military and Paramilitary Activities (Nicaragua v. United States), Judgment, I.C.J. Reports
1986, p. 108, para. 205.
374Ecuador is entitled as a matter of domestic law to prohibit aerial spraying of
pesticides in the manner now engaged in by Colombia, and it is entitled to hold
Colombia to a standard that prohibits that State from allowing activities which
will lead to the transboundary movement of pesticides from Colombia to
Ecuador, in circumstances that even Colombia accepts will expose the people to
toxic risks and consequent “irritationŽ. In the same manner that Ecuador has the
sovereign right to close its ports or its airspace if it so wishes, with due regard to
international law, a State has the sovereign right to apply higher standards of
environmental protection than its neighbouring countries and to be respected, free
of foreign intervention, in doing so.
5.14 Colombia seeks to trivialise Ecuador’s claim by portraying it as though
Ecuador were demanding the absolute freedom of its territory from any trace of
Colombian activity. This is not Ecuador’s case. As Colombia is well aware,
Ecuador’s action is prompted by the severity of the environmental and human
consequences suffered in its territory as a result of Colombia’s uncontrolled … or
inadequately controlled … activities. It is inappropriate for Colombia to invoke a
need to reconcile “conflicting interestsŽ by reference to considerations of
1033
“reasonableness and proportionalityŽ : Colombia has failed to respect the basic
principles of international cooperation and of due diligence with respect to
transboundary harm, as explained by Ecuador in Chapter 8 of its Memorial and
103CCM, Chap. 8, para. 8.35.
375Chapter 6 of its Reply. It is equally inappropriate for Colombia to apply its
internal laws and standards to the territory of Ecuador, or to accept that as a
matter of international law it is entitled to do so. If Colombia had provided
adequate information, if it had cooperated with Ecuador, if it had carried out a
proper transboundary environmental impact assessment (and a proper and
complete “Environmental Management PlanŽ), and if it had properly regulated
the spray flights (including, inter alia, by gathering and sharing proper flight data
and carrying out a proper and complete spray drift study), then it might be in a
better position to argue that it had not violated Ecuador’s sovereignty. But having
done none of these things, and having manifestly failed to provide an accurate
and complete account of its acts, Colombia has disabled itself from claiming that
its actions have respected Ecuador’s sovereignty.
5.15 Finally, in its Counter-Memorial, Colombia dismisses Ecuador’s claim of
violation of its territorial sovereignty as a result of transboundary harm to the
Ecuadorian people and environment by claiming that “[t]o describe the causing of
1034
such harm as a breach of sovereignty does nothing to further the analysisŽ .
For the reasons set out above, this is wrong: the duty to respect a State’s
sovereignty is a fundamental obligation of international law, and it is independent
from other international obligations and provides grounds for a specific cause of
103CCM, Chap. 8, para. 8.32.
376action in international law 1035. Its violation does not depend on proving the same
degree of harm as, for example, violations of fundamental human rights or
damage to the environment. Even if Colombia’s dismissive portrayal of the
spray’s harms were true, which they are not, minor irritations caused to a large
number of people in Ecuador, over extended periods of time and as a
consequence of the use of a pesticide that is prohibited from being used by aerial
spraying around the world, gives rise to a violation of a State’s sovereignty. The
Court has recognised this principle in numerous judgments, declaring the
violation of sovereignty to be a distinct violation of international law, even when
it is a consequence of acts which result in the violation of other or additional
international obligations 103.
5.16 Considering the brevity with which Colombia addresses in its Counter-
Memorial Ecuador’s claims of violation of its territorial sovereignty and integrity,
it has presented a notably lengthy response to Ecuador’s reference to the pertinent
allegations made by Australia and New Zealand in the Nuclear Tests cases.
Colombia compares the situation of transboundary harm resulting from French
nuclear tests with that arising from the transboundary movement of pesticides in
the present case:
103EM, Chap. 7, para. 7.7.
1036
See, e.g., Armed Activities on the Territory of the Congo (Democratic Republic of the Congo
v. Uganda), Judgment, I.C.J. Reports 2005, p. 227, para. 165.
377 “In any event the Nuclear Tests cases concerned long-distance
radioactive fallout from atmospheric nuclear testing, a very
different proposition than incidental and marginal spray drift from
lawful activities in the fight against illicit drugs. There are many
uncertainties about the effects of radioactivity, and no ‘safe’ dose.
The radioactive elements concerned have a long half-life and the
illnesses they appear to cause (the chain of causation is
undetectable and unpreventable) are initially difficult to detect and
expensive to treatŽ1037.
Colombia seems to be arguing that allegations of a violation of sovereignty
causing greater environmental harm would be more capable of being sustained in
the circumstances of the Nuclear Tests cases than in those of the present dispute.
This misses the point. Colombia accepts that the intrusion into the territory of a
State of a harmful substance can and does give rise to a violation of the obligation
to respect the sovereignty of a State. That is the key point. The distinction
between the consequences of radionuclides and the spray’s pesticide mixture is
one of degree, if it is anything, but it is not one of principle. There are many
uncertainties about the effects of the pesticides used by Colombia, and that is
precisely why their use is strictly controlled and why aerial spraying of pesticides
is banned in the European Union and elsewhere. In these circumstances, by
allowing such pesticides to be transported into the territory of Ecuador, Colombia
is failing to respect Ecuador’s sovereignty.
5.17 Colombia does not dispute that overflight of Ecuadorian territory by
aircraft involved in the aerial spraying would, in the absence of a treaty or other
103CCM, Chap. 8, para. 8.39.
378consensual arrangement, violate the sovereignty of Ecuador. This is a concession
1038
that Ecuador accepts . Colombia argues, however, that Ecuador has not given
any “particularised examples of overflightŽ, and for this reason there has been no
1039
violation . Ecuador is now in a position to provide the particularised examples
that Colombia refers to: in Chapter 2, there is set out the aerial spraying incidents
giving rise to violations of Ecuadorian sovereignty 1040. Ecuador recognises that
the number of overflights is small, but that does not diminish the significance of
the issue of principle: having regard to the likelihood of further violations,
Ecuador considers that there is a need for the Court to make clear that Colombia
is under a strict duty to prevent any overflights of the territory of Ecuador that
have not been authorised.
1038
CCM, Chap. 8, para. 8.42.
1039CCM, Chap. 8, para. 8.43.
1040See supra Chap. 2, para. 2.163, n. 320.
379 CHAPTER 6.
VIOLATION OF INTERNATIONAL ENVIRONMENTAL LAW AND
RESPONSIBILITY FOR TRANSBOUNDARY HARM Section I. Introduction
6.1 In its Memorial, Ecuador set forth detailed arguments with respect to
Colombia’s violation of its international obligations on prevention of
transboundary harm, environmental impact assessment and cooperation, and the
need for a precautionary approach. In its Counter-Memorial, Colombia tries to
evade its responsibility for causing transboundary harm by portraying the spray
programme as harmless and benign. Ecuador reiterates the arguments it made in
the Memorial, and responds in this Reply to Colombia’s arguments. Chapter 2 of
this Reply shows that Colombia has misrepresented the toxicity of the glyphosate-
based formulations used in the aerial spray programme. Diplomatic
correspondence with Ecuador and official U.S. and Colombian reports reveal that
the spray mixture contained chemicals that Colombia omitted to refer to in its
description. Various formulations of glyphosate with quite different
characteristics have been used at various times, in combination with other
chemicals. Given Colombia’s failure to provide a complete account or
description, and the ensuing uncertainty, it is impossible for Ecuador or the Court
to assess fully the risks posed by the chemicals used in the spray. It is clear,
however, that the spray mixture is toxic and significantly harmful to humans.
The nature and scale of Colombia’s spraying, as shown in Chapter 2, magnifies
the problem for communities living near the border with Colombia.
3836.2 Colombia argues that Ecuador has not proved significant harm. Chapter 3
of this Reply sets out further evidence of harm to people, crops, water supplies,
natural resources and the environment … all in Ecuador … resulting from
Colombia’s aerial spraying programme 1041. That evidence need not be repeated
here: it shows that aerial spraying along the border caused significant harm in
Ecuador prior to 2007, when Colombia ceased spraying within a 10 kilometre
buffer zone, and that further harm could be caused in the future if spraying closer
to Ecuador were resumed.
6.3 In its Counter-Memorial, Colombia claims that it had, at most, an
obligation to conduct an environmental impact assessment under the terms of the
applicable Colombian law and to keep the situation under review 1042. It argues
that the aerial spraying programme was adopted and implemented with all due
diligence 1043. And it denies that it failed to cooperate with Ecuador in the
1044
implementation of the aerial spraying programme .
6.4 Ecuador disagrees. Its detailed response is set out in subsequent sections
of this Chapter. In summary:
1041
See supra Chap. 3, Section I.
1042
Counter-Memorial of Colombia, Vol. I, Chap. 8, paras. 8.89-8.90 (29 Mar. 2010) (hereinafter
“CCMŽ).
1043
Ibid., Chap. 8, para. 8.60.
1044Ibid., Chap. 8, paras. 8.105-8.112.
384 a. Given the risk of toxic herbicide drifting into Ecuador and causing
harm, there was a duty under international law to carry out a prior
environmental impact assessment of transboundary effects.
b. Colombia did not carry out such an assessment before the spraying
operation was initiated.
c. Colombia’s aerial spraying caused significant harm to people,
property and the environment in Ecuador and there remains a risk of
further harm if the spraying is resumed within 10 kilometres of the
border.
d. Given the large scale of the spraying operations, and the uncertain
composition and effects of the chemicals in use, a heightened duty of
due diligence is called for, requiring a precautionary approach to
prevention of harm.
e. Colombia failed to exercise due diligence in authorising and
supervising the spraying activities, inter alia, by:
▯ allowing the use of inappropriate chemicals 1045;
▯ failing to ensure that spray1046nes operate at a speed and height
that will prevent spray drift ;
104See supra Chap. 2, paras. 2.17-2.64.
104See supra Chap. 2, paras. 2.88-2.107.
385 1047
▯ permitting inappropriate aircraft to be utilised ;
1048
▯ allowing night spraying ;
▯ paying insufficient attention to meteorological conditions,
1049
including temperature, humidity and wind conditions ;
▯ permitting an excessive rate of application having regard to the
areas being sprayed 1050;
▯ failing to ensure compliance by spray planes with the
1051
Environmental Management Plan (“EMPŽ) ;
▯ failing to give warning when spray operations were scheduled to
take place in border areas 1052;
1053
▯ failing to monitor the harmful effects of spraying .
f. Colombia has not cooperated with Ecuador in good faith as required
by customary international law, the 1988 Narcotics Convention, and
the 1992 UN Convention on Biological Diversity. Specifically, it has
failed to:
▯ consult Ecuador before initiating the border spraying programme;
▯ notify Ecuador of the composition of the spray mixture and of
planned spraying operations;
▯ undertake joint monitoring of the impact of the spraying operation.
104See supra Chap. 2, paras. 2.115-2.122.
1048
See supra Chap. 2, paras. 2.145-2.149.
1049
See supra Chap. 2, paras. 2.151-2.154.
105See supra Chap. 2, paras. 2.140-2.143.
105See supra Chap. 2, paras. 2.67-.2.73.
1052
See supra Chap. 3, paras. 3.7, 3.99, 3.141; Memorial of Ecuador, Vol. I, Chap. 3, paras. 3.-3.3,
3.17, 3.21, 3.25, 3.46 (28 Apr. 2009) (hereinafter “EMŽ).
1053
See infra paras. 6.45-6.71.
386 g. Finally, Colombia has not answered the need for adequate
precautionary measures to deal with the ongoing risk of significant
harm to Ecuador should the spraying programme along the border ever
resume.
6.5 However expressed, the central point of this part of the case is clear and
unambiguous: Ecuador is entitled in international law to expect Colombia to carry
out its spraying programme in a manner which assesses all the risks to Ecuador
and takes all necessary steps to prevent avoidable transboundary harm from
occurring. In Ecuador’s submission that means no spraying in border areas in
circumstances where significant harm to Ecuador or its people, property or
environment is likely to result. Only since 2007, when Colombia stopped
spraying within 10 kilometres of the border, has the problem of herbicide
deposition in Ecuador been tackled effectively; as regards the future, Colombia’s
halt to the programme does not address past illegalities, with their continuing
consequences. This simple fact shows that it is possible for Colombia to carry
out its spraying programme without the need to spray in close proximity to the
border. What Ecuador seeks now is a binding obligation on Colombia not to
resume spraying within 10 kilometres of the border.
387 Section II. The Applicable Law
A. G ENERAL INTERNATIONAL LAW
6.6 Colombia has a particularly narrow and conservative view of the
applicable law. The Parties differ markedly on this issue. Ecuador’s views as set
out in the Memorial and in this Chapter are grounded in principle and established
practise. They adopt and apply the Judgment of this Court in the Pulp Mills
Case 105. The significance of that Judgment for the present case is addressed
below.
6.7 Colombia argues that international environmental law is mainly treaty-
based, and that “[s]uch customary international law rules as exist in relation to the
environment are of a general and residual characterŽ 105. It appears to deny the
existence of any relevant norms of customary international law apart from the
general duty of due diligence referred to in the Court’s Advisory Opinion on the
1056
Legality of the Threat or Use of Nuclear Weapons . Specifically, it disputes the
existence of a requirement to carry out a prior environmental impact assessment
as to transboundary effects, or to ensure that members of the public in Ecuador
1054
Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010.
1055
CCM, Chap. 8, para. 8.6.
1056
Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons, I.C.J. Reports
1996, pp. 241-242, para. 29.
388who are potentially affected by the spray are informed in advance 1057. It
discounts the articles of the International Law Commission (“ILCŽ) with respect
to prevention of transboundary harm, saying that they do not reflect customary
international law, and in any event, do not apply because the threshold criterion
1058
of a likelihood of significant harm is not met . It denies that the precautionary
1059
principle has more than “adjectivalŽ significance .
6.8 Ecuador’s arguments with respect to violation of the law on transboundary
1060
harm were fully elaborated in the Memorial . As shown later in this Chapter,
Ecuador’s case on environmental assessment requires the Court to do no more
1061
than apply the general law as set out in the Pulp Mills case . No treaty
applicable between the Parties is required to substantiate Ecuador’s arguments in
this respect, nor does Ecuador seek to “incorporateŽ an indeterminate range of
otherwise inapplicable environmental treaties via Article 14 of the 1988 Narcotics
Convention, as Colombia alleges 1062. Ecuador does not accept Colombia’s
argument that the 1988 Narcotics Convention is either the principal source of
applicable law or that it functions as a lex specialis, displacing customary law and
1057
CCM, Chap. 8, paras. 8.67-8.88, 8.95-8.100.
1058Ibid., Chap. 8, para. 8.122(2).
1059Ibid., Chap. 8, para. 8. 57.
1060EM, Vol. I, Chap. 5.
1061
See infra Chap. 6, paras. 6.29-6.35. See also Case Concerning Pulp Mills on the River
Uruguay (Argentina v. Uruguay), Judgment, I.C.J. Reports 2010, especially paras. 101, 187, 197,
204-205.
1062CCM, Chap. 8, para. 8.18.
389other applicable treaties. While Ecuador relies on the 1988 Narcotics Convention
as an additional basis for applying general international law on transboundary
harm and applicable human rights treaties, it is not dependent on that Convention
to make its case. The Court has jurisdiction to apply general international law
and applicable treaties in this dispute, and Ecuador invites it to do so.
B. T HE 1988 UN C ONVENTION ON N ARCOTIC D RUGS
6.9 Colombia relies upon Articles 14, 24 and 25 of the 1988 Narcotics
Convention as applicable law 1064. Colombia claims that it has strictly complied
with all applicable treaties, but it cites in addition to the 1988 Narcotics
1065
Convention only the 1992 Biological Diversity Convention . Other treaties
1066
referred to by Ecuador are dismissed as inapplicable between the Parties ,
which misses the point that they may also be evidence of customary international
law or provide guidance in the interpretation of treaties which are applicable
between the Parties, including the 1988 Narcotics Convention. Colombia claims
to recognise the need to take account of human rights and the environment when
implementing the 1988 Narcotics Convention. It refers to the Declaration it made
on ratification and says that “Colombia’s concern in making that declaration was,
106EM, Chap. 4.
1064
CCM, Chap. 9, para. 9.1.
1065
Ibid.
106CCM, Chap. 8, para. 8.8.
390inter alia, to maintain a balance between criminalisation of coca cultivation and a
‘policy of alternative development, taking into account the rights of indigenous
1067
communities involved and the protection of the environment’Ž .
6.10 Colombia nevertheless argues that the 1988 Narcotics Convention
functions as a lex specialis excluding other rules of general international law
1068
relating to the environment or human rights . Ecuador cannot accept this
argument, which misconceives the function of lex specialis rules and the
character of the 1988 Narcotics Convention. The 1988 Narcotics Convention is
not a self-contained regime governing all aspects of relations between the Parties
with regard to drug eradication. To so hold would eviscerate the application of
other general rules of law between the Parties in a manner inconsistent with the
concept of a lex specialis and the proper interpretation of the 1988 Narcotics
Convention. In the present case, Ecuador submits that the correct approach, in
accordance with the wording in the 1988 Narcotics Convention, is to apply both
the Convention and other applicable rules and principles of international law not
incompatible with it.
6.11 Ecuador’s approach conforms to the treatment of lex specialis regimes by
the ILC and in decisions of this Court. As the ILC pointed out in its Report on the
106Ibid., Chap. 8, para. 8.17.
106Ibid., Chap. 8, paras. 8.13-8.18.
391Fragmentation of International Law, a rule may be “specialŽ in various ways,
either because it is a more specific application of a general rule, or because it
1069
modifies or sets aside the general rule . It does not necessarily follow, as
Colombia asserts in this case, that other more general rules are excluded or
trumped by a lex specialis. More usually, a lex specialis enables a court to locate
a specific rule or body of law within a broader set of rules whose content will
influence the interpretation and application of the lex specialis 1070. Ecuador
submits that this view of the relationship between special and general rules
applies aptly in the circumstances of the present case. As the ILC Study Group
notes, “preference was often given to a special standard because it not only best
reflects the requirements of the context, but because it best reflected the intent of
those who were to be bound by itŽ 1071. From this perspective, the lex specialis
doctrine is essentially a technique for interpreting and applying treaties. In
Ecuador’s view, explained in more detail below, that view seems more consistent
1069
United Nations General Assembly, International Law Commission, Fragmentation of
International Law: Difficulties Arising From the Diversification and Expansion of International
Law, Report of the Study Group of the International Law Commission, para. 88, U.N. Doc.
A/CN.4/L.682 (13 Apr. 2006).
1070Case Concerning the Gabþíkovo-Nagymaros Project (Hungary/Slovakia), Judgment, I.C.J.
Reports 1997, para. 132; Amoco International Finance Corp. v. Iran, US-Iran CTR 189, para. 112
(1987); Ambatielos Case (Greece v. United Kingdom), Preliminary Objections Judgment, I.C.J.
Reports 1952, p. 44; Southern Bluefin Tuna Cases, Provisional Measures, Judgment, ITLOS, Nos.
3-4 (2000). See generally Joost Pauwelyn, Conflict of Norms in Public International Law 385-
416 (Cambridge, 2003).
1071
United Nations General Assembly, International Law Commission, Report of the Study Group
on Fragmentation of International Law: Difficulties Arising from the Diversification and
Expansion of International Law, para. 12, U.N. Doc. A/CN.4/L.663 Rev.1 (28 July 2004).
392with the ordinary meaning of Article 14(2) of the 1988 Narcotics Convention, a
provision which Colombia reduces to insignificance in its reading of the text.
6.12 Article 14(2) does not set out special rules on human rights or
environmental protection. Rather, it makes only a general reference to the
existing law on those subjects, and thus serves to place an explicit limit on the
application of the 1988 Narcotics Convention, including Articles 24 and 25.
While in theory there may be circumstances in which the 1988 Narcotics
Convention modifies the law on transboundary harm if necessary to give effect to
its object and purpose, that is far from being the case here. It is neither necessary
nor reasonable to cause transboundary harm in Ecuador in order to facilitate
measures aimed at eradicating illegal drug crops or to give effect to what is in
substance a treaty on law enforcement cooperation.
6.13 Ecuador’s view of lex specialis regimes does not differ from that taken by
the Court. Ecuador invokes the Advisory Opinion on the Legality of the Threat or
Use of Nuclear Weapons 107. In that case, treaties and customary law relating to
the use of force, international humanitarian law, human rights law, and
international environmental law were relied on by various parties to the
proceedings. The Court did not decide the case on the basis of any one of these
107Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, I.C.J. Reports 1996, p.
240, para. 25.
393bodies of law exclusively, even if it did recognize the particular significance of
the jus in bello. Rather, it accepted that in the appropriate context the use of
nuclear weapons might engage some or all of the rules in question. Thus, human
rights law continued to apply in wartime, but “the test of what is an arbitrary
deprivation of life . . . falls to be determined by the applicable lex specialis,
namely the law applicable in armed conflict which is designed to regulate the
conduct of hostilitiesŽ 1073. Similarly, environmental obligations continued to
apply during an armed conflict and were relevant to assessing whether a
particular use of force was necessary and proportionate but, the Court “did not
consider that [environmental treaties] could have been intended to deprive a State
of the exercise of its right of self-defenceŽ 107.
6.14 In the present case, Ecuador does not seek to deprive Colombia of its right
to implement drug eradication programmes consistent with the 1988 Narcotics
Convention, but it does insist that this Convention cannot and need not be
interpreted so as to deny Ecuador all protection from transboundary harm, as
Colombia appears to argue. Ecuador’s fundamental point is that drug crop
eradication can proceed effectively pursuant to the 1988 Narcotics Convention
without causing significant transboundary harm to Ecuador. When such harm
occurs or is at risk of occurring, its legality must then be judged in accordance
107Ibid., p. 240, para. 25.
107Ibid., p. 242, para. 30.
394with the applicable rules of general international law on the environment and
human rights, as explicitly envisaged by Article 14(2) of the 1988 Narcotics
Convention.
6.15 Colombia recognizes the limits of its own argument. It accepts that “[t]he
natural interpretation of Article 14(2) is that general respect for human rights and
the environment is called for in the context of the required eradication
measuresŽ 1075. This is a concession that Ecuador is pleased to take. Yet,
Colombia does not say what it means by “general respectŽ. Later in the Counter-
Memorial, it again notes that Article 14(2) of the Convention requires the Parties
to take “due account of . . . the protection of the environmentŽ 107. It then says
that “this is subsumed under the general international law relating to
transboundary harm; a State which exercises due diligence in accordance with the
international law standard will thereby respect the protection of the
environmentŽ 1077. That is precisely Ecuador’s point: Colombia accepts that in
relation to the environment, Article 14(2) requires the Parties, inter alia, to apply
the applicable international law standard of due diligence. It follows inexorably
from the logic of Colombia’s own admission that a State that does not exercise
due diligence in accordance with the international law standard will thereby fail
107CCM, Chap. 8, para. 8.18.
107Ibid., Chap. 8, para. 8.119.
1077
Ibid.
395to “respect the protection of the environmentŽ and thus be in breach of Article
14(2) of the 1988 Narcotics Convention. That argument was fully developed by
Ecuador in its Memorial 1078, and is fully maintained.
6.16 Even if, arguendo, Ecuador is wrong in treating Article 14(2) as creating a
cause of action, it still follows that a State that does not exercise due diligence in
accordance with Article 14(2) will not be entitled to rely on the 1988 Narcotics
Convention as a defence to an alleged violation of customary international law on
transboundary harm. Colombia must exercise its powers with respect to drug
eradication on its territory in conformity with applicable international law on
transboundary harm and human rights 107. Colombia’s only answer to that
argument is to rely on Articles 24 and 25 of the 1988 Narcotics Convention. To
recall, these Articles provide:
“Article 24: A Party may adopt more strict or severe measures
than those provided by this Convention if, in its opinion, such
measures are desirable or necessary for the prevention or
suppression of illicit traffic.
“Article 25 : The provisions of this Convention shall not derogate
from any rights enjoyed or obligations undertaken by Parties to
this Convention under the 1961 Convention, the 1961 Convention
as amended and the 1971 ConventionŽ.
6.17 According to Colombia, neither Article 24 nor the 1961 Convention on
Narcotic Drugs or the 1971 Convention on Psychotropic Substances contain any
107EM, Chap. 8, paras. 8.24-8.37.
107See 1988 UN Convention on Narcotic Drugs, Arts. 2 and 14.
396provision “even arguablyŽ incorporating human rights, indigenous rights or
environmental obligations 1080. It goes on to assert that “[a] State accused of
disregard for those obligations in taking some measure could simply say that it
did so pursuant to the permission in Article 24, or the saving clause in Article
25Ž 1081. Colombia then invokes reliance on both provisions 1082.
6.18 It is implicit in Colombia’s argument that Articles 24 and 25 override
Article 14, and indeed every other article of the 1988 Narcotics Convention.
Ecuador does not accept Colombia’s interpretation of Articles 24 and 25. To do
so would give every individual party to that convention carte blanche to disregard
international human rights law and international environmental law, if in its
opinion, it were desirable to do so in order to prevent or suppress traffic in illicit
narcotics. Indeed, it would give a party to the Convention the right to override
any inconvenient part of international law. Presumably, if, in Colombia’s
opinion, it were necessary or desirable to spray toxic herbicides over Quito in
order to suppress illicit narcotics traffic, then Colombia would have a good
defence under Articles 24 and 25. Colombia’s reading of these articles is not
consistent with the object and purpose of the Convention, nor with the ordinary
meaning of the text.
1080CCM, Chap. 8, para, 8.13.
1081Ibid.
1082
Ibid.
3976.19 First, it needs to be recalled that according to Article 2, the object and
purpose of the 1988 Narcotics Convention is to promote cooperation between
States parties. It would be strange indeed if a convention on cooperation were to
be construed as empowering one State unilaterally to violate the rights of other
parties. Nothing in Article 2 supports such an interpretation. Article 2 provides:
“1. The purpose of this Convention is to promote co-operation
among the Parties so that they may address more effectively the
various aspects of illicit traffic in narcotic drugs and psychotropic
substances having an international dimension. In carrying out their
obligations under the Convention, the Parties shall take necessary
measures, including legislative and administrative measures, in
conformity with the fundamental provisions of their respective
domestic legislative systems.
2. The Parties shall carry out their obligations under this
Convention in a manner consistent with the principles of sovereign
equality and territorial integrity of States and that of non-
intervention in the domestic affairs of other States.
3. A Party shall not undertake in the territory of another Party the
exercise of jurisdiction and performance of functions which are
exclusively reserved for the authorities of that other Party by its
domestic lawŽ. (emphasis added).
6.20 Quite apart from the express reference to human rights and environmental
protection in Article 14, the text of Article 2 as highlighted above presupposes at
least three limits on the measures that may be taken to address illicit traffic more
effectively: conformity with fundamental provisions of national law (including
presumably constitutional protection of human rights and the environment);
consistency with sovereign equality, territorial integrity and non-intervention; and
non-exercise of jurisdictional functions reserved to other States. On a plain
398reading of the text, any measures adopted under Article 24 remain subject to the
limitations imposed by Article 2. The second of those limitations applies to the
present case: causing transboundary harm to Ecuador by spraying herbicides in
border areas is inconsistent with sovereign equality, respect for the territorial
integrity of Ecuador, and non-intervention in its domestic affairs 108. On that
reading, Article 24 does not give Colombia “permissionŽ to derogate from its
environmental obligations towards Ecuador, any more than it would sanction
human rights violations by Colombia in Ecuador. Article 2 thus reinforces the
argument that Article 14 constrains the measures parties may take pursuant to the
Convention, and requires them to respect the general law on human rights and the
environment.
6.21 Second, it is also necessary to look more closely at the wording of Article
24. When it refers to a party adopting “more strict or severe measures than those
provided by this ConventionŽ, the obvious intent is to allow parties to do more
within their own domestic jurisdiction. In common with other law enforcement
cooperation conventions, it provides a common minimum for action by all States
parties, but allows individual States to go further if they wish 108. The UN
108EM, Chap. 7.
1084
Other examples include Article 34(3) of the United Nations Convention on Transnational
Organised Crime, which provides: “Each State Party may adopt more strict or severe measures
than those provided for by this Convention for preventing and combating transnational organized
crimeŽ. U.N. Doc. A/55/383 (2000), reprinted in 40 I.L.M. 335 (2001). See also Article Art.
65(2) of the United Nations Convention Against Corruption, which provides: “Each State Party
399Commentary to the 1988 Narcotics Convention gives the example of limitation
periods for prosecution of offences, noting that “a party might provide instead
1085
that the prosecution of those offences would not be subject to any time-limitŽ .
However, nothing in the Commentary to the 1988 Narcotics Convention or its
predecessors of 1961 and 1971 suggests that the parties are thereby granted an
open-ended permission to derogate from international law or violate the rights of
other States, as Colombia appears to argue. The 1988 UN Commentary simply
notes that “[i]n the previous commentaries, it was pointed out that the article
permitted a party to adopt measures additional to those prescribed by the
Convention or to replace them by stricter or more severe measures than those
1086
provided for in the ConventionŽ .
6.22 Consider the effect of Colombia’s Article 24 argument if it were to be
applied to Article 17 of the same Convention. That article deals with illicit traffic
1087
at sea . Article 17 sets out significant limits on the power of States parties to
may adopt more strict or severe measures than those provided for by this Convention for
preventing and combating corruptionŽ. U.N. Doc. A/58/422 (2003), 43 I.L.M. 37 (2004).
1085Commentary on the United Nations Convention Against Illicit Traffic in Narcotic Drugs and
Psychotropic Substances 1988, p. 391, para. 24.3, U.N. Doc E/CN.7/590 (20 Dec.1988). ER, Vol.
IV, Annex 94.
1086
Ibid.
1087Article 17 provides:
ILLICIT TRAFFIC BY SEA
1. The Parties shall co-operate to the fullest extent possible to suppress illicit traffic by sea, in
conformity with the international law of the sea.
400interfere with navigation at sea or arrest of foreign vessels. It is intended to
protect the rights of freedom of navigation “in conformity with the international
2. A Party which has reasonable grounds to suspect that a vessel flying its flag or not
displaying a flag or marks of registry is engaged in illicit traffic may request the assistance of
other Parties in suppressing its use for that purpose. The Parties so requested shall render such
assistance within the means available to them.
3. A Party which has reasonable grounds to suspect that a vessel exercising freedom of
navigation in accordance with international law, and flying the flag or displaying marks of
registry of another Party is engaged in illicit traffic may so notify the flag State, request
confirmation of registry and, if confirmed, request authorization from the flag State to take
appropriate measures in regard to that vessel.
4. In accordance with paragraph 3 or in accordance with treaties in force between them or in
accordance with any agreement or arrangement otherwise reached between those Parties, the
flag State may authorize the requesting State to, inter alia:
a) Board the vessel;
b) Search the vessel;
c) If evidence of involvement in illicit traffic is found, take appropriate action with respect to the
vessel, persons and cargo on board.
5. Where action is taken pursuant to this article, the Parties concerned shall take due account
of the need not to endanger the safety of life at sea, the security of the vessel and the cargo or
to prejudice the commercial and legal interests of the flag State or any other interested State.
6. The flag State may, consistent with its obligations in paragraph 1 of this article, subject its
authorization to conditions to be mutually agreed between it and the requesting Party,
including conditions relating to responsibility.
7. For the purposes of paragraphs 3 and 4 of this article, a Party shall respond expeditiously to
a request from another Party to determine whether a vessel that is flying its flag is entitled to
do so, and to requests for authorization made pursuant to paragraph 3. At the time of
becoming a Party to this Convention, each Party shall designate an authority or, when
necessary, authorities to receive and respond to such requests. Such designation shall be
notified through the Secretary-General to all other Parties within one month of the
designation.
8. A Party which has taken any action in accordance with this article shall promptly inform
the flag State concerned of the results of that action.
9. The Parties shall consider entering into bilateral or regional agreements or arrangements to
carry out, or to enhance the effectiveness of, the provisions of this article.
10. Action pursuant to paragraph 4 of this article shall be carried out only by warships or
military aircraft, or other ships or aircraft clearly marked and identifiable as being on
government service and authorized to that effect.
11. Any action taken in accordance with this article shall take due account of the need not to
interfere with or affect the rights and obligations and the exercise of jurisdiction of coastal
States in accordance with the international law of the sea.
401law of the seaŽ, while enhancing the regime of enforcement through cooperation
with the flag State. If Colombia’s interpretation of Article 24 is correct, then any
party may disregard the limitations recognized by Article 17 if in its opinion it is
necessary or desirable to do so in order to suppress illicit drug traffic. If that is
the intended effect of Article 24, then why did the parties agree to adopt Article
17 108?
6.23 Colombia’s proposed reading of Article 24 lacks credibility and is
implausible. Colombia cannot say that Article 24 overrides human rights and
environmental protection obligations, but not the law of the sea, because nothing
in the wording of the text would support such a distinction. If the rights of other
States under the law of the sea are not overridden by Article 24, then the rights of
other States under international environmental law and human rights law are also
not overridden. These rights are expressly recognized in Articles 2 and 14. In
Ecuador’s submission, Article 24 can only be read as empowering States parties
to do more domestically, not to override international law or violate the territorial
sovereignty or rights of other States. Colombia remains bound by the limitations
imposed by Articles 2 and 14, as well as by general international law in regard to
the environment and human rights.
1088
For the negotiating history of Article 17 (referred to as draft article 12) see United Nations
Conference for the Adoption of a Convention against Illicit Traffic in Narcotic Drugs: Official
Records, Vol. I, pp. 27-28, 154-158; Vol. II, pp. 267-274, 308-314 (Vienna, 1988). ER, Vol. IV,
Annex 96.
4026.24 Similar comments apply to Article 25 insofar as it preserves rights and
obligations under the 1961 Convention on Narcotic Drugs and the 1971
Convention on Psychotropic Substances. Article 25 cannot have been intended to
override international law or permit violations of the rights of other States. That
much is clear from the text of the two earlier conventions.
6.25 Article 39 of the 1961 Convention as amended provides as follows:
“Notwithstanding anything contained in this Convention, a Party
shall not be, or be deemed to be, precluded from adopting
measures of control more strict or severe than those provided by
this Convention and in particular from requiring that preparations
in Schedule III or drugs in Schedule II be subject to all or such of
the measures of control applicable to drugs in Schedule I as in its
opinion is necessary or desirable for the protection of the public
1089
health or welfareŽ .
As in the case of Article 25 of the 1988 Narcotics Convention, the ordinary
meaning of this provision is to preserve the freedom of parties to take additional
domestic measures, not to derogate from international law.
6.26 Article 23 of the 1971 Convention provides: “[a] Party may adopt more
strict or severe measures of control than those provided by this Convention if, in
its opinion, such measures are desirable or necessary for the protection of the
1089United Nations Single Convention on Narcotic Drugs as amended by the 1972 Protocol
Amending the Single Convention on Narcotic Drugs, Art. 39, 976 UNTS 3, reprinted in 11 I.L.M.
804 (1972).
403public health and welfareŽ 1090. This wording is identical to Article 24 of the 1988
Narcotics Convention. It does no more than that Article. Colombia’s
interpretation of Article 25 of the 1988 Narcotics Convention is thus as
unsupported by the text and the object and purpose of the 1961 and 1971
Conventions as its interpretation of Article 24.
6.27 Simply preserving the rights of parties to the 1961 and 1971 Conventions
can thus have no effect on their obligations towards other States under general
international law. Those obligations are expressly preserved by Articles 2 and
14(2), and any relevant rules of international law applicable between the parties
must be taken into account when interpreting and applying Articles 2, 14, 24 and
25 of the 1988 Narcotics Convention in accordance with Article 31(3)(c) of the
1969 Vienna Convention on the Law of Treaties. This would include customary
international law on transboundary harm and environmental impact assessments
(“EIAŽ) 1091. The same argument applies equally to applicable human rights
treaties considered in the next Chapter.
1090
United Nations Convention on Psychotropic Substances (1971), 1019 UNTS 175, reprinted in
10 I.L.M. 261 (1971).
1091Iron Rhine Arbitration (Belgium/Netherlands), Arbitral Award, paras. 58-59 (2005); Case
Concerning the Gabþíkovo-Nagymaros Project (Hungary/Slovakia), Judgment, I.C.J. Reports
1997, pp. 77-78, paras 140-141; Case Concerning Pulp Mills on the River Uruguay (Argentina v.
Uruguay), Judgment, I.C.J. Reports 2010, p. 60, para. 204.
4046.28 In conclusion, whether under customary international law or Article 14(2),
Colombia’s obligations with respect to transboundary harm and EIA are not
overridden by Articles 24 and 25 of the 1988 Narcotics Convention, and these
articles provide no defence to Ecuador’s case on harm or the risk of harm caused
by aerial spraying of toxic herbicides, nor to its arguments with respect to EIA
and cooperation. They also provide no defence to the human rights arguments
1092
made in the Memorial and in the next Chapter of this Repl.
Section III. Failure to Carry Out a Prior Environmental Impact
Assessment
A. DUTY TO C ARRY O UT AT RANSBOUNDARY EIA
6.29 Given the obvious risk of significant transboundary harm posed by
spraying toxic herbicides close to inhabited areas of Ecuador, it is indisputable
that Colombia had an obligation in general international law to carry out a
transboundary EIA before spraying near those areas commenced in 2000. Indeed,
Colombia has not argued the contrary, and most of its discussion of earlier case-
law has been shown to be without merit following the Court’s 2010 Judgment in
Pulp Mills on the River Uruguay. In that case, the Court held that prior
assessment of transboundary impacts is not merely a treaty based obligation … as
109EM, Chap. 9; infra Chap. 7.
405Colombia maintains in its Counter-Memorial … but is a requirement of general
international law:
“In this sense, the obligation to protect and preserve, under Article
41 (a) of the Statute [of the River Uruguay], has to be interpreted
in accordance with a practice, which in recent years has gained so
much acceptance among States that it may now be considered a
requirement under general international law to undertake an
environmental impact assessment where there is a risk that the
proposed industrial activity may have a significant adverse impact
in a transboundary context, in particular, on a shared resource.
Moreover, due diligence, and the duty of vigilance and prevention
which it implies, would not be considered to have been exercised,
if a party planning works liable to affect the régime of the river or
the quality of its waters did not undertake an environmental
1093
impact assessment on the potential effects of such worksŽ .
6.30 The Court also held that “an environmental impact assessment must be
conducted prior to the implementation of a project. Moreover, once operations
have started and, where necessary, throughout the life of the project, continuous
1094
monitoring of its effects on the environment shall be undertakenŽ .
6.31 Colombia argues in its Counter-Memorial that “[a]t most, Colombia had
to conduct an assessment, under the terms of the applicable Colombian law, as to
whether its spraying program risked causing significant transboundary harm, and
1095
if so, what mitigation measures were appropriateŽ . Colombia’s position is
1093Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 60, para. 204 (emphasis added).
109Ibid., p. 61, para. 205 (emphasis added).
1095
CCM, Chap. 8, para. 8.89.
406inconsistent with that of the Court in the Pulp Mills case. For Colombia, the
obligation to conduct a prior assessment of transboundary impact arises, if at all,
from its own law, not general international law. Yet that is not what the Court
ruled in Pulp Mills.
6.32 Colombia then goes on to argue that it had no such obligation, because its
1993 Law on the Environment and the 1994 Regulatory Decree adopted under
that law did not require prior environmental authorization for the application of
pesticides already operating prior to that date 109. This is not a correct
appreciation of its own law. As the Rojas Report makes clear, under Law 99 of
1993, the 1994 and all subsequent aerial spray programmes required an
Environmental License, and thus an EIA:
“[A] strict legal analysis of the laws and regulations in force at that
time, can only lead to the conclusion that the spraying operations
of 1994 and subsequent programs, due to their particularities and
their scope, would have had to be distinguished from those
previously authorized, and therefore would have had to submit to
the regulatory regime of the Environmental License contained in
1097
Law 99 of 1993Ž .
Yet, in Colombia’s view, the 1993 law excludes the aerial spraying programme
from its requirements because it had already begun before 1993. This is simply
1096
Ibid., Chap. 4, paras. 4.10…4.11.
1097
Claudia Rojas Quiñonez, Esq., The Aerial Spray Program and Violations of Colombia’s
Domestic Laws Regarding the Environment and the Rights of Indigenous Peoples, para. 72 (Jan.
2011) (hereinafter “Rojas ReportŽ). ER, Vol. II, Annex 8. See also Letter from Cecilia Lopez
Montano, Minister of Environment, Republic of Colombia, to Nestor Humberto Martinez Neira,
Minister of Justice and Law, Republic of Colombia, p. 1 (20 Dec. 1994). ER, Vol. V, Annex 123.
407wrong. In the first place, after Pulp Mills, it can no longer be argued that there is
no obligation under general international law to conduct a transboundary EIA
prior to commencement of the activity in question. Second, as set out at length in
Chapter 4 of this Reply 1098, the argument is contradicted by Colombia’s own
Ministry of Environment, which for years demanded that the National
Antinarcotics Agency (“DNEŽ) carry out studies of the spray programme’s
environmental impacts. In 1999, the Ministry demanded that the DNE supply it
with “information in relation to the analysis and determination of the conditions
of exposure, whether the exposure was direct or indirect, and the possible
cumulative impacts, the latter with the aim of re-establishing an environmental
1099
risk assessment for the spraying activitiesŽ . When the DNE finally responded,
the Ministry of Environment concluded that the information given was
insufficient to assess environmental risk 1100.
6.33 The Ministry of Environment ordered the DNE to “develop within a
period of six months, for the areas affected by the spraying of glyphosate to
eradicate illicit crops, evaluations of environmental impactŽ that would facilitate
1098
See supra Chap. 4, paras. 4.19-4.29, 4.48.
1099Republic of Colombia, Ministry of Environment, Resolution No. 341, Adopting some
decisions in relation to the Program for the Eradication of Illicit Crops by Aerial Spraying with
Glyphosate, p. 2 (2001) (hereinafter “Colombian Ministry of Environment, Resolution No. 341 of
2001Ž). EM, Vol. II, Annex 14.
1100
Ibid., pp. 4-6. See also Rojas Report, op. cit., para. 78. ER, Vol. II, Annex 8.
408the development of mitigation and compensation measures 1101. In late 2001, the
Ministry of Environment was still requesting information that would allow for the
characterisation of sprayed areas and a determination of risks, and further ordered
the DNE to evaluate the toxicological risks associated with the chemicals
used 1102. The Ministry of Environment’s futile efforts to require the DNE to
assess the aerial spray programme’s environmental impacts are set out in detail in
Chapter 4 of this Reply, so there is no need to repeat them here. What is clear is
that as part of the authorisation process the Ministry of Environment required the
DNE to carry out environmental impact studies of the risks posed by the spray
programme, and that when the necessary information was not forthcoming, it
1103
imposed its own conditions on the spray programme . But neither the Ministry
of Environment nor any other State agency ever succeeded in making DNE (or
anyone else) carry out an EIA 1104.
6.34 In Pulp Mills, the Court noted that “it is for each State to determine in its
domestic legislation or in the authorization process for the project, the specific
content of the environmental impact assessment required in each case, having
1101Colombian Ministry of Environment, Resolution No. 341 of 2001, op. cit., Art. 2. EM, Vol.
II, Annex 14.
1102Republic of Colombia, Ministry of Environment, Resolution No. 1065, Environmental
Management Plan, Arts. 5, 10 (26 Nov. 2001). EM, Vol. II, Annex 15. See also Rojas Report,
op. cit., para. 78. ER, Vol. II, Annex 8.
1103See supra Chap. 4, paras. 4.19-4.29, 4.55-4.56, 4.61, 4.65. See also Rojas Report, op. cit.,
paras. 78-79. ER, Vol. II, Annex 8.
1104See supra Chap. 4, paras. 4.66, 4.75.
409regard to the nature and magnitude of the proposed development and its likely
adverse impact on the environment as well as to the need to exercise due
1105
diligence in conducting such an assessmentŽ . This paragraph makes two
important points. First, it implies that an EIA need not be specifically required by
law, but may be required as part of the authorisation or permitting process. What
matters is that some means is put in place to ensure that an EIA is carried out.
Even if Colombia is correct in saying that no EIA was required by the 1993 Law
on the Environment … which it is not … it is still the case that before authorising
the spray programme the Ministry of Environment was required to ensure that an
assessment of possible transboundary impacts be carried out. It attempted to
exercise that power, but the law enforcement agencies responsible for execution
of the aerial spraying programme refused to obey it, apparently with the backing
of higher authorities. Second, while the “specific contentŽ of each EIA is for the
State to determine, there must be an EIA and it must have regard to “the nature
and magnitude of the proposed development and its likely adverse impact on the
environmentŽ. In this way the Court has recognized that there are certain
minimum requirements that must be met. The Court cannot be understood as
saying that the content of an EIA … if any … is entirely a matter for the State to
decide in its sole discretion. The Court’s approach would be denied of any
practical effect if interpreted and applied to mean that a State could avoid its
110Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 61, para. 205.
410international obligation to carry out a transboundary EIA by not requiring one as
a matter of domestic law or authorization. On the contrary, the approach that
Ecuador recognises to be reflected in the Court’s approach has the merit of
according with the views of the ILC and the arguments of the parties based on the
Commentary. The 2001 ILC Commentary contains the following explanation:
“(7) The specifics of what ought to be the content of assessment is
left to the domestic laws of the State conducting such assessment.
For the purposes of article 7, however, such an assessment should
contain an evaluation of the possible transboundary harmful
impact of the activity. In order for the States likely to be affected to
evaluate the risk to which they might be exposed, they need to
know what possible harmful effects that activity might have on
them.
(8) The assessment should include the effects of the activity not
only on persons and property, but also on the environment of other
States. The importance of the protection of the environment,
independently of any ha1106o individual human beings or property
is clearly recognizedŽ .
6.35 It is apparent from the Commentary that whatever else may be required
by national law, international law requires at a minimum that an EIA assess
possible effects on people, property and the environment of other States likely to
be affected. If national law does not ensure that such an assessment is carried out
1106
United Nations General Assembly, Report of the International Law Commission on the Work
of its Fifty-Third Session (23 April–1 June and 2 July–10 August 2001), commentary to Article 7,
at pp. 402-405, U.N. Doc. A/56/10 (2001) (emphasis added, internal citation omitted). Article 7
provides: “Any decision in respect of the authorization of an activity within the scope of the
present articles shall, in particular, be based on an assessment of the possible transboundary harm
caused by that activity, including any environmental impact assessmentŽ. Ibid., p. 402.
411… for whatever reason … then there is inevitably a breach of this obligation. That
is the situation in which Colombia finds itself.
B. C OLOMBIA DID N OTC ARRY OUT A T RANSBOUNDARY EIA
6.36 The essential point in Ecuador’s case on EIA is that no assessment of the
risk of transboundary harm was conducted at any time prior to authorization or
implementation of Colombia’s aerial spraying programme; nor has it been carried
out subsequently. Chapter 4 sets out in detail the account of how the Colombian
Ministry of Environment and other agencies repeatedly tried but failed to compel
the DNE to carry out the necessary studies. Without exception, these demands
were frustrated by the organ of the Colombian State responsible for conducting
the aerial spraying (DNE) 110. Spraying along the Ecuador border began in
January 2000, at which point the Ministry of Environment was still trying to force
1108
the DNE to carry out the required EIA . It continued trying until the Minister
1109
of Environment resigned in 2003 .
6.37 Colombia has no case on EIA. Nothing in the Counter-Memorial shows
that Colombia carried out an EIA of possible transboundary effects prior to
authorising or implementing the border spraying programme. Indeed, Colombia
110See supra Chap. 4, Sections I and II.
1108
See supra Chap. 4, paras. 4.30,4.35.
1109
See supra Chap. 4, paras. 4.48-4.67.
412does not even purport to claim that it carried out an EIA. It complied neither with
the requirements of international law nor even its own national law. In particular
it failed to assess:
x the chemicals actually used in the spray programme;
x the likelihood of the spray drifting across the border;
x the likely impact on the receiving environment in Ecuador if
spray did drift across the border.
6.38 Colombia argues that spraying was approved in 1994 only “after carrying
1110
out relevant scientific studies and assessing the existing situationŽ . It says that
“[t]hese studies included experimental assessments of glyphosate applications in
the Natural Park at the Sierra Nevada in Santa Marta in the 1980sŽ 1111. The only
study listed at this point in the Counter-Memorial was undertaken in 1987. It is
reproduced in Annex 123, which indicates that “[t]his study has been undertaken
for the need of the Colombian National Police to establish the effects on the
jungle as a consequence of the intensive application (by means of spraying) of
1112
glyphosate used to destroy marijuana cropsŽ . It goes on to list the sites at
1110
CCM, Chap. 4, para. 4.8.
1111
Ibid., Chap. 4, para. 4.8, n. 253.
1112SGS (Societé Générale de Surveillance, S.A.) Colombia S.A., “Report of Contamination
Control for glyphosate application at the Sierra of Santa MartsŽ, Introduction (1987). CCM, Vol.
III, Annex 123.
413which tests were carried out, and the sampling scheme for soil, foliage and
rivers1113. But that is all.
6.39 Three omissions in the 1987 study are fundamental. First, in Chapter 2 of
this Reply Ecuador pointed out significant differences in the various spray
formulations used by Colombia. It is unclear which of these formulations, if any,
was used in the 1987 trial. Colombia does not say whether the spray used in 1987
consisted only of glyphosate, or was the same as one of the glyphosate-based
formulated products and other substances subsequently used along the border to
eliminate coca plantations. Was it Roundup Ultra? Or Roundup Export? Was
Cosmo-Flux added? Or POEA? We do not know and Colombia does not say.
Second, there is no reference in the study to any assessment of the likelihood of
drift or deposition of herbicides at some distance from the target. The only field
study conducted much later by Colombia on spray drift showed that glyphosate
mixed with Cosmo-Flux, one of the adjuvants used in Colombia’s spray
programme, is more prone to drift than any other glyphosate-based formulation
that was tested, yet Colombia refused to abandon the use of Cosmo-Flux 1114.
Third, the 1987 study makes no mention of possible transboundary impacts on
Ecuador, and it says nothing about possible effects on humans, animals or
1113
Ibid.
1114See supra Chap. 2, paras. 2.186-2.187; Las Palmas Ltda., Technical Department, Glyphosate
(10,4 l/ha) and Three Different Adjuvants, For Illicit Coca Crop (Erythoxylum spp.) Control,
Agronomic Efficacy Testing of Doses of Glyphosate in Illicit Crops: Final Report, p. 12 (July
2004). ER, Vol. III, Annex 15.
414property. It tells us only what might happen to glyphosate in the jungle in a
Colombian national park if one unidentified type of herbicide is directly sprayed
1115
on it months earlier .
6.40 Given these manifest deficiencies, it is impossible to read the 1987 study
as a transboundary EIA for the purpose of assessing the probable effects of
different glyphosate formulations drifting into inhabited areas of Ecuador.
6.41 Nevertheless, in October 1993 the Colombian Health Ministry and the
National Institute for Natural Renewable Resources and the Environment
(“IDERENAŽ) issued opinions approving the use of glyphosate for eradication of
coca and marijuana crops 1116. It is worth noting the terms in which IDERENA’s
opinion was given, because it does not support the rosy view of glyphosate …
1117
likening it to “baby shampooŽ … presented in the Counter-Memorial .
Reproduced in Annex 35 of Colombia’s Counter-Memorial, it says, inter alia:
“When the spraying process started in the Province of Huila, the
Antinarcotics Police established ten (10) parameters, among which
it is important to highlight the ones referring to environmental
1115
SGS (Societé Générale de Surveillance, S.A.) Colombia S.A., “Report of Contamination
Control for glyphosate application at the Sierra of Santa MartsŽ, Introduction (1987). CCM, Vol.
III, Annex 123.
1116
CCM, Chap. 4, para. 4.9. See also Note from the General Manager of INDERENA (National
Institute for Renewable Natural Resources and the Environment) to the Director of the National
Narcotics Directorate of Colombia (8 Oct. 1993) (hereinafter “Oct. 1993 letter from INDERENA
to DNEŽ). CCM, Vol. II, Annex 35; Note from the Colombian Health Minister to the Director of
the National Narcotics Directorate (11 Oct. 1993). CCM, Vol. II, Annex 36.
1117CCM, Chap. 7, para. 7.97.
415 aspects: the herbicide will not be sprayed on inhabited areas, fish
ponds, apiaries, poultry, or other animals, and on Special
Management Areas; spraying will not conducted in sites close to
water streams or sources; not to over flight (sic) water pipe
reservoirs, schools or other sites that pose risk to human health
and the environment.
Under these conditions, INDERENA ratifies the acceptance of the
action strategy set by the National Narcotics Council in its
communiqué dated 31 January 1992, pointing out the importance
that must be given to compliance with the specific and technical
parameters established for the eradication process of poppy crops
and that must be kept for eradication of coca and cannabis
cropsŽ1118.
6.42 It is Ecuador’s case that the spraying has not been carried out in
conformity with the parameters identified by IDERENA. Instead, toxic spray has
indeed been sprayed over, or drifted over, inhabited areas, fish ponds, poultry,
other animals, schools and other areas, where it “poses risk to human health and
the environmentŽ in Ecuador. It is notable that IDERENA foresaw the likely
consequences of aerial spraying in more vulnerable areas. But like the 1987
study referred to earlier, this opinion cannot be described as a transboundary EIA,
however concerned it may have been about the potential risks. Colombia does not
attempt to argue that it constitutes an EIA.
111Oct. 1993 letter from INDERENA to DNE. CCM, Vol. II, Annex 35 (emphasis added).
416 C. AN EMP ISN OT AN EIA
6.43 The authorisation to begin spraying in border areas was thus not based on
a transboundary EIA. Nevertheless, Colombia argues that the Environmental
Management Plan (“EMPŽ), which lays down the conditions for the spraying
programme, “is equivalent to an environmental impact assessmentŽ 111. That
might be convincing if an EMP met the main requirements of an EIA, but it does
not, as a comparison of Colombia’s own description of the two processes shows.
6.44 Colombian law provides that:
“The environmental impact assessment will include information
about the location of the project and the biotic and abiotic and
socio-economic elements of the milieu that may suffer
deterioration by the corresponding work or activity that for their
execution are required to have a license, and an evaluation of the
impacts they may cause. It will also include the design of the plans
for prevention, mitigation, correction, and compensation for
impacts and the environmental management plan for the work or
activityŽ112.
6.45 In contrast, Colombia defines an “environmental management planŽ in the
following terms:
“Environmental Management Plan: Is the plan that, in a detailed
way, establishes the actions required to prevent, mitigate, control,
compensate, and correct the possible negative environmental
effects or impacts caused in the implementation of a project, work
1119
CCM, Chap. 4, para. 4.10.
112Colombian Law 99 of 1993, Art. 57. CCM, Vol. II, Annex 32.
417 or activity. It includes follow-up, evaluation, and monitoring plans
1121
and contingency plansŽ .
6.46 While there is evidently some overlap between an EIA and an EMP, the
two processes are very different. Colombia acknowledges this. Paragraph 6.24
of the Counter-Memorial confirms that at the time of meetings with Ecuadorian
officials in 2003, “Colombia had only developed an Environmental Management
Plan, on the basis of the experience gained by prior experimental spraying
programmes and studies, and not an ‘Environmental Impact Assessment
1122
conducted prior to sprayings of Glyphosate’, as requested by EcuadorŽ .
6.47 Colombia’s Counter-Memorial nevertheless argues that the EMP adopted
in 2001 (nearly two years after spraying began near Ecuador) is the equivalent to
an EIA. This is a fallacy: an EMP is not and could not be an EIA 1123. An EIA is
necessary to identify risks to the human population, to agriculture and to other
socio-economic activities, to flora and fauna, and to the environment; and to
determine whether they are acceptable and under what conditions. An EMP
comes later in time, and in response to the risks identified by the EIA; based on
the risks identified in the EIA, the EMP establishes the operational parameters
and requirements for avoidance or mitigation of those risks. The EIA is therefore
112Colombian Decree 1753 of 1994, Art. 1. CCM, Vol. II, Annex 38.
112CCM, Chap. 6, para. 6.24.
1123
Ibid., Chap. 4, para. 4.10.
418the sine qua non for the EMP. Colombia’s EMP is made up of three specific
programmes, covering operations management, detection and spraying. It merely
sets parameters for the operation, including flight altitude, temperature, relative
1124
humidity, cloudiness, rainfall, type of nozzle, droplet size, etc . This is not an
EIA, whether under international law or under Colombian law.
6.48 Had there been a transboundary EIA, the EMP could and should have
been informed by it. Without the benefit of a properly conducted EIA, it is hardly
surprising that, among other things, the EMP fails to address the risk of drift
causing transboundary harm and does not consider the dangers presented by the
full range of chemicals in the spray mixture. Crucially, what the EMP appears to
lack is the key information provided by an EIA, even in Colombia, namely “the
location of the project and the biotic and abiotic and socio-economic elements of
the milieu that may suffer deterioration . . . and an evaluation of the impacts they
may causeŽ 1125. The lack of any information about the location of the project and
the impacts it may cause helps explain Ecuador’s objections to the spraying
programme: an EMP is simply not designed to address the possibility of
transboundary harm arising in Ecuador. It is focused only on how the spray
operation is conducted in Colombia.
1124
See Note Nº SARE-142 from the Director of the National Narcotics Directorate of Colombia
to the President of the Scientific and Technical Commission of Ecuador, para. 2.4. CCM, Vol. II,
Annex 13.
1125
Colombian Law 99 of 1993, Art. 57 (emphasis added). CCM, Vol. II, Annex 32.
4196.49 Thus, to confuse the EMP with a transboundary EIA is to miss the point
entirely. Whereas the EMP sets operational parameters, an EIA is “a national
procedure for evaluating the likely impact of a proposed activity on the
1126
environmentŽ . The UN Environmental Programme defines it as “an
examination, analysis and assessment of planned activities with a view to
1127
ensuring environmentally sound and sustainable developmentŽ . Colombia’s
own documentation shows that its EMP does none of these things.
6.50 Colombia attempts to support its argument on EIA by reference to various
other studies, including the studies delivered in 2005 and 2009 (the “Solomon
1128
StudiesŽ) . For reasons fully elaborated in Chapter 4, none of these studies can
1129
be treated as an EIA . Neither addressed the risks to Ecuador, and both relied
upon seriously flawed data when evaluating the risk of spray drift. In any event,
they were conducted long after spraying began 113. The conclusion that follows
inexorably from the history revealed in Chapter 4 is obvious: at no time, whether
before it began spraying in border areas or subsequently, has Colombia carried
out an EIA of potential transboundary impacts.
1126See United Nations Convention on Environmental Impact Assessment in a Transboundary
Context, Art. 1(vi) (1991), 1989 UNTS 310, reprinted in 30 I.L.M. 800 (1991).
11271987 United National Environmental Programme Goals and Principles of Environmental
Impact Assessment, preamble.
112CCM, Chap. 7, paras. 7.11-7.12.
112See supra Chap. 4, paras. 4.37-4.41, 4.45-4.46.
113See ibid.
4206.51 In conclusion, Colombia’s own evidence has shown that it did not carry
out a transboundary EIA. To plug this hole in its case, Colombia has attempted to
argue that no EIA was required, or alternatively to find an EIA equivalent in the
EMP. Of necessity, its argument fails because the EMP and later studies were not
designed to do what a transboundary EIA is meant to do … assess likely impacts
on other States. There has plainly been a violation of Colombia’s obligations in
general international law with respect to transboundary EIA.
Section IV. Failure to Act with Due Diligence to Prevent Transboundary
Harm
A. COLOMBIA S FAILURE TO PREVENT SIGNIFICANTH ARM
6.52 Colombia argues that there is no evidence of significant harm in
1131
Ecuador . Chapter 3 of this Reply sets out in detail the evidence of avoidable
and significant harm caused by Colombia’s deposition of toxic herbicides over
Ecuadorian territory. Chapter 3 lists, inter alia:
▯ effects on the health of the population, including damage to
eyesight, sickness, skin and throat irritation, vomiting and
diarrhea;
▯ destroyed or damaged crops and farm animals;
▯ effects on forests, water supplies and the environment;
▯ displacement and effects on the health and cultural life of
indigenous peoples;
113CCM, Chap. 7, para. 7.10.
421 ▯ cumulative effects on a fragile environment and subsistence
1132
lifestyle arising from repeated exposure to toxic spraying .
6.53 Contrary to Colombia’s assertions, the totality of the damage more than
satisfies the test of “significant harmŽ, a standard which Colombia accepts is the
threshold for obligations of transboundary harm prevention and risk management
1133
to arise . This is the term used by the ILC in the 1997 Convention on
1134
International Watercourses and the 2001 Articles on Prevention of
1135
Transboundary Harm . The Commentary to both texts notes that “significantŽ
1136
harm need not be serious or substantial but must be “more than trivialŽ . In the
present case, the health symptoms which Colombia perfunctorily dismisses
include eye irritation, respiratory distress, skin infections and gastroinsestinal
difficulties1137. None of these are trivial effects. Nor can the evidence of pollution
of essential water supplies, crop damage and loss of harvests, or harm to farm
animals, be classed as trivial. Some villagers have had to move away from polluted
1132
See supra Chap. 3, Sec. I.
1133
CCM, Chap. 8, para. 8.59.
1134Convention on the Law of the Non-Navigational Uses of International Watercourses, Art. 7
(1997), 36 I.L.M. 700 (1997).
1135International Law Commission, Draft Articles on the Prevention of Transboundary Harm
from Hazardous Activities, Arts. 1-4, 8-12 (2001) (hereinafter “Draft Articles on Prevention of
Transboundary HarmŽ).
1136
United Nations General Assembly, Convention On the Law of the Non-Navigational Uses of
International Watercourses, Report of the Sixth Committee Convening as the Working Group of
the Whole, p. 5, U.N. Doc. A/51/869 (11 Apr. 1997). United Nations General Assembly, Report
of the International Law Commission on the Work of its Fifty-Third Session (23 April–1 June and
2 July–10 August 2001), p. 152, paras. (4)-(7), U.N. Doc. A/56/10 (2001).
1137See supra Chap. 3, Section I, esp. para. 3.14.
422 1138
border areas . These effects are all documented in the Memorial and in Chapter 3
1139
of this Reply . Under human rights law, effects of the scale and kind inflicted on
Ecuador have been more than sufficient to establish violations of the right to life, the
right to private life, the right to property, and the right to health and a healthy
environment 1140. They are more than sufficient to satisfy the threshold standard of
significant harm in general international law. Colombia is responsible in
international law for its failure to prevent this harm, and Ecuador is entitled to
reparation as indicated in Chapter 8.
6.54 In response to Colombia’s assertion that it took “reasonable precautionsŽ
1141
when conducting its aerial spraying programme , it is only logical that the
reasonableness of the precautions allegedly taken depends on the risk of harm that
has been created. Colombia asserts that the herbicide was a Class III substance,
i.e., that it was only “mildly toxicŽ 1142. Ecuador has already shown in Chapter 2
1138See e.g., supra Chap. 3, paras. 3.83, 3.114. See also e.g., Declaration of Witness 10, 16 Jan.
2009, p. 4. EM, Vol. IV, Annex 198; Declaration of Witness 11, 16 Jan. 2009, p. 4. EM, Vol. IV,
Annex 199.
1139
EM, Chap. 6 ; supra Chap. 3, Section I.
1140Social and Economic Rights Action Centre and the Centre for Economic and Social Rights v.
Nigeria, ACHPR Comm. 155/96 (2002); Maya Indigenous Communities of the Toledo District v.
Belize, Inter-Am. C.H.R., Report No. 40/04, Case 12.053 (12 Oct. 2004); López Ostra v. Spain,
Judgment, ECHR, Series A No. 303-C (9 Dec. 1994); Guerra and Others v. Italy, Judgment,
ECHR, Reports of Judgments and Decisions 1998 I (19 Feb. 1998); Fadeyeva v. Russia [2005]
ECHR 376; Öneryildiz v. Turkey [2004] ECHR 657; Case of Taskin and Others v. Turkey,
Judgment, ECHR, paras. 113-119 (10 Nov. 2004); Tatar v. Romania, No. 67021/01 ECHR, para.
88 (2009); Budayeva, al. v. Russia, Judgment, No. 15339/02, ECHR (2008).
1141
CCM, Chaps. 7 and 8, paras. 7.187, 8.49.
1142
CCM, Chap. 8, para. 8.49.
423of this Reply that this is not true, and that Colombia has misrepresented the
composition of the compounds actually used at various times 114. The Counter-
Memorial’s claim that only two glyphosate-based products have been used
1144
(Roundup SL and GLY-41) is inconsistent with the evidence that Colombia
used a different product, Roundup Export, and discontinued its use only after the
U.S. Environmental Protection Agency gave it the highest possible toxicity rating
1145
and determined, among other things, that it causes permanent eye damage . It
is also inconsistent with the evidence that Colombia used Roundup Ultra as
well 114. Indeed, all of the compounds used in the aerial spraying programme are
significantly more toxic to human health than Colombia acknowledges. The
warnings found on manufacturers’ labels show clearly the risks to human health
1147
and legitimate agriculture and livestock created by using these chemicals .
They point to the need for the utmost care and caution. The addition of POEA and
Cosmo-Flux further exacerbates the problem that these chemicals are simply
inappropriate for spraying in inhabited areas. In many countries the aerial
spraying of chemicals such as those used by Colombia is banned altogether or
severely restricted 1148. The European Union directive on aerial spraying
1143See supra Chap. 2, paras. 2.17-2.63.
1144
CCM, Chap. 4, para. 4.50.
1145
See supra Chap. 2, paras. 2.18-2.23.
1146
See supra Chap. 2.
1147See supra Chap. 2, paras. 2.19-2.20, 2.27-2.29, 2.32-2.34, 2.37-2.41.
1148See supra Chap. 4, paras. 4.99-4.114.
424succinctly summarises why aerial spraying has to be severely limited and
stringently controlled:▯
“Aerial Spraying of pesticides has the potential to cause significant
adverse impacts on human health and the environment, in
particular from spray drift. Therefore, aerial spraying should
generally be prohibited with derogations possible where it
represents clear advantages in terms of reduced impacts on human
health and the environment in comparison to other spraying
methods, or where there are no viable alternatives, provided that
the best available technology to reduce drift is usedŽ 114.
6.55 For all these reasons and others reviewed in detail in Chapter 2 of this
Reply, it is correct to characterize aerial application of the glyphosate-based
formulations used by Colombia in inhabited border areas as “inherently
1150
hazardousŽ and Ecuador rejects Colombia’s claim that they are not . The aerial
1149
European Parliament and the Council of the European Union, Directive 2009/128/EC:
Establishing a Framework for Community Action to Achieve the Sustainable Use of Pesticides,
para. (14) (21 Oct. 2009). ER, Vol. IV, Annex 109.
1150
See supra Chap. 2, paras. 2.17-2.61; R. John Hansman, Ph.D. & Carlos F. Mena, Ph.D.,
Analysis of Aerial Eradication Spray Events in the Vicinity of the Border Between Colombia and
Ecuador from 2000 to 2008, p. 30 (Jan. 2011) (concluding that Colombia's aerial spraying
operations have routinely violated numerous operational parameters).ER, Vol. II, Annex 1;
Durham K. Giles, Ph.D., Spray Drift Modeling of Conditions of Application for Coca Crops in
Colombia, p. 48 (Jan. 2011) (concluding that meaningful quantities of spray drift are expected to
travel 10 kilometres or more). ER, Vol. II, Annex 2; Stephen C. Weller, Ph.D., Glyphosate-Based
Herbicides and Potential for Damage to Non-Target Plants Under Conditions of Application in
Colombia, p. 25 (Jan. 2011) (concluding that drift resulting from Colombia’s aerial spraying
operations would be expected to cause injury to plants at distances of up to 10 kilometres from the
site of application). ER, Vol. II, Annex 3; Charles A. Menzie, Ph.D. & Pieter N. Booth, M.S.,
Response to: “Critique of Evaluation of Chemicals Used in Colombia’s Aerial Spraying Program,
and Hazards Presented to People, Plants, Animals and the Environment in Ecuador,” As
Presented in: Counter-Memorial of the Republic of Colombia, Appendix, p. 41 (Jan. 2011)
(concluding that a buffer zone of at least 10 kilometres is warranted). ER, Vol. II, Annex 6;
Reinhard Joas, Ph.D., The Development of the 2009 European Union Pesticides Directive With
Particular Focus on Aerial Spraying, pp. 13-16 (Jan. 2011) (hereinafter “Joas ReportŽ)
425spraying of these chemical compounds has predictably and foreseeably caused
1151
significant harm in Ecuador . Moreover, there is no countervailing public
benefit that could justify imposing such a burden on the population living on
Ecuador’s side of the border. In these circumstances, the measures required to
prevent transboundary harm will necessarily be more demanding than Colombia
has so far recognized. Unfortunately, the measures taken by Colombia to prevent
transboundary harm have been lax and ineffective in the extreme.
6.56 Chapter 4 details how poorly regulated Colombia’s aerial spraying
programme is, especially when compared to generally accepted international
standards endorsed by the UN Food and Agricultural Organization and followed
or exceeded in many other countries 1152. Of itself, that is sufficient to show that
Colombia has not acted with due diligence to prevent significant harm in
Ecuador 1153. When the flight data records submitted in evidence by Ecuador
show that even Colombia’s already lax regulations are regularly flouted with
evident impunity, the argument that it took “reasonable precautionsŽ to prevent
1154
transboundary harm becomes wholly unsustainable . The combination of
(concluding that Colombia’s aerial spraying programme is inconsistent with European Union
legislation designed to prevent risks to human health and the environment). ER, Vol. II, Annex 7.
1151
See supra Chap. 3, Section I.
1152
Food and Agriculture Organization of the United Nations, Guidelines on Good Practice for
Aerial Application of Pesticides (2001). ER, Vol. IV, Annex 98.
1153
See supra Chap. 4, paras. 4.77-4.87, 4.99-4.114.
1154See supra Chap. 2, Sections II, III and IV.
426inappropriate chemicals, lax regulation and inadequate enforcement is fatal to
Colombia’s case on due diligence. It cannot credibly say that it has acted with the
due diligence expected of governments which carry out or permit hazardous
activities near an international border. When compared to the standard set by the
Court in the Pulp Mills case, Colombia’s failure to act accordingly is manifest.
B. DUE D ILIGENCE IN THE PULP M ILLSCASE
6.57 With regard to the prevention of transboundary harm, the Court held in
Pulp Mills:
“The Court points out that the principle of prevention, as a
customary rule, has its origins in the due diligence that is required
of a State in its territory. It is ‘every State’s obligation not to allow
knowingly its territory to be used for acts contrary to the rights of
other States.’ A State is thus obliged to use all the means at its
disposal in order to avoid activities which take place in its
territory, or in any area under its jurisdiction, causing significant
damage to the environment of another State. This Court has
established that this obligation ‘is now part of the corpus of
international law relating to the environment’Ž.
6.58 The obligation of prevention is not an obligation of result nor does it
require in all cases the elimination of all risk.To that extent Ecuador and
Colombia agree, and it is unnecessary to respond to Colombia’s arguments on the
point. But there is plainly an obligation of diligent conduct … to take measures
that are necessary in the circumstances to minimize or prevent significant
115Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 38, para. 101 (internal citations omitted).
427transboundary harm 1156. What measures are required will depend on the character
of the risk and the likely harm. Colombia cites passages to this effect from the
ILC commentary to its draft Articles on Prevention of Transboundary Harm 1157.
Ecuador does not disagree. The Parties are thus divided not about the principle of
prevention as such, but about the nature of the risk from aerial spraying of the
various compounds used by Colombia, and about the conduct required of
Colombia to prevent transboundary harm.
6.59 Ecuador argued and continues to argue that in the circumstances of this
case the standard of care that is necessary and readily achievable to reduce the
risk of significant transboundary harm is to refrain from spraying within 10
kilometres of the border since this is the only response that can eliminate the risk
1158
of transboundary harm to Ecuador . Colombia characterises this as an
1159
“absolute prevention thesisŽ , which it claims is not supported by either the
1160
ILC commentary or by this Court’s judgment in Gabþikovo-Nagymaros .
1156Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 55, para. 187: “The Court considers that the obligation laid down in Article 36 is
addressed to both Parties and prescribes the specific conduct of co-ordinating the necessary
measures through the Commission to avoid changes to the ecological balance. An obligation to
adopt regulatory or administrative measures either individually or jointly and to enforce them is
an obligation of conduct. Both Parties are therefore called upon, under Article 36, to exercise due
diligence in acting through the Commission for the necessary measures to preserve the ecological
balance of the riverŽ. (Emphasis added).
1157CCM, Chap. 8, para. 8.52.
1158EM, Chap. 8, paras. 8.27-8.31.
1159CCM, Chap. 8, paras. 8.53-56.
1160
Ibid., Chap.8, paras. 8.55-56.
428Colombia says that the “vigilance and preventionŽ required by the Court in that
1161
case “do not entail the elimination of all risk whateverŽ . That might be true if
the potential harm were not as significant as it is in this case, or if it did not reach
a level that amounted to a threat to human health and livelihood, as well as the
natural environment, or if it were impossible to eliminate the risk by other
reasonable means. But that is not the case here. Aerial spraying of glyphosate
formulations with toxic surfactants is a threat to human health and livelihood and
the environment in Ecuador. Eliminating the risk posed by aerial spraying is not
only possible using reasonable means; it has already been accomplished. The risk
was eliminated in 2007, when Colombia stopped aerial spraying in the border
areas. It could have been eliminated in 2000, by not spraying near the border, as
Ecuador has consistently sought.
6.60 To quote the Court’s Judgment in Pulp Mills: Colombia must use “all the
means at its disposal in order to avoid activities which take place in its territory,
or in any area under its jurisdiction, causing significant damage to the
1162
environment of another State . . .Ž . Ecuador is not asking for more than the
Court itself has described as the appropriate standard to be applied. In the present
case, as Ecuador has already shown in the Memorial, Colombia has not used “all
the means at its disposalŽ in order to prevent transboundary drift causing
116Ibid.
1162
Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 38, para. 101.
429significant harm. Nor has it taken seriously the need for an appropriately
precautionary approach to harm prevention. Yet it can do both: by making
permanent its suspension of aerial spraying within 10 kilometres of the border, in
effect since 2007. This Colombia has refused to do.
C. A PPLICATION OF THE PRECAUTIONARY P RINCIPLE
6.61 Colombia argues that the precautionary principle “may be seen as
providing guidance as to how States should conduct themselves in matters
concerning sustainable developmentŽ 1163. It goes on to claim that “[t]here is no
reason to think that it modifies the substantive law as concerns transboundary
1164
harmŽ . This is not a serious argument. The most authoritative version of the
precautionary principle was adopted by consensus in Principle 15 of the 1992 Rio
Declaration on Environment and Development 116. Principle 15 specifically
refers in this context to protection of the environment, not merely sustainable
development. It provides: “In order to protect the environment, the precautionary
approach shall be widely applied by states according to their capabilities. Where
there are threats of serious or irreversible damage, lack of full scientific certainty
1163
CCM, Chap. 8, para. 8.57.
1164
Ibid.
1165United Nations Conference on Environment and Development, Rio Declaration on
Environment and Development, Principle 15, U.N. Doc. A/CONF.151/26 Vol. I, reprinted in 31
I.L.M. 874 (1992).
430shall not be used as a reason for postponing cost-effective measures to prevent
environmental degradationŽ 1166.
6.62 A moment’s consideration of the many environmental treaties that refer to
the precautionary principle or precautionary approach will immediately show that it
1167
is not limited to “sustainable developmentŽ . As a general principle, Principle 15
certainly covers transboundary pollution, and to argue otherwise is not sustainable.
6.63 But even if Colombia’s argument with respect to the precautionary
principle and sustainable development were correct, there can be no doubt that
the kind of transboundary harm suffered by Ecuador amounts to an interference
1166
Ibid.
1167See e.g., United Nations Framework Convention on Climate Change, Art. 3, U.N. Doc.
A/AC.237/18 (Part II)/Add.1 (1992), reprinted in 31 I.L.M. 849 (1992); Convention on Biological
Diversity, Preamble, 1760 UNTS 79, reprinted in 31 I.L.M. 818 (1992) and Cartagena Protocol on
Biosafety to the Convention on Biological Diversity, 2226 UNTS 208, reprinted in 39 I.L.M.
1027 (2000); Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution on
Further Reduction of Sulphur Emissions, 2030 UNTS 122 (1994); Protocol to the 1979
Convention on Long-Range Transboundary Air Pollution on Heavy Metals, 2237 UNTS 4 (1998);
Protocol to the 1979 Convention on Long-Range Transboundary Air Pollution on Persistent
Organic Pollutants, 2230 UNTS 79 (1998); 1996 Protocol to the Convention on the Prevention of
Marine Pollution by Dumping of Wastes and Other Matter, Art. 3, reprinted in 36 I.L.M. 1 (1997);
Stockholm Convention on Persistent Organic Pollutants, Art. 1, reprinted in 40 I.L.M. 532 (2001);
Convention for the Protection of the Marine Environment of the Northeast Atlantic (“Paris
ConventionŽ), Art. 2, reprinted in 32 I.L.M. 1069 (1993); Convention on the Protection and Use of
Transboundary Watercourses and International Lakes, Art. 2(5), 1936 UNTS 269, reprinted in 31
I.L.M. 1312 (1992); Convention on Co-operation for the Protection and Sustainable Use of the
Danube River, Art. 2(4) (1994) available at: http://www.icpdr.org/icpdr-pages/drpc.htm;
Convention on the Protection of the Marine Environment of the Baltic Sea Area, Art. 3(2), 1507
UNTS 167, reprinted in 13 I.L.M. 546 (1974); Rhine Convention Against Pollution by Chlorides,
Art. 4, reprinted in 16 I.L.M. 265 (1976); United Nations Conference on Straddling Fish Stocks
and Highly Migratory Fish Stocks, Art. 6, reprinted in 34 I.L.M. 1542. See Patricia Birnie, Alan
Boyle and Catherine Redgwell, International Law and the Environment 152-164 (3rd ed., Oxford U.
Press, 2009).
431with or denial of the right to sustainable development. The Court has itself noted
how “[t]his need to reconcile economic development with protection of the
1168
environment is aptly expressed in the concept of sustainable developmentŽ .
As a result of Colombia’s aerial spraying of toxic herbicides Ecuador has suffered
water pollution, soil pollution, loss of crops and harm to animals, as well as
illness in the human population and damage to the environment in border areas.
The most obvious characteristics of unsustainable development include material
harm and a lack of material benefits for those most adversely affected 116. The
kind of damage suffered by Ecuador cannot be characterised as consistent with
sustainable development.
6.64 The UN Special Rapporteur on the Right to Health has also taken the view
that the precautionary principle is applicable in the present case:
“the Special Rapporteur’s preliminary view was that there was
credible and reliable evidence that the aerial spraying of
glyphosate along the border damages the physical and mental
health of people living in Ecuador. The Special Rapporteur’s
preliminary conclusion was that the evidence provided during the
mission was sufficient to call for the application of the
precautionary principle and that, accordingly, Colombia should not
1168
Case Concerning the Gabþíkovo-Nagymaros Project (Hungary/Slovakia), Judgment, I.C.J.
Reports 1997, p. 78, para. 140.
116Social and Economic Rights Action Centre and the Centre for Economic and Social Rights v.
Nigeria, ACHPR Comm. 155/96 (2002); Maya Indigenous Communities of the Toledo District v.
Belize, Judgment, Inter-Am. C.H.R., Report No. 40/04, Case 12.053 (12 Oct. 2004); Ilmari
Lansman et al. v. Finland, ICCPR Comm. No. 511/1992 (1996).
432 recommence aerial spraying in the 10 km border zone with
Ecuador . . .Ž1170.
6.65 Ecuador agrees with the UN Special Rapporteur. The obligations of due
diligence and transboundary EIA must be interpreted and applied in a
precautionary manner in order to minimise or eliminate the real and foreseeable
risk to Ecuador. This is hardly a controversial proposition. As a leading authority
has observed, “[t]he point which stands out is that some applications of the
principle, which is based on the concept of foreseeable risk to other states, are
1171
encompassed within existing concepts of state responsibilityŽ . For the same
reason, the ILC noted that the precautionary principle is already included in the
principles of prevention and prior authorisation, and in environmental impact
1172
assessment, “and could not be divorced therefromŽ . From this perspective, the
importance of the precautionary principle as a general principle is that it redefines
existing rules of international law on control of environmental risks and
conservation of natural resources and brings them into play at an earlier stage
than before. Like the UN Special Rapporteur on the Right to Health, Ecuador’s
position is that the requirements of due diligence in the circumstances of this case
1170Report of the Special Rapporteur on the Right of Everyone to the Enjoyment of the Highest
Attainable Standard of Physical and Mental Health, Paul Hunt: Preliminary Note on Mission to
Ecuador and Colombia, Addendum, para. 17, U.N. Doc. A/HRC/7/11/Add.3 (4 Mar. 2007). EM,
Vol. II, Annex 31.
1171Ian Brownlie, Principles of Public International Law 276 (6th ed., Oxford U. Press, 2003).
1172United Nations, General Assembly, Report of the International Law Commission on the Work
of its Fifty-Second Session, 1 May -9 June and 10 July - 18 August 2000, Supplement No.10, p.
128, para. 716, U.N. Doc. A/55/10 (2000).
433must be interpreted and applied with precaution in mind, given the known risks of
glyphosate, Cosmo-Flux, POEA and other chemical additives in Colombia’s
spray mixture 117, and the uncertainties involved in such a large-scale experiment
with human health and the environment. Colombia has given no answer to that
argument. It has neither acted with due diligence nor followed a precautionary
approach to the management of transboundary risk, as the following examples
show.
D. N O PRIOR N OTIFICATION
6.66 Although there have been tens of thousands of spray flights along or near
the border, save on one occasion in November 2004, no advance notification of
spraying operations likely to affect Ecuador has ever been provided to the
Government of Ecuador or to Ecuadorians living in or near the affected areas 1174.
Colombia asserts that “[a]s to advance notifications of individual missions,
Colombia was under no obligation, having regard to the evident security
concerns, to give such notificationŽ 1175. Colombia’s current EMP includes a
Communications Strategy which aims to “keep the society and community at
large informed nationally, regionally, and locally, especially the communities
117See supra Chap. 2, paras. 2.18-2.63.
1174
CCM, Chaps. 5 & 6, para. 5.39, 6.33.
1175
CCM, Chap. 6, para. 6.34.
434 1176
located in the areas of influence of the Glyphosate sprayingŽ . If the spraying
poses a risk for Colombians that can be mitigated by advance warning then it
equally poses a risk for Ecuadorians that could be mitigated by similar warnings.
Yet no mention is made of communication with Ecuador or with the communities
affected on the Ecuadorian side of the international frontier. Colombia sees such
notification as somehow “problematicŽ and seems to assume that what Ecuador
asks for is a public enquiry conducted by Colombia on Ecuadorian territory 117.
6.67 No doubt consultation with those likely to be affected would have been an
excellent policy, and Ecuador reserves the right to argue that such consultation
should have taken place 1178. But Ecuador’s argument is much simpler: what it
complains about is Colombia’s failure to provide those living on the Ecuadorian
side of the border with any advance warning of spraying operations. Whether
such warning is given directly or via the Ecuadorian government is immaterial.
In the circumstances of this case, there was a duty to ensure that advance warning
was given to those likely to be affected by spraying operations that posed a clear
1176
Resolution Nº 1054 of 30 September 2003 of the Ministry for the Environment of Colombia,
p. 190 (hereinafter “2003 Environmental Management PlanŽ). CCM, Vol. II, Annex 50.
1177
CCM, Chap. 8, para. 8.96.
1178
Notably, under Article 6 of ILO Convention 169. See infra Chap. 7, paras. 7.57 et seq.
435and known risk to public health 117. What is a risk on one side of the border is
equally a risk on the other.
E. F AILURE TO C OMPLY WITH THE INADEQUATE EMP
6.68 The EMP sets out the conditions under which the Colombian Ministry of
Environment authorised the aerial spraying programme. The point has already
been made earlier in this Chapter that no version of Colombia’s EMP directly
1180
addresses the risk of transboundary pollution affecting Ecuador . Nor has
Colombia ever consulted Ecuador regarding the sufficiency of the EMP to protect
Ecuador’s territory, people or environment. While the EMP purports to address
the environmental risks of glyphosate, it does not take into account the risks of
surfactants or other components of the spray mixture, which may present even
1181
greater human health and environmental concerns . The current EMP also
1182
permits higher flight altitudes and removes flight speed limits during spraying .
Both of these factors are important for limiting the amount of spray drift and
preventing the spray from spreading beyond the intended target areas 118. Finally,
1179Corfu Channel (United Kingdom v. Albania), Judgment, Merits, I.C.J. Reports 1949, p. 22;
Draft Articles on Prevention of Transboundary Harm, op. cit., Art. 8; EM, paras. 8.33-8.35, 8.55-
8.62.
118See supra Chap. 6, paras. 6.36-3.42.
11812003 Environmental Management Plan, op. cit., pp. 186 (proposing studies to evaluate the
impact of glyphosate on soil, water, and plants; does not address other chemical components of
the spray mixture). CCM, Vol. II, Annex 50.
118Ibid., p. 173.
118See supra Chap. 2, paras. 2.88-2.89, 2.98, 2.104-2.107.
436 1184
the requirement for an independent audit has been removed from the EMP .
These are fundamental failings. The rules governing the Colombian spray
programme are among the world’s most lenient. Some jurisdictions take the view
that aerial spraying of herbicides is simply too dangerous. They have banned the
practice 118. Those jurisdictions that do not ban spraying impose restrictions
1186
significantly stronger than those adopted by Colombia . This Court held in the
Pulp Mills case that the exercise of due diligence entails “a careful consideration
1187
of the technology to be usedŽ . In the present case, the problem is more the
failure to give careful consideration to the way the technology is used than the
technology itself, although at least one type of spray aircraft used by Colombia
was indeed unsuitable for the task, for reasons set out in the next paragraph.
Whatever the technology, a defective and inadequate set of operational
parameters cannot satisfy Colombia’s obligation of due diligence in international
law, even if the spray planes followed it to the letter in every respect.
6.69 But they did not or could not follow it, and the reality is even worse than
the inadequate EMP. Colombia claims to have taken “every careŽ to “ensure that
1184
2003 Environmental Management Plan, op. cit. CCM, Vol. II, Annex 50.
1185
Joas Report, op. cit., p. 8. ER, Vol. II, Annex 7; supra Chap. 4, paras. 4.101, 4.104.
1186See supra Chap. 4, paras. 4.104-4.114.
1187Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 65, para. 223.
437spraying occurs only on Colombian territoryŽ and that no drift results 1188. This is
nonsense. Colombia concedes that operational parameters in the EMP must be
1189
complied with in order to avoid excessive drift and off-target damage . The
Counter-Memorial reaffirmed the mandatory nature of the operational parameters
and specifically said they are necessary to ensure the protection of people, plants
1190
and animals in Ecuador . Far from ensuring that “every careŽ is taken,
1191
compliance with the EMP, lax as it is, has been poor . The spray planes have
violated the speed and height restrictions within 10 kilometres of the border tens
of thousands of times. Colombia itself admits that the higher the speed and the
greater the height, the farther the spray will drift from its intended target. As
explained in detail in Chapter 2, flight data records show that Colombian aircraft
repeatedly disregarded the requirements of the EMP, in a way that virtually
ensured that spray would reach Ecuador. The aircraft flew too high, too fast and
1192
too close to the border when spraying , resulting in a herbicide mist forming
and drifting over Ecuador and onto watercourses flowing through Ecuadorian
territory. The high speed military aircraft used for spraying operations (the OV-
10 Bronco) were particularly unsuited to the task 1193, and the pilots were often
1188
CCM, Chap. 7, para. 7.16.
1189
Ibid., Chap. 7, para. 7.17.
1190
Ibid., Chap. 4, para. 4.74.
1191See supra Chap. 2, paras. 2.72-2.73.
1192See supra Chap. 2, paras. 2.89-2.106.
1193See supra Chap. 2, paras. 2.115-2.122.
438operating in hostile conditions under fire. They had every reason to fly high and
fast to avoid being shot 1194. In these conditions, spray that should have fallen as
1195
large droplets disintegrated into a fine mist, more susceptible to drift .
Sometimes the aircraft operated at night, or in unfavourable wind or climatic
1196
conditions, increasing the risk of drift . On thousands of occasions their
1197
spraying significantly exceeded the prescribed application rate . There were
1198
repeated problems of ill-discipline and incompetence among the pilots . All of
these facts are set out in detail in Chapter 2 of this Reply.
6.70 In June 2003, the Ministry of Environment fined the DNE for violating
various emergency measures imposed by the Ministry in Resolution 341 of 2001,
including the immediate requirement to carry out environmental impact
studies 1199. In a subsequent ruling, the Council of State held that the DNE was
obligated to comply with Resolution 341:
“the guidelines stated by the environmental authorities should be
followed when illicit crops are being sprayed, and not even the
slightest deviation from these should be permitted, which means
that it is therefore necessary for permanent controls to be
1194
See supra Chap. 2, paras. 2.108-2.114.
1195See supra Chap. 2, paras. 2.97-2.98.
1196See supra Chap. 2, paras. 2.144-2.154.
1197See supra Chap. 2, paras. 2.123-2.133.
1198
See supra Chap. 2, paras. 2.173-2.185.
1199
Republic of Colombia, Ministry of Environment Resolution No. 670, Whereby a Sanction is
Imposed and Other Decisions Are Made (19 June 2003). EM, Vol. II, Annex 19. See supra
Chap. 4, para. 4.65.
439 undertaken, with continuous evaluations of any effects which
might begin to appear . . . . It is nevertheless advisable to order
the Ministry of the Environment to continue to comply strictly
with the Environmental Management Plan, and furthermore to not
stop carrying out studies with a view to obtaining even more
details of the effects of the chemical compound that is used in the
1200
spraying, with verification by the National Narcotics DivisionŽ .
The flight data records show that these orders from Colombia’s highest
administrative tribunal were regularly ignored by the agencies responsible for
execution of the aerial spraying programme … just as they previously ignored the
resolutions of the Ministry of Environment.
6.71 Colombia claims to have audited compliance with the EMP. Why then
has it not made the results of those audits available to the Court? The evidence
provided by the flight data records for 2004 through 2008 and reviewed in
Chapter 2 of this Reply shows clearly that the strict compliance ordered by the
1201
Council of State in 2004 did not occur . Moreover, as Chapter 2 of this Reply
points out, the Counter-Memorial did not present the Court with any of the data
Colombia claims to have recorded 120. Colombia did not submit any of the audit
reports. In contrast, the data recorded by the spray planes and obtained by
Ecuador from the U.S. Department of State is manifestly inconsistent with the
description of the aerial spraying programme provided in the Counter-
1200State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p.11 (19 Oct. 2004) (emphasis in original). ER, Vol.
V, Annex 151.
120See supra Chap. 2, Sections II-IV.
1202
See supra Chap. 2, paras. 2.5, 2.71, 2.82, 2.154.
440Memorial 120. The conclusion is obvious: the spray programme violates an order
imposed by Colombia’s highest administrative tribunal and Colombia does
nothing to enforce either the rules laid down in the EMP or the orders of its own
courts. This is not compatible with the reasonable standard of due diligence
required by international law as set out in the Pulp Mills case. As this Court said
in that case:
“It [due diligence] is an obligation which entails not only the
adoption of appropriate rules and measures, but also a certain level
of vigilance in their enforcement and the exercise of administrative
control applicable to public and private operators, such as the
monitoring of activities undertaken by such operators, to safeguard
the rights of the other party. The responsibility of a party to the
1975 Statute would therefore be engaged if it was shown that it
had failed to act diligently and thus take all appropriate measures
to enforce its relevan1204gulations on a public or private operator
under its jurisdictionŽ .
Colombia has manifestly not taken “all appropriate measures to enforce its
relevant regulationsŽ or exercised a “level of vigilance in their enforcementŽ as
required by the Court. It is ultimately responsible for that failure.
F. B UFFER ZONES
6.72 Colombia claims that it did not spray near Ecuador, or otherwise spray in
or near vulnerable areas, including areas of human settlement, water bodies and
120See supra Chap. 2, Sections II-IV.
1204
Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, p. 58, para. 197.
441other sensitive environments. The flight data outlined in Chapter 2 show that this
claim too is false 1205. Colombia has sprayed in very close proximity to highly
sensitive areas in Ecuador … reserves set aside to protect the vulnerable
communities of the Awá and Cofán indigenous peoples living on their traditional
lands on both sides of the border 120. Only since Colombia stopped border
spraying in 2007 has the problem of toxic spray deposition drifting into Ecuador
been tackled effectively. However, Colombia has made no commitment not to
resume border spraying. Ecuador thus remains at risk of suffering further
transboundary harm unless this temporary 10 kilometre buffer zone is made
permanent.
6.73 In this Chapter, as in the Memorial, Ecuador has identified various ways …
including observance of a 10 kilometre no-spray zone adjacent to the Ecuadorian
border … in which Colombia could and should have exercised greater diligence in
conducting the spraying operation so as to eliminate or reduce the harm and the
1207
risk of harm to Ecuador . To reiterate, Ecuador’s case is not that international
law or the obligation of due diligence require in all circumstances the elimination
of all risk. It argues only that necessary precautionary measures to prevent
significant harm or the risk of such harm must be taken. Even Colombia agrees
120See supra Chap. 2, Section IV.
120See supra Chap. 2, paras. 2.170-2.179.
1207
EM, Chap. 8, paras. 8.30-8.31.
442 1208
that precautions are necessary in order to protect Ecuador , yet it has failed to
ensure that they are followed by those agencies charged with conducting the
spray operation. International law requires at a minimum that Colombia assess the
risks to Ecuador, adopt a proper EMP consistent with its due diligence
obligations, and enforce its operational requirements. Colombia has done none of
these.
6.74 Given the large scale of the spraying operations, the uncertain
composition and effects of the toxic chemicals in use, and the known risks of
aerial spraying, Ecuador is entitled in international law to expect Colombia to
1209
take the necessary steps to prevent foreseeable harm from occurring . In
Ecuador’s submission, that has always meant no spraying in border areas in
circumstances where harm to Ecuador or its people is likely to result. It is not an
unreasonable request in the circumstances, as Colombia belatedly recognized in
2007. It could have been accepted right from the beginning when Ecuador first
made the request not to spray in border areas. At the very least, it should be made
permanent, if not by Colombia itself then by order of the Court.
1208Colombia’s Ministry of Foreign Affairs assured Ecuador on 20 December 2006 that the aerial
spraying “. . . is executed under the strictest technical measures which guarantee the protection of
the environment and human health, also preventing the sprayed mixture to reach Ecuadorian
territoryŽ. Diplomatic Note from the Colombian Foreign Ministry to the Ecuadorian Embassy in
Bogotá, 20 Dec. 2006. CCM, Vol. II, Annex 26.
1209
Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay), Judgment, I.C.J.
Reports 2010, pp. 38, 55, 58, 60-61, 65, paras. 101, 187, 197, 204-205, 223.
443 G. F AILURE TO C OOPERATE AND M ONITOR
6.75 The point of cooperation in an environmental context is to help the
parties prevent or mitigate transboundary harm. The ILC Commentary to Article
4 of its draft Articles on Prevention of Transboundary Harm makes the point
cogently:
“The principle of cooperation between States is essential in
designing and implementing effective policies to prevent
significant transboundary harm or at any event to minimize the
risk thereof. The requirement of cooperation of States extends to
all phases of planning and of implementation. Principle 24 of the
Stockholm Declaration and principle 7 of the Rio Declaration
recognize cooperation as an essential element in any effective
planning for the protection of the environment. More specific
forms of cooperation are stipulated in subsequent articles. They
envisage the participation of the State likely to be affected in any
preventive action, which is indispensable to enhance the
effectiveness of any such action. The latter State may know better
than anybody else, for instance, which features of the activity in
question may be more damaging to it, or which zones of its
territory close to the border may be more affected by the
transboundary effects of the activity, such as a specially vulnerable
ecosystemŽ 1210.
6.76 Ecuador has already set out in the Memorial its arguments on the
international obligation of transboundary cooperation to manage environmental
1210
United Nations General Assembly, Report of the International Law Commission on the Work
of its Fifty-Third Session (23 April–1 June and 2 July–10 August 2001), p. 396, U.N. Doc.
A/56/10 (2001).Article 4 of the ILC Draft Articles on Prevention of Transboundary harm
provides: “States concerned shall cooperate in good faith and, as necessary, seek the assistance of
one or more competent international organizations in preventing significant transboundary harm
or at any event in minimizing the risk thereofŽ. Ibid.
444risks1211. Colombia’s response to Ecuador’s arguments on non-cooperation is
twofold. It says firstly that it had no duty to cooperate, save under Article 14 of
1212 1213
the 1988 Narcotics Convention . Secondly, it argues that it did cooperate .
6.77 Colombia asserts that the 1988 Narcotics Convention is the “explicit legal
basisŽ for cooperation on matters relating to eradication of illicit drug crops 1214.
It claims that the 1988 Narcotics Convention is a lex specialis with respect to
transborder cooperation on drug eradication, but its reading of Article 14 of the
1215
Convention is curiously fragmented . It complains that Ecuador did not
cooperate with respect to eradication measures as provided for in Article 14(3),
while at the same time discounting the reference in Article 14(2) with respect to
fundamental human rights and taking due account of the protection of the
environment 121. Colombia ignores Article 2 entirely, although this article sets
out important conditions on the measures the parties may take. To recall, Article
2(2) provides that:
“The Parties shall carry out their obligations under this Convention
in a manner consistent with the principles of sovereign equality
and territorial integrity of States and that of non-intervention in the
domestic affairs of other StatesŽ.
1211
EM, Chap. 8, paras. 8.38-8.71
1212
CCM, Chap. 8, paras. 8.113-8.121.
1213
Ibid., Chap. 8, paras. 8.105-8.112.
1214Ibid., Chap. 8, para. 8.114.
1215Ibid., Chap. 8, paras. 8.113-8.121.
1216Ibid., Chap. 8, paras. 8.117-8.119.
4456.78 It seems obvious, as explained earlier in this Chapter, that the objectives
of the 1988 Narcotics Convention were carefully framed to ensure respect for the
territorial integrity of other States, including the human rights of their citizens,
and the protection of their environments 121. In that context the cooperation
envisaged by Article 14(3) cannot be interpreted so narrowly as to exclude the
human rights and environmental concerns referred to in Article 14(2) or the
limitations prescribed by Article 2. Put another way, Article 14(2) requires that
the general law on environmental protection, including transboundary
cooperation, must be taken into account when interpreting and applying Article
1218
14(3) . The latter article is not to be read in splendid isolation from the rest of
Article 14 or from Article 2, as Colombia prefers. In Ecuador’s reading of Article
14(3), “cooperation in eradication programmes along the common frontiersŽ
necessarily includes cooperation to prevent and mitigate the likely effects on
people and environmental damage in Ecuador caused by those programmes. Any
other interpretation would be inconsistent with the respect for Ecuador’s
territorial integrity required by Article 2.
6.79 Ecuador’s case with respect to the 1988 Narcotics Convention thus
remains that Colombia neither facilitated the exchange of scientific and technical
121See supra Chap. 6, paras. 6.10-6.28.
121Vienna Convention on the Law of Treaties, Art. 31(3)(c), 1155 UNTS 331, entered into force
27 Jan. 1980; Case Concerning Pulp Mills on the River Uruguay (Argentina v. Uruguay),
Judgment, I.C.J. Reports 2010, pp. 29, 60, paras. 62, 204.
446information as required by Article 14(3)(b), nor with respect to the hazards posed
by aerial spraying along the common frontier did it “seek to cooperateŽ as
required by Article 14(3)(c).
6.80 As regards cooperation in general international law, the Counter-
Memorial discusses the Lac Lanoux case and notes that the arbitral tribunal
rejected Spain’s argument that there existed any rule of international law giving
1219
one State a right of veto over activities in another State . Ecuador has not
argued that it has or should have a veto over spraying activities within Colombia.
6.81 What Ecuador maintains is that, because of the likely transboundary
effects in Ecuador, Colombia had a duty to cooperate by consulting Ecuador and
providing information about the chemical composition of the herbicide
compounds prior to commencing its programme of border spraying operations.
Thereafter, it should have given Ecuador due warning at an appropriate time
before each operation. It should also have cooperated with Ecuador in joint
monitoring. Ecuador reiterates that case law, multilateral and bilateral treaties, the
1988 Narcotics Convention, the ILC draft Articles on Prevention of
Transboundary Harm, and Principle 19 of the Rio Declaration, as well as
elementary considerations of humanity referred to in the Corfu Channel Case, all
point to the conclusion that neighbouring States have a duty in general
121CCM, Chap. 8, para. 8.108.
447international law to cooperate in order to control and minimize the risk of
1220
transboundary harm . They must give each other prior notice of the activity,
provide adequate information about the substances used, and the risks posed to
health, property or the environment. They must consult and negotiate in good
faith in order to identify means of preventing or minimizing the risk of
transboundary harm before it occurs 1221. Colombia did none of these things
before undertaking its programme of aerial spraying in border areas in January
2000. It acted entirely unilaterally and in breach of its obligations in general
international law. As detailed in Ecuador’s Memorial, Colombia did not respond
to repeated requests for information about the chemicals in use; nor did it give
1222
Ecuador advance notice of spraying, nor has it cooperated in joint monitoring .
6.82 Colombia’s own Environmental Code provides for “reciprocal and prior
noticeŽ with bordering States regarding actions taken in one State (by the
government or private parties) that may harm the environmental rights or interests
of another State. Such communication must be made with sufficient advance
notice so that the governments involved can address the situation 1223. Colombia
1220
EM, Chap. 8, para. 8.68.
1221Ibid., Chap. 8, paras. 8.63-8.70.
1222Ibid., Chap. 3, paras 3.6-3.44.
1223Republic of Colombia, National Code of Renewable Natural Resources and Environmental
Protection, Decree 2811 of 1974, Art. 10(b) (18 Dec. 1974) (“PART II: ENVIRONMENTAL
ISSUES OF INTERNATIONAL SCOPE OR INFLUENCE . Article 10: To prevent or solve
environmental problems and regulate the use of renewable natural resources shared with
neighbouring countries and, without prejudice to existing treaties, the Government will seek to
448nevertheless failed to consult Ecuador before authorising border spraying
operations even after it was alerted to Ecuador’s very real concerns 1224. Despite
repeated requests from Ecuador 1225, Colombia failed to identify the chemicals to
be used. It failed to notify Ecuador when and where spraying would take place.
Monitoring of effects on Ecuador’s side of the border plainly requires cooperation
by both States, yet that cooperation was never forthcoming. All of this made it
impossible for Ecuador to warn the border communities likely to be affected or to
monitor the spraying and its effects. Such elementary failures are not consistent
with good faith cooperation by Colombia.
complement the existing agreements or negotiate other agreements deemed appropriate . . . (b)
The reciprocal and prior notice of environmental changes or imbalances that can originate from
works or planned projects by the governments or the peoples of the respective countries, with
sufficient time in advance so that said governments can take the appropriate action when they
consider that their rights and environmental interests may be impairedŽ). ER, Vol. V, Annex 119.
1224
EM, Chap. 3, paras. 3.6-3.22; see also, supra Chap. 3, paras. 3.7, 3.99, 3.141.
1225EM, Chap. 3, paras. 3.6-3.22. Requests sent by Ecuador referring to the Environmental Impact
Assessment, the chemical compounds used, dates and locations of future spraying operations
include: 1) Diplomatic Note 12437-47 SP/DGA/DTANC, sent from the Ministry of Foreign
Affairs of Ecuador to the Embassy of Colombia in Ecuador (24 July 2000). EM, Vol. II, Annex
36. 2) Diplomatic Note 21085 SSN/DGST, sent from the Ministry of Foreign Affairs of Ecuador
to the Embassy of Colombia in Quito (16 Feb. 2001). EM, Vol. II, Annex 38. 3) Diplomatic
Note 47839 DGAF, sent from the Ministry of Foreign Affairs of Ecuador to the Embassy of
Colombia in Quito (18 Oct. 2002). EM, Vol. II, Annex 45. 4) Diplomatic Note 68934/2003-GM,
sent from the Ministry of Foreign Affairs of Ecuador to the Ministry of Foreign Affairs of
Colombia (23 Oct. 2003). EM, Vol. II, Annex 52. 5) Diplomatic Note 75204/2003-GM, sent
from the Ministry of Foreign Affairs of Ecuador to the Ministry of Foreign Affairs of Colombia
(21 Nov. 2003). EM, Vol. II, Annex 51. 6) Diplomatic Note 4820/2004-GM, sent from the
Ministry of Foreign Affairs of Ecuador to the Ministry of Foreign Affairs of Colombia (10 Feb.
2004). EM, Vol. II, Annex 55. 7) Diplomatic Note 15839/2004-GM-VM, sent from the Ministry
of Foreign Affairs of Ecuador to the Embassy of Colombia in Quito (10 Mar. 2004). EM, Vol. II,
Annex 59. 8) Diplomatic Note 20434/2003-GM, sent from the Ministry of Foreign Affairs of
Ecuador to the Ministry of Foreign Affairs of Colombia (31 Mar. 2004). EM, Vol. II, Annex 61.
4496.83 Colombia accepts that an assessment of possible risks would not be “a
one-off exerciseŽ but would involve “keeping the situation under continuous
1226
reviewŽ . In other words, it accepts a duty to monitor the effects of its spraying
programme. In 2004, Colombia’s Council of State ordered the Ministry of Social
Protection to “conduct studies geared towards determining the impact of the
1227
chemicals glyphosate, poea, and cosmoflux on the lives of ColombiansŽ . It
also ordered the DNE to “verify the effects of aerial fumigation using glyphosate,
plus poea, plus cosmoflux for illicit crop eradication on the environmental
elements, in the areas that it has selected from those that have been sprayed, by
way of sample, and its study will include sprayed areas over varying time periods.
During this process, it will continue to perform the audit required for monitoring
1228
the effects of fumigationŽ . Colombia claims that a “careful appraisalŽ is
conducted twice a year by qualified scientists with “full access to
informationŽ 1229. Yet the evidence set out in Chapter 2 of this Reply shows how
Colombia has entirely failed to monitor the spray operation’s compliance with its
own EMP requirements 123. The “strict complianceŽ with the EMP called for by
1231
the Council of State Judgment has not been met, yet Colombia appears to have
1226
CCM, Chap. 8, para. 8.89.
1227State Council of Colombia, Claudia Sampedro and Others, Judgment on Appeal From the
Administrative Tribunal of Cundinamarca, p. 2 (19 Oct. 2004). ER, Vol. V, Annex 151.
122Ibid., p. 11, para. 4.
122CCM, Chap. 8, para. 8.90.
123See supra Chap. 2, paras. 2.88-2.154.
1231
See supra Chap. 6, paras. 6.70-6.71.
450no idea that this is the case. Either Colombia has not been monitoring the
programme and does not know it is not working, or it knows but is unwilling to
admit the reality.
6.84 Against this background, it is impossible to view the diplomatic contacts
between the two sides as meaningful cooperation. Colombia’s failure to
cooperate is a violation of its obligations in general international law, a breach of
the 1988 Narcotics Convention, and of the 1992 Convention on Biological
Diversity, as detailed in Ecuador’s Memorial 1232. For Colombia, the border
spraying operation was in substance non-negotiable until 2007, when the
unilateral decision was taken to suspend spraying within 10 kilometres of
Ecuador. The suspension was and remains a helpful development, but it
represents a fragile truce rather than a permanent solution to the problem. The
Foreign Minister of Colombia reiterated his government’s position that the
suspension of aerial spraying near Ecuador is only temporary 123. Colombia has
rejected Ecuador’s repeated entreaties to make the suspension permanent and
binding: “Colombia was not in a position to make a commitment regarding the
123EM, Chap. 8, pp. 293-319.
123EM, Chap. 3, para. 3.72.
451fumigation question, nor could it predict what decisions would be made in the
1234
future regarding this issueŽ .
Conclusions
6.85 With respect to transboundary harm, the principal source of applicable
law is general international law and applicable treaties, including, but not limited
to, the 1988 Narcotics Convention. In accordance with the law of treaties, the
1988 Narcotics Convention must be interpreted and applied taking applicable
general international law into account and in accordance with Article 14(2).
6.86 The relevant general international law is to be found, inter alia, in the ILC
draft Articles on Prevention of Transboundary Harm and in judgments of this
Court, in particular the Court’s 2010 Judgment in Pulp Mills on the River
Uruguay. The threshold of significant harm required by Article 2(a) of the ILC
articles is more than met given the risk posed by aerial spraying of a chemical
compound, known to be toxic, in border areas adjacent to human habitation,
crops, domestic animals and water supplies. Colombia is wrong to say that
Ecuador claims to be “completely immuneŽ from incidental effects of activities
carried out in Colombian territory 123. Ecuador’s evidence shows that these
activities have caused significant harm to people, property and the environment in
123Ibid.
123CCM, Chap. 8, para. 8.122(3).
452Ecuador, and that the precautionary principle is applicable when applying the
obligation of due diligence to the circumstances of this case.
6.87 Colombia has failed to take appropriate or adequate precautionary
measures to prevent or mitigate significant harm. In particular, it failed to carry
out any EIA prior to commencing spraying border areas in January 2000, or
1236
subsequently . The EMP does not constitute an EIA under international law or
1237
even under Colombian law . The failure to conduct an EIA prior to
implementation of the border spraying programme, or at all, constitutes a
violation of general international law, in accordance with this Court’s 2010
Judgment in Pulp Mills on the River Uruguay, and is additionally a violation of
Articles 2 and 14 of the 1988 Narcotics Convention.
6.88 Colombia has also failed to conduct the aerial spraying in a manner
consistent with its obligation of due diligence. Aerial spraying of glyphosate-
based herbicides is inherently hazardous to human health and livelihood and the
environment in Ecuador. In the circumstances of this case, the required standard
of diligence is readily achievable … the risk of transboundary harm from aerial
spraying can be eliminated by not spraying near the border. The EMP sets out the
operational parameters for spraying, but it provides inadequate protection against
123Supra Chap. 4, Section I and II.
123Supra Chap. 6, paras. 6.44-6.51.
453transboundary impacts, and in any event it has not been complied with by the
planes conducting the spraying operation. The failure to take adequate preventive
measures when implementing the border spraying programme, or even to enforce
the EMP, constitute violations of general international law, in accordance with
this Court’s 2010 Judgment in Pulp Mills on the River Uruguay, and additionally
violate Articles 2 and 14 of the 1988 Narcotics Convention.
6.89 Colombia has failed to cooperate in managing the transboundary risk
posed by aerial spraying, in particular by failing to consult Ecuador before
authorising the spraying operation, by failing to provide information on the
composition of the spray, by failing to give advance warning of spraying
operations, by failing to cooperate in joint monitoring, and by failing to consult
and negotiate in good faith with Ecuador once alerted to Ecuador’s very real
concerns. The failure to cooperate is a violation of general international law, of
the 1992 Biological Diversity Convention, and of Articles 2 and 14 of the 1988
Narcotics Convention.
6.90 Finally, Colombia’s violations of its international obligations have
resulted in significant harm to persons, property and the environment in Ecuador.
It is responsible in international law for this harm and is under an obligation to
make reparation as set forth in Chapter 8 of this Reply.
454 CHAPTER 7.
VIOLATIONS OF HUMAN RIGHTS AND INDIGENOUS PEOPLES’
RIGHTS Section I. Overview
7.1 In its Memorial, Ecuador made three arguments with respect to the
violation of human rights by Colombia. First, it alleged that Colombia has
violated the rights of indigenous peoples in Ecuador, in contravention of the ILO
Convention No. 169 Concerning Indigenous and Tribal Peoples in Independent
Countries, Article 27 of the 1966 International Covenant on Civil and Political
Rights (“ICCPRŽ), and Article 21 of the American Convention on Human
1238
Rights . Second, it alleged that Colombia has violated the human rights of
persons in Ecuador, including the right to life, health, private life, food and water,
property, humane treatment, a healthy environment, and information, in
contravention, inter alia, of the 1969 American Convention on Human Rights
with the 1988 Additional Protocol, the International Covenant on Economic,
Social and Cultural Rights (“ICESCRŽ), the 1979 Convention on the Elimination
of All Forms of Discrimination Against Women (“CEDAWŽ), the 1989
1239
Convention on the Rights of the Child (“CRCŽ), and the ICCPR . Finally, it
argued that human rights obligations are incorporated in the 1988 Narcotic
1240
Convention via Article 14(2) and have been violated in the same way .
123Memorial of Ecuador, Vol. I, Chap. 9, paras. 9.18-9.36 (hereinafter “EMŽ).
1239
EM, Chap. 9, paras. 9.43-9.107.
1240
EM, Chap. 8.
4577.2 In its Counter-Memorial, Colombia declares that Ecuador’s claims
concerning human and indigenous rights are unsubstantiated and “unsustainableŽ
1241
under applicable treaties or under customary international law . In general, it
argues: (1) that the “factual underpinningŽ of the claims is inconsistent with the
scientific evidence, and that the evidence of harm submitted by Ecuador is
1242
insufficient and unreliable ; (2) that the aerial spraying is not targeted at people
1243
living across the border in Ecuador ; (3) that there is no allegation of death or
serious long-term illness 1244; and (4) that the effects of herbicide pollution are so
de minimis that Colombia has no obligations under international law which would
require it to respect human rights or the rights of indigenous peoples outside its
1245
own territory . Furthermore, Colombia contests Ecuador’s allegation that
violations in three interdependent fields of international law … namely
international environmental law, international human rights law, and international
protection of indigenous peoples … may constitute an “integrated, mutually
1246
reinforcing wholeŽ . In Colombia’s view, Ecuador is supposed to substantiate
each breach independently from the other in the three different fields. Finally,
with respect to the effects on indigenous peoples, Colombia makes the same
1241
Colombia Counter-Memorial, Chap. 9, para. 9.1 (hereinafter “CCMŽ).
1242
CCM, Chap. 9, para. 9.1.
1243
CCM, Chap. 9, paras. 9.13-9.14.
1244CCM, Chap. 9, para. 9.61.
1245CCM, Chap. 9, paras. 9.22, 9.63.
1246CCM, Chap. 9, paras. 9.4-9.8.
458arguments about evidence, territoriality of obligations, and targeting 124. It also
claims that Colombia has done nothing which could amount to a denial of the
1248
rights of indigenous peoples .
7.3 Following this overview, the Chapter is divided into five sections:
▯ In Section II, Ecuador rebuts Colombia’s contentions that its actions
have not affected the human rights of the population living on the
Ecuadorian side of the border. Contrary to the impression created in
Colombia’s Counter-Memorial, all concerned stakeholders in the
matters underlying the present dispute … with the exception of the
Government of Colombia in these proceedings … share the conclusion
that spray drift has crossed the border and caused significant risks and
significant harm, not only to the natural environment and crop fields,
but also … directly and indirectly … to the human population living
across the border on Ecuadorian territory. These consequences entail
serious human rights violations for which Colombia must bear
responsibility.
▯ Section III addresses the question of the so-called “territorialityŽ of
human rights and indigenous rights. It will be seen that Colombia,
124CCM, Chaps. 7 and 9, paras. 7.178-7.186, 9.13-9.14, 9.159, 9.164-9.167,
124CCM, Chap. 9, paras. 9.165, 9.169.
459 Ecuador and all other Latin American States share a regional legal
space aimed at securing human rights protection throughout the entire
region. This reality is incompatible with Colombia’s attempt to
exclude from international legal regulation the extra-territorial effects
of activities detrimental to human rights and indigenous peoples’
rights in Ecuador, and distinguishes the inapposite authorities on
which Colombia relies.
▯ Section IV responds to Colombia’s arguments on the special status of
indigenous peoples under international law. Ecuador shows that
significant harm has been caused to indigenous communities located
on both sides of the border and to their ability to sustain a traditional
subsistence existence. As with human rights generally, Ecuador
demonstrates that the obligation to respect indigenous rights is not
territorially limited and that specific targeting of indigenous peoples is
not required for there to be a violation of international law. These
communities should have been consulted and notified before spraying
began.
▯ In Section V, Ecuador presents its views on the interrelationship
between environmental protection and human rights. The conclusion
is that the inter-relationship between unlawful damage to the human
460 environment and breach of human rights obligations cannot be
ignored.
▯ Finally, Section VI sets out a summary of conclusions to be drawn
from the previous sections.
Section II. Colombia’s Violations of Human Rights
A. A ERIAL S PRAYING H AS SERIOUSLY INJURED AFFECTED INDIVIDUALS
7.4 Colombia systematically dismisses the probative value of the evidence
provided by Ecuador to support the breach of fundamental human rights, on the
1249
basis that it lacks “independently verified expert reportsŽ or that witness
1250 1251
statements provided are “vagueŽ or of “doubtful weightŽ . Ecuador has
1252
already replied thoroughly in Chapter 3 to such accusations . To support its
criticism of the evidence submitted by Ecuador of human rights violations, which
consumes nearly 100 paragraphs of response to Ecuador’s claims of human rights
violations, Colombia provides the repeated refrain that the witnesses are
1253
“vagueŽ . In Chapter 3 of this Reply, Ecuador has demonstrated, contrary to
Colombia’s attempt to claim otherwise in the Counter-Memorial, not only the
124CCM, Chap. 9, para. 9.75.
125CCM, Chap. 9, para. 9.124.
1251
CCM, Chap. 9, para. 9.77.
1252
See supra Chap. 3, paras. 3.23-3.27, 3.34-3.46.
125See e.g. CCM, Chap. 9, paras. 9.77, 9.131.
461specificity of the witness testimonies … in terms of the appearance of the spray
mist, and the impacts on their eyes, skin, respiratory and digestive systems, crops
and livestock … but also their remarkable consistency with one another, across
different time periods and geographic locations; with the spray flight data
obtained from the U.S. Government; with the effects of glyphosate-based
herbicidal sprays; and with the labelling warnings and instructions provided by
the manufacturers of the chemical products that Colombia uses in its spray
mixtures.
7.5 Colombia seeks also to dismiss the confirming value of the reports of the
UN Special Rapporteurs on the Right to Health, the Rights of Indigenous Peoples
and the Right to Food. Colombia states that the views of the Special Rapporteur
on the Right to Food are “in no way ‘authoritative’Ž 1254. It also dismisses the
reports from this Rapporteur and the Special Rapporteur on the Rights of
Indigenous People, claiming that they “are based on allegations of individuals and
have no scientific basisŽ 1255. Similarly, it asserts that the opinion of the Special
Rapporteur on the Right to Health that there is “‘credible, reliable evidence’Ž is
merely a matter of “express[ing] his viewsŽ and “adds nothing to what he was
1256
told by his informantsŽ and is not supported by any scientific evidence .
125CCM, Chap. 9, para. 9.91.
1255
CCM, Chap. 9, para. 9.93. See also CCM, Chap. 9, para. 9.109.
1256
CCM, para. 9.132.
462Chapter 3 of this Reply addresses the value of the mission carried out by the
1257
Special Rapporteur on the Right to Health . Among other things, this expert
took witness statements and gathered scientific evidence and engaged with
Colombian authorities. The findings of other UN appointed experts, such as the
Special Rapporteur on the Rights of Indigenous Peoples, are also of comparable
rigour and independence. They all confirm Ecuador’s arguments and evidence.
7.6 Colombia’s assessment of the work of UN Special Rapporteurs
misrepresents their independent and expert nature, and it also seeks to dismiss the
evidentiary value of their reports. As explained in Chapter 3 of this Reply, the
series of statements of local people, gathered at different times, in different
locations and by different experts, including those appointed by the United
Nations, matched with the flight path data, provide systematic and consistent
evidence of the risks and the harm caused to the people living in the border
areas1258.
7.7 The record shows that, since 2000, the living conditions of the populations
residing along the Ecuadorian side of the border have been significantly affected
by Colombia’s aerial sprayings. Ecuador has provided extensive evidence from
first-hand observers, corroborating the fact that the harm is directly linked to the
125See supra Chap. 3, paras. 3.65-3.70.
125See supra Chap. 3, Sections I and II, esp. paras. 3.7, 3.38.
463spraying of toxic herbicides which repeatedly took place in direct proximity to
their homes and property.
7.8 As Colombia’s arguments on the probative value of the evidence
submitted by Ecuador have already been rebutted in Chapter 3 of this Reply, it is
not necessary to address this matter further in this Chapter1259. However, Ecuador
is bound to reaffirm in the clearest possible terms its concerns as to the serious
human consequences of Colombia’s past actions relevant to the present dispute.
First, there is no doubt that tens of thousands of spraying operations have been
conducted in direct proximity to the border. The repeated sprayings in the border
region have been consistently mentioned in all witness statements submitted to
the Court and are corroborated by the flight data now available in these
proceedings 126. Equally relevant is that nearly all witnesses personally saw or
felt the spray drift towards the Ecuadorian side of the border.
7.9 Second, the record shows that the spray drift has caused serious health
problems. Two fundamental causes can be identified: the direct contact with the
spray drift deposited directly over individuals, and the more indirect effects
resulting from damage to food and water supplies and medicinal plants. With
respect to the first cause, many witnesses on Ecuadorian soil and in direct
125See supra Chap. 3, Sections I and II, esp. paras. 3.23-3.46.
126See supra Chap. 3, Section I.
464proximity with the planes saw the spray drift towards them, after which they felt
1261
the spray on their bodies . A clear example is provided by the statement of
Witness 32 who was in his canoe as the spray fell directly on him and the
1262
surrounding water body . In this respect, the information provided by the
witnesses has not been seriously challenged, and it cannot be. As regards the
second and more indirect cause, the previous Chapters of the Reply have already
clearly established the extent of environmental damage caused to the direct living
1263
environment of the affected populations . In short, the spray drift has given
rise to serious risks and destroyed significant amounts of the subsistence crops of
the affected individuals, causing food shortages. Drinking water has been
polluted and other food such as fish or farm animals have been poisoned before
being ingested by the local populations 1264.
7.10 Apart from the fact that they saw the spray drift falling on them and their
property, all the witnesses also observed the direct consequences on their
immediate surroundings. One witness explained how his subsistence crops died
1261See, e.g., Declaration of Witness 2, 16 Jan. 2009 (hereinafter “Witness 2 DeclarationŽ). EM,
Vol. IV, Annex 190; Declaration of Witness 4, 22 Dec. 2008 (hereinafter “Witness 4
DeclarationŽ). EM, Vol. IV, Annex 192; Declaration of Witness 6, 16 Jan. 2009 (hereinafter
“Witness 6 DeclarationŽ). EM, Vol. IV, Annex 194; Declaration of Witness 11, 16 Jan. 2009
(hereinafter “Witness 11 DeclarationŽ). EM, Vol. IV, Annex 199; Declaration of Witness 28, 17
Feb. 2009 (hereinafter “Witness 28 DeclarationŽ). EM, Vol. IV, Annex 212; Declaration of
Witness 33, 19 Feb. 2009 (hereinafter “Witness 33 DeclarationŽ). EM, Vol. IV, Annex 217.
1262Declaration of Witness 32, 19 Feb. 2009 (hereinafter “Witness 32 DeclarationŽ). EM, Vol.
IV, Annex 216.
1263See supra Chap. 3, Section I.
1264See e.g., supra Chap. 3, paras. 3.11, 3.82, 3.94-3.97; see also EM, paras. 9.70-9.74.
465 1265
from top down, and not as is usual from the roots upwards . Others explained
how the sprayings have affected the areas surrounding them, observing that the
effects were strongest on the Colombian side of the border, which was directly
targeted by the spray planes 126. On the Ecuadorian side of the border the damage
was still significant as corroborated by all witnesses, and the damage diminished
as the distance from the border grew. Finally, it must be stressed that the local
populations did not have any doubt as to the origin of their symptoms. The health
symptoms that they suffered were the same each time there were sprayings along
the border.
7.11 Turning to the health consequences, these have been amply described in
the written pleadings and well-documented through scientific, medical and
witness reports. All of the witnesses who have been directly affected have
declared under oath that they subsequently, although with different degrees of
intensity, suffered health effects which correspond to the symptoms associated
with the chemicals used in the spray mixture deposited by Colombia. These
1267 1268
include notably: eye irritation , skin rashes and bumps , headaches and
126Declaration of Witness 18, 15 Jan. 2009 (hereinafter “Witness 18 DeclarationŽ). EM, Vol. IV,
Annex 204.
1266See, e.g., Declaration of Witness 1, 16 Jan. 2009 (hereinafter “Witness 1 DeclarationŽ). EM,
Vol. IV, Annex 189; Declaration of Witness 3, 17 Jan. 2009 (hereinafter “Witness 3
DeclarationŽ). EM, Vol. IV, Annex 191; Declaration of Witness 10, 19 Jan. 2009. EM, Vol. IV,
Annex 198; Declaration of Witness 19, 17 Jan. 2009 (hereinafter “Witness 19 DeclarationŽ). EM,
Vol. IV, Annex 205.
1267
See, e.g., Declaration of Witness 5, 16 Jan. 2009 (hereinafter “Witness 5 DeclarationŽ). EM,
Vol. IV, Annex 193; Declaration of Witness 8, 16 Jan. 2009 (hereinafter “Witness 8
466 1269 1270 1271
dizziness , fever , and gastrointestinal difficulties . The eyewitness
testimonies are further corroborated by medical staff confronted with recurring
DeclarationŽ). EM, Vol. IV, Annex 196; Declaration of Witness 9, 16 Jan. 2009 (hereinafter
“Witness 9 DeclarationŽ). EM, Vol. IV, Annex 197; Declaration of Witness 12, 16 Jan. 2009
(hereinafter “Witness 12 DeclarationŽ). EM, Vol. IV, Annex 200; Declaration of Witness 17, 16
Jan. 2009 (hereinafter “Witness 17 DeclarationŽ). EM, Vol. IV, Annex 203; Declaration of
Witness 20, 16 Jan.2009 (hereinafter “Witness 20 DeclarationŽ), EM, Vol. IV, Annex 206;
Declaration of Witness 22, 16 Jan. 2009 (hereinafter “Witness 22 DeclarationŽ). EM, Vol. IV,
Annex 208; Declaration of Witness 37, 19 Feb. 2009 (hereinafter “Witness 37 DeclarationŽ). EM,
Vol. IV, Annex 220; Declaration of Witness 39, 19 Feb. 2009 (hereinafter “Witness 39
DeclarationŽ). EM, Vol. IV, Annex 222.
1268See, e.g., Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 2 Declaration, op.
cit. EM, Vol. IV, Annex 190; Witness 3 Declaration, op. cit. EM, Vol. IV, Annex 191; Witness 4
Declaration, op. cit. EM, Vol. IV, Annex 192; Witness 5 Declaration. EM, Vol. IV, Annex 193;
Witness 6 Declaration, op. cit. EM, Vol. IV, Annex 194; Declaration of Witness 7, 16 Jan. 2009
(hereinafter “Witness 7 DeclarationŽ). EM, Vol. IV, Annex 195; Witness 8 Declaration, op. cit.
EM, Vol. IV, Annex 196; Witness 9 Declaration, op. cit. EM, Vol. IV, Annex 197; Witness 11
Declaration, op. cit. EM, Vol. IV, Annex 199; Witness 12 Declaration, op. cit. EM, Vol. IV,
Annex 200; Declaration of Witness 13, 15 Jan. 2009 (hereinafter “Witness 13 DeclarationŽ). EM,
Vol. IV, Annex 201; Declaration of Witness 14, 17 Jan. 2009. EM, Vol. IV, Annex 202; Witness
17 Declaration, op. cit. EM, Vol. IV, Annex 203; Witness 18 Declaration, op. cit. EM, Vol. IV,
Annex 204; Witness 19 Declaration, op. cit. EM, Vol. IV, Annex 205; Witness 22 Declaration,
op. cit. EM, Vol. IV, Annex 208; Declaration of Witness 23, 16 Jan. 2009 (hereinafter “Witness
23 DeclarationŽ). EM, Vol. IV, Annex 209; Declaration of Witness 29, 16 Jan. 2009 (hereinafter
“Witness 29 DeclarationŽ). EM, Vol. IV, Annex 213; Declaration of Witness 30, 19 Feb. 2009
(hereinafter “Witness 30 DeclarationŽ). EM, Vol. IV, Annex 214; Declaration of Witness 31, 27
Feb. 2009 (hereinafter “Witness 31 DeclarationŽ). EM, Vol. IV, Annex 215; Witness 32
Declaration, op. cit. EM, Vol. IV, Annex 216; Witness 33 Declaration, op. cit. EM, Vol. IV,
Annex 217; Declaration of Witness 34, 19 Feb. 2009 (hereinafter “Witness 34 DeclarationŽ). EM,
Vol. IV, Annex 218; Witness 37 Declaration, op. cit. EM, Vol. IV, Annex 220; Declaration of
Witness 38, 19 Feb. 2009 (hereinafter “Witness 38 DeclarationŽ). EM, Vol. IV, Annex 221;
Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222.
1269See, e.g., Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 190; Witness 3 Declaration, op.
cit. EM, Vol. IV, Annex 191; Witness 5 Declaration, op. cit. EM, Vol. IV, Annex 193; Witness 8
Declaration, op. cit. EM, Vol. IV, Annex 196; Witness 9 Declaration, op. cit. EM, Vol. IV,
Annex 197; Witness 12 Declaration, op. cit. EM, Vol. IV, Annex 200; Witness 13 Declaration,
op. cit. EM, Vol. IV, Annex 201; Witness 17 Declaration, op. cit. EM, Vol. IV, Annex 203;
Witness 20 Declaration, op. cit. EM, Vol. IV, Annex 206; Witness 29 Declaration, op. cit. EM,
Vol. IV, Annex 213; Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218.
1270
See, e.g., Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 3 Declaration, op.
cit. EM, Vol. IV, Annex 191; Witness 4 Declaration, op. cit. EM, Vol. IV, Annex 192; Witness 7
Declaration, op. cit. EM, Vol. IV, Annex 195; Witness 11 Declaration, op. cit. Vol. IV, Annex
199; Witness 17 Declaration, op. cit. EM, Vol. IV, Annex 203; Witness 22 Declaration, op. cit.
EM, Vol. IV, Annex 208; Witness 29 Declaration, op. cit. EM, Vol. IV, Annex 213; Witness 34
Declaration, op. cit. EM, Vol. IV, Annex 218.
467surges of epidemics following each round of spraying that displayed the same
1272
pattern of symptoms .
7.12 This conclusion has also been reached by the UN Special Rapporteur on
the Right to Health, Mr. Paul Hunt, who expressed his serious concerns relating
to the human rights situation in the border regions affected by the sprayings, and
particularly the grave physical and mental health effects. In his opinion,
“There is credible, reliable evidence that the aerial spraying of
glyphosate along the Colombia-Ecuador border damages the
physical health of people living in Ecuador. There is also credible,
reliable evidence that the aerial spraying damages their mental
health. Military helicopters sometimes accompany the aerial
spraying and the entire experience can be terrifying, especially for
children. (Some children told me that, while they were in their
school, it was sprayed.)
This evidence is sufficient to trigger the precautionary principle.
Accordingly, the spraying should cease until it is clear that it does
not damage human health.
1271
See, e.g., Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 3 Declaration, op.
cit. EM, Vol. IV, Annex 191; Witness 4 Declaration, op. cit. EM, Vol. IV, Annex 192; Witness 5
Declaration, op. cit. EM, Vol. IV, Annex 193; Witness 7 Declaration, op. cit. EM, Vol. IV,
Annex 195; Witness 8 Declaration, op. cit. EM, Vol. IV, Annex 196; Witness 9 Declaration, op.
cit. EM, Vol. IV, Annex 197; Witness 11 Declaration, op. cit. EM, Vol. IV, Annex 199; Witness
12 Declaration, op. cit. EM, Vol. IV, Annex 200; Witness 17 Declaration, op. cit. EM, Vol. IV,
Annex 203; Witness 19 Declaration, op. cit. EM, Vol. IV, Annex 205; Witness 20 Declaration,
op. cit. EM, Vol. IV, Annex 206; Witness 22 Declaration, op. cit. EM, Vol. IV, Annex 208;
Witness 29 Declaration, op. cit. EM, Vol. IV, Annex 213; Witness 31 Declaration, op. cit. EM,
Vol. IV, Annex 215; Witness 32 Declaration, op. cit. EM, Vol. IV, Annex 216; Witness 33
Declaration, op. cit. EM, Vol. IV, Annex 217.
1272
See, e.g., Declaration of Dino Juan Sánchez Quishpe, 15 Jan. 2009. EM, Vol. IV, Annex 188;
Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223; Confederation of Indigenous
Nationalities of Ecuador (CONAIE) et al., Technical Report of the International Commission on
the Impacts in Ecuadorian Territory of Aerial Fumigations in Colombia, p. 17 (19-22 July 2001).
EM, Vol. IV, Annex 162.
468 It would be manifestly unfair to require Ecuador to prove that the
spraying damages human health because Ecuador does not have
access to essential information that is required to make that
assessment. For example, Ecuador does not know the precise
composition of the herbicide that Colombia is using. Thus,
Colombia has the responsibility to show that the spraying damages
1273
neither human health nor the environmentŽ .
Colombia challenges these conclusions by way of assertion, but provides no
evidence to contradict them 127. It still withholds the precise elements and
formulations of the various chemical mixtures it has aerially sprayed along or
near the border with Ecuador.
7.13 Colombia seeks to down play the serious effects of the spray mixture by
asserting that no evidence proving that affected populations have sought medical
attention has been submitted to the Court 127. This supposed lack of need for
medical attention would somehow underscore the benign nature of the chemical
spray’s effects on human health. Apart from the fact that medical records have
been submitted by Ecuador in the form of contemporaneous medical inquest
reports prepared by health professionals who examined victims in the days and
weeks following exposure to the spray, numerous witnesses mention that they
127U.N. Press Release, “U.N. Special Rapporteur on the Right to the Highest Attainable Standard
of Health, Paul Hunt, Ends Visit to EcuadorŽ (18 May 2007) (emphasis added). EM, Vol. IV,
Annex 185.
1274
CCM, Chap. 9, para. 9.132.
1275CCM, Chap. 7, paras. 7.133, 7.142. In paragraph 7.142, Colombia states the following: “In
spite of the severe and wide-spread symptoms allegedly suffered by such a large number of people
… a full-scale epidemic according to at least one of the witnesses … none of them seems to have
sought medical assistance, whether for themselves or for their childrenŽ. (Emphasis added).
469sought medical attention where available, even if the treatment they received was
rudimentary and not formally documented in the manner one may be accustomed
1276
to in the developed world . Thus, for instance, the affected populations in
1277
Mataje, Esmeraldas, visited the local nurse after the various spraying episodes .
One of them was so sick that she had to be transported to the Esmeraldas
1278
Hospital . Another witness explained how he had to carry sick people on his
1279
back for over five hours in order to reach the hospital at San Lorenzo .
Witnesses living within the Cofán-Bermejo Ecological Reserve recounted how
1280
they sought medical attention in the Health Centre General Farfán . A witness
in Puerto Escondido, Sucumbíos, had to bring her children to the San José Clinic
1281
in Lago Agrio . Another explained how the people in his community would
1276See, e.g., Witness 1 Declaration, op. cit. EM, Vol. IV Annex 189; Witness 2 Declaration, op.
cit. EM, Vol. IV, Annex 190; Witness 6 Declaration, op. cit. EM, Vol. IV, Annex 194; Witness 7
Declaration, op. cit. EM, Vol. IV, Annex 195; Witness 8 Declaration, op. cit. EM, Vol. IV,
Annex 196; Witness 19 Declaration, op. cit. EM, Vol. IV, Annex 205; Declaration of Witness 21,
16 Jan. 2009 (hereinafter “Witness 21 DeclarationŽ). EM, Vol. IV, Annex 207; Declaration of
Witness 27, 17 Feb. 2009 (hereinafter “Witness 27 DeclarationŽ). EM, Vol. IV, Annex 211;
Witness 30 Declaration, op. cit. EM, Vol. IV, Annex 214; Witness 32 Declaration, op. cit. EM,
Vol. IV, Annex 216; Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218; Declaration of
Witness 36, 19 Feb. 2009 (hereinafter “Witness 36 DeclarationŽ). EM, Vol. IV, Annex 219;
Witness 37 Declaration, op. cit. EM, Vol. IV, Annex 220; Witness 39 Declaration, op. cit. EM,
Vol. IV, Annex 222; Declaration of Witness 40, 20 Feb. 2009 (hereinafter “Witness 40
DeclarationŽ). EM, Vol. IV, Annex 223; Declaration of Witness 41, 20 Feb. 2009 (hereinafter
“Witness 41 DeclarationŽ). EM, Vol. IV, Annex 224.
1277See, e.g., Witness 30 Declaration, op. cit. EM, Vol. IV, Annex 214; Witness 32 Declaration,
op. cit. EM, Vol. IV, Annex 216; Witness 33 Declaration, op. cit. EM, Vol. IV, Annex 217;
Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218; Witness 37 Declaration, op. cit. EM,
Vol. IV, Annex 220; Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222.
1278
Witness 36 Declaration, op. cit. EM, Vol. IV, Annex 219.
1279
Witness 40 Declaration, op. cit. EM, Vol. IV, Annex 223.
1280See, e.g., Witness 27 Declaration, op. cit. EM, Vol. IV, Annex 211.
1281Witness 21 Declaration, op. cit. EM, Vol. IV, Annex 207.
470first go to the traditional doctor, thereafter the health promoter of the community
1282
trained in Western medicine, and if need be to the San Lorenzo Hospital . All
this medical assistance was sought in temporal proximity with the sprayings
occurring along Ecuador’s northern border. In most other cases, poorer families
could not afford to seek medical care, often hours away. This is particularly true
for the community of San Francisco 2, where the human rights damages were
particularly severe. Witness 11 expressed her despair when acknowledging the
lack of money to seek medical assistance 1283. In fact, the people of San Francisco
2 could only resort to traditional medicinal plant cures, the effects of which were
1284
impaired by the sprayings . This particular issue is addressed further below, in
the context of Ecuador’s claim relating to the violations of the rights of
indigenous peoples.
7.14 To complete this discussion on the human health effects of the aerial
spraying along Ecuador’s border, it is necessary to address what Colombia claims
to be inconsistencies or divergences among witness statements submitted to the
1285
Court . The Court should not be surprised by Colombia’s attack on the witness
statements, since they undermine Colombia’s case that there has been no harm
caused by its actions. If the witness statements are treated by the Court as
128Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224.
128Witness 11 Declaration, op. cit. EM, Vol. IV, Annex 199.
1284
See supra Chap. 3, para. 3.76-3.77. Witness 13 Declaration, op. cit. EM, Vol. IV, Annex 201.
1285
CCM, Chap. 9, paras. 9.77, 9.124, 9.131.
471probative … as Ecuador considers they should be … they constitute unchallenged
proof of the harms to human health, crops, livestock and the environment claimed
by Ecuador. Colombia has no choice other than to seek to discredit them, but it
has failed in this effort. It is not a sign of inconsistency, for example, that the
degree of harm to human health varies somewhat across time periods and
geographic locations. In fact, the contents of the spray mixture used by Colombia
varied, and some communities were hit more directly and more often than others
by the spraying. What makes the testimonies consistent, amongst other
considerations, is that the nature of the harms following spraying events is the
same … it always affects the eyes, skin, respiratory and digestive systems in
similar ways … even though the severity and duration of these symptoms felt by
different individuals sometimes varies.
7.15 Thus, in evaluating the validity of the different witness statements, the
Court is asked to take into account all relevant circumstances, most notably:
1. the fact that witnesses were affected in similar, even if not identical,
ways … this is logical given that some were in direct proximity of the
planes and the spray drift and thus felt the spray on their skin, while
others saw the drift progressively coming towards them and could not
escape, and yet others saw the activities from a distance of several
kilometres away but were nevertheless affected through damage to
their crops;
472 2. the fact that it is to be expected that different persons may react
differently even to the same dose of exposure … some may suffer
relatively mild symptoms, such as adult men in good health, others
may have stronger and especially severe reactions, such as older
persons, undernourished adults, and particularly infants with weaker
immune systems … an observation that is widely corroborated by the
witness statements where the special vulnerability of children is
frequently mentioned; and
3. the material resources available to each affected individual to react to
the sprayings differ … while some live closer to urban centres, or have
easier access to professional medical assistance, or even possess
subsistence crops in areas unaffected by the sprayings on which they
could continue to sustain their health … many of the witnesses could
not escape the direct consequences of the poisoning, or obtain medical
attention, thus being forced to rinse their bodies, to drink and cook
with contaminated water, to eat contaminated fish or subsistence
crops, or to self-medicate with contaminated medicinal plants.
7.16 It comes as no surprise that some witnesses claim to suffer from long-term
effects such as eye irritation. All these witnesses suffered from spraying
operations prior to 2005 when Colombia said it changed the composition of its
spray mix, switching from Roundup SL to GLY-41. These aspects have been
dealt with in detail in Chapter 2 of the present Reply 128. However, it must be
recalled that the U.S. Environmental Protection Agency recommended as early as
128See supra Chap. 2, paras. 2.31-2.42.
4732002 that Colombia abandon its use of the formulation then in use precisely
1287
because it could cause “irreversible eye damageŽ .
7.17 Finally, on the question of causation, Colombia argues that without any
conclusive scientific evidence no causal nexus can be established between the
1288
sprayings and the health effects reported in the witnesses’ sworn statements .
To be sure, no soil or water samples were collected immediately after spraying
events. Nor were there teams of scientists stationed in the border areas waiting
for these events so they could obtain such samples. Colombia itself made this
impractical by repeatedly failing to allow Ecuador or the border populations to
receive any advance notice of the sprayings. The remoteness of the affected
regions, and the lack of roads and other infrastructure made communication and
access to and from them especially difficult. Moreover, as Colombia well knows,
although the spray mixtures deposited along the border with Ecuador were highly
toxic, they were biodegradable in less than four weeks, meaning that no traces
1287United States Environmental Protection Agency, Office of Prevention, Pesticides, and Toxic
Substances, Report on Issues Related to the Aerial Eradication of Illicit Coca in Colombia:
Response from EPA Assistant Administrator Johnson to Secretary of State, p. 8 (Aug. 2002). ER,
Vol. III, Annex 45. And, indeed, Colombia acknowledges in a footnote of its Counter-Memorial
that it switched from Roundup SL to GLY-41 because of the inherent risk to human health, in
particular eye problems. Footnote 312 reads as follows: “As of 2005, with the purpose of avoiding
the possible effects associated with the use of the POEA surfactant contained in Roundup SL, in
particular the risk of eye irritation to workers in charge of preparing the mix at the operation sites,
the Colombian Government decided to use a new glyphosate-based formulated product, called
GLY 41Ž. CCM, para. 4.50, n. 312 (emphasis added).
1288CCM, Chap. 7, para. 7.141.
474would appear in any soil or water samples that might be collected even as little as
a month after spraying 1289.
7.18 What is remarkable on the question of causation is that all the evidence
that has been presented points in one direction. The fact that so many witnesses
and many more individuals have suffered from ill-effects merely serves to
highlight the extent of the damage produced and underscores the credibility of
Ecuador’s case. Now that Ecuador has obtained the spray flight data, it has been
able to connect temporally the sprayings with the harms testified to by the
witnesses in their immediate aftermaths and locations; this further corroborates
the witness testimonies. Also pointing in the same direction are the reports of the
various UN Special Rapporteurs; the scientific studies of the effects of glyphosate
and POEA on human health and crops; the labelling warnings and instructions
provided by the manufacturers of these products (pursuant to legal requirements
in States around the world); the banning of aerial spraying of pesticides in the
European Union and many other national jurisdictions, based on the inherent risks
of spray drift; and the spray flight data showing that Colombia’s spray pilots
1289
See e.g. Diplomatic Note DM/AL No. 25009, sent from the Ministry of Foreign Affairs of
Colombia to the Ministry of Foreign Affairs of Ecuador, p. 3 (14 July 2001) (“Its half-life in soil
is between 1 and 4 weeks at the most, and in tropical soils … such as ours … less than one week,
and then it is biodegradedŽ.). EM, Vol. II, Annex 42. See also Weller Report, p. 3. ER, Vol. II,
Annex 3; Charles A. Menzie, Ph.D. & Pieter N. Booth, M.S., Response to: “Critique of
Evaluation of Chemicals Used in Colombia’s Aerial Spraying Program, and Hazards Presented
to People, Plants, Animals and the Environment in Ecuador,” As Presented in: Counter-
Memorial of the Republic of Colombia, Appendix, pp. 25-26 (Jan. 2011) (hereinafter “Menzie &
Booth ReportŽ). ER, Vol. II, Annex 6.
475regularly violated … on tens of thousands of spray flights along or near the border
with Ecuador … the operational requirements (in terms of flight speed, altitude,
droplet size, application rate and time of day and weather conditions) that
Colombia itself deemed necessary to control spray drift.
7.19 The evidence is cumulative, in the sense contemplated by the Court when
it indicated, in its first case, that there exists “a series of facts linked together [that
1290
lead] logically to a single conclusionŽ . The symptoms observed are identical
throughout all affected areas, they correspond to those endured also on the
Colombian side of the border, they follow closely upon the precise locations of
spraying events, and they are thus clearly not the consequence of some oil
refining activity in one part of the country, or of the alleged “naturalŽ
deterioration of soil fertility as a result of so-called “slash and burnŽ
1291
agriculture . Independently of the question of the quantum of damages and
Ecuador’s capacity to substantiate all material damages invoked, which are
questions dealt with in Chapter 8, Ecuador must underscore that the persistent
repetition of spraying activities along the border notwithstanding clear objections
by Ecuador amounts to a serious breach of international human rights law
committed by Colombia, and the evidence in the record is sufficient to lead the
Court to declare such a finding.
129Corfu Channel (United Kingdom v. Albania), Judgment, Merits, I.C.J. Reports 1949, p. 18.
129CCM, Chap. 7, para. 7.161.
476 B. A ERIAL SPRAYING V IOLATES THE F UNDAMENTAL HUMAN RIGHTS OF
A FFECTED INDIVIDUALS
7.20 Colombia’s view of human rights law is peculiarly and unsustainably
narrow. It treats the right to life as if it were confined solely to cases of arbitrary
killing. This is not correct. The Inter-American Court of Human Rights has
unambiguously defined the right to life as a right that “includes not only the right
of every human being not to be deprived of his life arbitrarily, but also the right
that he will not be prevented from having access to conditions that guarantee a
1292
decent existenceŽ . It is the Inter-American Court of Human Rights, and not
1293
Ecuador’s willingness to “recycleŽ its arguments , that recognizes a connection
between a number of distinct human rights, such as the right to health and to food
and access to clean water with the right to a decent existence29. Ecuador has
demonstrated in its Memorial 129, and now even more clearly in Chapter 3, the
uniform pattern of harm caused by Colombia. As explained below, the States
Parties to the American Convention on Human Rights undertake to respect the
rights recognized herein, including the right to life, and thus including also the
right to a decent existence. For many people in border areas the cumulative effect
129Villagran Morales et al. v. Guatemala, Judgment, IACHR, Series C No. 77, para. 144 (19
Nov. 1999). See also Case of the Indigenous Community Yakye Axa v. Paraguay, Judgment,
IACHR, Series C No. 125, para. 161 (17 June 2005) (“the right to life . . . includes not only the
right of every human being not to be arbitrarily deprived of his life, but also the right that
conditions that impede or obstruct access to a decent existence should not be generatedŽ.);
UNHRC, General Comment 6 on Article 6, U.N. Doc. HRI/GEN/1/Rev.1, pp. 6-7 (29 July 1994).
1293
CCM, Chap. 9, paras. 9.63, 9.79.
1294
EM, Chap. 9, paras. 9.43-9.48.
129EM, Chap. 6.
477of persistent spraying of toxic herbicides has removed elements that provide basic
support for the enjoyment of a decent existence.
7.21 With regard to the right to a healthy environment, Colombia fails to see,
in contrast to its domestic legal system, that the right to a healthy environment is
closely related to the enjoyment of other fundamental rights. Colombia dismisses
the material contribution to the Inter-American system of the Yakye Axa Case,
which establishes the connection between the right to a healthy environment and
the right to a decent life, on the basis of its supposed non-extra-territorial
application. Colombia here confuses the normative content of the right with the
related State obligations to respect and ensure protection of those rights.
Colombia also falls into superfluous technicalities by arguing that the right
recognized in Article 24 of the Banjul Charter is not related to the right to a
healthy environment, simply because it is formulated as a right to a “satisfactory
environmentŽ 129. This is semantics that focuses on form not substance: the
decision of the African Commission on Human and Peoples Rights in the
Ogoniland Case shows that the objectives pursued by both formulations are the
same 1297.
129CCM, Chap. 9, para. 9.117.
1297EM, Chap. 9, paras. 9.80-9.82. See also Kaniye Ebeku, “The Right to a Satisfactory
Environment and the African CommissionŽ, 3 Afr. Hum. Rts. L. J. 149, 163 (2003); Justice C.
Nwobike, “The African Commission on Human and Peoples’ Rights and the Demystification of
Second and Third Generation Rights under the African CharterŽ, 1 Afr. J. Legal Stud. 129, 139
4787.22 With regard to the right to private life, Colombia decides to follow the
wording used by another regional human rights body. Referring to the López
Ostra Case in the European system, Colombia considers that, in order to affect
the right to private life, pollution needs to be “severeŽ and that in the present case
the pollution is minimal and hence cannot interfere with home or private life. But
the severity of pollution can only be measured by its actual or likely effects. In
the present case, the impact on the private and family life of those affected would
appear to be more harmful and significant over the long term than in López Ostra,
where the pollution caused foul odours rather than ill health or loss of crops and
1298
the basic means of subsistence . Colombia’s characterisation of the present
situation as “de minimis pollutionŽ grossly misrepresents the reality, which is far
(2005); Dinah Shelton, “Decision Regarding Communication 155/96 (Social and Economic
Rights Action Centre/Centre for Economic and Social Rights v. Nigeria),Ž 96 Am. J. Int’l. L. 937
(2002).
1298López Ostra v. Spain, Judgment, ECHR, Series A no. 303-C, paras. 49-51 (9 Dec. 1994):
“49. On the basis of medical reports and expert opinions produced by the Government or the
applicant (see paragraphs 18-19 above), the Commission noted, inter alia, that hydrogen sulphide
emissions from the plant exceeded the permitted limit and could endanger the health of those living
nearby and that there could be a causal link between those emissions and the applicant’s daughter's
ailment’s.
50. In the Court's opinion, these findings merely confirm the first expert report submitted to
the Audiencia Territorial on 19 January 1989 by the regional Environment and Nature Agency in
connection with Mrs. López Ostra’s application for protection of fundamental rights. Crown Counsel
supported this application both at first instance and on appeal (see paragraphs 11 and 13 above). The
Audiencia Territorial itself accepted that, without constituting a grave health risk, the nuisances in
issue impaired the quality of life of those living in the plant’s vicinity, but it held that this impairment
was not serious enough to infringe the fundamental rights recognised in the Constitution (see
paragraph 11 above).
51. Naturally, severe environmental pollution may affect individuals’ well-being and
prevent them from enjoying their homes in such a way as to affect their private and family life
adversely, without, however, seriously endangering their healthŽ.
479 1299
more serious for those affected than acknowledged by the Counter-Memorial .
If a lower level of harmful pollution qualifies as a violation of the right to private
life in López Ostra, then it follows that there is a comparable, if not more
egregious, violation of the right to private life on the present facts.
C. T ARGETING
7.23 It is not necessary for Ecuador to show that Colombia’s aerial spraying
explicitly “targetedŽ anyone in Ecuador 130. The human rights case law
concerned with pollution impacts on health and private life normally involves
1301
unintended and incidental consequences, and these are no less a violation . All
these cases have common features. First, there is some kind of nuisance … a
chemical plant, smelter, tannery, mine or waste disposal site, for example.
Second, there is a failure to take adequate preventive measures to control these
known sources of risk to life, health, private life or property. In none of the cases
are the victims “targetedŽ in any sense.
1299
CCM, Chap. 9, para. 9.142.
1300CCM, Chap. 9, paras. 9.13-9.14.
1301Maya Indigenous Communities of the Toledo District v. Belize, Judgment, Inter-Am. C.H.R.,
Report No. 40/04, Case 12.053, paras. 147-154 (12 Oct. 2004); Ilmari Lansman et al. v. Finland,
ICCPR Comm. No. 511/1992, para. 9.4 (1996); López Ostra v. Spain, Judgment, ECHR, Series A
no. 303-C, para. 58 (9 Dec. 1994); Guerra and Others v. Italy, Judgment, ECHR, Reports of
Judgments and Decisions 1998 I, para. 60 (19 Feb. 1998); Fadeyeva v. Russia [2005] ECHR 376,
para. 134; Öneryildiz v. Turkey [2004] ECHR 657; Case of Taskin and Others v. Turkey, Judgment,
ECHR (10 Nov. 2004), paras. 113-119; Tatar v Romania, no. 67021/01 ECHR (2009), para 88;
Budayeva, al. v Russia, Judgment, no. 15339/02, ECHR (2008).
4807.24 It is clear that Colombia has failed to act diligently in controlling and
monitoring the spraying operation, and it has been reckless in failing to take the
necessary measures to prevent transboundary effects. As detailed in Chapters 2,
4, and in the previous Chapter, inappropriate chemicals were used, aircraft ill-
suited for aerial spraying were employed, and the Environmental Management
Plan (“EMPŽ), which sets out the operating requirements for the spray planes,
was lax and inadequately enforced. These requirements were intended to protect
human health, food and water supplies, and the environment from harmful effects
of a highly toxic spray, including harmful effects in Ecuador. These are the kinds
of harmful impacts that the deposit of toxic herbicides will inevitably and
predictably cause if spraying is carried out carelessly or without regard for the
consequences, as has been the case. Moreover, whatever may be the position
within Colombia, it is also clear that Colombia took no steps to notify or warn
either the people residing in Ecuador’s border regions, or the Ecuadorian
authorities, in advance of the spraying. Those likely to be affected were thus
given no warning and no chance to take precautions to protect themselves from
the spray.
7.25 All of these failures by Colombia … failing to exercise diligent control
over the operations of the spray planes, failing to warn those likely to be harmed,
failing to enforce the conditions applicable to aerial spraying … are typical of the
cases on the harmful effects of pollution under all of the relevant human rights
481treaties, including the American Convention on Human Rights and the ICCPR, to
which both Ecuador and Colombia are parties. Ecuador has set out its case on all
of these issues in the Memorial and there is no need to repeat it here 1302. To
reiterate: none of the cases requires that those affected must have been “targetedŽ.
7.26 Nor can Colombia justify or excuse the spraying that has actually taken
place. The human rights case law demonstrates that States must balance the
interests of the community in eradicating drug plantations against the harm to
individual human rights 130. By failing to comply with or enforce its own EMP,
Colombia has carried out a spraying operation that disregards the harmful impact
on humans. The Inter-American Commission on Human Rights’ decision in Maya
Indigenous Community of the Toledo District v. Belize is instructive on this point.
The Commission found that:
“the State failed to put into place adequate safeguards and
mechanisms, to supervise, monitor and ensure that it had sufficient
staff to oversee that the execution of the logging concessions
would not cause1304ther environmental damage to Maya lands and
communitiesŽ .
In this respect, Colombia’s spraying programme has failed the most elementary
test of compatibility with or respect for fundamental human rights for exactly the
130EM, Chapter 9.
1303
Maya Indigenous Communities of the Toledo District v. Belize, Judgment, Inter-Am. C.H.R.,
Report No. 40/04, Case 12.053, para. 150 (12 Oct. 2004).
1304
Ibid., para. 147.
482same reasons that it fails the test of due diligence in the prevention of
transboundary harm.
Section III. The “Territoriality” of Human Rights Obligations
7.27 One of the central arguments in its Counter-Memorial is that Colombia is
under no international law obligation to respect and protect the human rights of
local populations and indigenous peoples living outside its territory or not falling
1305
under its jurisdiction . To justify this claim, Colombia relies heavily on the
Judgment of the European Court of Human Rights in Bankoviü v. Belgium 1306.
This argument is surprising since Bankoviü dealt with a situation in which the
alleged harm felt as a result of the violation occurred on the territory of a State
that was not a party to the European Convention on Human Rights. In the present
case, on the contrary, Colombia and Ecuador are both parties to the American
Convention, the ICCPR and ILO Convention No. 169. As a consequence, any
comparison with the situation of Yugoslavia at the time of the Bankoviü case is
misplaced. Moreover, as this Court observed in the provisional measures phase
of Georgia v. Russia, the question whether a human rights treaty applies extra-
territorially is essentially a question of interpretation, to be determined according
130CCM, Chap. 9, para. 9.15 et seq.
130CCM, Chap. 9, paras. 9.25-9.31.
483to the ordinary meaning and in light of the object and purpose of each treaty
1307
(Vienna Convention on the Law of Treaties, Article 31) .
7.28 As further elaborated below, Ecuador’s views on the territoriality of
human rights obligations in the context of two States within the Inter-American
juridical system and legal space are as follows:
▯ First, Colombia and Ecuador are part of a common legal space at the
regional level (un espace juridique commun) relating to the protection
of human rights and indigenous peoples.
▯ Second, from this it follows that Colombia and Ecuador are subject to
an international public order of human rights which does not allow for
loopholes in the effective protection of human rights for any part of
their populations.
▯ Third, Colombia is under an obligation not to frustrate the protection
of human rights in the regions adjacent to its borders.
130Case Concerning Application of the International Convention on the Elimination of All Forms
of Racial Discrimination (Georgia v. Russian Federation), Provisional Measures Order, I.C.J.
Reports 2008, p. 386, para. 109 (“Whereas the Court observes that there is no restriction of a
general nature in CERD relating to its territorial application; whereas it further notes that, in
particular, neither Article 2 nor Article 5 of CERD, alleged violations of which are invoked by
Georgia, contain a specific territorial limitation ; and whereas the Court consequently finds that
these provisions of CERD generally appear to apply, like other provisions of instruments of that
nature, to the actions of a State party when it acts beyond its territoryŽ.).
484In this context, Colombia’s position fails to appreciate the approach adopted in
the context of the ACHR, and other applicable treaties. Equally, as already
explained in the previous Chapter, one may also conclude that Colombia’s actions
to combat illicit drug crops without having regard for the human rights of
neighbouring populations amount to an abuse of rights under the 1988 Narcotics
1308
Convention and a violation of Articles 2 and 14(2) thereof.
A. THE SO -CALLED “T ERRITORIALITY Ž OFH UMAN R IGHTS OBLIGATIONS
7.29 To support its argument that it has no obligations under international law
to respect the human rights of neighbouring populations, Colombia resorts to a
narrow and literal interpretation of the jurisdictional clauses contained in three
principal human rights conventions. With respect to the American Convention on
Human Rights, Colombia interprets Article 1 as limiting Colombia’s obligations
to its own territory or to individuals under its jurisdiction, i.e., under its effective
control. Colombia draws these conclusions from its interpretation of the
Bankoviü decision of the European Court of Human Rights and from the
1309
Advisory Opinion of this Court in the Wall Case .
7.30 With respect to the case law of the European Court of Human Rights,
Colombia relies on the Bankoviü Case, in which the Court refused to exercise its
130See, e.g., supra Chap. 6, paras. 6.10-6.28, 6.76-6.84.
130CCM, Chap. 9, paras. 9.17-9.20, 9.25-9.32.
485jurisdiction because the contested acts … aerial bombing by NATO forces … could
not be considered as falling under the jurisdictional requirements contained in
Article 1 of the European Convention on Human Rights. Colombia asserts that
“the argument that State action could ‘affect’ human rights extraterritorially was
1310
explicitly rejected by the European Court of Human RightsŽ . That is not what
the Court in fact decided, having regard to its considerations about the European
legal space and the fact that Yugoslavia was not a part of it 1311. Moreover,
Colombia fails to realize that the Bankoviü Case turns on the particular facts, not
least that the military attacks were performed outside the regional scope of the
European Convention, as the Federal Republic of Yugoslavia was not a party to
the European Convention when the attacks occurred 1312. By contrast, Ecuador
and Colombia are parties to the American Convention and ILO Convention No.
169.
7.31 Moreover, the wording of the American Convention on Human Rights is
inconsistent with Colombia’s narrowly construed approach to territorial
interpretation. Article 1(1) of the American Convention provides that:
“The States Parties to this convention undertake to respect the
rights and freedoms recognized herein and to ensure to all persons
131CCM, Chap. 9, para. 9.26.
1311
Bankoviü et al. v. Belgium et al., Decision on Admissibility, Eur.Ct.H.R. Application No.
52207/99, para. 80 (2001).
1312
See ibid., para. 42.
486 subject to their jurisdiction the free and full exercise of those rights
and freedomsƒŽ.
The obligation to “respectŽ is not synonymous with the obligation to “ensureŽ.
Even if indigenous peoples within Ecuador are not subject to the jurisdiction of
Colombia, Colombia nevertheless has an obligation to respect their rights … an
obligation identical to the requirements of Article 14(2) of the 1988 Narcotics
1313
Convention … and Ecuador is entitled to invoke such respect. Despite the
difference in wording of the ICCPR, the same argument was accepted by the UN
Human Rights Committee. The Committee held in 1981 that “it would be
unconscionable to so interpret the responsibility under Article 2 of the Covenant
as to permit a state party to perpetrate violations of the Covenant on the territory
1314
of another state, which violations it could not perpetrate on its own territoryŽ .
That is precisely Ecuador’s argument.
7.32 The circumstances surrounding the present case thus fall within the terms
of Article 1 of the American Convention and Article 2(1) of the ICCPR, as well
as ILO Convention No. 169. To exclude from the protection offered by these
instruments acts committed in one Contracting State but which produce effects in
the territory of another Contracting State would also run counter to the object and
purpose of these instruments. In the Preamble of the American Convention, the
States Parties recognize “that the essential rights of man are not derived from
131See supra Chap. 6, paras. 6.10-6.28.
131Delta Saldias de Lopez v. Uruguay, ICCPR Comm. No. 52/1979, para. 12.3 (29 July 1981).
487one’s being a national of a certain state, but are based upon attributes of the
human personality, and that they therefore justify international protection in the
form of a convention reinforcing or complementing the protection provided by
1315
the domestic law of the American statesŽ . These States also expressed their
conviction that “the ideal of free men enjoying freedom from fear and want can
be achieved only if conditions are created whereby everyone may enjoy his
economic, social, and cultural rights, as well as his civil and political rightsŽ 1316.
Ecuador submits that “[e]xtraterritorial conduct of the States Parties contrary to
the observance of human rights enshrined in the Convention is hardly compatible
1317
with such reaffirmationŽ .
7.33 Colombia states that, in the interests of economy, it will not repeat, for
each of the human rights treaties individually discussed, its point that “various
human rights instruments, and in particular those which limit their scope to
persons ‘within’ or ‘subject’ to the jurisdiction of the Respondent State, do not
apply to the case of alleged injury caused incidentally by spray drift from lawful
activities of a State on its own territoryŽ 131. Nevertheless, it proceeds to do so.
1315American Convention on Human Rights, O.A.S. Treaty Series No. 36, 1144 UNTS 123,
preamble (18 July 1978).
1316Ibid. (emphasis added).
1317See L. Loucaides, “Determining the Extra-territorial Effect of the European Convention:
Facts, Jurisprudence and the Bankoviü CaseŽ in The European Convention on Human Rights:
Collected Essays, p. 77 (Leiden: Martinus Nijhoff Publishers, 2007).
1318CCM, Chap. 9, para. 9.49.
488However, when contesting the violation of each individual right claimed by
Ecuador, Colombia fails to demonstrate that such instruments limit their scope to
persons “withinŽ or “subjectŽ to the jurisdiction, or to otherwise prove its
point 131.
7.34 Although Colombia tries at length to argue that human rights obligations
do not apply to transboundary claims, this exercise is not carried out for each of
the nine rights on which Colombia purports to respond. It engages in rebuttal for
only five out of nine rights invoked by Ecuador. For the right to humane
treatment and the right to private life, Colombia is ready to assume that those
rights could be applicable to Colombia’s conduct 132; with regard to the right to
property and the right to information, Colombia makes no reference whatsoever
1321
to the issue of extra-territoriality .
7.35 Moreover, with regard to the rights to life, health, food, water and healthy
environment, for which it argues their non-extraterritorial application, Colombia’s
arguments are unpersuasive. Colombia states that most of these rights are
“essentially territorialŽ 132. This can only mean that at least some element of
these rights can be extra-territorial. To support its argument, Colombia cites
1319CCM, Chap. 9, paras. 9.49-9.50.
1320CCM, Chap. 9, paras. 9.130, 9.140.
1321
CCM, Chap. 9, paras. 9.119-9.125, 9.144-9.152.
1322
CCM, Chap. 9, paras. 9.64, 9.72, 9.85, 9.114.
489selectively from human rights treaties or from General Comments of the
1323
Committee on Economic, Social and Cultural Rights . It reproduces text which
1324
refers to the territorial State’s obligation to ensure those rightsowever,
Colombia generally fails to cite the relevant passages of these General
Comments 1325or articles of human right treaties, which formulate the obligation
of States to respect human rights or which declare the existence of the right under
international law. The existence of an obligation to ensure those rights by the
territorial State does not contradict the existence of an obligation to respect
human rights extra-territorially.
B. THE CONCEPTS OF A COMMON LEGAL SPACE AND AN INTERNATIONAL
P UBLIC ORDER OF H UMAN R IGHTS AT THER EGIONAL LEVEL
7.36 The preceding conclusions are reinforced by the fact that both Ecuador
and Colombia are Parties to the American Convention on Human Rights and
1323
CCM, Chap. 9, paras. 9.80, 9.87-9.89, 9.91. With regard to the right to food, Colombia cites
an excerpt of General Comment No. 12, which actually falls under the heading “implementation
at the national levelŽ. General Comment No. 12 has also other sections, such as those referring to
“normative contentŽ or to “obligations and violationsŽ, in addition to a section on “international
obligationsŽ, which refers expressly to the need to “respect the enjoyment of the right to food in
other countriesŽSee U.N. Committee on Economic, Social and Cultural Rights, General
Comment No. 12: The Right to Adequate Food (Article 11), U.N. Doc E/C.12/1999/5, para. 36 (12
May 1999) (“States parties should take steps to respect the enjoyment of the right to food in other
countriesƒŽ).
1324
CCM, Chap. 9, paras. 9.64, 9.72-9.73, 9.85-9.90.
1325
In addition to the example mentioned above on the General Comment on the Right to Food,
see also U.N. Committee on Economic, Social and Cultural Rights, General Comment 14: The
Right to the Highest Attainable Standard of Health, U.N. Doc. E/C.12/2000/4, para. 33 (11 Aug.
2000) (“The obligation to respect requires States to refrain from interfering directly or indirectly
with the enjoyment of the right to healthŽ.) (Emphasis in original).
490other relevant or applicable instruments. The object and purpose of the American
Convention is to reinforce the protection of human rights within the Americas.
As mentioned in the Memorial and clearly supported by scholars who have
1326
analysed the Inter-American human rights legal system , the American
Convention has instituted a common legal space (un espace juridique commun)
which does not allow for a vacuum in the protection of human rights within this
1327
geographical area . In that sense, the American Convention established an
international public order of human rights which prescribes that the human rights
of individuals within this common legal space must be respected. The concept of
common legal space was first developed in the case law of the European Court of
Human Rights, then adopted by the Inter-American Court of Human Rights. In
this respect, the Inter-American Court of Human Rights has adopted an extensive
interpretation of the obligations contained in the Inter-American Declaration of
Human Rights and the American Convention on Human Rights, and has resorted
frequently to direct references from the case law of the European Court of Human
Rights. As stated in Ecuador’s Memorial, both judicial institutions are
1326
EM, Chap. 9, para. 9.9.
1327Ibid. (quoting American Convention on Human Rights, Art. 1, O.A.S. Treaty Series No. 36,
1144 UNTS 123 (18 July 1978): “The States Parties to this Convention undertake to respect the
rights and freedoms recognized herein and to ensure to all persons subject to their jurisdiction the
free and full exercise of those rights and freedoms, without any discrimination. . . Ž).
491increasingly influenced by each other’s jurisprudence and a process of cross-
1328
fertilization characterizes the evolution of their case law .
1. The Inter-American Declaration of Human Rights
7.37 The extra-territorial application of the Inter-American Declaration of
1329
Human Rights was stressed in Alejandre and Others v. Cuba . The Cuban Air
Force was alleged to have shot down two civil aircraft in international air space
outside the territorial jurisdiction of Cuba. A claim was filed before the Inter-
American Commission on Human Rights by relatives of the victims on the basis
of Articles 1 (right to life) and 18 (right to a fair trial) of the Inter-American
Declaration of Rights and Duties of Man. The Commission said that it was
competent ratione materiae and ratione personae to hear the case. As to its
ratione loci jurisdiction, the Commission stated:
“The essential rights of the individual are proclaimed in the
Americas on the basis of equality and non discrimination, ‘without
distinction as to race, nationality, creed, or sex.’ Because
individual rights are inherent to the human being, all the American
states are obligated to respect the protected rights of any person
subject to their jurisdiction. Although this usually refers to
persons who are within the territory of a state, in certain instances
it can refer to extraterritorial actions, when the person is present in
the territory of a state but subject to the control of another state,
generally through the actions of that state's agents abroad. In
principle, the investigation refers not to the nationality of the
132EM, Chap. 9, para. 9.40.
132Armando Alejandre, Jr., Carlos Costa, Mario De La Pena, and Pablo Morales (Cuba), Inter-
Am. C.H.R., Report No. 86/99, Case 11.589, para. 23 (29 Sept. 1999).
492 alleged victim or his presence in a particular geographic area, but
to whether, in those specific circumstances, the state observed the
rights of a person subject to its authority and controlŽ 1330.
7.38 Likewise, in Saldaño v. Argentina, contrary to what was alleged by
1331
Colombia in its Counter-Memorial , the Commission declared:
“The Commission does not believe, however, that the term
‘jurisdiction’ in the sense of Article 1(1) is limited to or merely
coextensive with national territory. Rather, the Commission is of
the view that a state party to the American Convention may be
responsible under certain circumstances for the acts and omissions
of its agents which produce effects or are undertaken outside that
state’s own territory. This position finds support in the decisions
of European Court and Commission of Human Rights which have
interpreted the scope and meaning of Article 1 of the European
Convention for the Protection of Human Rights and Fundamental
Duties (European Convention). Article 1 of that instrument, on
which Article 1(1) of the American Convention was largely
patterned, stipulates that the high contracting parties ‘shall secure
to everyone within their jurisdiction the rights and freedoms
defined in Section 1 of this Convention’.Ž 133.
7.39 In the Saldaño case, the important point of direct relevance for Ecuador’s
argument is that the American Declaration was found to be applicable to “the acts
and omissions of its agents which produce effects or are undertaken outside that
state’s own territoryŽ 133. That is precisely the situation in the present
1330Ibid, para. 23 (emphasis added).
1331CCM, Chap. 9, paras. 9.40-9.43.
1332
Saldaño v. Argentina, Judgment, Inter-Am.C.H.R, Report No. 38/99, para. 17 (11 Mar. 1999)
(emphasis added).
1333
Ibid. (emphasis added).
493proceedings … the acts or omissions of agents of Colombia have produced harmful
effects outside Colombian territory. Consequently, the Declaration is implicated.
7.40 It should be further noted that in Coard v. United States, the Inter-
American Commission stated:
“While the extraterritorial application of the American
Declaration has not been placed at issue by the parties, the
Commission finds it pertinent to note that, under certain
circumstances, the exercise of its jurisdiction over acts with an
extraterritorial locus will not only be consistent with but required
by the norms which pertain. The fundamental rights of the
individual are proclaimed in the Americas on the basis of the
principles of equality and non-discrimination … ‘without
distinction as to race, nationality, creed or sex.’ Given that
individual rights inhere simply by virtue of a person's humanity,
each American State is obliged to uphold the protected rights of
any person subject to its jurisdiction. While this most commonly
refers to persons within a state’s territory, it may, under given
circumstances, refer to conduct with an extraterritorial locus
where the person concerned is present in the territory of one state,
but subject to the control of another state … usually through the
acts of the latter’s agents abroad. In principle, the inquiry turns
not on the presumed victim’s nationality or presence within a
particular geographic area, but on whether, under the specific
circumstances, the State observed the rights of a person subject to
its authority and controlŽ334.
7.41 If shooting down an aircraft in international airspace constitutes the
exercise of authority or control over the victims, then a fortiori spraying toxic
herbicides that foreseeably drift over the territory of another State will likewise
1334
Coard v. United States, Judgment, Inter-Am.C.H.R., Report No. 109/99, Case 10.951, para.
37 (29 Sept. 1999) (emphasis added).
494amount to the exercise of authority or control over the victims for the purposes of
applying the American Declaration extraterritorially.
2. Case Law of the Inter-American Court of Human Rights
7.42 What is true for the Inter-American Declaration is equally true for the
American Convention on Human Rights. The concept of a common legal space
prevailing among the States Parties to the Convention can be observed in the
Preamble to the American Convention, where it is stated that the signatory States
affirm “their intention to consolidate in this hemisphere, within the framework of
democratic institutions, a system of personal liberty and social justice based on
respect for the essential rights of manŽ 1335. In the Preamble, the signatory States
also recognize that:
“the essential rights of man are not derived from one’s being a
national of a certain state, but are based upon attributes of the
human personality, and that they therefore justify international
protection in the form of a convention reinforcing or
complementing the protection provided by the domestic law of the
1336
American statesŽ .
7.43 The Inter-American Court of Human Rights has retained a flexible vision
of its ratione loci jurisdiction embracing the entire space covered by all the States
133American Convention on Human Rights, preamble, O.A.S. Treaty Series No. 36, 1144 UNTS
123 (18 July 1978) (emphasis added).
133Ibid. (emphasis added).
495Parties to the American Convention on Human Rights. On that basis, the Court
affirms its jurisdiction over “the inter-American systemŽ as a whole.
7.44 In its second Advisory Opinion, The Effect of Reservations on the Entry
into Force of the American Convention on Human Rights, the Inter-American
Court of Human Rights spelled out the objective character of obligations arising
out of multilateral human rights treaties:
“The Court must emphasize, however, that modern human rights
treaties in general, and the American Convention in particular, are
not multilateral treaties of the traditional type concluded to
accomplish the reciprocal exchange of rights for the mutual benefit
of the contracting States. Their object and purpose is the
protection of the basic rights of individual human beings
irrespective of their nationality, both against the State of their
nationality and all other contracting States. In concluding these
human rights treaties, the States can be deemed to submit
themselves to a legal order within which they, for the common
good, assume various obligations, not in relation to other States,
but towards all individuals within their jurisdictionŽ1337.
7.45 More importantly, the Inter-American Court of Human Rights directly
referred to the notion of common public order developed by the European
Commission on Human Rights in Austria v. Italy, and endorsed it by affirming its
particular relevance with regard to the American Convention (“apply with even
1338
greater force to the American ConventionŽ) . Finally, the Inter-American
1337The Effect of Reservations on the Entry into Force of the American Convention on Human
Rights (Arts. 74 and 75), Advisory Opinion OC-2/82, IACHR, Series A No. 2, para. 29 (24 Sept.
1982) (emphasis added).
133Ibid. at paras. 29-31 (24 Sept. 1982).
496Court of Human Rights developed the concept of common legal space established
through the American Convention in its tenth Advisory Opinion, Interpretation of
the American Declaration of the Rights and Duties of Man:
“The evolution of the here relevant ‘inter-American law’ mirrors
on the regional level the developments in contemporary
international law and specially in human rights law, which
distinguished that law from classical international law to a
significant extent. That is the case, for example, with the duty to
respect certain essential human r1339s, which is today considered
to be an erga omnes obligationŽ .
7.46 Hence, the concepts of common legal space and public order have been
embraced by the Inter-American Court of Human Rights, which interprets its
mission so as to secure total protection of the human rights enshrined in the
American Convention throughout the entire region concerned, and to avoid a
vacuum where certain individuals cannot benefit from the system of protection.
3. The Legal Consequences Flowing from the Concepts of Common Legal Space
and Public Order of Human Rights
7.47 The situation where extra-territorial effects are produced within the
common legal space is totally different from the situation where the effects take
1340
place in a State which is not a member of that common legal space . To that
1339
Interpretation of the American Declaration of the Rights and Duties of Man within the
Framework of Article 64 of the American Convention on Human Rights, Advisory Opinion OC-
10/89, IACHR, Series A No. 10, para. 38 (14 July 1989).
1340Emmanuel Decaux distinguishes between: “la compétence extra-territoriale interne, celle
qu’un Etat exercerait à l’intérieur de l’espace commun des Parties, et la compétence extra-
497extent, the Bankoviü decision, relied on by Colombia, is simply not relevant to the
circumstances of the present case.
7.48 The American Convention on Human Rights contains two types of
obligations incumbent upon the Contracting States. On the one hand, the
Convention imposes positive obligations to secure the enjoyment of fundamental
human rights, and on the other, the Convention imposes negative obligations to
abstain from violating (or from failing to “respectŽ) the rights enshrined in the
Convention. Whereas Colombia seeks to exculpate itself by claiming that the
inhabitants in Ecuador’s border region are not specifically targeted by the
spraying of herbicide, Ecuador submits that Colombia is subject, at the very least,
to the negative obligation to abstain from infringing the human rights of the
individuals living in Ecuador. Colombia should have taken into account the
inherent risks of aerial spraying of toxic chemicals, and should have refrained
from such action along the border at least after the first official protests and
1341
requests for information expressed by Ecuador in July 2000 .
7.49 This position was also expressed by the UN Special Rapporteur on the
Right to Health, Mr. Paul Hunt, when he stated that “Colombia has a human
rights responsibility of international assistance and cooperation, including in
territoriale externe d’un Etat, en dehors de cet espace communŽ. E. Decaux, “Le territoire des
droits de l’hommeŽ, Liber Amicorum Marc-André Eissen, p. 69 (Bruxelles: Bruylant, 1995).
134See EM, Chap. 3, para. 3.6 et seq.
498health. Consequently, as a minimum, Colombia must not jeopardise the
enjoyment of the right to health in Ecuador. It must ‘do no harm’ to its
1342
neighbourŽ . Therefore, Colombia is wrong when it criticizes Ecuador’s
claims regarding violations of the American Convention, ICCPR, ICESCR, and
ILO Convention No. 169. Ecuador does not argue that Colombia has the
territorial obligation to fulfil the rights enshrined in these human rights
instruments, but Ecuador is of the firm view that, in a transboundary context,
Colombia has the negative obligation to refrain from frustrating the human rights
of populations living directly across the border in Ecuador.
7.50 As indicated in Chapter 6 of this Reply, these conclusions are also
consistent with the wording of Article 14 of the 1988 Narcotics Convention,
which provides that the “measures adopted shall respect fundamental human
1343
rightsŽ . Although Colombia is certainly entitled to combat drug cartels and to
eradicate illicit drug crops within its territory, such action can only be taken if it
respects the fundamental human rights, not only of its own population, but also of
those individuals affected by such measures in a transboundary context.
Therefore, Colombia should have refrained from aerial spraying in the border
region and should have adopted other measures indicated in Chapter 6 that would
134U.N. Press Release, “U.N. Special Rapporteur on the Right to the Highest Attainable Standard
of Health, Paul Hunt, Ends Visit to EcuadorŽ (18 May 2007) (emphasis added). EM, Vol. IV,
Annex 185.
134See supra Chap. 6, paras. 6.10-6.28
499have respected the environment and the human rights of the people living in
Ecuador.
7.51 Colombia’s interpretation of the 1988 Narcotics Convention constitutes an
abuse of right to the extent that Colombia believes that it can combat illicit drug
crops without having regard to the human rights of individuals living in the
1344
border regions across from its own territory . It would be wrong to presume
that the Contracting States to the 1988 Narcotics Convention intended to secure
respect for human rights only within the territory of the State adopting coercive
measures and not also in neighbouring States, if such measures have potentially
extra-territorial effects. This is all the more so where such effects are so serious
in their consequence.
7.52 Finally, it should be stressed that, by arguing that it has no obligations
under international law to respect the human rights of individuals living across
the border in Ecuador, Colombia renders the objective protection provided for in
the American Convention illusory. The Convention simply cannot be read as
permitting such violations to occur.
1344
The possibility of abuse of right, although not to be presumed, has been acknowledged by the
Court, and its predecessor, in various cases, including: Case of the Free Zones of Upper Savoy
and the District of Gex, Judgment, P.C.I.J. Series A/B N° 46, p. 167 (1932); Fisheries Case
(United Kingdom v. Norway), I.C.J. Reports 1951, p. 142.
500 Section IV. Breaches of the Rights of Indigenous Peoples
7.53 In its Memorial, Ecuador showed that the daily life of indigenous peoples
living on its side of the border had been particularly affected by Colombia’s aerial
spraying. Displacement of communities, interference with the use and enjoyment
of their property and traditional culture, and loss of access to and use of the
forests, including traditional medicinal plants, has all resulted from Colombia’s
1345
extensive spray programme . Ecuador claimed that Colombia’s actions in
these regards violated Article 27 of the ICCPR, Articles 4 to 7, 13 and 15 of ILO
Convention 169, and Article 21 of the American Convention 1346. In response,
Colombia makes the same arguments that it made in relation to human rights
violations. First, it denies that there is any evidence of harm resulting from aerial
1347
spraying activities taking place within Colombia . Second, it says that the
1348
aerial spraying is not targeted at indigenous peoples . Third, it says that
responsibility is territorial, and that Ecuador not Colombia is responsible in
international law for fulfilling obligations towards indigenous peoples in
Ecuador 1349. Finally, it asserts that Colombia has done nothing that could amount
to a denial of the rights of indigenous peoples 1350. Ecuador’s response to these
1345EM, Chap. 6.
1346EM, Chap. 9, paras. 9.18-9.36.
1347
CCM, Chap. 7, paras. 7.177-7.186.
1348
CCM, Chap. 9, para. 9.155.
1349
CCM, Chap. 9, paras. 9.159, 9.164-9.167, 9.169
1350CCM, Chap. 9, paras. 9.160, 9.165, 9.169.
501arguments is the same as its response to Colombia’s arguments on human rights.
However, before addressing these issues it is necessary to reiterate the special
status of indigenous peoples in international law and its relevance for the present
dispute.
A. THE SPECIAL STATUS OF INDIGENOUS P EOPLES UNDER INTERNATIONAL
LAW
7.54 The present dispute is of particular importance as it relates to violations of
the internationally protected rights of indigenous people, and not strictly
environmental harm and human rights violations in regard to the general
population. At least four different ethnic groups living along the Ecuadorian
border with Colombia have been seriously affected by the aerial sprayings: the
Awá, the Cofán, the Kichwa, and the Afro-Ecuadorian communities of
Esmeraldas. Colombia simply dismisses Ecuador’s claims based on indigenous
rights, believing that the subject merits only a short answer because of its
allegedly controversial nature. Far from being controversial, however, Colombia
has committed itself under international treaties and its own constitutional
provisions and enacted laws to respect the rights of all indigenous peoples. In
short, Colombia recognises through its own legal obligations that indigenous
peoples benefit from special protection under both international and national law:
both Ecuador and Colombia have committed themselves to promote the
502enjoyment of rights by indigenous communities, notably those living alongside
the common border.
7.55 The legal status of indigenous peoples has changed significantly over the
past four decades. The need to secure the living conditions of such human groups
has been widely acknowledged by the international community, through the
development of binding legal norms. This has been the case at the domestic level
of many States which afford special protection to indigenous groups … including
Colombia and Ecuador … and at the international level through the adoption of
international instruments recognizing the special rights of indigenous peoples. As
acknowledged by Colombia, both States involved in the present dispute are
parties to ILO Convention No. 169. This Convention provides for comprehensive
protection measures of all indigenous peoples … wherever they are located …
against interference by States. The Convention must be read in conjunction with
other fundamental human rights instruments, such as the Universal Declaration of
Human Rights, the ICESCR, the ICCPR, and other instruments on the prevention
of discrimination. The importance of the provisions of ILO Convention No. 169
was reaffirmed in 2007 through the adoption of the UN Declaration on the Rights
1351
of Indigenous Peoples .
135United Nations, General Assembly, Draft Resolution to the United Nations Declaration on the
Rights of Indigenous Peoples, U.N. Doc. A/61/L.67 (12 Sept. 2007).
5037.56 The ILO Convention provides for the protection of the fundamental
human rights of indigenous peoples and for the prohibition of discrimination.
Article 3(1) clearly indicates that “[i]ndigenous and tribal peoples shall enjoy the
full measure of human rights and fundamental freedoms without hindrance or
discriminationŽ. They benefit from the same level of international protection as
do all other human beings. In addition, however, the specific needs of indigenous
peoples have given rise to a recognition that special and additional measures of
international protection are needed. Thus, States are called upon to promote “the
full realisation of the social, economic and cultural rights of these peoples with
respect to their social and cultural identity, their customs and traditions and their
1352
institutionsŽ . Further, the Convention provides for special protection of
indigenous peoples’ property, cultures and environment 135, their social, cultural,
religious and spiritual values and practices 135, and the special importance of the
cultures and spiritual values of the peoples concerned in relationship with their
1355 1356
lands and territories and their natural resources .
7.57 Of utmost importance in this respect is the general obligation to consult in
good faith, through adequate procedures, and without discrimination, the
1352Article 2(2)(b).
1353
Article 4(1).
1354
Article 5(a).
1355Article 13.
1356Article 15.
504indigenous peoples who are likely to be affected by measures concerning their
immediate living environment. Article 6 of ILO Convention No. 169 is explicit
in stating that governments shall “consult the peoples concerned, through
appropriate procedures and in particular through their representative institutions,
whenever consideration is being given to legislative or administrative measures
which may affect them directlyŽ 135. This provision is not subject to any territorial
limitation. Equally, States adopting such measures are under the obligation to
“ensure that, whenever appropriate, studies are carried out, in co-operation with
the peoples concerned, to assess the social, spiritual, cultural and environmental
impact on them of planned development activitiesŽ 1358and such measures shall be
taken in co-operation with the peoples concerned in order to “protect and preserve
1359
the environment of the territories they inhabitŽ .
7.58 The fundamental importance of these obligations has been emphasized by
the Inter-American Court of Human Rights, in various cases cited in the
Memorial. Colombia has nothing to say about these cases. The Inter-American
Court has interpreted Article 21 of the American Convention on Human Rights
on the right to property in the light of the obligations contained in the ILO
1357Article 6(1) (emphasis added).
1358
Article 7(3).
1359
Article 7(4).
505Convention No. 169 1360. There is thus nothing controversial about the
international protection of indigenous peoples. In fact, it is widely recognized
that indigenous peoples live in very close inter-connectedness, indeed vital
1361
dependence, upon the ecosystem and the natural resources it contains .
Disruptions caused by aerial spraying have had especially strong impacts upon
the living conditions of affected communities, and therefore on the rights of
indigenous peoples.
7.59 The witness statements of affected indigenous communities corroborate
this. First, they focus more on the environmental impacts observed after the
spraying, whereas accounts of non-indigenous witnesses concentrate more on the
1362
harms caused to their crops and livestock . The detailed accounts of
1360Matter of Pueblo Indígena de Sarayaku Regarding Ecuador, Provisional Measures, IACHR,
Series E No. 21, para. 32(d) (17 June 2005); Case of the Mayagna (Sumo) Awas Tingni
Community v. Nicaragua, Judgment, IACHR, Series C No. 79, para. 83(d) (31 Aug. 2001); Case
of the Yakye Axa Indigenous Community v. Paraguay, Judgment, IACHR, Series C No. 125 paras.
127 and 130 (17 June 2005); Case of the Sawhoyamaxa Indigenous Community v. Paraguay,
Judgment, IACHR, Series C No. 146, paras. 117-119 (29 Mar. 2006); Case of the Saramaka
People v. Suriname, Preliminary Objections, Judgment, IACHR, Series C No. 172, paras. 92-93
(28 Nov. 2007).
1361
See e.g., Norman E. Whitten, Jr., Ph.D., Dr. William T. Vickers, Ph.D. & Michael Cepek,
Ph.D., Tropical Forest Cultural Ecology and Social Adaptation in the Ecuadorian Border Region
with Colombia (hereinafter “Whitten et al. ReportŽ) (Jan. 2011), pp. 3, 17, 20, 31, 49, 53. ER, Vol.
II, Annex 5.
1362Compare Witness 1 Declaration, op. cit. EM, Vol. IV, Annex 189; Witness 2 Declaration, op.
cit. EM, Vol. IV, Annex 190; Witness 4 Declaration, op. cit. EM, Vol. IV, Annex 192; Witness 5
Declaration, op. cit. EM, Vol. IV, Annex 193; Witness 7 Declaration, op. cit. EM, Vol. IV,
Annex 195; Witness 8 Declaration, op. cit. EM, Vol. IV, Annex 196; Witness 9 Declaration, op.
cit. EM, Vol. IV, Annex 197; Witness 13 Declaration, op. cit. EM, Vol. IV, Annex 201; Witness
17 Declaration, op. cit. EM, Vol. IV, Annex 203; Witness 18 Declaration, op. cit. EM, Vol. IV,
Annex 204; Witness 20 Declaration, op. cit. EM, Vol. IV, Annex 206; Witness 23 Declaration,
op. cit. EM, Vol. IV, Annex 209; Declaration of Witness 26, 17 Feb. 2009. EM, Vol. IV, Annex
210; Declaration of Witness 27, 17 Feb. 2009. EM, Vol. IV, Annex 211; Witness 30 Declaration,
506modifications in flora and fauna reflect the very close relationship of indigenous
1363
people with their land . Second, the Court will have noted that all witness
statements of indigenous peoples refer to the medicinal plants they have
traditionally relied on and how their use became ineffective, indeed dangerous,
1364
after the spray mixture had contaminated them . These two points naturally
lead to the third observation, namely that the statements of the indigenous people
show how severe the injuries they have had inflicted on them have been to their
1365
health .
op. cit. EM, Vol. IV, Annex 214; Witness 31 Declaration, op. cit. EM, Vol. IV, Annex 215;
Witness 32 Declaration, op. cit. EM, Vol. IV, Annex 216; Witness 33 Declaration, op. cit. EM,
Vol. IV, Annex 217; Witness 34 Declaration, op. cit. EM, Vol. IV, Annex 218; Declaration of
Witness 36, 19 Feb. 2009, op. cit. EM, Vol. IV, Annex 219; Witness 37 Declaration, op. cit. EM,
Vol. IV, Annex 220; Declaration of Witness 38, 19 Feb. 2009. EM, Vol. IV, Annex 221 with
Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222; Declaration of Witness 40 (hereinafter
“Witness 40 DeclarationŽ), 20 Feb. 2009, op. cit. EM, Vol. IV, Annex 223; Declaration of
Witness 41, 20 Feb. 2009 (hereinafter “Witness 41 DeclarationŽ). EM, Vol. IV, Annex 224
Declaration of Colombia Witness 2, 20 Feb. 2009 (hereinafter “Colombia Witness 2
DeclarationŽ), op. cit. EM, Vol. IV, Annex 226; Declaration of Colombia Witness 8, 4 Mar.
2009. EM, Vol. IV, Annex 231.
1363
See Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222; Witness 40 Declaration, op. cit.
EM, Vol. IV, Annex 223; Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224; Colombia
Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 226; Declaration of Colombia Witness 8, 4
Mar. 2009. EM, Vol. IV, Annex 231. See also Whitten et al. Report, p. 3. ER, Vol. II, Annex 5.
1364See, e.g., Witness 29 Declaration, op. cit. EM, Vol. IV, Annex 213; Witness 31 Declaration,
op. cit. EM, Vol. IV, Annex 215; Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224;
Colombia Witness 2 Declaration, op. cit. EM, Vol. IV, Annex 226; Declaration of Colombia
Witness 10, 5 Mar. 2009. EM, Vol. IV, Annex 233. See also Whitten et al. Report, pp. 22-24, 30,
34, 52 (describing the use of medicinal plants by communities in the border region, and their
particular vulnerability to contamination). ER, Vol. II, Annex 5.
1365See Witness 39 Declaration, op. cit. EM, Vol. IV, Annex 222; Witness 40 Declaration, op. cit.
EM, Vol. IV, Annex 223; Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224.
507 B. T HE E VIDENCE OF H ARM
7.60 Colombia approaches Ecuador’s arguments on violation of the rights of
indigenous peoples in the same way that it responds to the human rights
violations: it claims that evidence is supported by vague 1366and unscientific 1367
witness statements, and that indigenous peoples live in very poor and unhealthy
1368
conditions anyway . Chapter 3 of this Reply addresses each of these
arguments. The testimonies of indigenous witnesses are specific enough to match
up with, and be corroborated by, the data obtained from the U.S. Department of
State regarding the dates and locations of Colombia’s spray flights 1369. The
indigenous testimonies are also corroborated by independent reports, based where
appropriate on scientific data137. The significance of the fragile living conditions
of local indigenous communities impacted by Colombia’s aerial spraying
operations is also discussed in Chapter 3, and in the expert report of Dr. Whitten
et al. These anthropology experts, who are intimately familiar with the
communities of Ecuador’s border region, explain that:
“[O]ne unifying characteristic of the border communities is their
heavy dependence upon the natural environment for food, shelter,
medicine, spiritual practices and other critical aspects of their
livelihoods. . . the people of the border region are inextricably
136CCM, Chaps. 7 and 9, paras. 7.133, 7.134, 7.137, 7.177-7.180, 9.124.
136CCM, Chap 7, paras. 7.181-7.182.
1368
CCM, Chap. 7, paras. 7.183-7.185.
1369
See supra Chap. 2, paras. 2.163, 2.167, 2.169, 2.175, 2.178, 2.180-2.182.
137See supra Chap. 3, Sections I(B)(1) and I(D)(1).
508 linked to their natural environment and are thus extremely
vulnerable to environmental perturbationsŽ 1371.
One particular source of vulnerability is the dependence of many indigenous
communities on “swidden agricultureŽ or “shifting cultivationŽ systems, which
rely upon a “delicate balance between the natural forest ecosystem and crops
grown for human consumptionŽ 1372. Of course, when these systems are
devastated by an herbicide that kills all plants, the people that rely on them for
1373
daily sustenance suffer terribly .
7.61 Colombia claims that Ecuador “does not explainŽ how aerial spraying in
Colombia could threaten the lifestyle of indigenous peoples in Ecuador 1374. This
is surprising considering that Ecuador devotes in Chapter 6 of its Memorial an
entire section to harm to indigenous communities, which is complemented with
another section in Chapter 9 on violation of the rights of indigenous peoples. In
the Memorial, Ecuador provides detailed accounts from independent experts and
witness statements as to the manner in which aerial spraying has damaged or
destroyed the basis for peoples’ livelihoods in several indigenous communities,
killing crops and domestic animals, polluting water, causing health problems and
1371Whitten et al. Report, op. cit., p. 3. ER, Vol. II, Annex 5.
1372
Ibid., op. cit., p. 3; see also ibid., pp. 9-11, 21, 28, 36-36, 47.
1373
Ibid., op. cit., pp. 9-11, 21, 28, 36-36, 47.
1374CCM, Chap. 9, para. 9.154.
509psychological distress, and destroying medicinal plants and plants used for
1375
traditional rituals .
7.62 As explained in the Memorial, the special connection of indigenous
peoples with their environment is recognized by international treaties such as ILO
Convention No. 169, and by decisions of the UN Human Rights Committee and
most particularly of the Inter-American Court of Human Rights 1376. The seminal
decision in the Awas Tingni Case, the text of which was referred to in the
1377
Memorial , merits restatement, since Colombia has not yet addressed it
properly:
“Indigenous groups, by the fact of their very existence, have the
right to live freely in their own territory; the close ties of the
indigenous people with the land must be recognized and
understood as the fundamental basis of their cultures, their
spiritual life, their integrity, and their economic survival. For
indigenous communities, relations to the land are not merely a
matter of possession and production but a material and spiritual
element which they must fully enjoy, even to preserve their
1378
cultural legacy and transmit it to future generationsŽ .
1375
EM, Chap. 6, paras. 6.106-6.130.
1376
EM, Chap. 9, paras. 9.13-9.38.
137EM, Chap. 9, para. 9.31.
1378Case of the Mayagna (Sumo) Awas Tingni Community v. Nicaragua, Judgment, IACHR,
Series C No. 79, para. 149 (31 Aug. 2001).
510The recent 2010 decision in the Case of the Indigenous Community Xákmok
1379
Kásek confirms this approach .
7.63 In responding to the showing of breach of treaty provisions protecting the
rights of indigenous peoples, Colombia simply ignores the evidence provided by
Ecuador. It limits itself to generalities and vague assertions.
7.64 With regard to Article 27 of the ICCPR, when addressing the right of
minorities to “enjoy their own cultureŽ, Colombia considers it “absurdŽ that aerial
spraying can be the decisive cause of indigenous peoples abandoning their
culture 138. On the contrary, however, the report by expert anthropologists
submitted with this Reply explains how many of the indigenous groups affected
by the aerial spraying programme have resided in the border region since pre-
colonial times. In particular, the Cofán are the “earliest recorded inhabitants of
the regionŽ and all of their “currently titled lands are in their ancestral
territoryŽ1381. The Awá have lived in north-western Ecuador since the time of the
Spanish conquest, when they moved to their current territory in the rugged and
biodiverse rainforest region located in Ecuador’s Esmeraldas and Carchi
1382
provinces . Moreover, the indigenous peoples of this region maintain a daily
137Case of the Indigenous Community Xákmok Kásek v. Paraguay, Judgment, IACHR, Series C
No. 125, paras. 173-178 (24 August 2010).
1380CCM, Chap. 9, para. 9.161.
1381Whitten et al. Report, op. cit., p. 18. ER, Vol. II, Annex 5.
1382
Ibid., op. cit., p. 46.
511connection to their traditional lands and environmental resources as a source of
food, shelter, medicine, and as a basis for spiritual practices 1383. As explained in
the Memorial, the UN Human Rights Committee has recognized that disrupting
the ties of indigenous peoples with their land and community can threaten their
1384
way of life and culture . As evidenced in statements of Awá, Cofán and
Kichwa witnesses, the serious adverse effects of the aerial sprayings on their
1385
traditional, and often remote, communities, has weakened their balance and
has led on occasions to their members abandoning their homes 1386.
7.65 On the violation of Article 21 of the American Convention on Human
Rights (the right to property), it is mystifying that Colombia has chosen to ignore
the significant body of jurisprudence developed by the Inter-American Court of
Human Rights on the matter. Colombia appears to believe that Article 21 refers
only to the need for indigenous peoples to have their lands recognized by the
1383Ibid., op. cit., pp. 3, 17-39, 45-49, 53.
1384
EM, Chap. 9, paras. 9.18-9.21.
1385
See EM, Chap. 6, paras. 6.114-6.115; Declaration of Maria Blanca Chancosa Sanchez, 14 Jan.
2009. EM, Vol. IV, Annex 187; Witness 26 Declaration, op. cit. EM, Vol. IV, Annex 210;
Witness 27 Declaration, op. cit. EM, Vol. IV, Annex 211; Witness 28 Declaration, op. cit. EM,
Vol. IV, Annex 212; Witness 29 Declaration, op. cit. EM, Vol. IV, Annex 213; Witness 31
Declaration, op. cit. EM, Vol. IV, Annex 215; Witness 40 Declaration, op. cit. EM, Vol. IV,
Annex 223; Witness 41 Declaration, op. cit. EM, Vol. IV, Annex 224.
1386Witness 11 Declaration, op. cit. EM, Vol. IV, Annex 199; Witness 26 Declaration, op. cit.
EM, Vol. IV, Annex 210; Witness 27 Declaration, op. cit. EM, Vol. IV, Annex 211; Witness 28
Declaration, op. cit. EM, Vol. IV, Annex 212; Witness 29 Declaration, op. cit. EM, Vol. IV,
Annex 213.
512 1387
State . Instead, as underscored by the Inter-American Court of Human Rights
in the Awas Tingni Case, the concept of property in indigenous communities has
acquired a specific meaning, centred on the group and its close ties with the
land 138. The Inter-American Court has pronounced itself in a similar fashion in a
number of later cases, the most recent of which is the Case of the Indigenous
Community Xákmok Kásek. In its decision of August 2010, the Inter-American
Court stated that this community’s cultural identity was affected through lacking
its own land and natural resources. This resulted in a violation of the right to
1389
property .
7.66 Colombia has an obligation not to frustrate the rights of indigenous
peoples in Ecuador: the only issue in this case is evidentiary, whether it has failed
to satisfy that obligation. Does aerial spraying of toxic herbicides along and near
the border have a significant adverse impact on the rights of those affected in
Ecuador? Ecuador’s answer to that question is yes, for all the reasons set out in
Chapter 3 and especially in the reports of UN Special Rapporteurs on the Right to
Health, the Right to Food, and Rights of Indigenous Peoples, as summarised
1387
CCM, Chap. 9, para. 9.169.
1388EM, Chap. 9, para. 9.31 (quoting Case of the Mayagna (Sumo) Awas Tingni Community v.
Nicaragua, Judgment, IACHR, Series C No. 79, para. 149 (31 Aug. 2001)).
1389Case of the Indigenous Community Xákmok Kásek v. Paraguay, Judgment, IACHR, Series C
No. 125, paras. 173-178 (24 August 2010).
513 1390
above . Insofar as this evidence shows that indigenous peoples within Ecuador
are harmed by Colombia’s aerial spraying, that amounts to a denial of their rights.
C. TARGETING OF INDIGENOUS P EOPLES
7.67 As with human rights law generally, it is not necessary that the activity
that interferes with the rights of indigenous peoples should have been targeted
specifically at the victims in order to constitute a violation of their rights. The
case law dealing with interference with property, natural resources, and
traditional way of life does not support Colombia’s defence to Ecuador’s claims.
For example, in Maya Indigenous Community of the Toledo District v. Belize, the
Inter-American Commission on Human Rights accepted that logging concessions
threatened long-term and irreversible damage to the natural environment on
which the petitioners’ system of subsistence agriculture depended 1391. Loss of
topsoil would prevent forest regeneration, damaging water supplies and
1392
diminishing the availability of wildlife and plants . Citing the decision of the
African Commission on Human and Peoples’ Rights in Social and Economic
Rights Action Centre v. Nigeria, the Inter-American Commission concluded that
there had been violations of the petitioners’ right to property in their ancestral
139See supra Chap. 7, paras. 7.5, 7.12, 7.49.
1391
Maya Indigenous Communities of the Toledo District v. Belize, Judgment, Inter-Am.C.H.R.,
Report No. 40/04, Case 12.053, paras. 147-148 (12 Oct. 2004).
1392
Ibid. at para. 31.
514land1393. Its final order required Belize to repair the environmental damage, and
to take measures to demarcate and protect their land in consultation with the
1394
community . At paragraph 150 it noted:
“This Commission similarly acknowledges the importance of
economic development for the prosperity of the populations of this
Hemisphere. As proclaimed in the Inter-American Democratic
Charter, ‘[t]he promotion and observance of economic, social, and
cultural rights are inherently linked to integral development,
equitable economic growth, and to the consolidation of democracy
of the states of the Hemisphere.’ At the same time, development
activities must be accompanied by appropriate and effective
measures to ensure that they do not proceed at the expense of the
fundamental rights of persons who may be particularly and
negatively affected, including indigenous communities and the
environment upon which the1395pend for their physical, cultural
and spiritual well-beingŽ .
The important point here is that the development activities to which the
Commission refers were not “targetedŽ at the indigenous peoples whose rights
were affected.
7.68 Similarly, in Ilmari Lansman et al. v. Finland, the UN Human Rights
Committee held that
“A State may understandably wish to encourage development or
allow economic activity by enterprises. The scope of its freedom
to do so is not to be assessed by reference to a margin of
appreciation, but by reference to the obligations it has undertaken
in article 27. Article 27 requires that a member of a minority shall
1393
Ibid. at paras. 147-149.
139Ibid. at para. 197.
139Ibid., para. 150.
515 not be denied his right to enjoy his culture. Thus, measures whose
impact amount to a denial of the right will not be compatible with
the obligations under article 27. However, measures that have a
certain limited impact on the way of life of persons belonging to a
minority will not necessarily amount to a denial of the right under
article 27Ž139.
The Committee concluded that Finland had taken adequate measures to minimise
the impact on reindeer herding139. Once again the activities were not specifically
targeted at indigenous peoples, but the State nevertheless had an obligation to
take adequate measures to minimise their impact. The point in the present case is
that Colombia has not taken adequate measures … or indeed any proper measures
… to prevent or minimise collateral damage to the rights of those affected in
Ecuador, even though the spraying is not aimed or targeted at indigenous peoples
per se.
D. T ERRITORIALITY OF O BLIGATIONS T OWARDS INDIGENOUS PEOPLES
7.69 Ecuador’s response on the territoriality of obligations with respect to the
rights of indigenous peoples is the same as its response to Colombia’s arguments
on human rights in general. Ecuador does not argue that Colombia has to ensure
the rights of persons within Ecuador. However, within the common legal space
that both States occupy, Colombia has a negative obligation to refrain from
frustrating the rights of indigenous populations living across the border in
1396
Ilmari Lansman et al. v. Finland, ICCPR Comm. No. 511/1992, para. 9.4 (1996).
1397
Ibid., para. 9.7 (1996Compare Lubicon Lake Band v. Canada, ICCPR Comm. No.
167/1984, paras. 32.2-33 (1990) (finding that the impact of oil and gas extraction on the
applicants’ traditional subsistence economy constituted a violation of Article 27 of the ICCPR).
516Ecuador. Article 7 of ILO Convention No. 169, Article 21 of the American
Convention on Human Rights, Article 27 of the ICCPR, and Article 29 of the
2007 Universal Declaration on Human Rights are entirely consistent with
Ecuador’s position. Any other view would be inherently destructive of the
holistic protection of indigenous peoples, such as the Awá in Ecuador and
Colombia, whose territory and living space frequently straddle international
borders, as they do in the present case 1398. The particular cross-border
vulnerability of indigenous peoples was understood by the drafters of ILO
Convention No. 169; it requires States Parties to give special attention to their
needs as social groups, not simply as individuals when they happen to reside
within the territory of one or the other State. It is therefore not surprising that
there is no provision in the ILO Convention comparable to Article 2(1) of the
ICCPR. Assuming for the sake of argument that Colombia is correct in limiting
the obligations of States under the ICCPR to “individuals within its territory and
1399
subject to its jurisdictionŽ , there is plainly no textual, contextual or purposive
basis for applying the same territorial limitation to ILO Convention No. 169. On
the contrary, that Convention applies explicitly to “tribal peoples in independent
countriesŽ and “peoples in independent countries who are regarded as
139See Whitten et al. Report, op. cit., pp. 45-46. ER, Vol. II, Annex 5.
139CCM, Chap. 9, paras. 9.16-9.36.
517indigenousƒŽ 140. There is no reference to territory or jurisdiction as the criteria
for applying the Convention.
E. C ONSULTATION AND N OTIFICATION
7.70 Under Article 6 of ILO Convention No. 169, any State planning to adopt
measures that may detrimentally affect directly the living conditions or the health
of indigenous peoples must consult with the peoples concerned through
appropriate procedures. This applies to indigenous peoples living on both sides
of the border. No such prior consultation or prior notification has ever taken
place on the Ecuadorian side of the border. In fact, Colombia persistently repeats
the mantra that it is entitled to forcefully eradicate illicit crops and that it can only
successfully implement its programme under the cover of secrecy. Article 6(2)
stipulates that “the consultations carried out in application of this Convention
shall be undertaken, in good faith and in a form appropriate to the circumstances,
with the objective of achieving agreement or consent to the proposed
1401
measuresŽ . Leaving aside the question of the need for an agreement or
consent by potentially affected indigenous communities, the wording of this
provision leaves enough room for accommodating Colombia’s security concerns
1400
Article 1(1).
1401See also Colombian Law 21 of 1991, approving ILO Convention No. 169; Claudia Rojas
Quiñonez, Esq., The Aerial Spray Program and Violations of Colombia’s Domestic Laws
Regarding the Environment and the Rights of Indigenous Peoples, paras. 134-145 (Jan. 2011)
(regarding the incorporation of ILO Convention No. 169, and its obligations, into Colombian
domestic law) (hereinafter “Rojas ReportŽ). ER, Vol. II, Annex 8.
518with the necessity … even at very short notice … to alert local populations so that
they can take necessary measures to avoid direct contact with the spray.
7.71 Further, Article 7(3) of the ILO Convention states in very clear terms
that:
“Governments shall ensure that, whenever appropriate, studies are
carried out, in co-operation with the peoples concerned, to assess
the social, spiritual, cultural and environmental impact on them of
planned development activities. The results of these studies shall
be considered as fundamental criteria for the implementation of
these activitiesŽ.
It is not necessary to dwell on the question whether the spraying operations
qualify as “planned development activitiesŽ within the meaning of this provision.
It is clear in Colombia’s rhetoric that it seeks to achieve alternative development
goals through its aerial spraying programme. Moreover, it is readily apparent
from the plain meaning of Article 7(3) that it is concerned with securing
appropriate follow-up and monitoring mechanisms to ensure that government
activities or government-authorized activities in the territories of indigenous
peoples do not detrimentally affect the social or spiritual environments of such
communities.
7.72 It will not have escaped the attention of the Court that no such monitoring
reports have been submitted to the Court by Colombia in its Counter-Memorial,
or ever been made available to Ecuador, let alone studies carried out in
519cooperation with affected indigenous communities. This is not only true in
regard to the absence of monitoring reports on potential effects on Ecuadorian
territory, it is equally true for the lack of reports pertaining to the Colombian side
of the border. Not one report was presented by Colombia on monitoring activities
along its own border in direct proximity to its spraying activities. It is obviously
not enough to conduct impact studies or produce monitoring reports in regard to
totally disconnected regions somewhere else in Colombia. It is equally
insufficient to rely on random monitoring missions, because in order to be
scientifically valid they need to be carried out directly or shortly after the
sprayings take place. As such, Colombia is under a duty to assess the
transboundary impacts of its planned activities and to conduct regular monitoring
missions during the months where the sprayings occurred. It has failed to meet
this obligation.
7.73 Thus, in addition to the human rights violations suffered by the indigenous
people on the Ecuadorian side of the border, Colombia is in violation of its
international law obligations under ILO Convention No. 169. It should have
consulted and informed potentially affected indigenous communities … in the
present case including the representatives of the Awá, Cofán, Kichwa and Afro-
Ecuadorian communities … and conducted meaningful monitoring missions in
order to ensure that its actions did not go beyond what Colombia assumed to be
520the best-case scenario. These actions … or inactions … violate ILO Convention
No. 169.
7.74 The right of effective participation of indigenous communities in the
decision-making process when governmental measures are likely to affect them
has been expressly recognized by the Constitutional Court of Colombia on
1402
various occasions . The Court ruled in particular that the mechanism of prior
consultation of indigenous peoples is an integral part of the fundamental right to
1403
participate in the decision-making process . This finding has been repeatedly
1404
affirmed by Colombia’s Constitutional Court . Of particular importance for
1402
Republic of Colombia, Constitutional Court, Triviño et al., Judgment SU-039/97, p. 1 (3 Feb.
1997) The Court held the following: “[I]t is provided for, when the exploitation of natural
resources in indigenous territories is attempted, the communities’ participation in the ultimately
adopted decisions to authorize said exploitation. In this manner, the communities’ fundamental
right to preserve their referenced integrity is guaranteed and made effective through exercise of
another right that also possesses a fundamental nature, in terms of Article 40, paragraph 2 of the
Constitution, which is the right to the communities’ participation in the referenced decisionsŽ.
ER, Vol. V, Annex 128. See also Rojas Report, op. cit., paras. 134-145 (discussing indigenous
law in Colombia). ER, Vol. II, Annex 8.
1403Republic of Colombia, Constitutional Court, Triviño et al., Judgment SU-039/97, p. 1 (3 Feb.
1997) (“ƒ the participation of the indigenous communities in decision that may affect them
relating to the exploitation of natural resources, is noteworthy in that the mentioned participation
through the mechanism of consultation acquires the connotation of a fundamental right, since a
basic instrument is created to preserve the ethnic, social, economic and cultural integrity of the
indigenous communities and thus, to guarantee their continuing existence as a social groupŽ.) ER,
Vol. V, Annex 128. See also Rojas Report, op. cit., para. 147. ER, Vol. II, Annex 8.
1404
See, e.g., Constitutional Court of Colombia, Judgment T-428 (1992); Constitutional Court of
Colombia, Judgment T-405 (1993); Constitutional Court of Colombia, Judgment T-007 (1995).
For subsequent instances, see: Constitutional Court of Colombia, Judgment C-169 (2001). See
also Republic of Colombia, Constitutional Court, Urueta Rojas., Judgment C-418/02 (28 May
2002) (“For purposes of the resolution of the present process, it is relevant to highlight that
participation in itself reaches the level of a fundamental right which the State must assure and
facilitate for “allŽ, as an essential State aim, in the context of decisions that affect them and their
economic, political, administrative and cultural life. At the same time, participation is established
as an indispensable and irreplaceable tool for the effectiveness other recognized constitutional
521present purposes is the petition filed by the Organisation of Indigenous Peoples of
the Colombian Amazon (“OPIACŽ) seeking by way of temporary injunction the
protection of the right to life, identity, and cultural integrity, to free development
1405
of personality and due process affected by the aerial spraying program . The
Organisation protested against the fact that Colombia was carrying out its aerial
spraying programme without having consulted or notified the affected indigenous
communities. In this case, the Constitutional Court related in detail the evolution
in international law of the protection of indigenous peoples, the manner in which
indigenous rights have been incorporated into the Colombian legal order, and the
development of the Court’s jurisprudence in this respect.
7.75 With respect to the obligation of prior consultation and its relation to the
ILO Convention, the Colombian Constitutional Court ruled as follows:
“It is of particular importance in the present case to refer to ILO
Convention No. 169, specifically, the right of indigenous and tribal
peoples to participate in prior consultation in accordance with the
Constitution and constitutionality block and pursuant to the
dispositions in articles 93 and 94 of the constitutional system, not
only because the instrument containing the provision is from the
rights, whether or not they possess a fundamental nature. ƒ The indigenous communities’ right to
participation as a fundamental right is supported by Convention number 169, approved by Law 21
of 1991. This Convention is aimed at ensuring the rights of indigenous peoples to their respective
territories and the protection of their cultural, social and economic values, as a means to assure
their survival as human groupsŽ.). ER, Vol. V, Annex 142. See also, Rojas Report, op. cit., para.
147. ER, Vol. II, Annex 8.
1405Constitutional Court of Colombia, Judgment SU-383/03 (13 May 2003) (summarized in ILO,
Application of Convention No. 169 by Domestic and International Courts in Latin America: A
Casebook, pp. 87-97 (Geneva: ILO, 2009)). See also Rojas Report, paras. 130, 137, 150-155. ER,
Vol. II, Annex 8.
522 International Labour Organization and sets out the labour rights of
these peoples in article 53 of the Constitution but also i) because
the participation of the indigenous communities in decisions taken
in respect of the natural resources in their territories is set out in
article 330 of the Constitution, and this cannot be understood as
the negation of the right of these peoples to be consulted in other
aspects inherent to their subsistence as a recognisable community
pursuant to article 94 of the Constitution, ii) because said
Convention is the most recognized instrument against the
discrimination that is suffered by indigenous and tribal peoples, iii)
because the right of indigenous peoples to be consulted prior to
administrative and legislative decisions that directly affect them is
a measure of affirmative action that the international community
has adopted and recommended to combat the origins, causes,
forms and modern manifestation of racism, racial discrimination,
xenophobia, and the related forms of intolerance that affect the
indigenous and tribal peoplesŽ 140.
And the Constitutional Court went on to state:
“Thus, in line with ILO Convention No. 169, the consultations
ordered cannot be understood as a mere formality, given that
carrying them out in good faith means that the indigenous and
tribal peoples of the Colombian Amazon be informed about the
content of the Programme that is taking place in their territories in
order to obtain their consent on the impact of the measures1407their
habitat and on their cognitive and spiritual frameworkŽ .
7.76 It is thus clear that Colombia has not only failed to inform or consult the
Government of Ecuador or the populations living on the Ecuadorian side of the
1406Ibid. Translation taken from ILO, Application of Convention No. 169 by Domestic and
International Courts in Latin America: A Casebook, p. 94 (Geneva: ILO, 2009) (internal citation
omitted). See also Rojas Report, op. cit., paras. 150-155 (discussing this ruling). ER, Vol. II,
Annex 8.
140Ibid. at p. 96. See also ibid., paras. 150-155 (discussing this ruling). ER, Vol. II, Annex 8.
523border, but has manifestly also failed to apply ILO Convention No. 169 to its own
indigenous population 140.
7.77 Finally, with regard to transboundary cooperation, it is plain that
Colombia has also violated obligations deriving from a bilateral agreement of
November 2002 relating to transboundary cooperation in the so-called Zonas de
Integración Fronteriza (Border Integration Zones) 1409. In pursuance of a
Decision of the Andean Council of Foreign Ministers of June 2001 calling for the
establishment of border zones in which neighbouring States would cooperate with
1410
a view to promote sustainable development , Colombia proposed to Ecuador,
by way of a diplomatic note, that certain border zones be designated by both
1411
States . Ecuador replied positively and thus, by an exchange of notes
completed in November 2002, the provinces of Putumayo and Nariño in
Colombia as well as the provinces of Sucumbíos, Carchi and Esmeraldas in
Ecuador were designated as Zonas de Integración Fronteriza 1412. As will be
outlined further below, Colombia also acted in pursuance of its national
1408
See Rojas Report, op. cit., paras. 146-160, 169-173 (describing the Colombian government’s
non-compliance with internal regulations regarding indigenous rights). ER, Vol. II, Annex 8.
1409
Andean Community, Border Integration Zone Colombia-Ecuador, Diplomatic Notes
DM/DDF 44552 & 54679/02 GM/DGAF, Official Gazette of Agreements of Cartagena, No. 888
(21 Jan. 2003). ER, Vol. IV, Annex 101.
1410
Andean Community, Decision 501 Border Integration Zone in the Andean Community,
Official Gazette of Agreements of Cartagena, No. 680 (28 June 2001). ER, Vol. IV, Annex 100.
1411
Andean Community, Border Integration Zone Colombia-Ecuador, Diplomatic Notes
DM/DDF 44552 & 54679/02 GM/DGAF, Official Gazette of Agreements of Cartagena, No. 888
(21 Jan. 2003). ER, Vol. IV, Annex 101.
1412Ibid.
524legislation for the development of border regions, Law 191 of 1995, which
includes as a main objective the respect for human rights and the protection of
1413
indigenous peoples living in these regions . Colombia and Ecuador thus
agreed, as proposed by the above-mentioned Decision of the Andean Council of
Foreign Ministers, to cooperate in these border regions.
7.78 Set in this context, it is also significant that Colombia’s domestic law …
1414
including under its own Constitution … accord special recognition for the
needs of indigenous peoples, a point on which Colombia’s Counter-Memorial is
tellingly silent. In 1991, Colombia enacted the current Constitution, further
1415
strengthening the protection of indigenous rights . Since then, various laws
1416
have granted specific protection to the rights of indigenous peoples .
Significantly, Law 21 of 1991 approved ILO Convention No. 169 and
incorporated it into the national legal order 1417. The Colombian Constitutional
Court has since confirmed that the ILO Convention establishes a principal legal
1413Republic of Colombia, National Congress, Law 191 of 1995 (23 June 1995). ER, Vol. V,
Annex 125.
1414See Political Constitution of Colombia, Arts. 7, 68(5), 70(2), 246 and 330 (1991). See also,
Rojas Report, op. cit., paras. 134-145. ER, Vol. II, Annex 8.
1415See Rojas Report, op. cit., paras. 134-145. ER, Vol. II, Annex 8.
1416See ibid.
1417
Colombian Law 21 of 1991, approving ILO Convention No. 169. See Rojas Report, op. cit.,
paras. 134-145. ER, Vol. II, Annex 8.
525source for the domestic legislation relating to indigenous rights 1418. These laws
recognise greater autonomy for indigenous communities as regards the territorial
and political management of their lands and natural resources, and establish better
participatory rights in decision-making processes that are likely to affect their
interests and rights. It appears that these domestic norms were not complied with
1419
in authorising the aerial spraying programme . Relatedly, Colombia has
committed itself to providing particular attention and assistance to indigenous
peoples living in the border regions with Ecuador. In 1995, Colombia adopted
Law 191 of 1995, relating to its border regions: this sought, inter alia, to promote
the protection and development of indigenous peoples, and recognised a right of
contact with indigenous peoples living across the border in other neighbouring
1420
countries . If nothing else, this law recognises and reflects the transboundary
nature of the obligations imposed upon Colombia by ILO Convention No. 169, as
1418Republic of Colombia, Constitutional Court, Molina, Judgment C-401/05, paras. 17-22 (14
Apr. 2005) (holding, in conformity with Article 53(4) of the 1991 Constitution, “as a general
matter, all these [labour] conventions acquire the nature of legal, binding standards in internal
law due to the simple fact of their respective ratification, without the necessity of promulgating
new laws to incorporate its specific content into the country’s legal framework or its
development.Ž and that international human rights treaties that have been ratified by Colombia are
a principal source forming part of the so-called “constitutionality block” (“bloque de
constitucionalidadŽ.). ER, Vol. V, Annex 153. See also Rojas Report, op. cit., paras. 143-144.
ER, Vol. II, Annex 8.
1419
See Rojas Report, op. cit., paras. 146-160, 169-173 (describing the Colombian government’s
non-compliance with internal regulations regarding indigenous rights). ER, Vol. II, Annex 8.
1420
Republic of Colombia, National Congress, Law 191 of 1995, Arts. 3 and 5 (23 June 1995).
ER, Vol. V, Annex 125.
526well as the particular geographical, environmental, cultural and socioeconomic
1421
characteristics of the area , and contradicts Colombia’s territorial arguments.
7.79 It is submitted that, by implementing the programme of spraying toxic
pesticides affecting both the Ecuadorian and Colombian border provinces,
Colombia has violated its obligations created by the exchange of notes to
cooperate with Ecuador in the positive development of these border zones. Such
violation is further confirmed by Colombia’s domestic laws establishing special
protection for indigenous peoples.
Section V. Relationship Between the Protection of the Environment and the
Protection of Human and Indigenous Rights
7.80 Colombia’s argument that the interrelationship between environmental
law and human rights law has to be construed in the light of the principle of
specialty as applied by the ICJ in the Nuclear Weapons Case is legally
unfounded. According to Colombia, “[a] new set of norms and balances
concerning transboundary harm is not to be ‘deduced from the terms of the
Covenant itselfŽ, when international law already lays down the relevant
1421
Republic of Colombia, National Congress, Law 191 of 1995, Art. 4(c) (23 June 1995)
(defining Border Integration Zones as: “Those areas belonging to Border Departments, that
possess geographic, environmental, cultural and/or socioeconomic characteristics that advise joint
action and planning by border authorities in joint agreement with neighbouring countries, to take
actions that are suitable for promoting development and strengthening bilateral and international
exchangeŽ.) (emphasis added). ER, Vol. V, Annex 125.
527standardŽ 142. Thus, again according to Colombia, the content of human rights
norms in connection with environmental degradation cannot be determined
independently of the special rules of environmental law, or those relating to the
protection of indigenous peoples 1423. The extent to which Colombia’s purported
lex specialis rule relating to the right to life in cases of armed conflicts under
international humanitarian law can also be applied in times of peace, and relating
to the particular subject of environmental harm, is not apparent to Ecuador.
7.81 In fact, the present case is not concerned with the application of human
rights under two different legal regimes, i.e., in times of peace as opposed to
times of armed conflict, but with the complementary application of different types
of legal rules and institutions. International human rights instruments and
international environmental law and the rules protecting indigenous peoples take
into account the harmful impact of pollution on humans, and focus on the failure
of the State to take necessary preventive measures. As Ecuador has already
explained in the previous section of this Chapter, human rights and the law on
transboundary harm, for example, are not mutually exclusive, but rest on the
same foundations and must be applied in a mutually supportive way. Each of
these norms, together with those arising in respect of the protection of the rights
of indigenous peoples, is distinct and gives rise to an independent cause of action.
142CCM, Chap. 9, para. 9.7.
142Ibid.
5287.82 Moreover, as the Court has noted: “the environment is not an abstraction
but represents the living space, the quality of life and the very health of human
beings, including generations unbornŽ 1424. The interrelation between human
rights and environmental harm, and between the obligations to protect indigenous
peoples and human rights and environmental norms, has been reaffirmed on
several occasions by the Inter-American Court of Human Rights, most recently in
the Kawas-Fernández Case, where the Inter-American Court stated that:
“in accordance with the case law of this Court and the European
Court of Human Rights, there is an undeniable link between the
protection of the environment and the enjoyment of other human
rights. The ways in which the environmental degradation and the
adverse effects of the climate change have impaired the effective
enjoyment of human rights in the continent has been the subject of
discussion by the General Assembly of the Organization of
American States and the United Nations. It should also be noted
that a considerable number of States Parties to the American
Convention have adopted constitutional provisions which
expressly recognize the rights to a healthy environment. These
advances towards the development of human rights in the
continent have been incorporated into the Additional Protocol to
the American Convention on Human Rights in the Area of
Economic, Social and Cultural Rights ‘Protocol of San
Salvador’Ž 142.
1424
Legality of the Threat or Use of Nuclear Weapons, I.C.J. Reports 1996, pp. 241, para. 29.
142Case of Kawas-Fernández v. Honduras, Judgment, IACHR, Series C No. 152, para. 148 (3
Apr. 2009) (emphasis added, internal citations omitted). See also Matter of Pueblo Indígena de
Sarayaku Regarding Ecuador, Provisional Measures, IACHR, Series E No. 21 (17 June 2005);
Case of the Mayagna (Sumo) Awas Tingni Community v. Nicaragua, Judgment, IACHR Series C
No. 79 (31 Aug. 2001); Case of the Yakye Axa Indigenous Community v. Paraguay, Judgment,
IACHR, Series C No. 125, paras. 131 and 137 (17 June 2005); Case of the Sawhoyamaxa
Indigenous Community, Judgment, IACHR, Series C No. 146, paras. 118-121 and 131 (29 Mar.
2006); Case of the Saramaka People v. Suriname, Preliminary Objections, Judgment, IACHR,
Series C No. 172, paras. 121, 122, 126, 128 (28 Nov. 2007).
5297.83 Ecuador submits that the law concerning transboundary harm must be
interpreted and applied in the light of relevant human rights standards relating to
life, health, private life and property, among others, rather than mediated through
an alleged lex specialis rule. Ecuador submits that human rights obligations may
be particularly relevant in the context of transboundary harm: it is precisely in
such situations of conflicting sovereignties where the risk of human rights
violations in another jurisdiction is most likely. The same failings that lead to
Colombia’s failure to prevent transboundary harm in Ecuador have in this case
also led to a failure to respect human rights in Ecuador.
7.84 For all these reasons, Ecuador invites the Court to reject arguments
advanced by Colombia that seek to keep human rights and environmental
protection and the rights of indigenous peoples as distinct and separate fields
which need to be accommodated through an alleged lex specialis rule. Far from
presenting a normative conflict, these fields of law can and should be interpreted
and applied in a consistent and mutually supportive way.
Conclusions
7.85 In sum, Colombia’s attempt in the Counter-Memorial to challenge the
overwhelming case for human rights violations that Ecuador presented in the
Memorial fails for the following reasons:
530(1) The evidence adduced by Ecuador shows that drift from Colombia’s aerial
spraying of toxic herbicides has caused significant harm to people and their
property, as well as to their environment, in Ecuador. The failure to prevent these
harms is more than sufficient to constitute a violation of the right to life, health,
private life, food and water, property, humane treatment, a healthy environment,
and information, in contravention, inter alia, of the 1969 American Convention
on Human Rights with the 1988 Additional Protocol; the 1966 ICCPR and
ICESCR; the 1979 CEDAW; and the 1989 CRC.
(2) It is not necessary for Ecuador to show that Colombia’s aerial spraying is
“targetedŽ at anyone in Ecuador. The human rights case law concerned with
pollution impacts on health and private life shows that pollution normally
involves unintended and incidental consequences, but these impacts are no less
violations of human rights.
(3) Harm could have been prevented if Colombia had exercised the necessary
diligence in controlling the spraying operations and enforcing its own
environmental standards. The human rights case law demonstrates that States
must balance the interests of the community in eradicating illicit crops against the
harm to individual human rights. By failing to comply with or enforce its own
EMP, Colombia has carried out spraying operations in disregard of the harmful
impact on humans.
531(4) Human rights, the law on transboundary harm and the protection of the
rights of indigenous peoples are not mutually exclusive, but rest on the same
foundations and must be applied in a consistent and mutually supportive way.
The law concerning transboundary harm must be interpreted and applied in the
light of relevant human rights standards relating to life, health, private life, food
and water, property, humane treatment, a healthy environment and information,
rather than mediated through an alleged lex specialis rule.
(5) Colombia and Ecuador are part of a common legal space at the regional
level (un espace juridique commun) relating to the protection of human rights and
indigenous peoples. It follows that Colombia and Ecuador are subject to an
international public order of human rights which does not allow for loopholes in
the effective protection of human rights for any part of their population.
(6) The circumstances surrounding the present case fall within the terms of
Article 1 of the American Convention on Human Rights and Article 2(1) of the
ICCPR. To exclude from the protection offered by these instruments acts
committed in one Contracting State but which produce effects in the territory of
another Contracting State would run counter to the object and purpose of these
instruments.
(7) The daily life of indigenous peoples living on the Ecuadorian side of the
border has been particularly affected by Colombia’s aerial spraying.
Displacement of communities, interference with the use and enjoyment of their
532property and traditional culture, and loss of access to and use of the forests,
including traditional medicines, have all resulted from Colombia’s extensive
spray programme, in violation Article 27 of the ICCPR, Articles 4-7, 13 and 15 of
ILO Convention No. 169, and Article 21 of the American Convention.
(8) Colombia is in violation of its obligation under ILO Convention No. 169
to respect the rights of indigenous peoples, and to consult and notify indigenous
communities likely to be affected by aerial spraying of herbicides … in the present
case, including the representatives of the Awá, Cofán, Kichwa and Afro-
Ecuadorian communities. It is also in violation of its obligation to conduct
meaningful monitoring of the impacts on these communities.
(9) Finally, Colombia has violated obligations deriving from a bilateral
agreement of November 2002 relating to transboundary cooperation in the so-
called Zonas de Integración Fronteriza (Border Integration Zones).
533CHAPTER 8.
REMEDIES8.1 In Chapter 10 of the Memorial, Ecuador set out the consequences of the
responsibility and liability of Colombia for the multiple violations of international
law that have been occasioned by its acts and omissions in relation to the aerial
spraying programme commenced in 2000. Basing itself upon the ILC Draft
Articles on State Responsibility, and on the practise adopted by the International
Court of Justice, in its Memorial, Ecuador set out the principles underlying the
relief sought in its submissions, calling for (1) a declaration that Colombia has
1426
violated its international obligations , (2) an order of cessation and non-
1427
repetition of Colombia’s internationally unlawful acts , (3) the application of
1428
the principles governing reparation , (4) the application of the principles
1429 1430
governing compensation , and (5) the principle of satisfaction .
8.2 Colombia has chosen to respond only briefly to these arguments and
1431
submissions, in Chapter 10 of its Counter-Memorial . Colombia indicates in
the most general terms that Ecuador’s arguments are inadequate, and for this
reason it will only address “all issues of legal consequences, including
1426Memorial of Ecuador, Vol. I, Chap. 10, para. 10.2 (28 Apr. 2009) (hereinafter “EMŽ).
1427EM, Chap. 10, paras. 10.10-10.13.
1428EM, Chap. 10, paras. 10.14-10.25.
1429EM, Chap. 10, paras. 10.26-10.58.
1430
EM, Chap. 10, paras. 10.59-10.60.
1431
Counter-Memorial of Colombia, Vol. I, Chap. 10, paras. 10.1-10.6 (29 Mar. 2010) (hereinafter
“CCMŽ).
537 1432
quantificationŽ at a later stage in light of the Court’s “actual findings of factŽ .
Apart from this, Colombia sets out its reasons why it believes the Court should
not accede to Ecuador’s request that the Court should order Colombia to refrain
from further aerial spraying operations “at, near or across the border with
1433
EcuadorŽ . Ecuador will respond to each argument in turn.
8.3 As a preliminary matter, Ecuador notes that, as recently as 11 November
2010, Colombia reaffirmed that it would continue to maintain a buffer zone of 10
kilometres along the border with Ecuador, within which it would not engage in
1434
aerial spraying activity . This representation has now been made on numerous
occasions, and it is one on which Ecuador relies, not least as it reflects a
confirmation on the part of Colombia as to the reasonableness and proportionality
of the relief sought by Ecuador. However, Colombia’s representation is
insufficient in itself because it is neither a permanent nor a binding commitment
to refrain from conducting aerial spraying operations in close proximity to the
border. To obtain such a permanent and binding commitment … in light of
Colombia’s refusal to give one … Ecuador requires the intervention of the Court.
143CCM, Chap. 10, para. 10.7.
1433
CCM, Chap. 10, paras. 10.8-10.11.
1434
Republic of Colombia, Ministry of Foreign Affairs, Press Release (11 Nov. 2010). ER, Vol.
V, Annex 156.
538 Section I. The Purported Impropriety of Ecuador’s Approach to Relief
8.4 Colombia seeks to circumvent Ecuador’s claimed relief on the grounds
that Ecuador has not proven any material damage and is therefore not entitled to
any remedy at all 143. This is a classic bootstraps argument, and it suffers from a
number of basic and self-evident flaws. First, it proceeds on the erroneous basis
that Ecuador has not established any violation of any international legal
obligations, whether in respect of sovereignty, the prevention of harm to human
health and the environment, the violation of human rights and indigenous
peoples’ rights, or the violation of essential procedural requirements such as the
provision of information and the conduct of a prior environmental impact
assessment. If Colombia is wrong on that basic assumption … as Ecuador asserts
is self-evidently the case … then it follows inexorably that Ecuador is entitled to a
declaration of Colombia’s responsibility under international law and, in due
course, an assessment of liability. This is well-established and follows from the
ILC Articles, an instrument to which Chapter 10 of the Colombian Counter-
Memorial makes not a single reference. It is particularly appropriate in relation to
the procedural violations, which by definition cannot await a showing of actual
harm in order to have become applicable and to have been violated: it is sufficient
that such a declaration is an appropriate remedy in the face of the real risks that
exist, and that are largely recognised by Colombia.
143CCM, Chap. 10, para. 10.4.
5398.5 Colombia’s approach is to rewrite the basic rules of international law, by
ignoring established principles and the constant practise of the Court. As
Colombia well knows, it is usual and proper for the Court to be asked first to
identify a violation of an international legal obligation, and only then to
determine the consequences of that violation. This is exactly the approach sought
by Ecuador in its Application, in its Memorial and in its Submissions.
8.6 The second error into which Colombia falls is its patent inability to
distinguish between the identification of a violation, on the one hand, and the
quantification of loss, on the other hand. Colombia’s approach to Ecuador’s
request for relief is entirely premised on the alleged inability of Ecuador to
quantify in physical and monetary terms the totality of the harms that have
occurred. It is certainly true that Ecuador has not yet quantified the monetary
value of the harms it has suffered, but adopting the approach taken by the Court
in other cases it respectfully submits it has no need to do so at this stage of the
proceedings. In its Memorial, Ecuador has amply demonstrated that violations of
international legal norms have occurred and that these have had real and serious
consequences: the sovereignty of Ecuador has been violated; damage has been
caused to humans and to their property, including farms; and damage has been
1436
caused to the natural environment in Ecuador . These matters have been
143EM, Chaps. 5 and 7.
540 1437
addressed in detail in this Reply . In regard to quantification of the damages
incurred, Ecuador has taken the position that this is to be addressed at a later stage
1438
in the proceedings .
8.7 In adopting this approach Ecuador is following the tried and tested
practise of the Court, as reflected by way of example in the Nicaragua case and
the Case of Armed Activities in the Democratic Republic of Congo, both of which
were invoked in the Memorial but to which Colombia has offered no response 1439.
In neither of those cases was the Applicant State required to quantify in the first
phase of the proceedings the consequences of the violations or the precise
monetary valuation of the harms suffered.
8.8 Third, and most significantly, Colombia asserts that Ecuador has failed to
prove the “material elementŽ of what Colombia refers to as the “principal
1440
claimŽ . This is simply wrong, as the evidence before the Court shows.
Ecuador has proved violations of international legal obligations and it has also
proved that harmful consequences have been felt as a direct result of these
violations, and for which Colombia’s liability under international law is
1437See supra Chaps. 3 and 5.
1438
EM, Chap. 10, para. 10.49.
1439
EM, Chap. 10, paras. 10.3-10.5.
1440CCM, Chap. 10, para. 10.2.
541 1441
established . The extent of that harm and the quantification of its monetary
value where such compensation is due are properly matters for a later phase of the
proceedings.
8.9 In short, this is not about imposing on the Court any requirement to
“intuitŽ harm, as Colombia claims, or about any form of “confessionŽ that
1442
Ecuador is supposed to have made as to the adequacy or extent of its case . In
no recent case of which Ecuador is aware has the evaluation of the harm suffered
… in monetary terms … been quantified at this stage of the proceedings. Colombia
has fallen into confusion, or is wilfully mischaracterising Ecuador’s pleaded case,
or has simply run out of steam at the tail end of a lengthy and ambiguous pleading
that frequently fails to engage with the arguments made by Ecuador.
8.10 With regard to Ecuador’s claim for an order for non-repetition, Colombia
asserts that no such order should be made “in the absence of proof of any
wrongful act on the part of ColombiaŽ, the only wrongful act being “proof of
1443
damageŽ . Here again, Colombia melds the assessment of the violation of
Ecuador’s substantive international legal rights with the valuation of the harms
suffered, yet they are not one and the same thing. The object of the order for non-
144EM, Chaps. 5 and 7; see supra Chaps. 3 and 5-7.
1442
CCM, Chap. 10, para. 10.2
1443
CCM, Chap. 10, para. 10.4.
542repetition is to protect Ecuador from any further violation of its rights under
international law: those rights include, but are not limited to, the prevention of
further physical harm that is capable of being quantified in monetary terms.
Ecuador is also entitled to relief in relation to violations of sovereignty, the
protection of the aesthetic value of its natural environment, and the protection of
its people from the fears caused by the aerial spraying of toxic pesticides.
Whether or not these are susceptible to monetary valuation … and Ecuador
submits that they are … Colombia is not entitled to act in violation of international
norms that guarantee these rights and the values they reflect.
8.11 Finally, Colombia challenges Ecuador’s reliance on certain authorities
governing the identification of principles for the protection and valuation of
environmental and related harms. Colombia invites the Court to follow the
approach set forth in the single authority of the Trail Smelter case, which dates
back to the 1940s, and take no account of the more recent approach reflected in
the decisions of the UN Compensation Commission (“UNCCŽ) 1444. In this way,
Colombia wishes to inscribe itself in methodologies adopted more than six
decades ago, in the period before the advent of the modern rules of international
law pertaining to the rights of indigenous peoples, human rights and the
environment, and to ignore the totality of legal developments that have occurred
since then. This is telling, and also flatly inconsistent with the approach reflected
144CCM, Chap. 10, para. 10.6.
543in the domestic law of Colombia 1445. Colombia invites the Court to ignore,
amongst other approaches, that taken by the UNCC. Yet this is a pertinent
authority because it brings together the developments in international law and
practise over the past six decades. It reflects the approach taken by Ecuador in
these proceedings.
8.12 Colombia informs the Court that it will address “all issues of legal
1446
consequencesŽ of its unlawful acts at a later stage . This wholly inadequate
response wilfully mischaracterises the issues to be decided by the Court at the
present stage. Ecuador invites the Court to reject Colombia’s approach, and to
determine that Colombia has violated its international legal obligations, of both a
procedural and substantive character. The monetary consequences are properly to
be determined at a later stage, in accord with established international practise. In
this regard, Ecuador notes that in the recent Judgment in the Case Concerning
Ahmadou Sadio Diallo the Court adopted precisely the approach for which
Ecuador has argued: having found a violation of international law, the Court
ruled: (i) that reparation due “must take the form of compensationŽ; (ii) that the
parties should “engage in negotiation in order to agree on the amount of
compensation to be paidŽ; and (iii) failing agreement between the Parties within
1445See Claudia Rojas Quiñonez, Esq., The Aerial Spray Program and Violations of Colombia’s
Domestic Laws Regarding the Environment and the Rights of Indigenous Peoples, Sections I &
V(B) (Jan. 2011). ER, Vol. II, Annex 8.
144CCM, Chap. 10, para. 10.7.
544six months of the Judgment on the amount of compensation to be paid, the matter
“shall be settled by the Court in a subsequent phase of the proceedingsŽ and that
“a single exchange of written pleadings by the Parties would then be sufficient in
order for it to decide on the amount of compensationŽ 144.
Section II. Ecuador Has Not Attempted to Restrict Colombia’s Sovereignty
Over Its Territory
8.13 Having asserted that it will not engage in any assessment of the legal
consequences of its unlawful acts, Colombia proceeds in the second part of its
response to Ecuador’s remedial arguments to do precisely that. Colombia takes
exception to Ecuador’s request that it should not conduct aerial spraying “at, near
or across the border with EcuadorŽ 1448. The reasons it provides are wholly
unpersuasive.
8.14 Colombia reminds the Court that it has maintained a 10 kilometre “no
1449
sprayŽ zone along the Colombian side of the border . However, it is important
to note that this has been done by Colombia on a voluntary basis and without
1450
prejudice to its capacity to resume such spraying “as might be necessaryŽ .
1447
Case Concerning Ahmadou Sadio Diallo (Republic of Guinea v Democratic Republic of
Congo), Judgment, 2010, pp. 48-49, paras. 161-164.
1448
CCM, Chap. 10, para. 10.10.
1449
CCM, Chap. 10, para. 10.9.
1450Ibid.
545Such an approach provides inadequate guarantees that Ecuador’s rights under
international law will be protected.
8.15 Colombia asserts that it “does not claim any right to spray across the
borderŽ, that the spraying activity “must occur on Colombian territoryŽ and that
its only obligation is to meet a standard of due diligence in preventing drift across
the border 145. Colombia further asserts that “a 100m buffer zone from the
1452
boundary riverŽ is perfectly adequate to prevent spray drift reaching Ecuador .
These statements require a firm response. Colombia has voluntarily adopted a 10
kilometre buffer zone over the past three years. Ecuador is doing no more than
asking the Court to make that same buffer zone legally constraining upon
Colombia; Ecuador invites the Court to order that this buffer zone should be
permanently maintained. In this regard, a number of comments are justified.
8.16 First, Ecuador notes Colombia’s concession that no spray planes may
cross the border into the territory of Ecuador. The evidence in Chapter 2 shows
clearly that Colombian planes have on a small number of occasions crossed the
boundary with Ecuador whilst spraying 1453. Ecuador insists on its right to ensure
that this never happens, whether by design or by inadvertence. Accordingly,
1451
CCM, Chap. 10, para. 10.11.
1452Ibid.
1453See supra Chap. 2, para. 2.163,320 ; see also, R. John Hansman, Ph.D. & Carlos F. Mena,
Ph.D., Analysis of Aerial Eradication Spray Events in the Vicinity of the Border Between
Colombia and Ecuador from 2000 to 2008, p. 13 (Jan. 2011). ER, Vol. II, Annex 2.
546Colombia can have no objection to that part of the order Ecuador has requested
from the Court that concerns non-repetition of aerial intrusions.
8.17 Second, Ecuador notes the standard of due diligence to which Colombia
professes attachment. As described in the Memorial and in Chapter 2 of this
Reply 1454, Colombia has not met even this minimal standard, as spray has drifted
across the border and caused harmful effects. This is the direct result of
Colombia’s failure to exercise proper diligence in respect of the conduct of the
1455
spraying activities: the chemicals used are inappropriate ; the operational
requirements intended to control spray drift are lax by comparison with the vast
majority of other States 145, and even these requirements have been regularly
disregarded on tens of thousands of occasions, particularly in regard to the speed
of the planes, the height at which they dispense the spray, droplet size, application
rate, time of day, climatic conditions, etc., making it more likely that there will be
spray drift and that it will be transported many kilometres into Ecuador 1457.
8.18 Third, the history of aerial spraying on the border plainly demonstrates
that a 100 metre buffer zone is plainly inadequate: Colombian planes have
conducted spraying operations thousands of times more than 100 metres from the
1454See supra Chap. 2, paras. 2.65-2.202; EM, Chap. 5, Sec. III and Chap. 6.
1455See supra Chap. 2, paras. 2.18-2.63.
1456
See supra Chap. 4, paras. 4.99-4.114.
1457
See supra Chap. 2, paras. 2.88-2.154.
547border with Ecuador, yet toxic herbicides have drifted across the border well into
Ecuador and given rise to risks and caused significant harm to Ecuadorian people,
1458
crops, livestock, forests and the environment . For these reasons, Ecuador
maintains that a 10 kilometre buffer zone is the minimum acceptable parameter.
Conclusions
8.19 Contrary to the position adopted by Colombia, the Court’s task is not
limited to establishing that significant harm has occurred, although this is
certainly one of the matters for decision by the Court. In light of the hazardous
activity conducted by Colombia, and the patent risk of potential and irreversible
damage, the obligations of assessment, cooperation, consultation and provision of
information all arise in the context of a risk of potential harm. Accordingly, the
relief sought by Ecuador is appropriate in respect of these procedural violations,
in circumstances where Colombia has not met minimum international standards
or the requirements of its own domestic law: the Court can and should order a 10
kilometre buffer zone. Moreover, Colombia’s actions have caused measurable
and serious harm to people, crops, property and the environment and would do so
again if the spraying were to be resumed. Ecuador is entitled to all the relief it
has sought, and it is entitled to quantify the monetary damages it seeks to recover
in a later phase of the proceedings.
145See supra Chap. 3.
548SUBMISSIONS SUBMISSIONS
On the basis of the facts and law referred to above, Ecuador requests the Court to
adjudge and declare that:
(A) Colombia has violated its obligations under international law by causing or
allowing the deposit on the territory of Ecuador of toxic herbicides that have
caused damage to human health, property and the environment;
(B) Colombia shall indemnify Ecuador for any loss or damage caused by its
internationally unlawful acts, namely the use of herbicides by aerial dispersion,
and in particular:
(i) death or injury to the health of any person or persons arising from the
use of such herbicides;
(ii) any loss of or damage to the property or livelihood of such persons;
(iii) violation of the human rights of such persons;
(iv) violation of the special rights of indigenous peoples;
(v) environmental damage or the depletion of natural resources;
(vi) the costs of monitoring to identify and assess future riss to public
health, human rights and the environment resulting from Colombia’s
use of herbicides; and
(vii) any other loss or damage;
(C) Colombia shall:
(i) respect the sovereignty and territorial integrity of Ecuador;
(ii) respect the human rights of Ecuadorian nationals;
(iii) respect the special rights of indigenous peoples in Ecuador;
(iv) tae no action to harm the natural environment in Ecuador;
551(v) forthwith, tae all steps necessary to prevent, on any part of its
territory, the use of any toxic herbicides in such a way that they could
be deposited onto the territory of Ecuador; and
(vi) prohibit the use, by means of aerial dispersion, of such herbicides
within 10 ilometres of the border with Ecuador.
31 January 2011
____________________________
Mr. Diego Garcia Carrión
Agent of the Republic of Ecuador
552 Certification
I certify that the annexes are true copies of the documents referred
to and that the translations provided are accurate.
__________________________________
Mr. Diego Garcia Carrión
Agent of the Republic of Ecuador
553 LIST OF FIGURES
FOLLOWING
PAGE
Figure 2.1 Excerpt From GLY-41 Label 47
Figure 2.2 Warning Symbols From GYL-41 Label … 48
Human Health
Figure 2.3 Warning Symbols From GYL-41 Label … 48
Animals & Environment
Figure 2.4 Flight Speed of Spray Events Within 83
10 Kilometres of Ecuador’s Sucumbíos
Province (2000 - 2008)
Figure 2.5 Breakup of Spray Droplets As They 84
Encounter Wind Turbulence
Figure 2.6 Altitude of Spray Events Within 10 89
Kilometres of Ecuador’s Sucumbíos Province
(2000 - 2008)
Figure 2.7 Effects of Altitude and Flight Speed on Spray 92
Drift
Figure 2.8 Photograph of AT-802 Spray Plane 97
Figure 2.9 Photograph of OV-10 Aircraft 100
Figure 2.10 Time of Day of Spray Events (2000 - 2008) 116
Figure 2.11 Density of Parallel Spray Lines Within 147
10 Kilometres of Ecuador’s Border During a
Single Month (September 2002)
Figure 2.12 Cumulative Impact of Drift From Multiple 147
Spray Lines
Figure 2.13 Distance From Spray Events Adjacent To 153
Ecuador’s Sucumbíos Province (2000 - 2008)
555Figure 2.14 Distance From Spray Events Adjacent To 15 3
Ecuador’s Esmeraldas Province (2000 - 2008)
Figure 3.1 Spray Events Within 10 Kilometres of 167
Ecuador’s Sucumbíos Province
(December 2000 - February 2001)
Figure 3.2 Spray Events Within 10 Kilometres of 182
Ecuador’s Sucumbíos Province (August -
September 2002) & Spray Events Within
10 Kilometres of Ecuador’s Sucumbíos
Province (November - December 2004)
Figure 3.3 Spray Events Within 10 Kilometres of 191
Ecuador’s Sucumbíos Province
(November 2001 - October 2002)
Figure 3.4 Spray Events Within 10 Kilometres of 216
Ecuador’s Esmeraldas Province (August -
September 2000)
Figure 3.5 Warning Symbols From GLY-41 Label … Fish 221
and other aquatic organisms
Figure 3.6 Spray Events Within 10 Kilometres of 225
Ecuador’s Esmeraldas Province (March 2004
- December 2005)
Figure 3.7 Spray Events Within 10 Kilometres of 225
Ecuador’s Esmeraldas Province
(February 2007)
Figure 3.8 Coca Cultivation Density in the Andean 255
Region, 2006
556 LIST OFANNEXES
VOLUME II
EXPERT REPORTS
Annex 1 R. John Hansman, Ph.D. & Carlos F. Mena, Ph.D., Analysis of
Aerial Eradication Spray Events in the Vicinity of the Border
Between Colombia and Ecuador from 2000 to 2008 (Jan. 2011)
Annex 2 Durham K. Giles, Ph.D., Spray Drift Modeling of Conditions of
Application for Coca Crops in Colombia (Jan. 2011)
Annex 3 Stephen C. Weller, Ph.D., Glyphosate-Based Herbicides and
Potential for Damage to Non-Target Plants Under Conditions of
Application in Colombia (Jan. 2011)
Annex 4 Henri Balslev, Ph.D., The Vulnerability of the Ecuador-Colombia
Border Region to Ecological Harm (Jan. 2011)
Annex 5 Norman E. Whitten, Jr., Ph.D., Dr. William T. Vicers, Ph.D. &
Michael Cepe, Ph.D., Tropical Forest Cultural Ecology and
Social Adaptation in the Ecuadorian Border Region with Colombia
(Jan. 2011)
Annex 6 Charles A. Menzie, Ph.D. & Pieter N. Booth, M.S., Response to:
“Critique of Evaluation of Chemicals Used in Colombia’s Aerial
Spraying Program, and Hazards Presented to People, Plants,
Animals and the Environment in Ecuador,” As Presented in:
Counter-Memorial of the Republic of Colombia, Appendix (Jan.
2011)
Annex 7 Reinhard Joas, Ph.D., The Development of the 2009 European
Union Pesticides Directive With Particular Focus on Aerial
Spraying (Jan. 2011)
Annex 8 Claudia Rojas Quiñonez, Esq., The Aerial Spray Program and
Violations of Colombia’s Domestic Laws Regarding the Environment
and the Rights of Indigenous Peoples (Jan. 2011)
557 VOLUME III
REGULATIONS & TECHNICAL REPORTS
Annex 9. Italian Republic, Legislative Decree No. 194 (17 Mar. 1995)
Annex 10. Spray Drift Tas Force, A Summary of Aerial Application
Studies (1997)
Annex 11. Sweden, Environmental Code 808 (last amended 2009) (1998)
Annex 12. A. J. Hewitt et al., Development of the Spray Drift Tas Force
Database for Aerial Applications, 21(3) ENVIRONMENTAL
TOXICOLOGYAND CHEMISTRY (2002)
Annex 13. C. N. Boutin, C. Elmegaard and C. Kjaer, “Toxicity Testing
of Fifteen Non-crop Plant Species with Six Herbicides in a
Greenhouse Experiment: Implications for Ris AssessmentŽ,
in Ecotoxicology. 13:349…369 (2004)
Annex 14. French Republic, Decree On the Use of Products Mentioned in
Article L.253-1 of Rural Code (5 Mar. 2004)
Annex 15. Las Palmas Ltda., Technical Department, Glyphosate (10,4
l/ha) and Three Different Adjuvants, For Illicit Coca Crop
▯(U\WKR[\OXP▯VSS▯▯▯&RQWURO▯▯$JURQRPLF▯(I¿FDF\▯7HVWLQJ▯RI▯
Doses of Glyphosate in Illicit Crops: Final Report (July 2004)
Annex 16. Republic of Slovenia, Act on Plant Protection Products
(9 Sept. 2004)
Annex 17. United Kingdom, Department for Environment, Food and Rural
Affairs, Code of Practice For Using Plant Protection Products
(2006)
Annex 18. Costa Rica, Executive Decree No. 34202-MAG-S-MINAE-
MOPT-G-MSP (21 May 2007)
Annex 19. Nova Scotia Environment and Labour, Media Backgrounder:
Herbicide Management (July 2007)
558Annex 20. Republic of Estonia, Plant Protection Products Act, entered into
force 1 May 2004, amended 1 July 2008
Annex 21. Austrian Federated State of Vorarlberg, Ordinance on Plant
Protection Products, LGB1.Nr. 18/2008 (2008)
Annex 22. Australian Pesticides and Veterinary Medicines Authority
(APVMA), AVPM Operating Principles in Relation to Spray
Drift Risk (15 July 2008)
Annex 23. American Society of Agricultural and Biological Engineers,
6SUD\▯1R]]OH▯&ODVVL¿FDWLRQ▯E\▯'URSOHW▯6SHFWUD▯▯$16,▯$6$(▯
S572.1 (Mar. 2009)
Annex 24. French Republic, Rural and Maritime Fishery Code, Art. L253-
1(V) (2010)
Annex 25. Government of Sasatchewan, Ministry of Agriculture, 2010
Guide to Crop Protection (2010)
Annex 26. Federal Republic of Germany, Federal Institute of Ris
Assessment, Health Assessment Report POE-tallowamines
(6 Sep. 2010)
Annex 27. United States Roundup Ultra Label
Annex 28. Colombia GYL-41 SL Label and Safety Data Sheet
VERIFICATION & OBSERVATION REPORTS
Annex 29. Organization of Indigenous Nations of Colombia (ONIC),
Evaluation of the Fumigations in Colombia: Destruction of
Rural Areas from Plan Colombia (Aug. 2002)
Annex 30. Colombian Medical Inquests (Sept. 2002)
Annex 31. Ecuadorian Medical Inquests (Sept.- Nov. 2002)
Annex 32. Association of American Jurists, et al., 5HSRUW▯RQ▯9HUL¿FDWLRQ▯
Mission, “Impacts in Ecuador of Fumigations in Putumayo
Department under Plan Colombia” (Oct. 2002)
559Annex 33. Letter from Victor Velasco Tapia, Government of Sucumbios,
Republic of Ecuador, to Lourdes Luque, Minister of Health,
Republic of Ecuador (Oct. 2002)
UNITED STATES GOVERNMENT DOCUMENTS
Annex 34. Memorandum from Peter P. Trent, INL/RM/ASD, PSC Bogota,
to Grant Harden,INL/RM/ASD, COR (4 July 1996)
Annex 35. Memorandum from David Johnson, INL/C/ASD, to Grant
Harden, INL/C/ASD (14 Nov. 1996)
Annex 36. Memorandum from David Johnson INL/C/ASD to Grant
Harden, INL/C/ASD (12 Dec. 1996)
Annex 37. Memorandum from Tim Doty, COR, INL/RM/AS to Dyncorp,
PSD Manager (28 May 1997)
Annex 38. Memorandum from Tim Doty, COR, INL/RM/AD, to Dyncorp,
PSD Manager (Aug. 1997)
Annex 39. Memorandum from Stephen H. Harris, INL/RM/AD, to
Dyncorp (21 June 1999)
Annex 40. Aviation Resource Management Inspection of Air Wing
Colombia Site (23 March 2000)
Annex 41. U.S. Department of Agriculture, Agricultural Research
Service, &RORPELD▯&RFD▯9HUL¿FDWLRQ▯0LVVLRQ▯$SULO▯0D\▯▯▯▯▯▯
(7 July 2001)
Annex 42. Memorandum from Stephen H. Harris, COR, DoS/INL/A, to
Dyncorp, PSD Manager (Dec. 2001)
Annex 43. United States Congressional Research Service, Andean
Regional Initiative (ARI): FY2002 Assistance for Colombia and
Neighbors (14 Dec. 2001)
Annex 44. Memorandum from Lowell E. Neese, SAA, DoS/
INL/A (Bogota), to Stephen H. Harris, COR, DoS/INL/A
(13 Mar. 2002)
560Annex 45. 8QLWHG▯6WDWHV▯(QYLURQPHQWDO▯3URWHFWLRQ▯$JHQF\▯▯2I¿FH▯RI▯
Prevention, Pesticides and Toxic Substances, Report on Issues
Related to the Aerial Eradication of Illicit Coca in Colombia,
Response from EPA Assistant Administrator Johnson to
Secretary of State (19 Aug. 2002)
Annex 46. Memorandum from Lowell Neese, Senior Aviation Advisor,
DoS/INL/A (Colombia), to Paul O’Sullivan, COR, DoS/INL/A
(21 Apr. 2003)
Annex 47. Memorandum from Lowell Neese, Senior Aviation Advisor,
DoS/INL/A (Colombia), to Paul O’Sullivan, COR, DoS/INL/A
(9 May 2003)
Annex 48. Memorandum from David A. Campbell, COR, DoS/INL/A, to
Dyncorp, PSD Manager (Feb. 2004)
Annex 49. Memorandum from Stephen H. Harris, COR, DoS/INL/A, to
Dyncorp, L 55 (Undated)
Annex 50. United States Department of State, Evaluation Summary
Technical Operations, G 111 (Undated)
Annex 51. Memorandum from Michael J. Kenna, INL/RM/AD, Senior
Aviation Advisor, to Steve Harris (COR) and George Arzente,
INL/RM/AD/COR, L 14 (Undated)
VOLUME IV
NEWS ARTICLES
Annex 52. Steve Salisbury, “Pray and Spray: SOF With Coe-Bustin’
BroncosŽ, SOLDIER OF FORTUNE (July 1998)
Annex 53. Tod Robberson, “2 U.S. Pilots Die on Colombian Anti-Narcotics
MissionŽ, DALLAS MORNING NEWS (Dallas, 29 July 1998)
Annex 54. Larry Rohter, “To Colombians, Drug War is Toxic EnemyŽ,
THE NEW YORK TIMES (New Yor, 1 May 2000)
561Annex 55. “The Void of the FumigationsŽ, EL TIEMPO
(Bogotá, 28 May 2000)
Annex 56. “Mayor Denounces FumigationsŽ, EL UNIVERSO
(Guayaquil, 22 Aug. 2000)
Annex 57. “In Mataje the Implementation of Plan Colombia Causes First
RavagesŽ, LA HORA (Quito, 18 Sept. 2000)
Annex 58. “44 Affected by the FumigationsŽ, EL COMERCIO
(Quito, 22 Oct. 2000)
Annex 59. “No To Fumigation: GovernorsŽ, EL TIEMPO
(Bogotá, 15 Jan. 2001)
Annex 60. Juan Forero, “No Crops Spared in Colombia’s Coca
WarŽ, THE NEW YORK TIMES (New Yor, 31 Jan. 2001)
Annex 61. “The Drama of FumigationsŽ, EL UNIVERSO
(Guayaquil, 10 July 2001)
Annex 62. “Fumigation DisputeŽ, EL TIEMPO (Bogotá, 22 July 2001)
Annex 63. “Colombia Drug Czar to Keep SprayingŽ, THE NEW YORK
TIMES (New Yor, 31 July 2001)
Annex 64. “Colombia Denounces Indiscriminate Spraying in PutumayoŽ,
EL COMERCIO (Quito, 10 Jan. 2002)
Annex 65. “Binational Meeting of Indigenous Communities„ Plan
Colombia terrorizes the communitiesŽ, LA HORA (7 Apr. 2002)
Annex 66. “Between Faith and FumigationsŽ, EL TIEMPO
(Bogotá, 10 May 2002)
Annex 67. “Another Controversy Over FumigationŽ, EL COMERCIO
(Quito, 9 July 2002)
Annex 68. “Hunger and Misery from FumigationsŽ, EL UNIVERSO
(Guayaquil, 7 Sept. 2002)
562Annex 69. “Farmers Against FumigationsŽ, EL UNIVERSO
(Guayaquil, 19 Sept. 2002)
Annex 70. “Ecuadorians Demand CompensationŽ, LA HORA
(Quito, 26 Sept. 2002)
Annex 71. “Glyphosate Affects Crops in SucumbíosŽ, EL COMERCIO
(Quito, 8 Oct. 2002)
Annex 72. “Requesting an End to FumigationsŽ, EL TIEMPO
(Bogotá, 10 Oct. 2002)
Annex 73. “Fumigations Cause Concern in PutumayoŽ, EL COMERCIO
(Quito, 10 Nov. 2002)
Annex 74. “Glyphosate RainŽ, EL TIEMPO (Bogotá, 25 Feb. 2003)
Annex 75. “Spray Program on Indigenous Territories Is StrugglingŽ,
EL TIEMPO (Bogotá, 28 Apr. 2003)
Annex 76. “We Will Continue To Fumigate While I Am PresidentŽ,
EL TIEMPO (Bogotá, 30 June 2003)
Annex 77. “Anti-drug plane shot down, U.S. saysŽ, CHICAGO TRIBUNE
(Chicago, 23 Sept. 2003)
Annex 78. “Billiard Shot on Three FrontsŽ, EL TIEMPO
(Bogotá, 13 Nov. 2003)
Annex 79. “Suarez new Environment Minister as Rodriguez QuitsŽ,
BUSINESS NEWS AMERICAS (14 Nov. 2003)
Annex 80. “EU criticises Colombia on rightsŽ, BBC
(London, 22 Jan. 2004)
Annex 81. “More Refugees As a Result of Fumigations Along the BorderŽ,
EL UNIVERSO (Guayaquil, 8 Feb. 2007)
Annex 82. “Colombia Announces Ceasing of Fumigations to
Ease Relations with QuitoŽ, EL UNIVERSAL.COM
(Caracas, 9 Feb. 2007)
563Annex 83. “Colombia Sprayed Within 1 m of the BorderŽ,
EL UNIVERSO (Guayaquil, 10 Feb. 2007)
Annex 84. “Fishermen in Esmeraldas Fear Spraying with Glyphosate
Affects MangrovesŽ, EL UNIVERSO (Guayaquil, 20 Feb. 2007)
Annex 85. “Putumayo: Governor Denounces FumigationsŽ, HOY
(Quito, 29 July 2007)
Annex 86. “A Constitution Appeal Is Ordered In San Jorge: A Judge
Recognizes the Ris of Glyphosate FumigationsŽ, ENTRE
RIOS ENTRE TODOS (Entre Rios, 13 Apr. 2009)
Annex 87. Crytstal Gammon, “Weed Killer Kills Human Cells:
6WXG\▯,QWHQVL¿HV▯'HEDWH▯RYHU▯µ,QHUW¶▯,QJUHGLHQWV´▯▯
ENVIRONMENTAL HEALTH NEWS (22 June 2009)
Annex 88. “Santa Fe: A Ruling In Favor of LifeŽ, RENACE (4 Jan. 2010)
Annex 89. “It Is Warned that Fumigations Are Being Carried Out in Paraná
Despite Them Being ProhibitedŽ, LAVOZ (9 Jan. 2010)
Annex 90. “Fumigations with Glyphosate Is Not Permitted on the the
PeninsulaŽ, BARILOCHE2000 (Bariloche, 10 Feb. 2010)
Annex 91. Shane Romig, “Argentina Court Blocs Glyphosate Spraying
Near Rural TownŽ, DOW JONES NEWSWIRES (21 Mar. 2010)
Annex 92. “Colombian Government Violates Pact and Fumigates with
GlyphosateŽ, VOCES.ORG (10 Nov. 2010)
Annex 93. “Colombia Fumigates AgainŽ, LA HORA (Quito, 11 Nov. 2010)
MULTILATERAL ORGANISATION DOCUMENTS
Annex 94. Commentary on the United Nations Convention Against
,OOLFLW▯7UDI¿F▯LQ▯1DUFRWLF▯'UXJV▯DQG▯3\VFKRWURSLF▯6XEVWDQFHV▯
1988, U.N. Doc. E/ CN.7/590 (20 Dec. 1988)
564Annex 95. European Union, Council Directive Concerning the Placing
of Plant Protection Products on the Market, 91/414/EEC
(15 July 1991)
Annex 96. United Nations, Conference for the Adoption of a Convention
DJDLQVW▯,OOLFLW▯7UDI¿F▯LQ▯1DUFRWLF▯'UXJV▯▯2I¿FLDO▯5HFRUGV (1994)
Annex 97. Organization for Economic Co-Operation and Development
(OECD), Activities to Reduce Pesticide Risks in OECD and
Selected FAO Countries, Part I: Summary Report,
OCDE/GD(96)121 (1996)
Annex 98. Food and Agriculture Organization of the United Nations,
Guidelines on Good Practice for Aerial Application of
Pesticides (2001)
Annex 99. European Parliament, Resolution on Plan Colombia and Support
for the Peace Process in Colombia, EUR. PARL. DOC.
B5-0087 (1 Feb. 2001)
Annex 100. Andean Community, Decision 501 Border Integration Zone
LQ▯WKH▯$QGHDQ▯&RPPXQLW\▯▯2I¿FLDO▯*D]HWWH▯RI▯$JUHHPHQWV▯RI▯
Cartagena, No. 680 (28 June 2001)
Annex 101. Andean Community, Border Integration Zone Colombia-
Ecuador, Diplomatic Notes DM/DDF 44552 & 54679/02 GM/
'*$)▯▯2I¿FLDO▯*D]HWWH▯RI▯$JUHHPHQWV▯RI▯&DUWDJHQD▯▯1R▯▯▯▯▯▯
(21 Jan. 2003)
Annex 102. Report of the Special Rapporteur on the Situation of Human
Rights and Fundamental Freedoms of Indigenous People, Mr.
Rodolfo Stavenhagen, Mission to Colombia, U.N. Doc.
E/CN.4/2005/88/Add.2 (10 Nov. 2004)
Annex 103. Commission of European Communities, The Impact Assessment
of the Thematic Strategy on the Sustainable Use of Pesticides,
SEC(2006) 894 (12 July 2006)
Annex 104. Commission of European Communities, A Thematic Strategy on
the Sustainable Use of Pesticides, Technical Annex, SEC(2006)
895 Final (12 July 2006)
565Annex 105. Organization for Economic Co-Operation and Development
(OECD), Joint Meeting of the Chemicals Committee and The
Woring Party on Chemicals, Pesticides and Biotechnology,
Report of the OECD Pesticide Risk Reduction Steering Group:
The Second Risk Reduction Survey, ENV/JM/MONO (2006)14
(19 July 2006)
Annex 106. European Commission, EU Policy for a Sustainable Use of
Pesticides: The Story Behind the Strategy (2007)
Annex 107. Paul Hunt, UN Special Rapporteur on the Right to the Highest
Attainable Standard of Health, Closing Remarks to the Press,
Quito, Ecuador (18 May 2007)
Annex 108. 8QLWHG▯1DWLRQV▯2I¿FH▯RQ▯'UXJV▯DQG▯&ULPH▯▯Coca Cultivation in
the Andean Region, A Survey of Bolivia, Colombia, Ecuador and
Peru (June 2007)
Annex 109. European Parliament and the Council of the European Union,
Directive 2009/128/EC: Establishing a Framework for
Community Action to Achieve the Sustainable Use of Pesticides
(21 Oct. 2009)
Annex 110. 8QLWHG▯1DWLRQV▯2I¿FH▯RQ▯'UXJV▯DQG▯&ULPH▯▯World Drug Report
2010 (2010)
OTHER
Annex 111. Guillermo Fernández-Soto, La Ilusión Posible: Un Testimonio
Sobre La Política Exterior Colombiana (Grupo Editorial Norma,
2004)
Annex 112. Freedom of Information Act Correspondence With United States
Environmental Protection Agency Including Roundup Export
Label (3 Mar. 2009)
Annex 113 Letter from Rebecca L. Pusas, Counsel to Government
RI▯(FXDGRU▯▯WR▯8QLWHG▯6WDWHV▯'HSDUWPHQW▯RI▯6WDWH▯2I¿FH▯RI▯
Information Programs and Services (3 Apr. 2009)
566Annex 114 /HWWHU▯IURP▯0DUJDUHW▯3▯▯*UDIHOG▯▯'LUHFWRU▯▯2I¿FH▯RI▯,QIRUPDWLRQ▯
Programs and Services, United States Department of State to
Rebecca L. Pusas, Counsel to Government of Ecuador
(13 Nov. 2009)
Annex 115. Letter from Rebecca L. Pusas, Counsel to Government of
Ecuador, to A. Harold (Hal) Eisner, United States Department of
6WDWH▯2I¿FH▯RI▯,QIRUPDWLRQ▯3URJUDPV▯DQG▯6HUYLFHV▯
(19 Feb. 2010)
Annex 116. )D[▯IURP▯$▯▯+DUROG▯(LVQHU▯▯2I¿FH▯RI▯,QIRUPDWLRQ▯3URJUDPV▯
and Services, United States Department of State to Rebecca L.
Pusas, Counsel to Government of Ecuador
(12 Mar. 2010)
Annex 117. United States District Court of the District of Colombia, Arias,
et al. v. Dyncorp, et al., Quinteros, et al. v. Dyncorp, et al.,
Declaration of Redacted Witness (2 June 2010)
Annex 118. Freedom of Information Act Correspondence With United States
Environmental Protection Agency Including Email from Stephen
J. Wratten, Monsanto Company, to Jay Ellenberger, United
States Environmental Protection Agency (30 Oct. 2010)
VOLUME V
COLOMBIAN GOVERNMENT DOCUMENTS
Annex 119 Republic of Colombia, National Code of Renewable Natural
Resources and Environmental Protection, Decree 2811 of 1974
(18 Dec. 1974)
Annex 120 Republic of Colombia, Ministry of Agriculture, Study Proposal
for the Environmental Management of the National Park Areas
of Sierra Nevada de Santa Marta and Buffer Zones Affected by
Marijuana Crops and their Destruction by Aerial Spraying with
Glyphosate (31 July 1986)
567Annex 121 Republic of Colombia, Ministry of Agriculture, Technical
Commission, 6SHFL¿FDWLRQ▯RI▯WKH▯7HUPV▯RI▯5HIHUHQFH▯IRU▯
Environmental Research in the Sierra Nevada of Santa Marta
Affected by Marijuana Crops and Spraying with Glyphosate
(1986)
Annex 122 Republic of Colombia, Colombian Agriculture and Livestoc
Institute, ICA Concepts Regarding A Report Issued by the
Environmental Audit Techeca Ltda. (1994)
Annex 123 Letter from Cecilia Lopez Montano, Minister of Environment,
Republic of Colombia, to Nestor Humberto Martinez Neira,
Minister of Justice and Law, Republic of Colombia
(20 Dec. 1994)
Annex 124 5HSXEOLF▯RI▯&RORPELD▯▯2I¿FH▯RI▯WKH▯2PEXGVPDQ▯▯0HHWLQJ▯
Minutes (17 Feb. 1995)
Annex 125 Republic of Colombia, National Congress, Law 191 of 1995
(23 June 1995)
Annex 126 Republic of Colombia, Ministry of Environment, Legal
Department, Order No. 558A (13 Aug. 1996)
Annex 127 Republic of Colombia, Ministry of Environment,
Order No. 557A (13 Aug. 1996)
Annex 128 Republic of Colombia, Constitutional Court, Triviño et al.,
Judgment SU-039/97 (3 Feb. 1997)
Annex 129 Letter from Guillermo Acevedo Mantilla, Subdirector of
Environmental Licenses, Ministry of Environment, Republic
of Colombia, to Ivon Alcala Arevalo, Director, National Drug
Directorate, Republic of Colombia (8 Oct. 1998)
Annex 130 Letter from Guillermo Acevedo Mantilla, Subdirector of
Environmental Licenses, Ministry of Environment, Republic
of Colombia, to Ruben Olarte Reyes, Director, National Drug
Directorate, Republic of Colombia (13 Nov. 1998)
Annex 131 Republic of Colombia, Ministry of Environment, Division of
568 Licenses, Technical Report No. 419.99 (21 Dec. 1999)
Annex 132 Republic of Colombia, Ministry of Environment, Division of
Environmental Licenses, Order No. 599 (23 Dec. 1999)
Annex 133 Republic of Colombia, Ministry of Environment, Division of
Environmental Licenses, Order No. 143 (29 Mar. 2000)
Annex 134 Letter from Medardo Galindo Hernandez, Ombudsman,
Republic of Colombia, to Juan Mayr Maldonado, Minister of the
Environment, Republic of Colombia (24 July 2000)
Annex 135 Republic of Colombia, National Narcotics Council, Resolution
No. 005 (11 Aug. 2000)
Annex 136 Republic of Colombia, Ministry of Environment, Division of
Environmental Licenses, Technical Report No. 589, Evaluation
of the Additional Information Provided by the National
Narcotics Directorate (20 Dec. 2000)
Annex 137 Comptroller General of the Republic of Colombia, Appointed
Comptroller for the Environment, Special Audit of the Policy for
Eradication of Illicit Crops (July 2001)
Annex 138 Republic of Colombia, Advisor on the Plan Colombia Illicit
Crop Eradication Program, Certain Toxicological and Technical
Considerations For Aerial Spraying With Glyphosate on Illicit
Crops, Bogotá, Colombia (9 July 2001)
Annex 139 Republic of Colombia, Ministry of Environment, Resolution No.
1066 (26 Nov. 2001)
Annex 140 Letter from Juan Mayr Maldonado, Minister of the
Environment, Republic of Colombia, to Gabriel Merchan
Benevides, Director General of the National Drug Directorate,
Republic of Colombia (2002)
Annex 141 Republic of Colombia, Ministry of Environment, Resolution No.
108 (31 Jan. 2002)
Annex 142 Republic of Colombia, Constitutional Court, Urueta Rojas.,
Judgment C-418/02 (28 May 2002)
569Annex 143 Comptroller General of the Republic of Colombia, Plan
Colombia: Third Evaluation Report (Aug. 2002)
Annex 144 Letter from Francisco Santander Delgado, Director General,
Corponariño, Republic of Colombia, to Maria Cecilia
Rodriguez, Minister of the Environment, Republic of Colombia
(26 Sept. 2002)
Annex 145 5HSXEOLF▯RI▯&RORPELD▯▯2I¿FH▯RI▯WKH▯2PEXGVPDQ▯▯National
Ombudsman Resolution No. 26, Human Rights and International
+XPDQLWDULDQ▯/DZ▯LQ▯WKH▯&RQWH[W▯RI▯$UPHG▯&RQÀLFW▯DQG▯
Fumigation of the Coca Crops in the Province of Putumayo (9
Oct. 2002)
Annex 146 5HSXEOLF▯RI▯&RORPELD▯▯2I¿FH▯RI▯2PEXGVPDQ▯▯The
Implementation of the Strategy of Aerial Eradication of Illicit
Crops With Chemicals, From a Constitutional Perspective (April
2003)
Annex 147 Republic of Colombia, Administrative Tribunal of
Cundinamarca, Claudia Sampedro and Others, Judgment (13
June 2003)
Annex 148 Republic of Colombia, Ministry of Environment, Resolution No.
0670, Whereby a sanction is imposed and other decisions are
made (19 June 2003)
Annex 149 Government of Colombia, Ministry of Agriculture, Resolution
No. 03759 (16 Dec. 2003)
Annex 150 Note SARE-142, sent from the National Directorate of
Narcotics of the Ministry of Interior and Justice of Colombia to
WKH▯3UHVLGHQW▯RI▯WKH▯7HFKQLFDO▯6FLHQWL¿F▯&RPPLVVLRQ▯RI▯(FXDGRU▯
(14 Apr. 2004)
Annex 151 State Council of Colombia, Claudia Sampedro and Others,
Judgment on Appeal From the Administrative Tribunal of
Cundinamarca (19 Oct. 2004)
Annex 152 Comptroller General of the Republic of Colombia, Plan
Colombia: Fifth Evaluation Report (Dec. 2004)
570Annex 153 Republic of Colombia, Constitutional Court, Molina, Judgment
C-401/05 (14 Apr. 2005)
Annex 154 Diplomatic Note No. DDH 58003 from the Colombian Foreign
Ministry to the Executive Secretary of the Inter-American
Commission on Human Rights (18 Sept. 2005)
Annex 155 Republic of Colombia, Position Statement by Colombia to the
%LQDWLRQDO▯6FLHQWL¿F▯DQG▯7HFKQLFDO▯&RPPLVVLRQ▯LQ▯5HODWLRQ▯
to the Destruction in Colombia of Illicit Crops in the Frontier
Zones with Ecuador (8 June 2007)
Annex 156 Republic of Colombia, Ministry of Foreign Affairs, Press
Release (11 Nov. 2010)
571
Reply of Ecuador