Counter-Memorial of Colombia

Document Number
17548
Document Type
Date of the Document
Document File
Document

INTERNATIONAL COURT OF JUSTICE
____________________________________________

CASE CONCERNING
AERIAL HERBICIDE SPRAYING
(ECUADOR v. COLOMBIA)

COUNTER-MEMORIAL OF THE
REPUBLIC OF COLOMBIA

VOLUME I

29 MARCH 2010II TABLE OF CONTENTS

Chapter 1. ECUADOR’S CASE: AN OVERVIEW

A. The Court’s Jurisdiction over the Dispute and the Admissibility
of Ecuador’s Claims........................................................................
....................1
(1) J URISDICTION UNDER THE PACT OF B OGOTÁ ........................................... 2
(2) J URISDICTION UNDER THE 1988C ONVENTION ........................................3
(3) I NADMISSIBILITY OF E CUADOR ’S CLAIMS ON BEHALF OF

C OLOMBIAN NATIONALS .......................................................................
.. 8

B. Ecuador’s Case as Disclosed by the Memorial ................................................16

C. Colombia’s Response on the Merits ................................................................20

D. The Structure of this Counter-Memorial ..................................................... 34

PART I – THE FACTUAL MATRIX

Chapter 2. THE COLOMBIA-ECUADOR BORDER REGION

A. The Sectors of the Border Region .............................................................. 39
(1) T HE P ACIFIC SECTOR .......................................................................
...... 40

(2) T HE A NDEAN SECTOR .......................................................................
....42
(3) T HE A MAZONIAN SECTOR .....................................................................42

B. Social Conditions in the Provinces of Sucumbíos and Esmeraldas .................45
(1) B ASIC PROBLEMS .......................................................................
...........45
(2) P ROVINCE OF SUCUMBÍOS ..................................................................... 53

(3) P ROVINCE OF ESMERALDAS ................................................................... 55
(4) C OLOMBIAN COOPERATION WITH ECUADOR .......................................... 57

C. Conclusi......................................................................
.........................5

III Chapter 3. THE FIGHT AGAINST DRUGS IN COLOMBIA

A. An Overview of the Drug Problem in Colombia..............................................61

B. Political and Social Impacts of Illicit Crops and Drug Trafficking in

Colombia ....................................................................
....................................... 65

C. Economic Impacts of Illicit Crops and Drug Trafficking in Colombia............68

D. Environmental Impact of Illicit Crops and Drug Trafficking in

Colombia ....................................................................
....................................... 70
(1) D ESCRIPTION OF RESOURCES.................................................................70
(2) L OCATION OF ILLICIT CROP..................................................................74
(3) T HE COCAINE PRODUCTION CHAIN AND ITS ENVIRONMENTAL

EFFECTS .......................................................................
........................... 76

E. International Support for the Fight against Drugs in Colombia .......................83
(1) U NITED STATES’SUPPORT AND PLAN C OLOMBIA .................................. 84
(2) S UPPORT FROM THE EUROPEAN U NION AND INDIVIDUAL

COUNTRIES FORPLAN C OLOMBIA .......................................................... 85
(3) S UPPORT FROM THE UNITED NATIONS :LLICITCROPS
I NTEGRATED M ONITORING SYSTEM (SIMCI) ........................................ 86

(4) S UPPORT FROM THE ORGANIZATION OF AMERICAN STATES:
ITER -AMERICAN D RUG ABUSE C ONTROL COMMISSION
(CICAD) ..............................................................
................................ 88

F. Colombia’s Obligation to Take Action against Illicit Crops............................94

G. Conclusions ............................................................
........................................... 98

Chapter 4. THE PROGRAM FOR THE ERADICATION OF ILLICIT

CROPS BY AERIAL SPRAYING WITH GLYPHOSATE
) G I C E P (

A. The Need for Aerial Spraying........................................................................
.103

B. Scientific Assessments Supporting the Aerial Spraying Program..................107
(1) S TUDIES LEADING TO THEENVIRONMENTAL
M ANAGEMENT PLAN .......................................................................
....107
(2) S UBSEQUENT STUDIES:T HECICAD PROCESS .....................................110

IV (3) C ONCLUSIONS .......................................................................
............... 112

C. Implementation of the PECIG Program..........................................................112
(1) T HE SUPERVISORY FRAMEWORK – ENVIRONMENTAL M ANAGEMENT
PLAN .......................................................................
............................. 113
(2) O VERALL SUCCESS OF THE PROGRAM ..................................................117

(3) A LTERNATIVES TO AERIAL SPRAYING .................................................. 119

D. Technical Aspects of Aerial Spraying............................................................121
(1) T HE SPRAYING MIXTURE ......................................................................
122
(2) P ROCEDURES FOLLOWED IN THE PECIG PROGRAM .............................131

(3) M INIMIZING DRIFT .......................................................................
........136

E. Conclus.i...s..................................................................
.......................1

Chapter 5. THE DISPUTE WITH ECUADOR

A. Overview of Diplomatic and Other Exchanges 2000 to 2004........................143

B. Ecuador Confirmed in December 2004 that There Were No Adverse Effects
Due to the Sprayings........................................................................
...............167

C. The Position of the Parties as from 2005 to 2008...........................................180

D. Conclusions ............................................................
......................................... 203

Chapter 6. CERTAIN DISPUTED QUESTIONS OF FACT

A. Introduc.t....................................................................
.......................2

B. Alleged Failure to Disclose the Formula........................................................210
(1) P UBLICLY AVAILABLE COLOMBIAN DOCUMENTS ................................. 211
(2) C OMMUNICATIONS BY C OLOMBIA TO ECUADOR

(2000-2004)...........................................................
.............................. 214
(3) C OLOMBIAN ENVIRONMENTAL A UDIT ................................................. 219

VC. Alleged Failure to Deliver an Environmental Impact

Assessment ....................................................................
.................................. 220

D. Alleged Failure to Notify Ecuador of Spraying Missions ..............................224

E. Conclusions........................................................................
.............................226

PART II – SCIENTIFIC AND LEGAL ISSUES

Chapter 7. THE DAMAGE ALLEGED BY ECUADOR

A. Overv.i......................................................................
...........................2

B. The Factual Prerequisites for Ecuador’s Case................................................234
(1) T OXICITY OF THE SPRAY MIXTURE ....................................................... 235
(2) S PRAY DRIFT .......................................................................
................. 240
(3) E XPOSURE OF E CUADORIAN RESIDENTS ............................................... 249

(4) E XISTENCE OF A CAUSAL LINK ............................................................250
(5) C ONCLUSION .......................................................................
................. 254

C. The Scientific Evidence Concerning Glyphosate-Based Sprays....................254
(1) A LLEGED EFFECTS ON HUMAN HEALTH ................................................ 255

(2) A LLEGED EFFECTS ON FAUNA ..............................................................270
(3) A LLEGED EFFECTS ON SOIL AND IMPACTS ON NON -TARGET CROPS ......277
(4) A LLEGED EFFECTS ON WATER RESOURCES ........................................... 280
(5) A LLEGED EFFECTS ON AIR QUALITY ..................................................... 283
(6) E CUADOR ’S SCIENTIFIC EVIDENCE :THE M ENZIE REPORT .................... 283

(7) A N EXPERT EVALUATION OF THE M ENZIE R EPORT ..............................288
(8) C ONCLUSION FROM THE SCIENTIFIC EVIDENCE ....................................289

D. The Categories of Injury Alleged by Ecuador................................................290
(1) A LLEGED INJURIES TO INDIVIDUAL LIFE AND HEALTH .......................... 290

(2) A NIMALS ........................................................................
.....................319
(3) L AWFUL CROPS .......................................................................
............. 323
(4) N ATIVE FLORA AND THE ENVIRONMENT ..............................................326
(5) T HE “SPECIAL HARM ” TO THE INDIGENOUS COMMUNITIES ................... 333

E. Conclus.i....s..................................................................
.......................3
VIChapter 8. ECUADOR’S CLAIMS FOR TRANSBOUNDARY
INJURY OR HARM

A. Introduc.t....................................................................
.......................3

B. The Applicable Law........................................................................
................343
(1) T REATY AND CUSTOM IN INTERNATIONAL ENVIRONMENTAL
LAW .......................................................................
..............................343
(2) T HE SIGNIFICANCE OF THE 1988N ARCOTICS CONVENTION ................346
(3) T HE 2001 ILC DRAFT ARTICLES ON PREVENTION OF

TRANSBOUNDARY HARM FROM HAZARDOUS ACTIVITIES ...................... 353

C. Ecuador’s Claim of Breach of Sovereignty....................................................360
(1) T RANSBOUNDARY EFFECTS AS A PER SE BREACH .................................360
(2) O THER ALLEGED BREACHES OF SOVEREIGNTY ..................................... 365

D. Ecuador’s Claim concerning Non-Prevention of Transboundary Harm.........366
(1) P REVENTION OF TRANSBOUNDARY HARM ............................................ 367
(2) O BLIGATIONS OF COOPERATION ........................................................... 377
(3) B REACH OF ARTICLE 14(2) OF THE 1988N ARCOTICS
C ONVENTION .......................................................................
...............409

E. Conclus.i...s..................................................................
.......................4

Chapter 9. ECUADOR’S CLAIMS BASED ON HUMAN AND

INDIGENOUS RIGHTS

A. Introduc.t....................................................................
.......................4

B. Alleged Breaches of Human Rights................................................................418
(1) R ELATIONSHIP OF HUMAN RIGHTS AND ENVIRONMENTAL
CLAIMS .......................................................................
......................... 418

(2) A PPLICABILITY OF HUMAN RIGHTS OBLIGATIONS TO
TRANSBOUNDARY CLAIMS ................................................................... 421
(3) H UMAN RIGHTS OBLIGATIONS AND ARTICLE 14(2)OF THE 1988
N ARCOTICS C ONVENTION ................................................................... 444

(4) T HE SCOPE OF HUMAN RIGHTS ALLEGEDLY BREACHED ....................... 446
VIIC. Alleged Breach of Indigenous Rights.............................................................492
(1) E CUADOR ’S RELIANCE UPON INDIGENOUS RIGHTS ..............................492

(2) C OLOMBIA ’S RESPONSE ON THE FACTS ................................................493
(3) A LLEGED BREACH OF INDIGENOUS TREATY RIGHTS ............................. 494

D. Conclusi.o ..............................................................................................5

Chapter 10. THE REMEDIAL SITUATION

A. Ecuador’s Proleptic Approach to Injury.........................................................505

B. Ecuador’s Attempt to Restrict Colombia’s Sovereignty over its
Own Territory........................................................................
.........................510

C. Conclus.i.o..............................................................................................5

SUBMISSIONS .......................................................................
.................................. 513

APPENDIX 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. (2009) [EM, Vol. III]
Annex 158........................................................................
...........................................515

LIST OF ANNEXES .......................................................................
.......................... 553

LIST OF DOCUMENTS DEPOSITED WITH THE REGISTRY ...................... 573

VIII COUNTER-MEMORIAL OF THE

REPUBLIC OF COLOMBIA

Chapter 1

ECUADOR’S CASE: AN OVERVIEW

1.1 This Memorial is filed in accordance with the Court’s

Order of 30 May 2008 and responds to the Memorial submitted
by Ecuador on 28 April 2009.

A. The Court’s Jurisdiction over the Dispute and the
Admissibility of Ecuador’s Claims

1.2 The present dispute concerns Ecuador’s objections to

Colombia’s programme of aerial spraying of illicit coca crops,

and the supposed transboundary impact of that program on

Ecuador, its residents and on the environment.

1.3 In its Memorial, Ecuador seek s to found the jurisdiction

of the Court over the present dispute upon two bases, namely:

(a) Article XXXI of the American Treaty on Pacific
Settlement (“the Pact of Bogotá”); and

(b) Article 32 of the United Nations Convention

Against Illicit Traffic in Narcotic Drugs and

1 EM, paras. 4.1, 4.3.
2 EM, Vol. II, Annex 1.
1 Psychotropic Substances 1988 (“the 1988 UN

Narcotics Convention”). 3

(1) JRISDICTION UNDER THE P ACT OF B OGOTÁ

1.4 As to the Pact of Bogotá, Colombia has been a party

since 1968. For its part, Ecuador signed it on 30 April 1948,
subject to the following reservation:

“The Delegation of Ecuador, upon signing this
Pact, makes an express reservation with regard to
Article VI and also every provision that
contradicts or is not in harmony with the
principles proclaimed by or the stipulations

contained in the Charter of the United Nations,
the Charter of the Orga nization of American
States, or the Constitu tion of the Republic of
Ecuador.”

Article VI of the Pact provides that the obligation to settle

disputes through access to the Court:

“may not be applied to matters already settled by
arrangement between the pa rties, or by arbitral
award or by decision of an international court, or
which are governed by agreements or treaties in
force on the date of the conclusion of the present

Treaty.”

1.5 The effect of Ecuador’s reservation upon signature
would have been to expand th e Court’s jurisdiction to the

matters deliberately excluded by Article VI, including matters

“already governed by agreements or treaties in force”. But

3 1582 UNTS 164; E/Conf. 82/15; EM, vol. II, Annex 3.
2(quite apart from other problem s with Ecuador’s reservation,

including the reference to its Constitution as overriding the

Pact), a State cannot unilaterally expand the scope of another

State’s consent to jurisdicti on. Thus the validity of the

reservation was doubtful.

1.6 In the event, Ecuador waited almost 50 years to ratify the

Pact, and when it did so, by notification of 7 May 2008, it paid

no attention to the reservation it had already made. In its

Memorial, although referring to the fact that a reservation was
made upon signature, Ecuador simply states that the reservation

is “of no relevance to this case: 4in particular it does not resile

from the reservation. In the circumstances that occurred, the

requirement expressed in Ar ticle 23(2) of the Vienna

Convention on the Law of Treatie s has not been satisfied, and
the reservation not having been “formally confirmed by the

reserving State” must be regarded as having been withdrawn.

Colombia calls on Ecuador to confirm that this is indeed the

case.

(2) JURISDICTION UNDER THE 1988 C ONVENTION

1.7 Ecuador asserts that the 1988 Convention “impos[es]

obligations in relation to the respect for sovereignty and

territorial integrity, protection of the environment and respect

4 EM, para. 4.5.
3 5
for fundamental human rights,” and that the consequence of

Articles 2 and 14 of the 1988 Convention is that the Convention
“imposes obligations that cover the entire subject matter of the

dispute that is before the Court”. 6 In this context it refers to:

(1) Article 2 of the 1988 Convention, in particular
Article 2(2), which 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 States”; and

(2) Article 14(2), which imposes an obligation upon
the Parties to take measures to “prevent illicit

cultivation of and to erad icate plants containing
narcotic or psychotropic substances”, and goes
on to provide that “The measures adopted shall
respect fundamental human rights and shall take

due account of traditional licit uses, where there
is historic evidence of such use, as well as the
protection of the environment.” 8

1.8 Elsewhere in the Memorial, Ecuador argues:

(1) “By allowing these deposits caused by aerial
spraying in border areas, Colombia has violated
its international obligations to respect the
territorial sovereignt y of Ecuador. These

obligations arise under gene ral international law.
They also arise by operation of specific treaties,

5
6 EM, para. 4.18.
7 EM, para. 4.18.
8 EM, para. 4.19.
EM, para. 4.20.
4 including in particular [Article 2 of the 1988
Narcotics Convention]” 9

(2) that the obligation to prevent transboundary harm

is “imported directly into these proceeding10 by
Article 14 of the 1988 Convention, and that
“Colombia has undertaken aerial spraying […] in

a manner which fails to respect fundamental
human rights and prot ect the environment,
thereby violating, inter alia, Article 14(2) of the
1988 Narcotics Convention […]” 11

(3) that “there is a continuing and necessarily

evolving duty to protect Ecuador from
environmental harm in accordance with the terms
of Article 14(2)” of the 1988 Convention; 12

(4) that Article 14(2) of the 1988 Convention “is
intended to incorporate the relevant requirements

of international environmental law […] [and]
with respect to human rights”; 13

(5) that the alleged failure by Colombia “to protect
the environment of Ecuador from the effects of

its aerial spraying activities” “represents a breach
of the 1988 Narcotics Convention” and of
Colombia’s “obligation under Article 14(2) to
14
respect fundamental human rights”;
(6) that “Colombia has undertaken aerial spraying in

a manner that violates Ar ticle 14(2) of the 1988
Narcotics Convention”; 15

9
EM, para. 7.1. However, later in Chapter 7, Ecuador does not
explicitly plead breach of Article 2 of th e 1988 Convention, see in particular
10, para. 7.22-7.23; although cf. EM, para. 1.19.
11 EM, para. 8.1.
EM, para. 8.2.
12 EM, para. 8.74.
13 EM, para. 8.73.
14 EM, para. 8.79.
15 EM, para. 8.81; see also ibid., para. 8.84, alleging that
“Colombia did not cooperate as required by international law, nor did it take
measures to guarantee respect for fundamental human rights or protection of

5 (7) that obligations to protect human rights are
“imported directly into the present dispute by the
1988 Narcotics Convention”, 16 and that, by
reason of the aerial spraying, not only is

Colombia alleged to have violated a number of
obligations owed to Ecuador under international
human rights treaties “but additionally or

alternatively th171988 Narcotics Convention,
Article 14(2)”; and

(8) that Colombia has “violated the 1988 UN
Narcotic Drugs Convention, insofar as human
rights obligations are in corporated thereunder by
virtue of Article 14(2)”. 18

1.9 Ecuador’s arguments in this regard suggest that the 1988

Convention imposes certain obligati ons in relation to territorial

sovereignty, transboundary harm and fundamental human rights.
As will be demonstrated in later Chapters, these arguments

cannot be accepted. 19 On its true inte rpretation, the 1988

Convention merely makes refere nce to those concepts in

defining and limiting the substantiv e obligations of the States

Parties under the 1988 Convention, in particular the obligation

to take measures to prevent illi cit cultivation of and to eradicate

plants containing narcotic or psychotropic substances under

Article 14.

the environment as required by Article 14(2) of the 1988 Narcotics
16nvention.”
17 EM, para. 9.11.
18 EM, para. 9.11
19 EM, para. 9.109.
See Chapters 8 and 9 below.
61.10 Thus, the 1988 Convention is nonetheless central to the

dispute, since it actually requi res Colombia, by means of its
choosing, to engage in the campaign against illicit drugs,

including the eradic ation program, and i ndeed calls on other

States to cooperate with Colombia in that regard.

1.11 Article 14 is entitled “Meas ures to Eradicate Illicit
Cultivation of Narcotic Plants and to Eliminate Illicit Demand

for Narcotic Drugs and Psychotropic Substances”. Article 14(1)

preserves the rights and obligations of parties to the earlier drugs
conventions of 1961 and 1971: the 1988 Convention is to be “no

less stringent” than these earlier conventions. Article 14(2) then

provides:

“Each Party shall take appropriate measures to
prevent illicit cultivation of and to eradicate
plants containing narcotic or psychotropic
substances, such as opium poppy, coca bush and

cannabis plants, cultivated illicitly in its territory.
The measures adopted shall respect fundamental
human rights and shall take due account of
traditional licit uses, where there is historic
evidence of such use, as well as the protection of
the environment.”

1.12 On its face, Article 14(2) does not impose independent

obligations as regards fundamental human rights or protection
of the environment. Rather, it is concerned with imposing an

obligation upon States to adopt measures to prevent illicit

cultivation of plants contai ning narcotic or psychotropic

7substances, subject to the qualification that such measures
respect or take into account the considerations mentioned.

1.13 It is Colombia’s position that the aerial spraying

program, far from being in breach of Article 14(2), is actually an

implementation of it. This has material consequences for the

case, in that if this is so, Ecuador itself was under an obligation,

under Article 14(3)(c), to coope rate in its implementation,
whereas in fact it has done nothing of the sort. But at the level

of jurisdiction, there is plainly a dispute between the parties as

to the interpretation and application of the 1988 Convention.

(3) INADMISSIBILITY OF ECUADOR ’S CL20MS ON BEHALF OF
C OLOMBIAN NATIONALS

1.14 In Chapter 9 of its Memorial , Ecuador alleges breaches

of a whole catalogue of interna tional human rights obligations

(including the rights to life, health, food, water, a healthy

environment, private life, propert y and information) as well as

alleging violation of the rights of indigenous peoples.

1.15 In setting out those allegations and its arguments as to its

claim for compensation, Ecuador’s Memorial is fundamentally

unclear as to the group or groups of persons to which the claims

20
Issues of admissibility can be raised at the merits phase: Avena and
Other Mexican Nationals (Mexico v United States of America), ICJ Reports
2004, p. 28, para. 24. See also Application of the Convention on the
Prevention and Punishment of the Crime of Genocide (Bosnia and
Herzegovina v. Serbia and Montenegro), Judgment of 26 February 2007,
para. 101.
8relate and in relation to which declarations of violation or

compensation is sought. In partic ular, it is not clear whether it

purports to bring claims only on behalf of its own nationals.

Clarity as to the scope of Ecuador’s claims in this regard is not

assisted by its tactic of requesting the Court first to declare a

violation of the relevant obligations while reserving the right to

provide proof of the damage alle gedly suffered in a later phase

of the proceedings. 21

1.16 In setting out its claims of violation of human rights and

the rights of indigenous peoples in Chapter 9 of the Memorial,

Ecuador utilizes a variety of formulations to refer to the

individuals in relation to whom the various violations are

alleged to have occurred, including:

• “the indigenous Awá, Cofán and Kichwa peoples and of
22
the Afro-Ecuadorian communities in Esmeraldas”; “the
23
local indigenous populations”; “indigenous peoples”,

“the affected indigenous peoples”, “indigenous

communities” and “indigenous communities in the
24 25
region”; “indigenous peoples [...] in border areas”;

“indigenous communities who live, farm and hunt in the
26
affected areas” ; and “indigenous communities,

21
22 EM, paras. 9.110 and 10.58.
EM, para. 9.22.
23 EM, para. 9.28.
24 EM, para. 9.29(a)-(e).
25 EM, para. 9.36.
26 EM, para. 9.37.

9 27
including those in Ecuador” (rights of indigenous

peoples);

• “indigenous communities in the areas affected by the
28
sprayings” and “those living in affected border areas of

Ecuador” (right to life);

30
• “local communities in Ecuador”; “the affected

populations in Ecuador”; 31“indigenous peoples living

along the border” and “farming communities in the

region”; and “the Ecuadorian population along the

33
border”; (right to health);

• “communities in the areas affected by the aerial

34 35
spraying”; and “local communities”; “local
36
communities in Ecuador”; and “farmers and indigenous
37
peoples” (right to food);

• “[l]ocal communities in the border area between
38
Colombia and Ecuador”, and “affected farmers and

indigenous peoples [...] the most vulnerable populations

27
28 EM, para. 9.38.
EM, para. 9.47.
29 EM, para. 9.48.
30 EM, para. 9.51.
31 EM, para. 9.51.
32
33 EM, para. 9.55.
34 EM, para. 9.56.
EM, para. 9.60.
35 EM, para. 9.61.
36 EM, para. 9.64.
37 EM, para. 9.65.
38
EM, para. 9.70.
10 39
living along the Ecuador-Colombia border” (right to

water);

40
• “the concerned Ecuadorian population” and “farming
41
and indigenous communiti es in the border area”; and

“peoples’” (right to a healthy environment);

43
• “people living in the border area with Colombia”; and

“farmers” (right to property);

45
• “the Ecuadorian population in the border region”,

(right to humane treatment);

46
• “local communities” in the border area; and “local

communities”, “farmers and indigenous peoples” 47 (right

to private life);

48
• “the local population in the affected areas”; “those

likely to be affected by the aerial sprayings”; 49

“inhabitants in the border area” and “the population in

the border area” 50 and “those likely to be affected when

39
40 EM, para. 9.74
EM, para. 9.83.
41 EM, para. 9.83.
42 EM, para. 9.84.
43 EM, para. 9.86.
44
45 EM, para. 9.86.
46 EM, para. 9.89.
EM, para. 9.95.
47 EM, para. 9.99.
48 EM, para. 9.103.
49 EM, para. 9.104.
50
EM, para. 9.106.
11 spraying was due to take place” 51(right to information);

and

52
• “local inhabitants in Ecuadorian territory”, “the
53
affected populations”, and “indigenous peoples”.

1.17 As regards Ecuador’s discussion of remedies, in Chapter
10 of its Memorial, Ecuador alleges in general terms that:

“Colombia’s actions have caused grave,
continuing and long-lasting harms to Ecuador: to

its sovereignty, to its people and property,
including indigenous peoples , and to its
environment”. 54

1.18 Ecuador goes on to make clear that it seeks declarations

of Colombia’s alleged breaches of international law, cessation

of those alleged breaches, assurances and guarantees of non-

repetition, as well as compensation in relation to certain of the

breaches.

1.19 At the outset of its discussion of the harm allegedly

suffered, Ecuador alleges that the spraying has “caused damage

and injury to human health, including illness and death among

the people who inhabit the border region ”, alleges that “the

health of people in the border region has been adversely

51 EM, para. 9.107.
52 EM, para. 9.108.
53 EM, para. 9.108.
54 EM, para. 10.2 (emphasis added).
55 EM, para. 10.2.
56 EM, para. 10.9 (emphasis added).

12affected by pollution of freshwat er supplies used for drinking,

cooking and bathing” 57and makes reference to the “crops upon

which indigenous and local communities in Ecuador’s border
58
region depend”.

1.20 Chapter 10 of Ecuador’s Memorial is contradictory as to

the precise scope of Ecuador’s claim for compensation as

regards death or injury to hea lth. One paragraph suggests that

the claims in this regard should be understood as relating to

damage suffered by “local communities and indigenous people

in the border region of Ecuador affected by the aerial herbicide

spraying”, a position which would appear to cover literally “any
59
person or persons”. However, a few paragraphs later, Ecuador

specifies that it seeks compensation “in respect of the loss of life

and damage to the health of its nationals caused by the herbicide
60
spraying”.

1.21 Things are no clearer as regards the claim for

compensation in relation to “loss of or damage to the property or
livelihood or human rights of su ch persons”. Ecuador alleges

that Colombia’s aerial spraying has “damaged property of local

and indigenous communities in the border region, and adversely

affected their livelihoods”, 61 that the sprayi ng has caused crop

57 EM, para. 10.9 (emphasis added).
58 EM, para. 10.9 (emphasis added).
59 EM, para. 10.30 (emphasis added); and see also EM, para. 10.29.
60 EM, para. 10.34 (emphasis added).
61 EM, para. 10.35 (emphasis added).

13damage which has had a “significant adverse impact on the

livelihoods and food security of the local people,” and that, as

a consequence, it is “entitled to full compensation in respect of

this loss of or damage to property, including crops and domestic

animals, and the costs imposed by the displacement of farmers
63
and affected other individuals”.

1.22 Ecuador’s Final Submissions to the Court as regards

compensation go substantially further than the arguments

developed in the body of its Memorial. Paragraph (B) of the

Submissions requests the Court to order that

“Colombia shall indemnify Ecuador for any loss
or damage caused by its in ternationally 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 sp ecial rights of
indigenous peoples;

(v) environmental damage or the
depletion of natural resources;

(vi) the costs of monitoring to identify
and assess future risks to public health,

62 EM, para. 10.36 (emphasis added).
63 EM, para. 10.37 (emphasis added).
14 human rights and the environment
resulting from Colombia’s use of
herbicides; and
64
(vii) any other loss or damage;”.

1.23 Although the terms of sub-paragraphs (i), (ii), (v) and

(vii) track the terms of Ecuador’s development of its arguments

as to compensation contained in Chapter 10 of the Memorial, set
out above, sub-paragraphs (iii) and (iv) (relating, respectively, to

the alleged violations of human rights and alleged violations of

the rights of indigenous peoples) ar e not mentioned at all in the

context of compensation in Chapter 10. Ecuador merely

reserves its right to supplement the heads under which it claims
65
compensation at a later date.

1.24 Thus the way in which Ecuador’s claims are put is far

from consistent. Although certain passages of the Memorial are

phrased extremely broadly, and mi ght be read as implying that

Ecuador is bringing claims in respect of the harm allegedly

suffered by all individuals in the border region alleged to have

been affected by the aerial sprayi ng, whether in Ecuador or in

Colombia, other passages (and in particular its claim for relief)
imply that Ecuador is bringing cl aims on behalf of, and seeks

compensation in respect of, allege d injury to all persons on its

64 EM, p. 412 (emphasis added).
65 EM, para. 10.58. See also para. 9.110.
15territory, 66including Colombian nationa ls and any nationals of

third States. In fact, Ecuador has filed a num ber of witness

statements made by Colombians living in the border area, 67

Although elsewhere in the Memo rial as a general matter it

disclaims any intention “to press claims on behalf of the people
68
of Colombia”, it appears to be pre ssing claims on behalf of

some individual Colombians.

1.25 There is no basis for it to do so. It is well-established

that a State cannot bring clai ms for compensation against

another State in respect of nationa ls of the latter State. That
69
rule, articulated in the Mavrommatis case, is confirmed by

modern practice as well. The ILC Articles on Diplomatic

Protection sought to develop certain exceptions to the

nationality of claims, but in all circumstances retained the rule
70
that no claim could be brought against the State of nationality.

B. Ecuador’s Case as Disclosed by the Memorial

1.26 In its Memorial of 28 Apr il 2009, Ecuador argues that

Colombia’s aerial spraying program has “caused extensive,

66
See e.g. EM, para. 5.3 “Ecuador’s claims are based solely on harms
caused by Colombia’s aerial spraying programme within Ecuador, namely to
people, plants, animals and the natural environment on the Ecuadorian side
67 the common border” (emphasis added).
EM, vol. IV, Annexes 225-233.
68 EM, para. 5.101; although cf. the more ambiguous terms of EM,
para. 5.3.
69 Mavrommatis Palestine Concessions (Greece v. U.K. ) P.C.I.J.
Reports, 1924, Series A, No. 2, p. 12.
70 See ILC Articles on Diplomatic Protection, arts 5(3), 7, 8(3) &
commentary thereto, in ILC Report 2006, A/61/10, ch. IV.

16long-lasting and widespread harm to Ecuador, to its people and
71
to its environment”. Ecuador’s case is predicated on a

conclusion of fact, expressed in such phrases as: “the toxic

mixture used in the aerial fumi gations has polluted water, killed

wild and domestic animals, destroyed forest and decimated
72
crops”. “Colombia’s actions have had a particularly

devastating impact on Ecuador’s rich, protected environment, on

plants, animals and wildlife, as well as on the communities that
are dependent on the long term we ll-being of that environment.

… The fragile equilibrium prevailing between these

communities and their environment, which is a constitutive part

of their specific culture, has b een severely endangered and, in

some cases, destroyed; this has forced indigenous and other

local residents to abandon th eir areas of settlement.” 73 “The

spraying of toxic chemicals on the border area has severely

disrupted the lives of local communities over many years, to the
74
point that their lives have been transformed.” “Colombia’s

aerial spraying of herbicides has destroyed or damaged

thousands of hectares of valu able crops in Ecuador, with

particularly devastating effect s on short-cycle crops and the

subsistence crops upon which indigenous and local communities
75
in Ecuador’s border region depend.” Overall, “the harms

occasioned by Colombia’s fumigations constitute an integrated,

71 EM, para. 1.5.
72 EM, para. 9.83.
73 EM, para. 1.10.
74
75 EM, para. 9.95.
EM, para. 10.9.
17mutually reinforcing whole that have undone the very fabric of
76
life in the border region”.

1.27 According to Ecuador, this is above all, then, a case

about allegations of very serious transboundary harm allegedly
caused by conduct of Colombia through a program ostensibly

designed to combat the production process of cocaine, an illicit

drug, on Colombian territory.

1.28 The catalogue of alleged damage is a long one, but it is

sufficient to focus on deaths and long-term injury as set out in

the fact witness statements anne xed to the Memorial. There is

uncertainty as to the identity of the victims, and in most cases

names are not given. But it appears from those witness

statements that a significant number of deaths of individuals

resident within Ecuador, mostly children but also some adults,
are attributed to exposure to aerial spraying. 77 Also attributed to

the spraying are birth deformities (Annex 162), miscarriages

(Annex 169, Annex 225), impaired sight (Annex 200),

continued headaches and di zziness (Annex 203), facial

disfigurement (Annex 218), as well as shorter term problems
78
(vomiting, allergies, diarrhoea, flu-like symptoms). These are

76 EM, para. 6.6.
77 See EM, vol. IV, Annexes 161, 162, 166, 169, 194, 199, 200, 201,
204. Because of the vagueness of details and likely overlaps, it is not

possible to say how many individuals are involved in these allegations, or
indeed how many of these deaths are a ttributed by Ecuador to the spraying
78ogram.
For further analysis see Chapter 7 of this Counter-Memorial.
18serious allegations; but it is also said that the spraying caused

the death of a veritable catacomb of cows, horses and other farm

animals. 79

1.29 Other documents relied upon by Ecuador are even more

extreme. According to a report by Acción Ecológica, as of June
2001:

“100% of the people living in the border have
suffered poisoning from the sprayings with
RoundUp Ultra in a 5-km band, and 89% if the
band is extended to 10 km.” 80

1.30 There is not a trace of eviden ce that any proportion of

persons living kilometres from the area where the spraying took

place could possibly have been affected by it – still less 89% of

the population. Far from being evidence of breach by Colombia

of any obligation in relation to transboundary harm, the report

by Acción Ecológica – if evidence of anything – is evidence of
an epidemic unrelated to aerial spraying.

1.31 It is true that Ecuador also pleads causes of action that

are not dependent on the proof of serious (or indeed any)

transboundary harm. In particular:

(1) Under the rubric of “sovereignty” and “territorial
integrity” it appears to assert a doctrine of

79 See in particular below, paras.7.149-7.152.
80 See EM, Vol. IV, Annex 161, p. 11.
19 absolute immunity, i.e. that zero amount of drift
of the spray mix should cross the border. 81

(2) It argues that Colo mbia did not conduct an
environmental impact assessment in relation to

the aerial spraying program, whether in
Colombia or specifically in relation to possible
impacts in Ecuador. 82

(3) It argues that Colombia failed to cooperate with
Ecuador in investigati ng possible impacts of
aerial spraying, and failed to provide relevant
83
information.

(4) It suggests, with little or no detail, that there have
been cases of overflight over Ecuadorian
territory.84

But these claims are subsidiary to Ecuador’s principal position,

which is that Colombia has persistently caused serious and

substantial harm to Ecuador and its people in breach of

applicable norms of international law.

C. Colombia’s Response on the Merits

1.32 Colombia’s position, as demonstrated in this Counter-
Memorial, is straightforward. It is simply that there is no

reliable evidence that any of th e damage alleged has actually

occurred, still less that it is attributable to aerial spraying; the

evidence is to the contrary. In this context, it must be stressed

that the onus of proof both of da mage to its territory and of

81
82 See EM, paras. 10.13, Submissions, para. (C)(v).
83 See EM, paras. 8.41-8.44.
84 See EM, paras. 8.38-8.70.
E.g., EM, para. 6.3.
20causation is on Ecuador as the Applicant and moving party in

these proceedings.

1.33 But Colombia does not rely only on the burden of proof.

If the deaths of humans, large farm animals, etc had actually

occurred in Ecuador on this scale as a result of the aerial

spraying, that fact could – and should – be proved beyond a

shadow of a doubt. There would be forensic reports, coronial

inquests, detailed medical evidence supported by qualified
85
experts, etc. None of this has been produced. The same is true

of the environmental devastation allegedly caused by spray drift.
There should be concrete evid ence, including photographs, of

“thousands of hectares of valu able crops” lying dead in the

fields, official government repor ts, substantiated reports by

internationally recognised environmental and other scientific

experts, etc. Again, none of this has been produced. Instead,

Ecuador has produced a small num ber of anonymised witness

statements, alleging in vague and unverifiable terms a range of

harms which occurred between 7 and 9 years previously. It has
also commissioned one new scie ntific report (hereafter the

Menzie Report 8) which focuses almost entirely on

85
By contrast, an affidavit by the Director of the Lago Agrio hospital
– “the only hospital for the communities in Sucumbíos near the border with
Colombia” (EM, Vol. IV, Annex 188, p. 2) – states that in the period since
1999 he “generally saw the most extreme cases” (Ibid. , p. 3). Dr Sánchez
makes no reference to any deaths or long-term impairment. He produces no
medical evidence as to the causes of the “headache, vomiting and skin
86oblems” which he did observe.
C.A. Menzie, P.N. Booth & S. B.K. Driscoll, “Evaluation on
Chemicals used in Colombia’s Aerial Spraying Program and Hazards
21potentialities, is written in complete disregard of likely dose

rates and actual exposures, and is not based on any scientific
87
field work.

1.34 The position described in th e previous paragraph is
confirmed by the following considerations:

(1) The aerial spraying program carried out by Colombia
is not confined to the border region. It is a

Colombia-wide program. Exactly the same

procedure, equipment and materials are used when

spraying in Colombian territory close to the border

with Ecuador as in the rest of Colombia. If mere
drift of the spraying mixture across the border into

Ecuador had caused the cat alogue 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 large animals, environmental

devastation, economic collapse. Nothing of the sort

has occurred. No substantiated complaint of death or

serious harm to human heal th has been presented in
Colombia since the inception of the program. No

Presented to People, Plants, Animals, and the Environment in Ecuador”
(April 2009), EM, Vol. III, Annex 158.
87 For an analysis of the Menzie Report see below, paragraphs 7.95-
7.104.
22 substantiated case has been presented of death to

large mammals (e.g., cows and horses). There have

been occasional cases of ha rm to lawful crops – for

which, if duly established, compensation has been
88
paid. But these were cases of direct spraying, not

of spray drift over hundreds of metres, still less
kilometres.

(2)The spray mixture (published and well-known) is

based on commercially-available widely-used
chemicals which are lawful in Colombia and in many

other countries, including Ecuador. Glyphosate, the

active ingredient, is incapable of causing most of the

damages, in particular to humans and animals,

alleged by Ecuador, when us ed as in the Colombian

spray program. This fact is established by a series of
scientific studies, prepar ed independently of the

present proceedings. These studies were conducted

under the auspices of the Inter-American

Commission of Drug Abus e Control (CICAD), an
89
organ of the OAS based in Washington. Te

conclusion of the first of these studies (CICAD I), in
2005, was as follows:

“The risk assessment concluded that
glyphosate and Cosmo-Flux® as used in

88 See below, paragraph 4.22.
89 See further paras. 3.54-3.58 and 4.15-4.19 for the origin and
outcomes of the CICAD work.
23 the eradication program in Colombia did

not present a significa nt risk to human
health. Estimated acute worst-case
exposures in humans via all routes were
less than doses of concern, even for
chronic responses. In the entire cycle of

coca and poppy production and
eradication, human health risks associated
with physical injury during clear-cutting
and burning and the use of pesticides for
protection of the illicit crops were judged

to be more important than those from
exposure to glyphosate.
For the environment, risks from

the use of glyphosate and Cosmo-
Flux® to terrestrial animals were
judged to be small to
negligible.”90

The conclusion of the second set of studies (CICAD

II), published in 2009, was as follows:

“Overall, the risks to sensitive
wildlife and human health from
the use of glyphosate in the

control of coca (and poppy)
production in Colombia are small
to negligible, especially when
compared to the risks to wildlife
and humans that result from the

entire process of the production of

90 Annex 116: CICAD, K.R. Solomon, A. Anadon, A.L. Cerdeira, J.
Marshall & L-H. Sanin, “Environmental and Human Health Assessment of
the Aerial Spray Program for Coca and Poppy Control in Colombia”, OAS,
Washington, D.C., 31 March 2005, p. 11 (hereafter referred to as CICAD I).
24 cocaine (and heroin) in
Colombia.” 91

(3)Moreover, Ecuador – for the brief moment when it

was prepared to consider the issue on the basis of

actual, contemporary field evidence – acknowledged

that this was so. In mid-2004, the Ecuadorian

Commission on Atomic Energy took water samples

in both zones, concludi ng that “no Glyphosate
92
residues were found” in the waters tested. On 15

October 2004, the Presidents of the two countries

declared that “they were pleased with … the delivery

of studies conducted in Colombia on the possible

effects of the sprayings with glyphosate on illicit
crops.” 93 On 29 December 2004, the Ecuadorian

Foreign Ministry summarized the situation as

follows:

“There is an environment of calm; daily
activities are conducted normally; the
locals who were interviewed are in good

91
Annex 131-A, CICAD II: K.R. Solomon, E.J.P. Marshall & G.
Carrasquilla, “Human Health and Environmental Risks from the Use of
Glyphosate Formulations to Control the Production of Coca in Colombia:
Overview and Conclusions”, in Journal of Toxicology and Environmental
Health, Part A, 72:914-920, 2009, at p. 919. The set of 9 papers published in
this Journal will be referred to collectively as CICAD II (Annex 131, Annex
131-A to Annex 131-I).
92 Annex 80: Press Bulletin Nº 388 of the Ecuadorian Foreign
Ministry, “No glyphosate residues exist in the waters of the rivers of the
Sucumbíos Province”, 25 June 2004; and Annex 81: Press Bulletin Nº 480 of
the Ecuadorian Foreign Ministry, “No glyphosate residues were found in
Esmeraldas, border with Colombia”, 26 August 2004.
93
Annex 17: Joint Declaration from the Presidents of Colombia and
Ecuador, Esmeraldas, Ecuador, 15 October 2004, p. 2.
25 health; animals and crops are in order.
There has been no violation of the

Ecuadorian airspace, nor has th94e been
any displacement of persons.”
(4) This is detailed in Chap ter 5. However, the point is

that for most of the period from December 2004 to

the present time (all but 14 months) during which

aerial spraying was carried out, it has not been

conducted close to the border, as a result of without
prejudice abstention on Colo mbia’s part of spraying

within 10 km from the border. The relatively few

missions conducted during the period when this

suspension was not in place cannot possibly have
caused the serious harm alleged by Ecuador. If no

such harm had been caused by the end of 2004, it

was never caused.

1.35 It is to be noted that this Court is not the only forum

before which these issues are be ing litigated. A large group of
Ecuadorian nationals (initially more than 3000, though already

reduced by a third), plus three Ecuadorian provinces (Carchi,

Esmeraldas, Sucumbíos), are plaintiffs in two joined

proceedings before a United States District Court for the District
of Columbia under the Alien Tort Claims Act. Their claims are

brought against DynCorp, the United States corporation which

actually carries out the spraying operation under contract with

94 Annex 84: Memorandum of the Foreign Ministry of Ecuador, 29
December 2004.

26 95
the United States Department of State. It is alleged that

DynCorp is complicit in a spraying program unlawful under

international law. Twenty test pl aintiffs have been selected for

the purposes of a tria l of fact and law. The proceedings are

public and the court file is publicly accessible. The names of all
the plaintiffs are likewise public. 96

1.36 The problem of reliability of affidavit evidence of

alleged injury is highlighted by the following incident in the

Dyncorp case before the United Stat es District Court for the

District of Columbia.

1.37 Although in the present procee dings Ecuador wishes to

protect the identity of its witnesses, it makes an exception for
97
Mr Victor Mestanza. He is said to live at Puerto Mestanza, on

the bank of the San Miguel River which constitutes the border.
He and seven members of his fa mily are plaintiffs in the

Dyncorp case. In the course of de positions in that case, it

emerged that allegations by fi ve members of the Mestanza

family that they suffered persona l injury as a result of spraying

95 See Arias v. Dyncorp, 517 F. Supp. 2d 221 (2007).
96 At one point there were 3292 named plaintiffs in the two joined
cases. Of these, 681 were dismissed without prejudice by order of 16
September 2009, and 590 were dismissed with prejudice by order of 12
January 2010. The latter group of 590 dismissed plaintiffs were “given
repeated opportunities to provide the requested information about the
location of their exposure and their alleged damages, but have failed to do

so”, in violation of “multiple explicit court orders”: See Annex 155: Arias v.
Dyncorp, --- F. Supp. 2d ---, 2010 WL 94563 (DDC, 2010, Roberts J). That
97aves, for the moment, 2018 plaintiffs, including the 20 “test plaintiffs”.
See EM, paras. 6.67, 6.85, 6.92, 6.93, 6.95.
27were untrue: at all relevant times, they were resident at

Guayaquil, 400 kms away. In consequence, their personal

injury claims were withdrawn, and three of them were dismissed
98
with prejudice from the case. The point is that the five

members of the Mestanza family in question gave statements

indistinguishable in tone and generic content to those filed by
99
Ecuador in vol. IV of its Memorial. Their statements may

appear to be true and telling; th ey relate to a location virtually
on the border, not kilometres away. Yet they are conceded to

have been fabricated.

1.38 Against this background, Co lombia would make five

points.

100
• First, it is impossible to tell from the anonymous

affidavits produced by Ecuador – any more than one

can tell from the named affidavits of the Mestanzas –

what actually happened seven or nine years ago. It is

only by a proper forensic procedure – or by

98
See Annex 153: Arias v. Dyncorp , Plaintiffs’ Motion to dismiss
Three Individual Plaintiffs, 23 Decem ber 2009; Annex 154: Defendants’
Response, 6 January 2010. The extent of misrepresentation is detailed at
ibid., pp. 3-10. These documents are publicly accessible on the District
Court’s file.
99 See ibid., pp. 5-8 for a summary of the sworn statements now
retracted.
100 After the Memorial had been filed, Ecuador made an offer through
the Registrar to Colombia to disclose the names of witnesses on a restricted
basis. The inference was that Colombian officials generally could not be
trusted not to interfere with witnesses. Rather than allow such an inference
to stand, Colombia has preferred to receive the witness statements

anonymously. (It notes, however, that the plaintiffs in the Dyncorp litigation
have not sought anonymity: see para.1.35 above.)
28 unequivocal independent evidence – that the truth

can be established. In the present case there is no

unequivocal independent evidence of harm, as

distinct from mere assertion.

• Second, an even more crucial question in the present

case is whether such harm as may have occurred was
caused by the aerial spraying in Colombia. The fact

that someone says he saw planes in the distance and

that certain harms were suffered at the time does

nothing to establish causation – even if it were true.
It has to be shown that the spraying caused the

injury.

• Third, when reputable scientists confirm that a spray
mixture “did not present a significant risk to human

health”, a case that depends in major part on the

proposition that serious l ong-term health problems

were caused is directly challenged; it is challenged in
its essentials. Ecuador’s case, to repeat, is that

Colombia “caused extensive, long-lasting and

widespread harm to Ecuador , to its people and to its
101
environment”, and the core of its case concerns

lasting personal injury and perhaps even death. If
that core fails, the whole case is fatally impaired.

101 EM, para. 1.5.

29 • Fourth, as will be shown in Chapter 2, the two
Ecuadorian provinces c oncerned are amongst the

poorest and most deprived regions of Ecuador. Child

mortality throughout these provinces is abnormally

high; sanitation is poor; access to medical services is

limited. According to Ecuador itself, “nearly one-

third of all the residents in rural areas of Esmeraldas,
Carchi and Sucumbíos, in cluding children, suffered

from chronic malnutrition” in 2001. 102 Thths

nothing to do with the spray program.

• Health problems in the border area may however

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 environment. The production of coca paste for
drug trafficking in this re gion is intimately linked to

the activities of unlawful armed bands. Clearing of

jungle to plant coca is immensely destructive: it has

been estimated that from 1990-2004, approximately
103
345,233 hectares of forest were cleared. Te
pesticides used in coca cultivation, and the chemicals

102 EM, para. 2.23.
103 Annex 131-D, CICAD II: R.A. Brain & K.R. Solomon,
“Comparison of the Hazards posed to Amphibians by the Glyphosate Spray
Control Program versus the Chemical and Physical Activities of Coca
Production in Colombia”, in Journal of Toxicology and Environmental
Health, Part A, 72:937-948, 2009, at p. 945.
30 used in the production proces s in laboratories in the

border area, do much more damage to humans, to

flora and fauna, than the spray mixture could ever

do. Again, this point is established beyond doubt by

independent scientists. As part of CICAD I, a two-

tier hazard assessment was undertaken of the
toxicology of substances used in the production of

cocaine and heroin. The authors concluded:

“many of the substances used in cocaine
and heroin production and refining are
potentially hazardous to human and
environmental health. Comparatively,
several of the short-listed pesticides are

considerably more toxic to humans and
non-target organisms in the environment
than glyphosate (plus Cosmo-Flux®).
Most of the more hazardous pesticides

were found to be insecticides, which are
toxic to mammals and other wildlife, as
well as to insects. [I]f used improperly,
particularly in the production of coca and
heroin, these compounds have the

potential to present significant hazards to
human and environmental health, much
more so than the hazards identified for
glyphosate as used in the eradication of
104
the illicit crops.”

104 Annex 118, CICAD I: R.A. Brain, A.N. Crossan, L. Smith, K.R.
Solomon, “The Toxicology of Substances used in the Production and
Refining of Cocaine and Heroin: A Ti er-Two Hazard Assessment” (CICAD
OAS, Washington, 31 July 2005) Executive Summary, p. v. The full Report
with appendices is available at:
http://www.cicad.oas.org/Desarrollo_Alternativo/ENG/Projects%20By%20C
ountry/Colombia/OAS_CICAD_Tier_2_Hazard_Assessment_July_2005%5
B1%5D.pdf (last visited 10 March 2010).
311.39 In short, if the harms enumerated by Ecuador occurred –

and the extent to which they did so cannot be determined on the

very limited evidence put forward by Ecuador – there are very
good explanations for the cause of those harms which have

nothing to do with aerial spraying.

1.40 Ecuador has sought to reformulate its arguments in terms

both of human rights standard s and of putative standards

concerning the rights of indigenous peoples. These claims will

be dealt with in Chapter 9 of this Counter-Memorial. But the
core point is that a faile d claim for causing serious

transboundary harm to persons and the environment is not

improved by recasting it in terms of human or indigenous rights.
Colombia was, no doubt, obliged to consider potential impacts

on its neighbours, and on the environment, in formulating and

implementing the aerial spraying program. But if it in fact did

no harm in the course of its lawful conduct, and no evidence has
shown otherwise, if it gauged th e risks in a reasonable manner,

as will be demonstrated below, it is not necessary of the Court to

analyse considerations of observance of human rights standards.

1.41 As to the per se violations of international law on which

Ecuador relies (paragraph 1.33 a bove), Colombia would make
the following points:

(1) The doctrine of “absolute immunity” relied upon
by Ecuador does not repres ent international law:

32 the world is not divided into hermetically sealed

boxes. The obligation is to take all reasonable
105
measures not to cause significant harm.

(2) As to assessment, Colombia assessed the aerial

spraying program at th e time it was introduced,

and continued to do so. Its assessments, as

reflected in the Environment Management Plan,

accorded with Colombian law in force at all

relevant times, and have since been vindicated by
106
the two CICAD reports.

(3) Colombia did in fact cooperate with Ecuador in

investigating possible impacts of aerial spraying,
107
and it did provide re levant information.

Ecuador’s real complaint is that Colombia did

not agree with its position against aerial spraying

– but Colombia was under no obligation to

accept Ecuador’s views. On the contrary both
States were and are bound by the 1988 United

Nations Convention agains t Illicit Traffic of

Narcotic Drugs and Psychotropic Substances, in

particular Article 14. If there has been any

failure of cooperation in the present case, it is a

105 See further, paragraphs 8.34-8.40 and 8.46-8.61.
106
See further, paragraphs 4.8-4.19, 4.23-4.29. 6.26-6.29, 6.38, 7.12,
7.18-7.22, 7.29, 7.52-7.54, 7.58-7.64,7.75-7.77, 7.86, 7.91-7.93, 7.171,
10774-7.175.
See further, paragraphs 5.2-5.44.
33 failure by Ecuador to cooperate with Colombia in

the suppression of trafficking in illicit drugs.

(4) As to overflight, Colomb ia does not assert any

right of overflight for aircraft engaged in
spraying operations.

D. The Structure of this Counter-Memorial

1.42 This Counter-Memorial is divided into two parts. Part I
(chapters 1-6) establishes the context of the dispute and deals

with some key issues of fact.

• Chapter 2 describes the border region and gives a

succinct account of its socio-economic difficulties.

• Chapter 3 explains the basis for the fight against the

drug trade in Colombia, and the substantial

international backing for Colombia’s stance.

• Chapter 4 describes in some detail how the aerial

spraying program (Program for the Eradication of
Illicit Crops by Aerial Spraying with Glyphosate

(PECIG)) is carried out.

• Chapter 5 gives a full account of the exchanges

between the two States in relation to the dispute

since 2000, correcting the many deficiencies of

Ecuador’s Memorial in that regard.

• Chapter 6 disposes of th ree of Ecuador’s main

factual allegations against Colombia.
341.43 Against this background, Part II (chapters 7-10) deals

with Ecuador’s legal and scientific case.

• Chapter 7 details the scientific evidence concerning

aerial spraying with glyphosate, and analyses the

allegations of injury made by Ecuador and its
witnesses.

• Chapter 8 deals with Ecuador’s legal claims for
transboundary injury or harm.

• Chapter 9 deals with Ec uador’s claims based on
human and indigenous rights.

• Chapter 10 deals briefly with the remedial situation.

There follow Colombia’s submissions.

1.44 Attached to this Counter -Memorial is an Appendix,

consisting of an expert report by Dr Stuart Dobson, plus a

separate volume of 156 documentary annexes.

3536 PART I

THE FACTUAL MATRIX

37Figure 2.1 Colombia-Ecuador Border Region

38 Chapter 2

THE COLOMBIA-ECUADOR BORDER
REGION

2.1. Colombia and Ecuador formed part of a single State until
1830, when the federation of Gran Colombia was dissolved.

The land boundary between the two countries was fully and

definitively fixed by the Suárez-Muñoz Vernaza Treaty of 15
108
July 1916 and subsequent demarcation agreements. The

boundary extends for 717 kilometres from the Pacific Ocean in
the west, to the mouth of the Güepí River on the right bank of

the Putumayo River in the Amazonian region. It is shown on

Figure 2.1, opposite.

A. The Sectors of the Border Region

2.2. The frontier zone between Colombia and Ecuador is

divided into three sectors with clearly differentiated social,

economic and geographical features: the Pacific sector, the

Andean sector and the Amazonian sector. For present purposes
the relevant provinces are those of Nariño and Putumayo on the

Colombian side, and the provinces of Esmeraldas, Carchi and

Sucumbíos on the Ecuadorian side.

10C75.
39 (1) T HE PACIFIC SECTOR

2.3. The Pacific sector is formed of flat or slightly undulating
terrain, with large areas of rainforest extending from the coast to

the foothills of the Andes. In the territory of Colombia the

Pacific sector comprises the province of Nariño; in Ecuador, the

province of Esmeraldas. There are two ports in the area:

Tumaco in Colombia and Esmeraldas in Ecuador.

2.4. The inhabitants of the Ecuadorian province of

Esmeraldas include Afro-Ecuadorians and a number of

indigenous groups. The economy of the province is based on

the export of shrimp and bananas, as well as cocoa, tobacco and
coffee. Other important activit ies are logging, cultivation of

African oil palm ( Elaeis guineensis), and provision of support

and services to the trans-Andean pipeline. Despite a wealth of

natural resources, the inhabitants of the Pacific sector suffer
109
severe poverty, and difficult social conditions prevail.

2.5. The Colombian part of the Pacific sector of the border

extends for some 17,299 square kilometres, that is, slightly over

half the total area of Nariño province. The largest part of the

local population is mixed race, followed by Afro-Colombians,
who make up approximately 19% of the population. Indigenous

peoples make up approximately 11%. The main economic

activity is agriculture, with particular emphasis on the

109 This is conceded in EM, paras. 2.22-2.26. See also below, paras.
2.13, 2.15-2.19.
40cultivation of corn, beans and co coa. Cattle ranching, fishing
and mining are also significant economic activities in the region.

2.6. Tumaco is an important fishing centre, particularly for

tuna and shrimp, carried out mo stly by vessels of Ecuadorian

flag under contract from Colo mbian entrepreneurs who have

moved their seat of business to the city of Manta, Ecuador.

2.7. During the 1990s, the cultivation of illicit crops in

Colombia expanded considerabl y. The Nariño Province was

affected in particular. Following the implementation of the

Illicit Crops Integrated Monito ring System –SIMCI, for its

Spanish acronym in 1999, it was possible to establish, for

instance, that between Ma rch 1999 and August 2000, coca
110
cultivation in the province increased by 136%. Today, Nariño
produces 24% of the total coca crop in the country. 111 The

tightening of narcotics enforcem ent in the Caribbean has re-

directed the activities of illegal armed groups and drug

traffickers to the Pacific Ocean, and, for this reason, the Pacific

sector of Colombia has become a principal staging area and

export route for illicit narcotics.

110 Annex 103: UNODC, “Colombia Coca Cultivation Survey 2003”,
June 2004 (hereafter, Colombia Coca Cultivation Survey 2003), p.15.
111 Annex 108: UNODC, “Colombia Coca Cultivation Survey 2008”,
June 2009 (hereafter, Colombia Coca Cultivation Survey 2008), p.13.
41 (2) T HE A NDEAN SECTOR

2.8. The Andean Sector of the Colombia-Ecuador border is

formed by the Andes mountain range, with cold and temperate

climates and a predominantly mountainous te rrain with

elevations of up to 4,500 meters above sea level. The Andean
Sector extends southeast for some 200 km, up to the edge of the

Amazon basin. In Colombia, the sector falls within Nariño

province; in Ecuador it comprises part of Carchi province.

2.9. Two important centres of trade and commercial
exchange are located on either si de of the border in this sector:

Ipiales in Colombia and Tulcán in Ecuador. The Andean Sector

is principally agricultural. Its crops and products include fruit,

maize, oat, barley, wheat, sugarcane, cocoa, yucca, potatoes and
coffee. The Pan-American Highway, which runs the length of

South America, passes through the area.

2.10. The geographical and climatic characteristics of the
Andean sector are not generally favourable for the production of

illicit crops.

(3) T HE AMAZONIAN SECTOR

2.11. The Amazonian sector, formed chiefly by the provinces

of Putumayo in Colombia and Sucumbíos in Ecuador, is
traversed by the Putumayo River, a tributary of the Amazon, and

by the San Miguel River, a tributary of the Putumayo. The area

42is partially covered by rainforest; land in the region is mainly

devoted to conservation purposes, followed by forestry and

agro-forestry. However, in cer tain areas, over-use has resulted

in significant environmental damage. Growers of illicit crops

are especially responsible, as they typically do not employ the

soil management techniques appropriate to the local
112
environment. Destruction of forests by coca cultivators
113
presents a threat to fauna of the region as well.

2.12. Putumayo province in Colombia is geographically

divided into higher, middle and lower Putumayo. The principal

population group approximately 23%, consists of persons of

mixed race; 21% of the populati on is indigenous, and 6% Afro-

Colombian. A road connects Pu erto Asís, a river port on the

Putumayo River, with Mocoa, the provincial capital. From

Mocoa, the road continues northwa rds. In the other direction,

the road connects Puerto Asís with Ecuador by means of a

bridge across the San Miguel River.

2.13. In Ecuador, Sucumbíos, despite being that country’s
114
richest oil province, has extremely high poverty levels,

112UNOA Dlte,rntveelopment in the Andean Area: Technical
Guide, (date unknown), pp. 21-22. Available at:
http://www.unodc.org/documents/alternative-development/0981448.pdf (last
visited 10 March 2010).
113 Maria D. Álvarez, “Illicit crops an d bird conservation priorities in
Colombia,” (2002) 16(4) Conservation Biology 1086.
114
Annex 77: Sucumbíos Provincial Government Homepage (History,
Political Organization), p. 2.
43various endemic diseases, and a scarcity of basic utilities and

services. 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 sa nitation and health services. As

a consequence, the quality of life in the isolated communities of

the Ecuadorian Amazon is extremely poor, and the presence of
the Ecuadorian State often scarcely felt. 115

2.14. In 2000, the province of Putumayo accounted for 40% of

the coca crop of Colombia—i.e., 30% of the total coca crop of

the world. 116 The principal protectors of coca cultivation—and

its principal benefactors—were the illegal armed groups

operating in the region. As a result of measures taken under

Plan Colombia, however, by 2008 the province accounted for

only 12% of the coca crop in Colombia, down to 5.8% of the
117
global total. Suppression of narco tics cultiva tion in the

province has contributed significantly to the overall progress of

the counter-narcotics progra mme in Colombia: in 2000

Colombia produced 74% of th e world’s coca crop, whereas by
118
2008 the country’s share had dropped to 48%.

115 EM, paras. 2.22-2.26. See also ibid. , paras. 5.162, 6.1, 6.21, 6.22,
6.37, 6.50, 6.52, 6.86, 6.107, 6.119, 6.132, 9.86.
116 Annex 104: UNODC, “Colombia Coca Cultivation Survey 2004”,
June 2005 (hereafter, Colombia Coca Cultivation Survey 2004), pp. 15, 19.
117
118 Annex 108, Colombia Coca Cultivation Survey 2008, pp. 13, 17.
Ibid., p. 17.
44 B. Social Conditions in the Provinces of Sucumbíos and
Esmeraldas

(1) B ASIC PROBLEMS

2.15. Ecuador’s Memorial alleges that the social and economic

situation in the Ecuadoran provinces of Sucumbíos and

Esmeraldas has “changed dramatically” as a result of aerial
sprayings carried out in Colo mbian territory during certain

periods and in certain areas has. 119 In truth, the provinces of

Esmeraldas and Sucumbíos have been the victims of systemic

neglect by the Government of Ecuador; their present difficulties

are a continuation of long-term is olation and official failure to

address basic needs. 120 The discovery of oil has enriched the

central government, but the frontier provinces have realized

little, if any, benefit from this development. Two UN Special

Missions noted in their reports following visits to Ecuador in

2004 and 2006, at that country’s request, that the border region
121
has great potential and natural riches. As Ecuador itself
122
acknowledges in its Memorial, however, this is in stark

contrast to the high levels of poverty, scarcity of basic social

services, and institutional, technical and financial weakness that

119 See EM, para. 2.6.
120 Ministry of Foreign Affairs of Ecuador, Plan Ecuador, Scenario of
the Plan (hereinafter: “Plan Ecuador”), pp. 5-6. Available at:
http://www.resdal.org/ultimos-documentos/plan-ecuador07.pdf (last visited
10 March 2010).
121
122 EM, Vol. II, Annex 27, p. 19; Annex 28, p. 14
EM paras. 2.22 (footnote 33), 2.26 (footnote 36).
45characterize this area. 123 Moreover, it is far from clear that the

introduction of infrastructure to support the oil extraction

industry has off-set the deleteriou s effects of that industry on

public health.

2.16. According to the Ecuadorian Government, the

percentage of persons below the poverty line in Ecuador in 2001
124
(assessed on the basis of Unmet Basic Needs ) was 61.6%. Of

the 22 provinces of Ecuador, Su cumbíos had the second highest

poverty level, with 86.2% of its population below the poverty
125
line. Esmeraldas also had a level of poverty bordering 80%.

Conditions in the two provinces, relative to the rest of the

country, were depressed before Colombia commenced the aerial

spraying programme. In the period 2005-2006 the levels of

poverty in Sucumbíos and Esmeraldas did not deteriorate, as

claimed by Ecuador; they in f act improved somewhat, to 71%

123 See e.g., EM, paras. 2.22-2.26. Also, EM, Vol. II, Annex 27, pp.
17-19; Annex 28, pp. 14, 24.
124 Under the Unmet Basic Needs (or “deprivation indicators”) method,

a household is defined as poor when it lacks access to education, health,
nourishment, housing, urban utilities or services and/or Employment
opportunities (structural poverty). The method was adopted by the United
Nations Economic Commission for Latin America and the Caribbean
(ECLAC), in the early 1980s: Expert Group on Poverty Statistics,
Compendium of Best Practices in Poverty Measurement (Rio de Janeiro: Rio
Group, September 2006) pp. 101-120, 123.
125 Annex 78: “Social Development and Poverty in Ecuador, 1990-
2001”, Chapter 8, Poverty Profile acco rding to Unmet Basic Needs”, Social
Front’s Technical Secretariat Information and Analysis Unit, Government of
Ecuador, Social Report 2003, Quito, December 2003, p. 10.

46 126
and 64.2%, respectively. The observed improvement by no
means shows that the border provinces of Ecuador have

resolved their long-term social and economic crisis. It does

however contradict Ecuador’s contention that the aerial spraying

programme in Colombia correlates to a worsening of conditions

there. The social and economic crisis of the Ecuadoran border

provinces is a long-term, structural crisis, and it is due to the

failures of the Ecuadoran government, not to any alleged effects
incidental to the spraying programme.

2.17. Ecuador would also attr ibute the environmental

dimension of the regional crisis to Colombia’s anti-narcotics

effort. Yet the main causes of damage to the environment in the

border region are well-known. E xploitation of forestry in the

region has been indiscriminate, and widespread environmental
damage has resulted. 127 The other major economic sector in the

provinces, hydrocarbon extraction, likewise has not been subject

to appropriate environmental controls. The Ecuadorian

hydrocarbon industry has caused significant harm to the local

ecosystem. Any incidental eff ects of anti-narcotics spraying—

and Colombia will show that the spraying programme has had

no adverse environmental e ffects in Ecuador—pale in

126
Annex 94: “The Living Conditions of Ecuadorians: Results of a
Survey of Living Conditions – Fifth Round. Poverty and Inequality”,
National Institute for Statistics and Censuses of Ecuador, Quito, 2009, p. 10.
127 Annex 106: “Report of the Unite d Nations Technical Preliminary
Mission to propose studies on the impact of the aerial sprayings and
complementary actions in the northern border of Ecuador”, Quito, Ecuador,
April 2006, pp. 4, 15, 16.
47comparison to the massive and indiscriminate damage resulting

from Ecuador’s own economic activities in the exploited

provinces. Pollution and enviro nmental degradation in the

region is a significant problem, and the failure of the Ecuadorian

government adequately to regulate exploitation of local
128
resources has been the cause.

2.18. During its visit to the north ern border of Ecuador in

2006, the United Nations’ Preliminary Technical Mission

observed that the border region in general suffers from a lack of

access to potable water supplies, ha s little infrastructure for the

disposal of sewage and solid residues, and that the local
population has no education with regard to sanitation. 129 All of

those factors contribute to the critical epidemiological profile of

the area, characterized by the prevalence of illnesses such as

acute respiratory infections, acute diarrhoeal disease, illnesses

common to humid, tropical clim ates, malnutrition and skin

conditions.

2.19. According to the Ecuadorian Ministry of Public Health,

the principal causes of morbidity in the provinces of Sucumbíos

and Esmeraldas in 2007 were acu te respiratory infections and

acute diarrhoeal diseases. These are the same diseases which
constitute the principal causes of morbidity in all twenty-two

128 CCM, Annex 129. See also, EM, Vol. II, Annex 27 p. 19, Annex 28
p. 14.
129 See EM, Vol. II, Annex 28, pp. 341, 342.
48provinces of Ecuador. 130 Provinces which Ecuador must

concede are not in any way affected by aerial spraying display

much the same disease profile as Sucumbíos and Esmeraldas;

and no prima facie reason exists to attribute the epidemiological

situation in Sucumbíos and Esmeraldas to aerial spraying. In
any event, as will be demons trated, there is no causal link

whatever between aerial sprayi ng and the two principal causes

of morbidity in the border provinces.

2.20. What does distinguish the bor der provinces from other
parts of Ecuador is the residence there of various criminal

groups. The social and economic conditions of the provinces of

Sucumbíos and Esmeraldas have combined with the lack of an

effective State presence to create conditions favourable to illicit

activity. Criminal groups, especi ally guerrillas, avoiding more
stringent law enforcement, have fled Colombia and settled in the

region. They have brought with them the cultivation, processing

and sale of coca crops, as well as trafficking in arms, explosives

and munitions.

2.21. These groups, using the bases they have established in

Ecuador, have launched attacks against military personnel and

carried out kidnapping, extortion and killing of civilians in

Colombian territory.

130 Annex 92: “Ten Main Morbidity Causes per Province”, Public
Health Ministry of Ecuador, Epidemiology sub-process, Quito, 2007, p. 1.
492.22. Also, over the years, weapons, explosives, ammunition

and chemical precursors involve d in the processing of illicit
drugs have been smuggled into Colombia across the common

border with Ecuador.

2.23. This cross-border illicit trade was confirmed by the
Commission on Transparency and Truth for the Angostura Case

(Comisión de Transparencia y Verdad sobre el Caso de

Angostura), set up by the President of Ecuador in March 2009.
The Commission was established to investigate the facts relating

to the attack by Colombian armed forces on an encampment of

Colombian (FARC) guerrillas in Ecuadorian territory, located a

few hundred meters from the border between both countries at
the Putumayo River, on 1 March 2008, and to investigate links

between Ecuadorian officials, ill egal organizations involved in

drug trafficking and the Colombian guerrillas. The Commission
was made up of persons from diverse national backgrounds

chosen by the Ecuadorian Government itself.

2.24. The final Report of the Co mmission, delivered to the
President of Ecuador on 10 December 2009, clearly describes

the serious situation in the northern region of Ecuador bordering

Colombia. The report notes the presence of illegal armed
groups and drug traffickers engaged in criminal activities.

Excerpts of the Report read as follows:

“Colombian illegal armed groups, in fulfilling
their political-military project, have

50 systematically violated [Ecuador’s] national
sovereignty and territorial integrity, through
armed actions to attack military detachments and
patrols with the purpose of seizing weapons and

equipment (in subversive lingo, ‘recovery of
weapons’ that belong to the people). And, they
have used the national territory to set up

laboratories for processi ng drugs, engendering a
highly tense and risky atmosphere for the
national and foreign population living in the
border zone, seriously a ffecting both Ecuador’s
131
social as well as national security.”

2.25. The Report also refers to encampments and drug

laboratories maintained by Co lombian guerrillas and drug

traffickers in Ecuadorian territo ry. It quotes a statement by the

President of Ecuador to the effect that “it is... clear that all over

that area, the FARC have a larg e intelligence support network
132
from the civilian society.”

2.26. The Report posits that “[t]he low effectiveness of

military and police efforts that operate on the basis of human

intelligence for the location, capture, elimination and destruction

of encampments, allows [the Commission] to infer the

131
Informe de la Comisión de Transparencia y Verdad [Report of the
Commission on Transparency and Truth for the Angostura Case], Quito,
December 2009, p. 65.
Available at:
http://www.diario-
expreso.com/media/ediciones//20091211/actualidad/documentos/10122009_
132132.pdf (p. 65) (last visited 10 March 2010)
Ibid. p. 65.
51infiltration of militia who are in charge of giving early warning

to the FARC.” 133

2.27. Subsequently, the Report reads as follows:

“As stated above, the efforts carried out for the
control of drug trafficking and smuggling are still
insufficient. Despite the controls… smuggling
and drug trafficking continue to increase.” 134

2.28. Among the Report’s conclusions, the Commission

recalls the prevailing situation in the border area where it

conducted its investigation, co ncerning the existence of

laboratories for cocaine processing:

“23. There is a lack of control in the import and
destination of precursors for the processing of
drugs. Ecuador has thus ceased to be merely a

transit country in order to135art becoming a place
for [drugs] processing.”

133 Informe de la Comisión de Transparencia y Verdad [Report of the
Commission on Transparency and Truth for the Angostura Case], Quito,

December 2009, p. 65.
Available at:
http://www.diario-
expreso.com/media/ediciones//20091211/actualidad/documentos/10122009_
114132.pdf (p. 65) (last visited 10 March 2010)
134 Ibid., at p. 95. Available at:
http://www.diario-
expreso.com/media/ediciones//20091211/actualidad/documentos/10122009_
114405.pdf (pp. 95, 129) (last visited 10 March 2010)
135 Ibid., at p. 129.

52 (2) P ROVINCE OF SUCUMBÍOS

2.29. As noted above, two Ecuadorian provinces are

principally concerned in Ecuador’s claims. Colombia turns first

to consider the province of Sucumbíos. The majority of the

inhabitants of Sucumbíos live in homes not serviced by public

utilities. Only 13.9% have acce ss to potable water, and only

26.7% are connected to sewage sy stems. The electricity grid

covers only a fraction of the provi nce, while waste disposal is
136
precarious and well below the national average.

2.30. Sanitary provisions in the province are therefore wholly
137
inadequate. Together with the scarcity of health services, this

has resulted in high rates of re spiratory infection and acute

diarrhoeal diseases. These, in turn, are the main factors

contributing to the province having one of the highest morbidity
138
rates in the country.

2.31. Hydrocarbon exploration and exploitation activities in

Ecuador, particularly in the provinces of Sucumbíos and

Orellana – the country’s largest oil producing provinces – have

resulted in serious environmen tal damage, and have worsened

the public health situation in the border area. The adverse

136 Ministry of Foreign Affairs of Ecuador, Plan Ecuador, 2007, p. 5.
Available at: http://www.resdal.org/ultimos-documentos/plan-ecuador07.pdf
(last visited 10 March 2010)
137 Annex 90: “Health baseline of the Ecuadorian Northern Border
Provinces”, Public Health Ministry of Ecuador, Pan-American Health

138anization (PAHO), Quito, Ecuador, 2006, p. 27.
Annex 92.
53effects of the hydrocarbon industr y on public health have been

the basis of legal actions before United States’ courts. 139

Claimants there contended that oil companies involved in the

region have not used appropriate technologies to mitigate

environmental harm, with the result that dangerous pollutants

have been introduced into rivers, lakes, marshes and reservoirs.

Some of the pollution has been gradual, such as leaching from
140
separation ponds. There also are frequent oil spills, events

responsible for the sudden introduction of vast quantities of

chemicals known to be damaging to human health and the
141
environment. Several indigenous communities have

complained to the Ecuadorian Government about the

environmental damage which has re sulted from the activities of
142
the oil industry in the region.

139
The procedural history of one set of such actions is summarized in
Aguinda v. Texaco, Inc., 303 F.3d 470, 473-476 (2d Cir.2002) WL
2002524696.
140 Annex 129: M. San Sebastián & A-K. Hurtig, “Oil exploitation in
the Amazon Basin of Ecuador: a public health Emergency”, Pan Am J Public
Health,15(3):205-207, 2004.
141 Ibid. As regards adverse health e ffects related to oil extraction

activities in Amazon Basin of Ecuador (that includes the Sucumbíos
142vince), see p. 208.
For example, the Cofán territory in Ecuador “...was atomized and
fragmented as of 1967 with the start of oil exploitation in the region”. In
Annex 76: “Ai’cofan indigenous nation”, National Institute for Statistics and
Censuses of Ecuador (date unknown), pp. 1, 3. The Council for the
Development of the Nationalities and Peoples of Ecuador ( Consejo de
Desarrollo de las Nacionalidades y Pueblos del Ecuador) has also pointed

out the difficulties encountered by the Cofan indigenous communities in
Ecuador “facing heavy pollution due to the frequent oil spills as a result of
the extraction operations”. In their territory “there are also logging activities
and some men work as day labourers in the settlers’ farms”. In Annex 93:
“Cofan indigenous nation”, Webpage of the Council for the Development of
the Nationalities and Peoples of Ec uador, 6 November 2008. A similar
54 (3) P ROVINCE OF ESMERALDAS

2.32. The Province of Esmeraldas is located on the north-east

coast of Ecuador. Despite enormous potential in terms of

natural resources, it suffers from various serious socio-economic

problems that have impaired its development, notably

deforestation. Some experts indicate that as early as 2010 there

will be no primary forests remaining, except in nature reserves,
143
which are themselves threatened. According to the FAO, in

1960 the province of Esmeraldas had some 15 million hectares

of Ecuador’s wood reserves. Today those reserves do not exceed

200,000 hectares, as one of the main activities in Esmeraldas has
144
been illegal and indi scriminate logging. According to the

German Agency for Technical Cooperation (GTZ), the forests

of Esmeraldas constitute the source of most of the wood on the
145
Ecuadorian national market.

situation confronts the Siona community, which has suffered “the loss of
territory due to the opening of roads for the use of oil companies” and
“pollution caused by oil exploitation”. In Annex 75: Siona indigenous
nation, Web page of the Council for the Development of the Nationalities
and Peoples of Ecuador (date unknown), pp. 1, 3.
143 Annex 79: “Model Project for the Planting, Production and
Sustainable Agro-industry of Guanábana (Soursop)”, Federation for the
Integral Development of Peasants and Afro-descendants of Esmeraldas
(FEDICAE), Esmeraldas, January 2004, p. 4.
144
145 Ibid., p. 4.
Report of the Consulting Project ECU/99/017/UNDP, Diagnosis of
the current situation of the Choco ecoregion in the Esmeraldas Province:
Direct causes, authors and underlying cau ses of the loss of biodiversity and
degradation of environmental biodiversity, Quito, 27 May 2005, p. 51.
Available at:
http://www.bibliotecaonu.org.ec/files/Diagnostico%20de%20la%20situacion
%20actual%20Choco...ambiental.pdf (last visited 10 March 2010)

552.33. Sewage and waste collection pose a serious problem for

public health; infant mortality rates are high and above rural
146
national averages. Likewise, chronic malnutrition in the
147
province’s rural areas runs at 23% of the population.

2.34. The provincial government of Esmeraldas reported in

2005 that:

“…tropical diseases such as malaria, dengue and
leishmaniasis are frequent, as are sexually
transmitted diseases (STD), HIV/AIDS and
diseases related to deficient sanitation conditions.

This is explained by several factors, among them,
the situation of poverty, the low coverage of
public utilities, the lack of attention to the canton

from the central level, deficiencies in the
coordination between actors in the local sphere,
bad self-care habits of the population, among
others.” 148

2.35. The proportion of the population without proper health

service coverage in rural areas is very large. That situation

results mainly from the lack of roads and scarcity of transport

that deprive the population of access to the facilities of the

health system.

146 Annex 100: United Nations Children's Fund (UNICEF), “Baseline
of the Province of Esmeraldas, Final Consulting Report”, April 1998, pp. 16,

147 22.
148 Annex 106, p. 24.
Annex 91: “Esmeraldas Health and Environment Program
(SYMAE)”, Municipality of Esmeraldas, 1 August 2006, p. 2.
56 (4) C OLOMBIAN COOPERATION WITH ECUADOR

2.36. Before 2006 Ecuador did not have a public health

surveillance system in place in the border area that allowed any

adequate follow-up on the alleged impacts of pesticides.

2.37. In a meeting between the Foreign Ministers of Colombia

and Ecuador in December 2005, Colombia offered to cooperate

on this subject:

“5. The Colombian Foreign Minister offered to
place at Ecuador’s disposal, her country’s

successful experiences in projects aimed at
strengthening public health surveillance on the
use of pesticides, with a view to the preparation
of a joint project that would be submitted to the

PAHO-WHO under the technical cooperation
scheme.” 149

2.38. As a result of a bilateral technical and scientific
150
cooperation agreement betw een the two countries, in

December 2007 Ecuador implemented a Public Health
151
Surveillance System of Int oxications due to Pesticides. The

149 Joint Communiqué of the Meeting of Colombia-Ecuador Foreign
Ministers, Quito, 7 December 2005, nu mbered para. 5 (EM, Vol. II, Annex

72). Ambassador Claudio Cevallos, Director-General for Ecuadorian Border
Affairs with Colombia of the Foreign Ministry, during a Meeting of the
Bilateral Technical Commission on Health, alluding to the Joint
Communiqué, “emphasized Colombia’s experience in pesticide
management.” (See Annex 22: Minutes of the Bi-national Technical
Commission on Health, Atacames, Esmeraldas, 2-3 March 2006).
150 Annex 23: Health and Environment integrating borders, Proposal for
Technical Cooperation between Countries Colombia-Ecuador:
Epidemiological Surveillance, Pesticides, and Water Quality, 2007.
151 “For Ecuador it is very important to... establish at a national level,
the Public Health Surveillance System for Pesticide Intoxication and the

57project was jointly financed by the Governments of Colombia

and Ecuador together with th e PAHO/WHO. No mention is

made of this initiative in Ecuador’s Memorial.

C. Conclusions

2.39. Contrary to Ecuador’s assertions, 152 Colombia’s aerial

spraying programme is not the cause of the very real and serious

social and economic problems of the provinces of Esmeraldas

and Sucumbíos, nor even a contri butory factor. The provinces

are plagued by chronically precari ous infrastructure, long-term

neglect by central government auth orities, and rampant misuse
of its natural resources, especially in the oil and forestry sectors.

The mechanisms by which such factors undermine public health

and degrade the environment ar e well-known. Their impact in

Esmeraldas and Sucumbíos has not been limited to the

immediate border area but has been felt throughout the

provinces, affecting all the communities of the region, including
the indigenous population.

2.40. Furthermore the population have long suffered from high

poverty levels, social problems a nd serious difficulties in terms

of access to basic services and ut ilities (including in particular

Single Notification Record for Intoxications due to pesticides which
Colombia has implemented several years ago”. In Annex 29: Minutes of the
Bilateral Evaluation Meeting. Health and the Environment Integrating
Borders: “Strengthening of Surveillance in Public Health, Pesticides, and
Water Quality”, Ipiales, Nariño, 12 December 2007.
152 EM, paras. 6.3-6.7.
58water, sewage and health services). As Colombia will

demonstrate, the situation in Esmeraldas and Sucumbíos has
nothing to do with aerial sprayi ng operations car ried out in

Colombia but, instead, originates in conditions within Ecuador

pre-dating Colombia’s anti-narcotics programme.

2.41. The leading causes of morbidity in the provinces of

Sucumbíos and Esmeraldas (as we ll as in the rest of Ecuador)

are acute respiratory and diarrheic diseases. The aerial spraying
program carried out by Colombia over portions of its territory

neighbouring Ecuador began only in 2000 and was carried out

along certain segments of the border during circumscribed
periods of time. Additionally, aer ial spraying was suspended in

the border for an entire year – December 2005 to December

2006 – and a new suspension remain s in place since February

2007. As will be seen, the aerial spraying could not have caused
the adverse effects complained of by Ecuador.

2.42. The absence of effective public authority in the region
also fosters the presence of il legal armed groups. The presence

of such groups in Ecuador hampers the efforts of Colombian

authorities and facilitates the trafficking of illicit substances,

chemical precursors and weapons.

2.43. Since 2006 Colombia offered cooperation for the

implementation of a public health surveillance system, which

59started to be implemented a year later with the financial support

of Colombia, the WHO and PAHO.

60 Chapter 3

THE FIGHT AGAINST DRUGS IN COLOMBIA

A. An Overview of the Drug Problem in Colombia

3.1. Drug-trafficking in Colombia started to become a major

problem in the mid-1970s due to the increase in marijuana

consumption in the United States and Europe. 153 Taking

advantage of Colombia’s geog raphic location and complex

terrain, armed groups and drug traffickers at first established

illicit crops on small plots of land located in remote areas.

Subsequently, the crops progressively came to cover ever larger

areas that were protected by th e illegal armed groups against
154
government eradication efforts.

3.2. During the 1980s, the drug cartels and mafia gained
strength and began their viol ent and blood-ridden conflict for

control of drug production and distribution in the United States

153 United Nations Office on Drugs and Crime (UNODC), A Century of
International Drug Control, Vienna, 2008, pp. 65-66. Available at:
http://www.unodc.org/documents/data-and-
analysis/Studies/100_Years_of_Drug_Control.pdf (last visited 10 March
2010).
154 Annex 99: General Assembly Resolution S-20/2, “Political
Declaration”, 10 June 1998, UN doc. A/S-20/4, p. 3. The 1998 Political
Declaration expressed deep concernabout “the links between illicit drug

production, trafficking and involvement of terrorist groups, criminals and
transnational organized crime” (§10), and was alarmed “by the growing
violence resulting from links between illicit production of and illicit
trafficking in arms and drugs” (§11).
61and Europe. 155 They built clandestine air strips and processing

laboratories in remote areas of the Colombian rainforests; they

devised routes for the provision of coca paste from Bolivia and

Peru; they ensured the provision of chemical precursors from

industrialized countries for drug processing; finally, they

succeeded in infiltrating and compromising law enforcement

authorities both in the distribution centres, as well as in certain

neighbouring States, which were used not only as necessary

transport routes but also as places for the exchange of drugs for

arms and explosives.

3.3. World consumption of cocaine increased throughout the
156
1980s. In the 1990s, consumption fe ll for a time, then started
157
to increase again in 1996. In 1999, estimates suggested that
158
some 13 million people abused cocaine worldwide. The

largest numbers of cocaine user s were found in North America

(45% of the world total), followed by Western & Central Europe
159
(24%) and Latin America (19%).

3.4. From the mid to late 1990s, there was an unprecedented

growth in the areas affected by illicit crops in Colombia,

coupled with an increase in th e political and social problems

155
Op. cit. UNODC, A Century of International Drug Control, p. 67.
156 Ibid., pp. 66, 67, 83.
157 Ibid., p. 71. See, also, United Nations Office for Drug Control and
Crime Prevention (ODCCP), Global Illicit Drug Trends, (New York, 1999),
p. 109. Available at: http://www.unodc.org/pdf/report_1999-06-01_1.pdf.
158SDee, CP, Global Illicit Drug Trends, 1999, p. 93, 95.
159 Ibid., pp. 109-110.

62linked to drug trafficking in those regions. That phenomenon

reached a peak in 2000, when an integrated anti-drugs policy
160
was adopted.

3.5. The expanding trend reached the point where, despite all
the efforts undertaken, in 2000 Colombia was the world’s

largest coca and cocaine producer, with 77% of the worldwide

crop, 80% of coca leaf production, and 79% of cocaine

manufacturing potential. 161

3.6. Colombia has been one of the countries most affected by

drug mafias. No other single nation has suffered or sacrificed as

much in the fight against th e world drug problem. With

financial resources obtained as a result of activities related to

cocaine and heroin, the illegal armed groups and drug mafias

undertook a growing campaign of kidnappings, murders,
coercion and threats, as well as attacking civilians (regardless of

income, activity or political affiliation), infrastructure and public

authorities. As a result, the tr aditional economy of several

regions where the illicit crops were grown was undermined and

the entire country was seriously threatened. 162

160 Annex 57: “Plan Colombia Progress Report 1999-2005”, National
Planning Department of Colombia, Office of Justice and Security, September
2006, pp. 8; 10-12; 25-27; 33-40.
161 Annex 102: United Nations ODCCP, “Global Illicit Drug Trends
2001”, New York, 2001, p. 67.
162
Three presidential candidates (Luis Carlos Galán, Carlos Pizarro and
Bernardo Jaramillo), one Attorney General (Carlos Mauro Hoyos), two
ministers of justice (Rodrigo Lara and Enrique Low), two senior journalists
(Guillermo Cano, director of El Espectador , Raúl Echavarría, deputy editor
633.7. In 1988, addressing the United Nations Conference for

the Adoption of a Convention agains t Illicit Traffic in Narcotic

Drugs and Psychotropic Substances, the Colombian Minister of

Justice, Guillermo Plazas Alcid said:

“[N]o country in the worl d ha[s] paid as high a
price as Colombia in the fight against drug abuse

and illicit trafficking. One by one, Ministers of
State, judges of the Supr eme Court, officials in
the armed forces and police, members of
intelligence units, soldiers and journalists ha[ve]
all fallen as an intimidated nation raised

anguished voices for protection from the scourge.
Ha[s] the world forgotten the burning up of
Colombia’s Palace of Justice which housed
senior judges and law o fficials? ... Colombia’s

tough experience in fighting the problem ha[s]
left lesions in the political and social fabric of the
country... Timely and ade quate treatment of the
problem of drug abuse a nd illicit trafficking
should be given maximum priority at the national
163
and international levels.”

3.8. The fight against drugs in Colombia, including the fight
against the drug cartels and mafi as, has also had an enormous

cost for the country as a whole, which remains unparalleled in

in chief of Occidente), and hundreds of judges and policemen have been
killed. A commercial plane with 111 passengers was bombed in the air after
its take off at Bogotá’s airport; the Headquarters of the National Security
Agency (Departamento Administrativo de Seguridad, DAS) and main offices
of El Espectador newspaper were also bombed.
163 United Nations Information Service, directly transcribed from [the]
“Statement by Justice Minister of Colo mbia and Minister of State for Home
Affairs of Nepal address Conference to adopt Convention against Illicit Drug
Trafficking”, 4th Plenary Meeting, 1988, UNIS/NAR/262, in: Op. cit.
UNODC, A Century of International Drug Control, pp. 67, 100.
64proportion to any other State. Successive administrations in

Colombia have waged a relentless fight against this scourge,

devoting enormous resources that might have otherwise been

employed for social investment and to promote the economic
development of the country.

3.9. Colombia is the second mo st biologically diverse

country in the world and the cu ltivation and processing of illicit
drugs in Colombia has had particularly serious consequences for

the environment. Studies have shown that the establishment of

illicit crops and their processing are the worst cause of
164
deterioration of the country’s environment.

3.10. These adverse impacts are spe lled out in more detail in

the following Sections.

B. Political and Social Impacts of Illicit Crops and Drug
Trafficking in Colombia

3.11. It has been long recognized that illicit drugs fuel political
instability. In Colombia, part icularly during the 1980s and

1990s, presidential candidates, ministers, judges, officers,

soldiers and corpsmen of the Army and the National Police,

diplomats, priests, homemakers , journalists and thousands of

other Colombians from all sorts of backgrounds, engaged in
diverse economic activities and of all levels of means, fell

victim to criminals associated with drug trafficking and

164 Annex 116, CICAD I, pp. 11-12.

65terrorism. The drug mafia tried to infiltrate different segments

of the State and its executive, le gislative and judicial branches.

Millions of citizens were victims of generalized fear and threats
165
or extortion.

3.12. The violence Colombia has had to endure has largely

been financed by the resources obtained by illegal armed groups

from the cultivation, production an d trafficking of illicit drugs.
The symbiosis between drug- trafficking and terrorism

strengthens criminal activities such as killings, attacks on

civilian and governmental infr astructure, kidnappings and

extortion perpetrated by illegitimate armed organizations. These
activities would not be viable without the financial resources

provided by drug-trafficking. It is no coincidence that most of

the areas where the majority of the coca crops are located are

under the influence of illegal armed groups.

3.13. In fact, drug trafficking has been the main cause for the

high levels of violence in Colomb ia in the last two and a half

decades. Drug trafficking has in creased overall crime, caused

congestion in the criminal ju stice system, and gradually
transformed and corrupted moral values in favour of crime.

3.14. Drug trafficking has permeated the illegal armed groups

and has been the key factor in the increase in size of guerrilla

165 Annex 97: UNDCP, “Economic and Social Consequences of Drug
Abuse and Illicit Trafficking”, Technical Series 0101, 1998, p. 39.

66and paramilitary movements. Between 1980 and 2000, the

guerrillas grew by twenty-fold. In 1980, the FARC had only 10
Fronts and 980 members. Twenty years later, the FARC had 63

Fronts and 16,492 members. There is a parallel between the

increase in illegal crops and the growth in the number of

guerrilla members in Colombia, starting in 1991. These figures

explain the incidence of drug trafficking in the nation’s levels of

violence and the reason why combating drug trafficking became

a crucial issue for Colombia. Fortunately, this trend has been

reversed in recent years as a result of the implementation of

“Plan Colombia” and the “Democratic Security Policy.” The
guerrilla groups have been signifi cantly reduced and the overall

security of the country has improved; 13,533 members of the

FARC demobilized between August 2002 and December

2009. 166 Homicide rates and terrorist acts dropped 45% and
167
98%, respectively, between 2002 and 2008.

3.15. The Government of Colombia continues its struggle for

the stability of the State and within that framework, for the
security, peaceful existence and health of millions of youths and

166
Presidency of the Republic of Colombia,Communications Office,
Demobilizations, 2010. Available at:
http://www.presidencia.gov.co/resultados/english/05_demobilized/demobil…
167on_100209.pdf (last visited 10 March 2010)
Presidency of the Republic of Colombia, Ten Achievements:
Security Based on Democracy, 2009. Available at:
http://www.presidencia.gov.co/resultados/english/documents/AVANsecurity
_091019.pdf (last visited 10 March 2010)
67adults in many other countries. Domestically, this effort has

garnered massive popular support.

C. Economic Impacts of Illicit Crops and Drug
Trafficking in Colombia

3.16. Illicit crops have had se rious adverse impacts on
Colombia’s economy. Although only a fraction of the enormous

illicit profits make their way back to Colombia – most of the

money remains in the United States and Europe, and other

consumer countries – the sudden peaks in revenue generated by

drug trafficking greatly destabil ized the country’s finances by
affecting its production capacity and deviating resources

towards illicit enterprises and away from the construction or

strengthening of industries, th ereby affecting the country’s
168
economic growth.

3.17. The illicit drug industry distor ts internal consumption,

investment and import patterns, since drug traffickers and their

associates invest in equipm ent and non-productive goods that
enable them to launder their illicit profits. Thus, drug

trafficking not only has an adverse impact on domestic industry

but also on the export sector since large influxes of foreign

currency cause the value of domestic currency to rise artificially,

to the serious detriment of exports. The testimony in 2003
before the United States’ Senate Foreign Relations Committee

168 Annex 97, pp. 27-29.

68of a former State Department of ficial with first-hand expertise

on Colombian affairs is telling in this respect:

“Only a fraction of the immense revenue of drug

trafficking returns to Colombia. Most drug
money remains in the United States. Still, the
impact of those ill-gotten gains has been
enormous. Analysts at various times using
different methodologies ha ve estimated the total

flow of earnings back to the country as low as
$1.5 billion and as high as $3 billion, i.e., not
much more than one percent of GDP, but all
conclude that the net effect on the Colombian
economy has been negative and disastrous. The

rise in narcotics traffick ing is closely related to
the rise of criminality in Colombia and that of
course was closely related to the rise in violence.
One study points out that criminality leads to a
misallocation of resources and a drop in national

productivity amounting to roughly one percent of
the GDP. Other studies correlate the rise of
violence with the drop of investment from 1980
onward and blame that for taking two percent
points off GDP growth of the period.” 169

3.18. The adverse effects of drug tr afficking and illicit crops

are not only reflected in the private sector, but also have a

significant impact on Colombia’s national budget. Between

1978 and 2006 direct government spending by Colombia in the

fight against drugs has consistent ly increased. In that period,

169 Testimony before the Senate Foreign Relations Committee,
“Economy of Colombia”, Statement by Phillip McLean, Senior Fellow and
Deputy Director, Center for Strategic and International Studies, 29 October
2003.
Available at: http://csis.org/files/media/csis/congress/ts031029mclean.pdf
(last visited 10 March 2010), pp. 4-5.
69the total amounted to 13.3 trillion pesos at 2006 values, equal to
170
US$ 4,781 billion.

D. Environmental Impact of Illicit Crops and Drug
Trafficking in Colombia

3.19. Illicit crops and their pro cessing constitute the most

critical factor contributing to environmental degradation in

Colombia due to the deforestation of large areas of the

Amazonian tropical rainforests a nd Andean forests in order to

establish coca and opium poppy cr ops. Further, significant

pollution of the ecosystem is caused by the quantities of

pesticides, chemical precursor s and solvents used in the

cultivation and processing of illic it drugs and the disposal of
171
residues and waste products as a result of such activities.

(1) D ESCRIPTION OF RESOURCES

3.20. Colombia has an area of 1,141,748 sq km, i.e., roughly

the size of France and Spain combined. Further, Colombia is

one of the world’s richest countries in terms of species diversity

170
Annex 63: “Expenditures of the Colombian State in the fight against
drugs, 2005-2006”, National Narcotics Directorate and National Planning
171artment, Bogotá, 2008, p. 4.
Annex 127: R.E. Ramos C., J.P. Ramos B., Environmental
Assessment of the Impact of coca cr ops and the processing of coca leaf,
Universidad de los Andes, Bogotá, Colombia, 2002, pp. 1, 5, 10-11
(Conclusions No. 5, 6, 7 and 8); See also Annex 116, CICAD I, p. 15, Table
1. “Pesticides used in the production of coca”; also, “The most obvious
characteristic of coca cultivation is the degradation of ecosystem” in Annex
131-H, CICAD II: J.D. Lynch, S.B. Arroyo, “Risks to Colombian amphibian
fauna from cultivation of coca ( Erythroxylum coca ): A geographical

analysis” in Journal of Toxicology and Environmental Health , Part
A,72:974-985, 2009, p. 975.
70per unit area and is second only to Brazil in overall species

numbers; although its land area accounts for only 0.77% of the

total surface of the Earth, it is home to approximately 10% of
172
the Earth’s species of terrestrial plants and animals. Itis

therefore one of the seventeen mega-diverse countries of the
173
world. Colombia is one of the largest environmental hubs in

the planet, one of the top ten co untries in terms of areas of

primary forest, and it has a potential annual fresh water

production – in cubic kilometres – greater than India or the

continental United States.

172 J.A. McNeely, K.R. Miller, W.V. Reid, R.A. Mittermeier and T.B.
Werner, Conserving the World's Biological Diversity, (IUCN, Gland,
Switzerland; WRI, CI, WWF-US, and the World Bank, Washington, D.C.,
1990), p. 93.

Available at:http://www-
wds.worldbank.org/external/default/WDSContentServer/WDSP/IB/2000/01/
15/000178830_98101901395728/Rendered/PDF/multi_page.pdf
173st visited 10 March 2010)
Cancun Declaration of the Group of Like-Minded Megadiverse
Countries, 18 February 2002. Available at:
http://www.lmmc.nic.in/Cancun%20Declaration.pdf (last visited 10 March
2010). For a definition of the term “M egadiversity countries”, see Glossary
of Biodiversity Terms of the United Nations Environment Programme –
World Conservation Monitoring Center (UNEP-WCMC): “The small
number of countries, located largely in the tropics, which account for a high

percentage of the world's biodiversity by virtue of containing very large
numbers of species.”
Available at: http://www.unep-wcmc.o rg/reception/glossaryM-R.htm (last
visited 10 March 2010). According to UNDP, biodiversity is essential in
maintaining high levels of quality in the top soil; it helps regulate the
composition of the atmosphere and is crucial to maintaining climatic
stability, due to its role in stabilizing carbon levels. In aquatic ecosystems,
biodiversity helps purify and maintain ph levels. Biodiversity also helps
protect the forest against the cold. Some 99% of all the pests that affect
agriculture are controlled by other organisms, whose survival depends on the
preservation of biodiversity.

713.21. The Colombian territory in cludes ecosystems of such

variety as the Amazon (tropical ra inforest), the Plains (Llanos),

the Andes mountain range, the Cari bbean and the Pacific coasts

and islands. The Amazon rainforest, responsible for 15% of the

planet’s oxygen supply, covers 40 % of South America. Within
Colombia, the Amazon rainforest constitutes 35% of the

national territory. Colombia has between 45,000 and 55,000

plant species, 174representing around 16% of all plant species.

3.22. In terms of animal species, Colombia is the country with

the most birds in the world, with 1,865 bird species – 66

endemic – equal to 18% of the world’s total. It is also the

richest in amphibians, with 733 species – 400 of them endemic

– 13% of the world’s diversity. It is home to the world’s second

largest variety in butterflies, about 3,200 species. It has 2,000

species of freshwater fish, corre sponding to 10% of the world’s
diversity, and 524 species of rept iles (6%). As for mammals,

471 species are found in Colomb ia, with 32 endemic species,

i.e., the fifth most diverse count ry in that respect. Combined

figures for reptiles, birds, amphibians and mammals, add up to a

total diversity of 3,540 species. 175

174 D.G. Campbell & H.D. Hammond, (eds.), Floristic Inventory of
Tropical Countries, The New York Botanical Garden, NY, 1985.
175 Alexander von Humboldt Research Institute on Biological
Resources; CHM Colombia, Mecanismo de Facilitación del Convenio sobre
Biodiversidad Biológica, Biodiversity in Colombia , Available at:

http://www.humboldt.org.co/chmcolombia/biodiversidad.htm (last visited 10
March 2010); Tropical Butterflies Biodiversity Project, Colombia, Available
at: http://www.andeanbutterflies.org/colombia_sp.html (last visited 10 March
2010); Alexander von Humboldt Research Institute on Biological Resources,
723.23. Colombia has a system of National Natural Parks

covering a total area of 106,230 km 2, 10% of the national

territory. In the last five year s, 5 new national parks have been

created. In total, there are 257 protected natural zones in

Colombia, with a combined area larger than Belgium, Denmark

176
and the Netherlands together.

3.24. Colombia is furthermore renowned for its sound

environmental practices. It was the first country in the region to

adopt a code on natural resour ces in 1974, and to create a
177
national environmental system in 1993, that has been a

landmark reference for other countries in the region. Thus,

Colombia was ranked 9th out of 149 countries on the Yale and

Columbia University 2008 Environmental Performance
178
Index.

Fish of the Colombian Andes, p. 11. Available at:
http://www.humboldt.org.co/humboldt/homeFiles/inventarios/peces_01_01.p
df (last visited 10 March 2010).
176 Annex 151: Colombian Association of Regional Autonomous

Corporations (ASOCARS), “Shared Responsibility: The World Drugo
Problem from a Green Perspective”, Periódico Virtual, Issue N 14, 2008, p.
2.
177 Colombian Decree N o 2811 of 18 December 1974, issued by the
President of Colombia (Official Jour nal No 34.243 of 27 January 1975)
Available at:
http://www.secretariasenado.gov.co/senado/basedoc/decreto/1974/decreto_2
811_1974.html (last visited 10 March 2010); Annex 32: Colombian Law 99

17822 December 1993.
The 2008 Environmental Performance Index (EPI) ranks 149
countries on 25 indicator s tracked across six established policy categories:
Environmental Health, Air Pollution, Water Resources, Biodiversity and
Habitat, Productive Natural Resources, and Climate Change. Available at:
http://epi.yale.edu/countryscores (last visited 10 March 2010).
73 (2) L OCATION OF ILLICIT CROPS

3.25. In general terms, illicit coca and opium poppy crops are

located precisely in highly valuable forest ecosystems, including
179
Amazon soils and the And ean biodiversity hotspot. This is

explained by the fact that growers tend to select areas isolated

from urban centres, with abundant plant biomass and the

presence of plentiful water bodie s in order to set up crops and
180
laboratories and deposits of chemicals.

3.26. Illicit crops are typically lo cated in areas with a large

presence of illegal armed groups that hamper the authorities’

activities and often provide “security services” to the crops and

processing laboratories. 181

3.27. In terms of the impact on the environment, the most

troublesome aspect is the presence of illicit crops in natural

reserves, designed to preserve biodiversity against damage from

human activities. In 2008 nearly 3,450 hectares of Colombia’s

National Natural Parks were dest royed to make way for coca

crops, 182as illustrated by the map on the following page ( Figure

3.1), taken from the UNODC Colombia Coca Survey 2009. 183 

179 Annex 116, CICAD I, p. 9.
180 Annex 40: “Environmental Impact Caused by Chemical Substances,

Illicit Crops and Related Activities”, National Narcotics Directorate of
181ombia, Strategic and Research Division, 2000, pp. 5, 6.
Annex 40, p. 6. See also Annex 108, Colombia Coca Cultivation
Survey 2008, p. 70.
182 Annex 108, Colombia Coca Cultivation Survey 2008, p. 19.
183 Ibid., p. 18.
74 Figure 3.1 National Parks and coca cultivation in Colombia
(UNODC, Colombia Coca Cultivation Survey 2008, June 2009, p. 18)

75 (3) T HE COCAINE PRODUCTION CHAIN AND ITS
ENVIRONMENTAL EFFECTS

3.28. According to the findings of the second phase of the

scientific study conducted und er the auspices of the

Organization of American States, through CICAD (“CICAD

II”), “the degradation of eco systems associated with the

production of coca and its proces sing into cocaine paste and

then into cocaine hydrochlorid e, constitutes one of Latin
184
America’s most important current environmental issues”. In

Colombia, the most obvious environmental effect of coca
185
cultivation is the clearance of forests. In this regard, “the

tropical rain forests constitute the largest biome in Colombia,

though over 11 of the original 44 million hectares have been
186
lost”.

3.29. The cocaine production chain consists of different

stages, each of them damaging to the environment: clearance of

existing vegetation; the planti ng of crops; the setting up of

184 Annex 131-D, CICAD II, R.A. Brain et al. (2009), p. 945, referring
to the following studies: Armstead, L. 1992, “Illicit narcotics cultivation and
processing: The ignored environmental drama” Bull. Narcot. 44:9–20; Viña,
A., Echavarria, F. R., and Rundquist, “S atellite change detection analysis of

deforestation rates and patterns along the Colombia–Ecuador border”, Ambio
185118–125, Washington, D.C., 2004.
UNODC, Coca Cultivation in the Andean Region - A survey of
Colombia, Bolivia and Peru, June 2006, p. 19.
Available at: http://www.unodc.org/pdf/andean/Andean_report_Part2.pdf
(last visited 10 March 2010).
186 Annex 131-D, CICAD II, Brain et al. (2009), p. 945. Biome is “a
major portion of the living environment of a particular region (such as a fir
forest or grassland), characterised by its distinctive vegetation and
maintained by local climatic conditions.” At: http://www.unep-
wcmc.org/reception/glossaryA-E.htm (last visited 10 March 2010).

76clandestine laboratories and cocai ne factories; and the actual
drug processing.

3.30. The most commonly used tec hnique to clear the land is

often characterized as “slash and burn” 187consisting of the

cutting, felling, and subsequent bu rning of forests. Generally,

primary forests are felled by means of chainsaws, which leads to

increased affected areas due to the dragging effect of larger
felled trees on medium and smaller trees. On the other hand, the

burning of the forest is not a c ontrolled action since barriers to

prevent fire from expanding are either not created at all, or are

so small that they allow the flames to spread to other forest

areas, causing the indiscriminate burning of many more hectares

than are actually required for growing illicit crops.

3.31. One hectare of land devoted to coca cultivation produces

on average 7.7 kilos of pure processed cocaine hydrochloride

per year, 188one gram being the measure of a dose. On average,

coca growers clear an area four times larger than that required

for the crops themselves in order to set up clandestine

laboratories, or because ofuncontro lled forest fires. Therefore,

for each 7,700 doses of pure processed cocaine hydrochloride, 4
hectares or 40,000 square meters of forest are felled. In other

187
“The farmers deforest from the canopy to the leaf litter in order to
prepare that land for a coca plantation, leaving the habitat completely
destroyed by physical activities” in Annex 131-H, CICAD II, p. 975.
188 UNODC, World Drug Report 2006 , p. 239. Available at:
http://www.unodc.org/pdf/WDR_2006/wdr2006_chap3_cocaine.pdf (last
visited 10 March 2010).
77words, for each gram of cocaine consumed, more than 5 square

meters of tropical forest are clear cut.

(a) Lands deforested for crops

3.32. As is well-known, the destruction of an ecosystem
includes the removal of native vegetation coverage, with the

consequent loss of hydrological regulation capabilities; the

degradation of micro flora, micro fauna and as yet unknown

genetic potential; the displacement of endemic species; soil

erosion and an irreparable lo ss of biodiversity. The normal

clearance method result in air po llution from smoke and entails

a considerable increase in CO 2emissions which contribute to
189
climate change. As the 2005 study by CICAD (“CICAD I”)
explains:

“...[T]he clear-cutting of forests for the purposes
of coca and poppy production reduces
biodiversity, contribute s to the release of

189 “Easily the most visible environmen tally destructive effect of coca
and poppy cultivation is deforestation... The immediate effect of
deforestation is the reduction of natural habitat and subsequent reduction in
the bio-diversity of the region. A secondary effect of the deforestation
derives from the typical method of preparing an area for cultivation through a

slash and burn procedure. This burning is the major source of air pollution in
the jungle...” in American Univ ersity, “Colombia Coca Trade”, Trade and
Environment Database (TED) Case Studies , November 1997, pp. 2.
Available at: http://www1.american.edu/TED/colcoca.htm (last visited 10
March 2010); for definitions of reservoir and greenhouse gases sink, see
Glossary of Terms of the United Nations Framework Convention on Climate
Change.
At: http://unfccc.int/essential_background/glossary/items/3666.php#top (last
visited 10 March 2010).
78 greenhouse gases, increas es the loss of soil
nutrients, and promotes erosion of soils.” 190

3.33. Nearly 200,000 hectares of Colombian natural forest are

deforested yearly to grow co ca crops for cocaine production.

Over the last 20 years, over 2.2 million hectares – an area the

size of Slovenia and equal to half the territory of the

Netherlands or Switzerland – has been deforested for coca
191
crops. According to CICAD I, “r eturn to the conditions of

tropical old-growth forest that existed prior to clear-cutting and
burning may take hundreds of years”. 192

(b) Chemicals used

3.34. The adverse impact of illicit crops on the environment

does not end with the destruction of the forests. Several of the

agrochemicals used for growing and processing coca fall under

toxicological categories I (Extre mely Toxic) and II (Highly

Toxic), and resist biodegra dation. Among them are those
193
known as the “dirty dozen”. For example, one of the
chemicals used in processing cocaine is nitric acid which can

cause death in humans and aquatic organisms and affects

190
191 Annex 116, CICAD I, p. 14.
192An151x.
Annex 116, CICAD I, p. 11.
193 Among the chemicals used in illic it crops are those considered
worldwide as the so-called “Dirty Doze n” that have been banned in most
industrialized countries. In addition to Paraquat, the use of Lindano for
pediculosis control is common; pentachlorophenol is used as a pesticide and
Parathion, a potent herbicide with action ranging from pest control in food
crops to insect extermination. In Annex 40, p. 12.

79soils.194 It is estimated that growers of coca crops in Colombian

territory apply five million litres of herbicides, pesticides and

foliar fertilizers on a yearly basis. Overall, in order to transform

coca plants into cocaine, it is estimated that every hectare of

coca crops requires the use of approximately 127 kilos/ha of

solid precursors, 447 litres/ha of liquid precursors and 400
195
litres/ha of water.

3.35. Most of the chemical precursors used in the processing

of illicit crops are smuggled into Colombian territory across its

land borders, contrary to international instruments regulating the
196
trade and distribution of such substances. Contributing to the

criminal atmosphere of the borders is the alliance between the

drug trafficking cartels and armed groups in order to smuggle

chemical precursors for drug pr ocessing into Colombia and

smuggle out refined illicit drugs with the purpos e of sending

them to the United States and Europe.

194 Annex 118, CICAD I, Toxicology of Substances Used in the
Production and Refining of Cocaine and Heroin: A Tier-Two Hazard
Assessment (2005), p. 27.
195 Annex 127, p. 5. See also , Speech of the Colombian Vice-

President, Francisco Santos, at the University of Califor nia, Los Angeles
(UCLA), The environmental impact of cocaine, 25 February 2009. Available
at:
http://newsroom.ucla.edu/portal/ucla/electronicplay.aspx?Fid=64241&id=E0
C5478 (last visited 10 March 2010).
196 Articles 12 and 13 of the UN 1988 Narcotics Convention; Annex
145: Andean Community Decision 602 of December 2004, Andean
Regulation for the Control of chemical substances used in the illegal
manufacture of narcotic drugs and psychotropic substances; Annex 137:
Andean Cooperation Plan for the Control of Illegal Drugs and Related
Offenses, Andean Community, Decision 505 of April 2001.

80 (c) Other effects

3.36. Further along in the chain of the processing of illicit
crops, once the first harvest is obtained coca growers must turn

their harvested leaves into more transportable and more easily

concealed packages. Again, they resort to isolated areas – very

often in the tropical rainforests or other sensitive areas – where

they set about establishing processing labs in order to produce
197
coca paste or base.

3.37. Clandestine laboratories are always located near water

sources that are essential both in order to dispose of the great

amount of chemicals used during the processing of coca leaves,

and so as to be able to subm erge containers of precursor

chemicals in water in order to keep them cool and prevent their

evaporation in the warm climate of the rainforest. Since the
containers used for storing ch emicals are not airtight, they

inevitably leak noxious substances into the rivers. In terms of

chemical waste, in the producti on of 1 kg of coca paste, 625 kg

of solid waste are generated, 1.9 litres of sulphuric acid and 1.25

litres of ammonia are released into the environment, and nearly

197
For further information on the pr ocess of growing coca and cocaine
production, and its deleterious effects on the environment, see Annex 118:
CICAD, R.A. Brain et al., “The Toxicology of Substances Used in the
Production and Refining of Cocaine and Heroin: A Tier-Two Hazard
Assessment”, OAS, Washington. D.C., 31 July 2005 (part of CICAD I).
Complete document available at:
http://www.cicad.oas.org/Desarrollo_Alternativo/ENG/Projects%20By%20C
ountry/Colombia/OAS_CICAD_Tier_2_Hazard_Assessment_July_2005%5
B1%5D.pdf (last visited 10 March 2010).
81200 litres of water are contaminated. 198 Despite the high levels

of toxicity of these products, when working intensively with

pesticides coca growers use no protection, thus exposing

themselves to great health risks.

3.38. Additionally, the opening of rudimentary access trails;

the construction of facilities – namely clandestine laboratories –

for the processing of the coca leaves and extraction of coca
paste or base; the establishmen t of dumps for the storage of

chemical substances; and occasionally, the establishment of

narcotics refining facilities, dr ying zones and sleeping quarters

required for the processing of illicit crops, all add up to deepen
the negative impact on these highly sensitive ecosystems.

3.39. From all of the above, the severity of the negative

environmental impact of illicit coca crops and cocaine

processing on the ecosystems of any country, and those of
Colombia in particular, are clea rly borne out. Therefore, in

promoting eradication programs against illicit crops and other

efforts in the fight against the trafficking of illicit narcotics,

Colombia is not only complying with an international mandate,
but is also aiming at protecting the environment as the heritage

of mankind.

198 Annex 127, p. 11.

82 E. International Support for the Fight against Drugs in
Colombia

3.40. The noxious effects of illicit coca cultivation and

production in Colombia described above were for a long time

accompanied by the indifference of the international community

as a whole. Certain States in particular did not adopt effective

measures in response to the exponential increase in

consumption, nor did they seek to curtail the sale and trafficking

of chemical precursors, products and solvents involved in drug

processing, or the illicit trafficki ng of arms and explosives used
by terrorists. Others harboured the ringleaders and sheltered the

money and assets derived from these criminal activities.

Starting in 1987, at Colombia’s initiative, the notion of

“collective” 199or “shared responsibility” 200for the eradication

199 General Assembly Resoluti on 39/142, 14 December 1984;
A/RES/39/142, para. 3: “The eradication of trafficking in narcotic drugs is
the collective responsibility of all States, especially those affected by
problems relating to illicit production, trafficking or abuse.” (emphasis
added).
200
The term “shared responsibility”, began appearing consistently as
such in all relevant United Nations documents and instruments as of 1987-
1988; See e.g., General Assembly Resolution 41/127, 4 December 1986;
A/RES/41/127: “...the eradication of this scourge calls for acknowledgement
of shared responsibility in combating simultaneously the problems of illicit
demand, production, distribution and marketing...” (emphasis added).
Colombia, at the time under the strain of the scourge of drug trafficking, had
actively participated in the hemispheric and international discussions on the
issue, a landmark of which was the ministerial-level International Conference
on Drug Abuse and Illicit Trafficking held in Vienna from 17 to 26 June
1987, attended by representatives from 138 States: see e.g. United Nations,
Declaration of the International Conference on Drug Abuse and Illicit
Trafficking and Comprehensive Multidisciplinary Outline of Future Activities

in Drug Abuse Control (1988) (Available at: http://www.nzdl.org/cgi-
bin/library.cgi?e=d-00000-00---off-0cdl--00-0----0-10-0---0---0direct-10---4-
83of trafficking in narcotic drugs began to be discussed: it was
201
embraced in the 1988 UN Narcotics Convention.

(1) U NITED S TATES ’ SUPPORT AND P LAN C OLOMBIA

3.41. Since 1999, Colombia and the United States, on the basis

of the principle of shared responsibility, have concluded a series

of agreements – widely known as “Plan Colombia” – in order to
fight against illicit drugs and organized crime. The aim of these

agreements is to strengthen Colombia’s institutional capacities

and to improve the socio-econo mic situation of the most

vulnerable segments of the population by offering alternatives to

the production of illicit drugs.02

3.42. Within this framework, Plan Colombia comprises 10

strategies: Economic, Fiscal a nd Financial, Peace, National

Defence, Judiciary and Human Rights, Anti-Narcotics,

Alternative Development, Social Participation, Human

Development and Internationa l Cooperation. Colombia’s
program for the eradication of illicit crops has received funding

------0-1l--11-en-50---20-about---00-0-1-00-0-0-11-1-0utfZz-8-
00&a=d&c=cdl&cl=CL4.162&d=HASH6756f63121ab31e2e8bfbc (last
visited 10 March 2010)). See also “Shared Responsibility: Colombia’s
proposal against illicit drugs”, Commonwealth Health Ministers Book, 2007,
pp. 54-57 (available at:
http://www.unu.edu/events/files/2008/Santos_SharedResponsibility.pdf (last
visited 10 March 2010)),
201 1582 UNTS 164; E/Conf. 82/15; EM, vol. II, Annex 3, 10th

preambular paragraph: “Recognizing that eradication of illicit traffic is a
responsibility of all States and that, to that end, co-ordinated action within the
202mework of international cooperation is necessary”.
Annex 57, p. 8.
84from Plan Colombia, as part of one of the Plan’s components,

namely, the “Fight against the world drug problem and

organized crime”. 203

3.43. Resources for the implementation of Plan Colombia have
204
been provided by both countries. Using those resources, the

Colombian armed forces and National Police were modernized,

equipped and trained in order to improve their counter-narcotic
205
capabilities. During its initial phase, a significant reduction

of 46.4% in the number of hectares planted with coca crops was
206
achieved.

3.44. Plan Colombia was initially conceived for a period of 6

years, but it has been progre ssively extended with certain

variations up to the present time.

(2) S UPPORT FROM THE E UROPEAN U NION AND INDIVIDUAL

COUNTRIES FOR PLAN COLOMBIA

3.45. In addition to the support from the United States, support

for “Plan Colombia” also came from the European Union as

well as other countries like Canada and Japan. In a first meeting

in London on June 2000 “the support to the strategies

contemplated in “Plan Colombia” by the European Union

203An57ex , pp. 7-8.
204Ibid.
205 Ibid., p. 10-11.
206 Ibid., p. 12.

85started to become a reality.” 207Further meetings took place in

Madrid, 208Bogotá, 209and Brussels, 210where more than US$ 1.5

billion dollars were committed to programs associated with

“Plan Colombia.”

3.46. The strong support of the European Union to the fight

against drugs in Colombia can be evidenced in the intervention

made by Gun-Britt Andersson, th e then State Secretary for

Development Cooperation, Migrati on and Political Asylum of

the European Union, during the Brussels meeting in April 2001:

“The internal armed c onflict and the production

and traffic of illegal drugs are intertwined. The
European Union has the determination to
continue to condemn and combat the production

and traffic of illegal drugs under the principle of
shared responsibility. It is necessary that all
parties are involved in joint efforts with the view

of finding viable alterna tives to drug production.
The production and commerce of drugs can never
be justified with ec onomic arguments nor
arguing concerns about poverty.” 211

(3) S UPPORT FROM THE U NITED N ATIONS : LLICIT C ROPS

INTEGRATED M ONITORING SYSTEM (SIMCI)

3.47. In June 1998, during the 20 thGeneral Assembly Special

Session devoted to Drugs, UNODC set up an Illicit Crop

207 G. Fernández de Soto, La Ilusión Posible, Ed. Norma, Bogotá, 2004,
p. 99.
208 Ibid., p. 101.
209 Ibid., p. 105.
210 Ibid., p. 107.
211 Ibid., p. 108.

86Monitoring Programme (ICMP) in Vienna. Through this

program, which makes extensive use of satellite imagery,
member States are granted support in designing and

implementing a system for monitoring the extent and evolution

of illicit crops.12

3.48. The Illicit Crops Integrated Monitoring System – SIMCI,

from its Spanish acronym – is implemented with the support of

the ICPM. SIMCI derives from an agreement concluded

between the Colombian Governme nt and the United Nations in

1999, with the purpose of detectin g the illicit crops present in
Colombian territory by processing and interpreting satellite

imagery, and publishing an annual survey of such crops in

Colombia on that basis. The statistics produced also cover

potential cocaine production a nd comparative analyses with
213
previous years.

3.49. The work resulting from this project has allowed the

relevant Colombian authorities to collect reliable information on
coca crops in order to improve eradication operations, as well as

to finalise studies and research activities on issues relating to

such crops.

212
UNODC Illicit Crop Monitori ng Programme (ICMP). At:
http://www.uncosa.unvienna.org/uncosa/en/directory/unodc/index.html (last
213ited 10 March 2010).
Annex 95: Description of the Illicit Crops Integrated Monitoring
System (SIMCI), p. 2.
87 (4) S UPPORT FROM THE O RGANIZATION OF AMERICAN
S TATES: NTER -AMERICAN D RUG ABUSE C ONTROL COMMISSION
(CICAD)

3.50. In light of the serious global and regional situation posed
by the problem of production and consumption of illicit drugs,

the Inter-American Drug Abuse Control Commission – CICAD

from its Spanish acronym – wa s established in 1986 as an

agency of the Organization of American States (OAS) as a

specialized policy forum for matte rs relating to illicit drugs in

the Americas.

3.51. CICAD has become the most important body in the

Inter-American context in the fight against the drug problem and
has been paid particular attention from all States of the

continent. Its tasks cover a wide range of activities including

follow-up on drug consumption, the related social and economic

problems, prevention and rehab ilitation, as well as providing

basic information to institutions dealing with the problem in

each State in the hemisphere.

3.52. The OAS, in particular through CICAD, has cooperated
with the Colombian Government . CICAD provided technical

assistance in order to establih a national “Observatory on

Drugs”,214based on the Drug Information System of Colombia

214
Annex 114: Annual Report of the Inter-American Drug Abuse
Control Commission (CICAth to the General Assembly of the Organization
of American States at its 34ular Session, 17-20 November 2003,
88(SIDCO), set up in 1996 by the Colombian Government. The

Observatory, the activities of which enjoy CICAD’s continuing

support, is entrusted with consolidating domestic information on

the abuse of lawful substances and the production, trafficking,

offer and consumption of illicit drugs and related crimes,
enabling its distribution both domestically and internationally.

CICAD has also sponsored national and international

conferences and meetings on vari ous aspects pertaining to the

drug problem, including the activities of maritime drug

trafficking and those that take place in coastal waters and zones
215
and port facilities.

3.53. CICAD has provided cons ultative assistance to

Colombia’s National Plan for Alternative Development; for an

important programme of socio-economic development as an

alternative to coca cultivation aimed at benefiting a Cofán
indigenous community and the indigenous localities of Valle del

Guamuez and San Miguel in Putumayo province; 216also, for

projects in the areas of agri culture and cattle raising, fish-

breeding, and the sale of dairy products.

Montreal, Canada, 3 May 2004, p. 26. OEA/Ser.L/XIV.2.34
CICAD/doc.1264/03 rev. 2.
215 Annex 112: Annual Report of the Inter-American Drug Abuse
Control Commission (CICAD) to the General Assembly of the Organization
nd
of American States at its 32 Regular Session, 12-15 November 2001,
Caracas, Venezuela, 3 April 2002 , pp. 7-8. OEA/SER.L/XIV.2.30
216AD/doc.1123/01 rev.1.
Ibid., p. 13.
893.54. On 4 February 2004, Colombia and the Secretary-

General of the OAS entered into a Memorandum of

Understanding, pursuant to which a group of scientists, under

the auspices, funding and supervision of CICAD, carried out an

independent scientific study on th e possible risks or effects on

human health and the environment of the use of the herbicide

glyphosate for the control of coca and poppy illicit crops in

Colombia. 217 The results of the st udy (known as CICAD I),

discussed at Chapters 4 and 7 below, were presented to the OAS

Secretary-General and the OAS Permanent Council. 218

217
Annex 113: 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 the environment, 4

218ruary 2004. th
3t5e t Regular Session of CICAD, Washington, D.C., 27-30
April 2004, a progress report on the first phase of CICAD’s study was
presented (Annex 115, Session Documents, p. 2). Likewise, during 37 th
Regular Session of CICAD, Santo Domingo, Dominican Republic, 26-29
April 2005, the conclusions of the first phase of CICAD’s study were
presented. (Annex 117: 37 thRegular Session of CICAD, Santo Domingo,

Dominican Republic, 26-29 April 2005, Document 1421). Also: “...A study
of the environmental and human health impact of aerial spraying with the
glyphosate herbicide on coca and poppy crops in Colombia was completed
and the findings officially presented to the Government of Colombia, and to
the scientific community and the general public through CICAD’s website.”
In Annex 120: Annual Report of the Inter-American Drug Abuse Control

Commission (CICAD) to the Gtheral Assembly of the Organization of
American States at its 36 Regular Session, Santo Domingo, Dominican
Republic, 4-6 June 2006, p. 13. The Colombian Vice-Minister furnished all
the members of the Permanent Counc il with a copy of CICAD I at the
Council’s Session of 9 January 2007, where the issue of the sprayings near
the border was examined (see OAS, CP /INF.5432/07, p. 4, available for
download at: http://www.oas.org/cons ejo/Documents%20INF2007.asp (last

visited 10 March 2010). The Pe rmanent Council in cluded the Vice-
Minister’s statement in its 2006-2007 Report to the General Assembly
(Annex 121: Annual Report of the Permanent Council to the General
90Likewise, the results were o fficially communicated to the

Ecuadorian Government by Colombia, under cover of a

diplomatic Note dated 6 May 2005. 219

3.55. Ecuador argues that this study was “Colombia's

unilateral initiative” and alleges that it was not informed or
220
afforded an opportunity to participate in the study. However,

the CICAD study had been commissioned in order to analyse

the effects of the aerial sprayings in Colombia . In any case,

acknowledged by Ecuador in its Memorial, a copy of the first
221
CICAD study was sent to it.

3.56. The work of CICAD also contemplated a second phase,

which was conducted pursuant to a further Memorandum of

Understanding concluded on 23 May 2006 between Colombia

and the OAS Secretary-General with a view to obtaining further

in-field and laboratory confirmation of the results of the first
222
phase, and dealing with certain unresolved issues. Asinthe

first phase, the study was carried out by an independent team of

Assembly of the Organization of American States 2006-2007, pp. 33-34.
OAS, AG/doc.4698/07).
219
EM, Vol. II, Annex 67 (Diplomatic Note DAA/CAL 23927 of 6
May 2005, from the Ministry of Foreign Affairs of Colombia to the Ministry
220Foreign Affairs of Ecuador).
221 EM, para. 3.36.
EM, para. 3.47.
222 Annex 119: 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, 23
May 2006.

91scientists. CICAD II involved several environmental and health

issues including, among others, the influence of spray

procedures and conditions on spra y drift, and the assessment of

toxicity of the spray mixture to amphibians. Mention of the

CICAD II study’s scope and progress was included in CICAD’s

2007 Report to the OAS General Assembly, 223approved during

rd
CICAD’s 43 Regular Session held on 30 April-2 May 2008,
224
which Ecuador attended.

3.57. Once completed, most of the conclusions of the follow-

up study were presented to CIC AD by the independent experts

in November 2008, and an explan ation of the study’s scope as

well as mention of its completion were again included in

223 Annual Report of the Inter-American Drug Abuse Control
Commission (CICAD) to the General Assembly of the OAS at its 38 th

Regular Session, OAS Document OEA/Ser.L/XIV.2.43, CICAD/doc.1656/08
rev.1, 2 May 2008, pp. 14-15. Available at:
http://www.cicad.oas.org/AnnualReports/2007/AnnualReportCICAD-2007-
eng.pdf (last visited 10 March 2010). The pertinent excerpt reads as follows:
“Colombia: Study on the Effects of Aerial Glyphosate Spraying
Since 2005, CICAD’s scientific evaluation team has been working on an
independent study, undertaken at the request of the governments of
Colombia, the United States and the United Kingdom, to measure the impact

of aerial spraying of coca fields in Colombia on human health and the
environment. In follow-up studies in 2007, CICAD’s team carried out
research analyzing the following components:
[...]
• Analysis of the drift from aerial spraying with Glyphosate
and Cosmo-Flux, as employed in the Colombian
eradication program. Tests duplicating wind conditions
on the Ecuadorian-Colombian border took place in a wind

224 tunnel at the University ofrdueensland in Australia.”
See Final Report of the 43 Regular Session of the CICAD, 30
April-2 May 2008, Washington, D.C. OAS Document OEA/Ser.L/XIV.2.43,
CICAD/doc. 1672/08 rev 1, 29 July 2008, p. 9 (section 13 - Participants).
Available at: http://www.cicad.oas.org/apps/Document.aspx?Id=675 (last
visited 10 March 2010).
92 225
CICAD’s 2008 Report to the OAS General Assembly, and
th
approved during CICAD’s 45 Regular Session held on 6-8
226
May 2009, which Ecuador also attended. The OAS General

Assembly adopted CICAD’s 2008 Annual Report on 4 June

2009. 227 The study, th e results of which are extensively

discussed in this Counter-Memor ial, was published in the peer-

reviewed Journal of Toxicology and Environmental Health. 228

225
Annex 122: Annual Report of the Inter-American Drug Abuse
Control Commission (CICAD) to the General Assembly of the OAS at its
39thRegular Session, San Pedro Sula, Honduras, 2-3 June 2009. OEA/Ser.G,
CP/doc.4395/09 corr. 1, 26 May 2009, p. 12. The pertinent excerpt reads as
follows:

“Colombia: Study on the Effects of Aerial Glyphosate Spraying
In 2005, CICAD’s scientific evaluation team presented the results of an
independent study, undertaken at the request of the governments of
Colombia, the United States and the United Kingdom, to measure the impact
of aerial spraying of coca fields in Colombia on human health and the
environment. Although no association between spraying and human

reproduction was found, the team proposed to carry out additional studies to
identify possible risk factors associated with other human activities or the
environment. The independent scientific evaluation team that CICAD hired
in 2006 presented most of its findings of the follow-up study on the human
health and environmental evaluation of the aerial spraying to control coca
and poppy crops in Colombia in late 2008. The completed study, which

consisted of several technical articles, was submitted for consideration in the
peer-reviewed scientific periodical Journal of Human and Environmental
Toxicology...”
226 See Final Report of the 45 thRegular Session of the CICAD, 6-8
May 2009, Washington, D.C. OAS Document OEA/Ser.L/XIV.2.45,
CICAD/doc. 1742/09, 9 July 2009, p. 11 (section III - Participants).

Available at: http://www.cicad.oas.o rg/APPS/Document.aspx?Id=800 (last
227ited 10 March 2010).
OAS Document AG/RES. 2493 (XXXIX-O/09). Available at:
www.oas.org/dil/AG-RES_2493-2009_eng.doc (last visited 10 March 2010).
228 Annex 131, Annex 131-A to 131-I (CICAD II) [9 articles]:
Journal of Toxicology a nd Environmental Health, Part A Current

Issues, Volume 72, Numbers 15 & 16, Taylor & Francis, London,
2009.
933.58. The study, in its field phase, was carried out by the

research team in different areas within Colombia. This is in

sharp contrast to the Menzie Report, submitted by Ecuador as its

main scientific evidence (EM, Annex 158). Indeed, only one of

the Exponent team members that prepared the Menzie Report

visited the border zone in Ecuador and that person merely made
229
field observations and collected testimonies, without taking

any samples or conducting actual field tests.

F. Colombi a’bsligatono Take Action against Illicit
Crops

3.59. In accordance with the provisions of the international

instruments in force concerning the world drug problem, and, in

particular, the 1961 Single Convention on Narcotic Drugs as

amended by its 1972 Protocol, 230the 1971 Convention on

Psychotropic Substances, 231 the 1988 United Nations

Convention against Illicit Traffic in Narcotic Drugs and

Psychotropic Substances, 232and other related instruments,

229
230 EM, Vol. III, Annex 148, pp. ix, 1.
976 UNTS 105 (consolidated version). For the 1961 Convention as
originally adopted, see 520 UNTS 151; for the 1972 Protocol, see 976 UNTS
3.
231 1019 UNTS 175.
232 “As of March 2008, 183 countries were parties to the 1988
Convention (i.e., 95% of all United Nations Member States, having more
than 99% of the world’s total population). Non-parties to the Convention are
just three countries in Africa (Equatorial Guinea, Namibia and Somalia), one
country in Asia (Timor Este), one country in Europe (Holy See), and seven
island countries in the Oceania regio(Kiribati, Marshall Islands, Nauru,

Palau, Papua New Guinea, Solomon Islands, Tuvalu).” Op. cit., UNODC, A
Century of International Drug Control, p. 68.
94Colombia is obliged to take effective measures to suppress the

production and consumption of illicit drugs.

3.60. The 1988 United Nations Convention against Illicit
Traffic of Narcotic Drugs and Psychotropic Substances, 233the

very instrument invoked by Ecuador in its Application before

the Court, was inspired by the need to tackle the scourge of

illicit drugs in a global fashion, setting a framework of State
obligations in that regard. As stated in the Preamble to the 1988

Convention, the States Parties

“Deeply concerned by the magnitude of and

rising trend in the illicit production of, demand
for and traffic in narcotic drugs and psychotropic
substances, which pose a serious threat to the
health and welfare of human beings and
adversely affect the economic, cultural and
political foundations of society;

[…]

[Recognized] the links betw een illicit traffic and

other related organized criminal activities which
undermine the legitimate economies and threaten
the stability, security and sovereignty of States;

[…]
[Recognized] also that illicit traffic is an

international criminal activity, the suppression of
which demands urgent a ttention and the highest
priority;

[…]

233 1582 UNTS 164; E/Conf. 82/15; EM, Vol. II, Annex 3.

95 [Recognized] that eradication of illicit traffic is a
collective responsibility of all States and that, to
that end, co-ordinat ed action within the
framework of international co-operation is

necessary...”

3.61. These concerns find concrete expression in Article 2 of

the 1988 Convention, which provides:

“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 ob ligations 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.”

3.62. It is in that cont ext that Article 14 of the Convention
stipulates that:

“Each Party shall take appropriate measures to

prevent illicit cultivation of and to eradicate
plants containing narcotic or psychotropic
substances, such as opium poppy, coca bush and
cannabis plants, cultivated illicitly in its territory.
The measures adopted shall respect fundamental
human rights and shall take due account of
traditional licit uses, where there is historic

evidence of such use, as well as the protection of
the environment.”

3.63. As acknowledged in the 1998 United Nations “Action
Plan on International Cooperation on the Eradicati on of Illicit

96 234
Drug Crops and on Alternative Development”, “[w]hen there

is organized criminal involve ment in the illicit drug crop

cultivation and drug producti on, the measures such as

eradication, destruction of illicit drug crops and arrests … are
235
particularly appropriate.” Moreover, as also acknowledged
by the Action Plan, even when alternative development projects

are successful, some growers and processors are not likely to

abandon production voluntarily simply because more lucrative

opportunities exist, or because pressure may be exerted by

illegal armed groups; such growers must see that there is a risk
236
associated with pursuing the illicit cultivation of drug crops.

Accordingly, the Action Plan stip ulates that “[i]n areas where

viable alternative sources of income already exist, law

enforcement measures are requi red against persistent illicit
237
cultivation of narcotic crops.”

3.64. Colombia’s obligation to combat illicit crops is

paralleled by the obligation of all States to cooperate in the fight

against the world drug problem . Thus, the 1998 UN Action

Plan affirmed in its Preamble that “the fight against illicit drugs

must be pursued in accordance with the provisions of the

international drug control treaties , on the basis of the principle

234 General Assembly Resolution S-20/4-E, 10 June 1998 (Measures to
enhance international cooperation to counter the world drug problem: Action
Plan on International cooperation on the Eradication of Illicit Drug Crops and

235Alternative Development); A/S-20/14, pp. 19-22.
236 Ibid., §29.
237 Ibid., §27.
Ibid., §30.
97 238
of shared responsibility”. The 1988 UN Convention sets

forth in its preamble the principl e of collective responsibility of

all States, that is “not only, or even especially, those affected by

problems relating to illicit produ ction, traffic or abuse” to
239
cooperate in the suppression of illicit drugs. Thus, “the
preamble places the Convention in the category of multilateral

treaties responding to general concerns or interests affecting all

States within the internati onal community in a similar

manner”. 240 For this reason international organizations such as

the UN and the OAS have reaffirmed their commitment to seek

to eradicate illicit drugs and have given strong support to

Colombia’s own struggle against them.

G. Conclusions

3.65. Colombia is the country that has most decisively

maintained the struggle agains t the world drugs problem. In

fulfilling its obligations to the international community in this

regard, it has contributed to th e security and health of youths

and adults around the world, while ensuring the stability of the

State. But in this struggle Colombia has had to pay a very high

cost both in human lives and economic resources.

238 General Assembly Resolution S-20/4-E, 10 June 1998 (Measures to
enhance international cooperation to counter the world drug problem: Action
Plan on International cooperation on the Eradication of Illicit Drug Crops and
on Alternative Development); A/S-20/14, Preamble, first para.
239
United Nations, Commentary to the United Nations Convention
against Illicit Trafficking in Narcotic s and Psychotropic Substances (New
240k, 1999), E/CN.7/590, para. 0.21.
Ibid.
983.66. The scourge of illicit drugs has affected Colombia in a
variety of ways over the years. Financing of illegal armed

groups with the revenue of drug tr afficking, political instability,

kidnapping and murder of hundreds of prominent national

figures and thousands of comm on citizens, and the pervasive
influence of drug trafficking in many segments of the country’s

political, economic and social life seriously affected Colombia’s

institutional structure.

3.67. The progressive environmen tal deterioration and the

adverse impact on the Colombian population caused by the

cultivation and production of il licit substances are just as
serious. As described above, Colombia is the second most

biologically divers e country in the world and hosts a

considerable part of the Amaz on rainforest, the planet’s lung
and home of several indigenous communities. However, it is

precisely in the most biologically diverse zones of the country

that illicit crops are found, given that the growers seek places

with dense vegetation and plentifu l water supplies from natural
sources, in order to keep the cr ops and clandestine laboratories

from being easily detected and to dispose of the chemicals used

in the processing of alkaloids. As a consequence these illegal
activities seriously imperil the subsistence and cultural

development of the communities located in the areas where the

illicit crops are planted.

993.68. Illicit crops and their processing are the most critical

factors for environmental damage in Colombia, given the
deforestation and forest burning involved in setting them up,

and the highly toxic nature of the chemicals used during the

growing of the crops and their processing – many of which are

smuggled across the borders. The damage caused by clear
cutting the forest is such that it can take over a hundred years for

a single affected hectare to be re stored to its original state.

Hence, in taking steps to eradicat e illicit crops, Colombia is not
only fulfilling its international obligations to fight against the

world drug problem, but it is also endeavouring to protect both

its population and the environment.

3.69. A number of States and inte rnational organizations have

cooperated with Colombia in its fight against illicit drugs. In

particular, the United States’ Government has contributed funds
to Plan Colombia, the first phase of which achieved a reduction

of 46.4% of the areas affected by illicit crops. The second phase

is still ongoing and is producing results that are just as effective.

The European Union has likewise cooperated with Colombia in
various ways in this struggle. The United Nations, through

SIMCI, has assisted Colombia to obtain reliable information as

to the location of coca crops for the purposes of its eradication
efforts. The OAS has taken part in several projects, including

the Colombian National Drug Observatory and through CICAD

has conducted studies on the eff ects of illicit crops and on the
impact of the eradication program by means of aerial spraying.

1003.70. Colombia’s fight against illicit drugs is aimed at
fulfilling its obligations under inte rnational law in relation to

illicit drugs derived mainly from the Single Convention on

Narcotics Drugs of 1961, the 1971 Convention on Psychotropic

Substances and the United Nati ons Convention Against Illicit
Traffic in Narcotic Drugs and Psychotropic Substances of 1988.

Ecuador is also bound by these obligations.

3.71. It is particularly important for Colombia to be able to

rely on decisive cooperation from bordering and neighbouring

countries in its strenuous fi ght against the production and

trafficking of illicit drugs. Otherwise, the hefty costs and
sacrifices the country has had to endure will have been useless.

101102 Chapter 4

THE PROGRAM FOR THE ERADICATION OF
ILLICIT CROPS BY AERIAL SPRAYING WITH

GLYPHOSATE (PECIG)

A. The Need for Aerial Spraying

4.1. Following the implementation of satellite detection

techniques in 1999, 241it became possible to identify the location

of illicit crops with greater accuracy. In 2000 the Colombian

provinces of Nariño and Putumayo bordering Ecuador

harboured 46% of the total area of illicit crops de tected in the

country, a remarkable figure sin ce the sum of th eir respective
242
territories amounts to only 5% of the nati onal territory. As

recalled above, in 2000, Colomb ia produced some 77% of the
world’s coca leaf: the significance of the two border provinces

to the world drug supply chain at that time is obvious.243

4.2. As the planted areas grew progressively more extensive,

plantations were increasingly locat ed in remote areas and were
244
protected and even operated by illegal armed groups. Indeed,

in Nariño and Putumayo – which up until that time had been

peaceful – there was an alarming increase in killings,

241
242 See above, paras. 3.48-3.49.
243 Annex 103, Colombia Coca Cultivation Survey 2003, p. 15.
244 See Annex 102, p. 67.
See para. 3.1 and note 154 above.
103kidnappings, and attacks against civilians, the Police and the
Army. Those responsible were illegal armed groups as well as

drug traffickers faced with obstr uction by the State of their

illegal trade.

4.3. As a consequence of that situation, as well as the

opposition of the growers of the illicit crops to eradication

efforts, in 2000 the Colombian Government increased the use of
aerial spraying as the principa l method of eradication. In

particular, it decided to impl ement the program now known as

the Program for the Eradication of Illicit Crops by Aerial

Spraying with Glyphosate (PECIG, from its Spanish acronym)

in all affected portions of its territory, including the provinces of
245
Nariño and Putumayo.

4.4. At the same time, the Colombian Government continued

manual eradication efforts in t hose areas of the country where

security conditions permitted. However, this is extremely

245
The Commentary to Article 14 of the 1988 UN Narcotics Convention
reads as follows: “There are several el ements and technologies available for
use in eradication operations, including manual and mechanic uprooting and
manual or aerial spraying [...] the use of agents to eradicate such plants will
depend on a series of factors such as geographic location, climate,
topographic features and prevailing socioeconomic conditions, to which it
shall correspond.” United Nations, Commentary to the United Nations
Convention against Illicit Trafficking in Narcotics and Psychotropic
Substances (New York, 1999), para. 14.27.
104dangerous, due to attacks on the manual eradicators and
246
accompanying personnel.

4.5. Thus Colombia’s eradication strategy continues to be

threefold. Eradication is carried out through the following

methods:

(1) Forced manual eradication where security

conditions permit and in sensitive areas such as

natural parks;

(2) voluntary manual eradication, as part of
alternative development initiatives; and

247
(3) aerial spraying of larger and remote areas.

4.6. Aerial spraying continues to be the most effective

method for large-scale eradication of illicit crops. It continues

to be carried out in Colombian territory in conformity with the

246 Despite the security measures provided by the National Defence

Ministry, skirmishes with the illegal armed groups, antipersonnel mines,
attacks and accidents have taken the lives of eradicators and Police and Army
personnel involved in or accompanying manual eradication tasks. Thus, in
the period between 2006 and 2009, 39 civilian manual eradicators were killed
and 149 injured while performing manual eradication tasks. Also, 69
policemen were killed and 197 injured. In contrast, duringaerial spraying
operations in the period 2001-2009, 8 policemen have died and 18 have been
injured. See Annex 72: Note N ° 0958 ARECI/JEFAT from the Anti-
Narcotics Direction of the Colombian National Police (DIRAN) to the
Colombian Foreign Ministry, 18 February 2010, p. 2; and Annex 73: Note N°
20103291383181 from the Presidential Agency for Social Action and
International Cooperation ( Acción Social ) to the Colombian Foreign
Ministry, 23 February 2010. Also, see further below, para. 4.37.
247
Annex 66: Report by the National Narcotics Directorate (DNE),
2010, p. 2.
105strict procedures laid down by the Colombian legal system and

with due respect for the environment.

4.7. This Chapter gives a full account of the origins,

implementation and outcomes of the PECIG program. It
provides the necessary factual and technical basis for the

consideration, in Part II of this Counte r-Memorial, of the

scientific and other evidence (Chapter 7) and of the legal issues

raised by Ecuador’s claim (Chapters 8 and 9). The account in

this Chapter is supported and supplemented by the following

annexed reports of the relevant Colombian departments or

agencies involved in the program:

• Report by the Anti-Narcotics Direction of the

Colombian National Police (DIRAN); 248

• Report by the Colombian Agriculture and Livestock
249
Institute (ICA);

• Report by the Ministry for the Environment, Housing

and Territorial Development; 250

• Summary of Activities carried out by the Ministry of

Social Protection and the National Health Institute
251
(INS);

248 Annex 67: Report by the Anti-Narcotics Direction of the Colombian
National Police (DIRAN), 2010.
249
Annex 65: Report by the Colombian Agriculture and Livestock
250titute (ICA), 2010.
Annex 70: Report by the Ministry for the Environment, Housing and
Territorial Development, 2010.
106 • Report by the National Narcotics Directorate

(DNE). 252

B. Scientific Assessments Supporting the Aerial
Spraying Program

(1) S TUDIES LEADING TO THE ENVIRONMENTAL

M ANAGEMENT PLAN

4.8. In 1992, after carrying out re levant scientific studies 253

and assessing the existing situation, the Government of

Colombia, through the Nati onal Narcotics Council ( Consejo

Nacional de Estupefacientes ) – the highest domestic authority

for such matters – authorized controlled fumigation with

glyphosate, initially for poppy cr ops in limited parts of the

Colombian territory. 254

251
Annex 68: Report by the National Health Institute (INS),
“Summary of Activities Carried Out by the Ministry of Social Protection and
the National Health Institute with Rega rd to the Program for the Eradication
of Illicit Crops with Glyphosate Herbicide – PECIG”, 2010.
252An66.x
253 These studies included experime ntal assessments of glyphosate
application in the Natural Park at the Sierra Nevada in Santa Marta in the
1980s. SGS Colombia S.A., under contract to the National Narcotics
Directorate, analyzed the results of several soil and water samples following

the application of glyphosate on the Sierra Nevada in Santa Marta; it found
no traceable contamination due to glyphosate. See Annex 123: SGS (Societé
Génerale de Surveillance, S.A.) Colombia S.A., “Report on Contamination
Control for glyphosate application at the Sierra of Santa Marta”, 1987, pp. 2,
5, 6, 7, 8, 12.
254 Annex 31: Communiqué of the National Narcotics Council of
Colombia to public opinion on the eradication of illicit poppy crops, 31
January 1992.

1074.9. In October 1993, both the Health Ministry and the

National Institute for Natural Renewable Resources and the
Environment (INDERENA, the predecessor to the present

Ministry for the Environment) issued a favourable opinion in

relation to the use of controlled aerial spraying with glyphosate

herbicide for the eradication of coca and marijuana crops. 255

4.10. In December 1993, the Law on the Environment was

passed. In 1994, a regulatory decree adopted under that Law

provided that activities – includi ng the aerial spraying of illicit

crops – that had been authorized and begun prior to that date
could continue, but that the relevant environmental authority

could require the establishment of an Environmental

Management Plan. In accordance with the provisions

establishing a transitional regime, that Plan is equivalent to an
256
environmental impact assessment.

4.11. The new legal regime expressly set out the activities

requiring prior environmental authorization; the application of
pesticides was not included among them. However, the

National Government, going beyond the legal requirements,

regarded it as appropriate to require an Environmental

255
Annexes 35 and 36: Notes of 8 and 11 October 1993, signed by the
General Manager of INDERENA (National Institute for Renewable Natural
Resources and the Environment) and the Health Minister, respectively,
authorities in charge of preserving the environment and protecting the health
256the population.
Annex 32: Colombian Law 99 of 1993, Article 57; and Annex 38:
Colombian Decree 1753 of 1994, Articles 1 and 38.
108Management Plan (EMP), taking the new regulations into

account.

4.12. Subsequently, the Ministry for the Environment took
steps with a view to establishing the EMP for the PECIG

program. Thus, in 1996 it issued the terms of reference for the

environmental study to be carried out by the National Narcotics

Directorate.

4.13. The Ministry for the Envi ronment 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 formally adopted by

Resolution 1065 of 2001. However, environmental

management programs are meant to be dynamic instruments.

As a result of further experiences acquired in the PECIG

program and the management of its EMP, the latter was
amended by Resolution 1054 of 2003, 257which inter alia

included other government agencies in its execution.

4.14. To summarize, at the time the Colombian Government
began conducting aerial spraying op erations in the provinces of

Nariño and Putumayo in 2000, it did so in the framework of the

legislation in force with regard to health and the environment.

In particular it drew up the EMP. That plan, together with the

257 Annex 50: Resolution Nº 1054 of 30 September 2003 of the
Ministry for the Environment of Colombia.

109amendment adopted in 2003, has governed the PECIG up to the

present time.

(2) S UBSEQUENT STUDIES :T HE CICAD PROCESS

4.15. In 2001, the Colombian Government decided to request

CICAD, an impartial, specialized agency of the Organization of

American States, concerned with drug-related issues in whose
258
work both Colombia and Ecuador participate, to carry out a

scientific assessment of the impact on human health and the

environment of the use of glyphosate for the control of coca and
poppy crops in Colombia. The Un ited States and the United

Kingdom (although the latter was not an OAS Member) made

similar requests.

4.16. Following a careful process to establish the terms of

reference, agreement was reached with CICAD and a

Memorandum of Understanding wa s signed in February 2004,
259
by Colombia and the Secretary General of the OAS. The
study to be undertaken was intend ed to deal with the use of

glyphosate in the entire Colombian territory and not only its use

in the border area with Ecuador.

4.17. On 15 October 2004, during a meeting in Esmeraldas,

the President of Colombia informed the Ecuadorian President of

258
httepe://www.cicad.oas.org/EN/ AboutCICAD.asp (last visited 10
259ch 2010).
See above, para. 3.54.
110 260
the study Colombia had reque sted CICAD to carry out. In

addition to information concerning the study’s progress and

completion, which were transmitted to all CICAD members
261
throughout, the results of the study were officially
262
communicated to Ecuador on 6 May 2005. The Scientific and
263
Technical Commission analyzed the results of the study.

4.18. On 23 May 2006 a second Memorandum of

Understanding was concluded be tween Colombia and the OAS

Secretary-General with the purpose of conducting a second

phase of the study (CICAD II). 264 This was completed in 2008

and published in 2009.

4.19. Despite the fact that the Ecuadorian Government on

several occasions was invited to take part in the CICAD
265 266
study, it declined to participate at all. However, the

Ecuadorian Government was kept informed of its progress and

completion in the course of CI CAD regular sessions, as well as

through CICAD’s reports to the OAS General Assembly. 267

260 Annex 17, p. 2.
261 See above, para. 3.54-3.58.
262
263 EM, Vol. II, Annex 67.
264 EM, Vol. III, Annex 153.
See above, para. 3.56.
265 Annex 24: Diplomatic Note DM/VRE/DPM/CDR/DDA/CAL Nº
44664 from the Colombian Foreign Minister to the Ecuadorian Foreign
Minister, 5 September 2006. Annex 25: Diplomatic Note Nº 39064/06-
VM/SSNDF/DGRFC from the Ecuadorian Foreign Minister to the CICAD

266cutive Secretary, 20 September 2006.
267An25.x
See above, para. 3.56-3.58.
111 (3) C ONCLUSIONS

4.20. The PECIG program was based on investigations and
reports concerning potential imp acts of aerial spraying, going

back to the INDERENA opinion of 1993. 268 These studies have

been substantially amplified and developed since 2001, most

notably through the two CICAD studies of 2005 and 2009. This

has occurred without challengi ng, still less undermining, the

overall positive appraisal of the PECIG program and its

implementation, which the Colombian Government had already
adopted in 2000. These later scientific studies will be examined

in detail in Chapter 7.

C. Implementation of the PECIG Program

4.21. The PECIG eradication program is set out and
implemented under a series of Resolutions, including

Resolutions 001 of 1994 and 013 of 2003, which govern the

involvement of the different Colombian agencies responsible for

268 Annex 35: 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 October
1993.
112the program. 269 In particular, the program is carried out on the

basis of the EMP of 2001 as amended. 270

4.22. Furthermore, in 2001 the Colombian Government

introduced a compensation program to deal with any claims as

to the harmful effects that might incidentally occur in relation to

lawful crops in the immediate vicinity of sprayed illicit crops.271

In this regard, between 2002 and 2008, 117 individuals who had

addressed complaints concerning damages to lawful crops were

compensated. 272 Many more complaints have been rejected for

want of satisfactory proof of harm or for other reasons.

(1) T HE SUPERVISORY FRAMEWORK –E NVIRONMENTAL
M ANAGEMENT P LAN

4.23. The Environmental Management Plan is a set of rules

and procedures that must be followed and observed by all the

agencies that directly or indirect ly take part in the Program for

the Eradication of Illicit Cr ops by Aerial Spraying with

Glyphosate (PECIG). Its purpose is to ensure that the

269 Annex 37: Resolution Nº 001 of 11 February 1994 of the National
Narcotics Council of Colombia; Annex 41: Resolution Nº 005 of 11 August
2000 of the National Narcotics Council of Colombia; Annex 49: Resolution
Nº 013 of 27 June 2003 of the National Narcotics Council of Colombia. See
also Annex 66, p. 1; Annex 67, p. 1.
270
EM, Vol. II, Annex 15 (Resolution Nº 1065 of 26 November 2001
of the Ministry for the Environment of Colombia); modified by Resolution
Nº 0099 of 31 January 2003 of the Ministry for the Environment of Colombia
(CCM, Annex 48); CCM, Annex 50.
271 Annex 43: Resolution Nº 017 of 4 October 2001 of the National
Narcotics Council of Colombia; Annex 61: Resolution Nº 008 of 2 March
2007 of the National Narcotics Council of Colombia.
272 Annex 67, p. 11.

113implementation of the PECIG eradication program is in

accordance with the environmental provisions in force, to verify

the effectiveness of the spraying of illicit crops and to assess any
273
effects on surrounding areas.

4.24. The EMP sets out measures for the inspection,

verification and control of spraying operations, industrial

security at sites used in sp raying operations, management of
solid waste and residual water, and environmental monitoring

through the collection and analysis of soil and water samples. It

also foresees a public hea lth program, a program of

communications and social management, and a contingency

management plan. 274

4.25. The State agencies that take part in the implementation

of PECIG are responsible for compliance with the EMP. 275 As

for the verification of the effec tiveness of aerial spraying, other

entities are involved as observers, including, among others, the

Ministry of Social Protecti on, the Colombia n Agricultural

Institute, the Attorney General’s office, and the office of the
276
Public Prosecutor.

273 EM, Vol II., Annex 15; CCM, Annex 50.
274 EM, Vol II., Annex 15; CCM, Annex 50.
275 Annex 65, pp.1-3; Annex 66, pp. 2-4; Annex 67, p. 1; Annex 68, pp.

276; Annex 70, p.2.
EM, Vol II, Annex 15; CCM, Annex 50.
1144.26. The Ministry for the Environment oversees the

implementation of the EMP and verifies compliance with the

guidelines and duties foreseen in it. Two reports per year are

submitted to the Ministry, which may issue rulings on the

activities carried out by the agencies involved in the
277
implementation of the PECIG program.

4.27. Although Ecuador annexed the text of Resolution 1065
of 2001 to its Memorial, it failed to translate the passage

concerning the verification carried out by Colombia’s Ministry

for the Environment. That verification is carried out through a

technical visit to a spraying base, in order to assess the level of

compliance by the relevant authorities involved in the execution

of the PECIG with the operational parameters as well as norms

relating to industrial safety, storage of spray mix substances and

final disposal of residues at the spraying base. 278

4.28. The PECIG program is also overseen by an external

technical audit, contracted y early through public tender with
279
resources provided by the National Narcotics Council. The

277
EM, Vol II, Annex 15 (Resolution Nº 1065 of 26 November 2001 of
the Ministry for the Environment of Colombia); modified by Resolution Nº
0099 of 31 January 2003 of the Ministry for the Environment of Colombia
(CCM, Annex 48) and CCM, Annex 50. See also, Annex 70, p. 2.
278 Annex 45: Resolution Nº 1065 of 26 November 2001 of the
Ministry for the Environment of Colombia, Section 3 of the ‘whereas’ part,
Technical Visit.
279 Annex 37, Resolution Nº 001 of 11 February 1994 of the National

Narcotics Council of Colombia (Article 7), modified by Resolution Nº 005 of
11 August 2000 of the National Narcotics Council of Colombia (Annex 41),
115technical auditors conduct visits to the operation sites and the

spraying zones, as well as to the offices of the agencies in

charge of implementing the PECI G program, in order to verify

that their activities abide by the procedures and guidelines

contained in the Environmental Management Plan. For

instance, in order to verify the environmental monitoring

provided for in the Plan, the auditors take counter-samples of
280
soil and water. Likewise, at the oper ation sites, the audit
verifies whether industrial security measures are being

implemented as well as whether there is compliance with the

rules as to solid waste and residual water management.

4.29. The auditing company submits quarterly reports 281of its

evaluations and assessments, with recommendations for any

outstanding corrective actions to be taken. The results of the

external technical audit are submitted to the National Narcotics
282
Council annually.

revoked by Resolution Nº 013 of 27 June 2003 of the National Narcotics
Council of Colombia (Annex 49). S ee also Appendix 1 to Report by the
National Narcotics Directorate (DNE), 2010 (Annex 66).
280 Resolution Nº 013 of 27 June 2003 of the National Narcotics
Council of Colombia, Article 5 (Annex 49); Annex 59: Addendum Nº 1 to
Tender Nº 02 of 2007, with the purpose of “Contracting the Audit to the
Program for the Eradication of Illicit Crops by Aerial Spraying with
Glyphosate Herbicide”, National Narcotics Directorate of Colombia, pp. 1, 6-
7.
281 Appendix 1 to Annex 66.
282
Annex 49: Resolution Nº 013 of 27 June 2003 of the National
Narcotics Council of Colombia (Article 5).
116 283
4.30. In its Memorial, Ecuador refers to a 2003 decision of

the Administrative Tribunal of Cundinamarca – a province in
the middle of Colombia, more than 900 km from the border with

Ecuador – which ordered the su spension of the aerial spraying

program. But on appeal, the Counc il of State found that there

was no cause for ordering the suspension, since a measure of

that type could lead to the weakening of the State and at the

same time, to the strengthening of the illegal armed groups

financed with the revenues of drug trafficking. Although in its

judgment the Council of State did not entirely dismiss the

adverse effects allegedly caused by the sprayings and stressed
the need for continuing contro l over the sprayings, it also

stressed that these effects were not as serious as was claimed by

the plaintiff.84

(2) O VERALL SUCCESS OF THE PROGRAM

4.31. As the 2009 World Drug Report shows, despite the

enormous difficulties encountered in the process, eradication

and interdiction efforts in Colombia have succeeded to the point

that the area affected by illicit crops in the country underwent a

50% decrease as compared with 2000. As a result only 48% of
the world’s coca crop is currently produced in Colombia, in

contrast to 77% in 2000. 285 According to the most recent

283
284 EM, para. 2.49
Annex 54: Council of State of Colombia, Judgment on appeal from
285 Administrative Tribunal of Cundinamarca, 19 October 2004 (excerpts).
Annex 109: United Nations Office on Drugs and Crime (UNODC),
“World Drug Report 2009”, Vienna, 2009, p. 64; Annex 102, p. 67.
117figures from the United States’ Department of State, between

2007 and 2008 there was a 29% decr ease in the areas cultivated
286
with coca.

4.32. As to cocaine production potential, whereas in 2000 79%

of the world’s total could be allocated to Colombia, that figure

decreased to 51% in 2008, with a significant impact on reducing
287
worldwide figures. Cocaine production potential in Colombia

in 2004 was estimated at 640 tons , while in the year 2008 the
288
UNODC placed it at 430 tons, reflecting a 33% decrease; a

39% decrease between 2007 and 2008 alone, according to the

United States’ Department of State. 289

4.33. In the border area with Ecuador, in 2000, the province of

Nariño produced 6% of coca crops in Colombia, while the

province of Putumayo produced 40 %; together, they produced

34% of the world’s coca crops. 290 Between 2000 and 2005 a

69.7% decrease of the cultivated area was registered in the two

provinces. 291

286
Annex 152: Embassy of the United States in Bogotá, Fact Sheet
2008, “Cocaine production and cultivation: Colombia”, 6 November 2009,
enclosure to Press Item: “Official U.S. Colombia Survey Shows Sharp Drop
in Coca Cultivation and Cocaine Production”.
287 Annex 109, pp. 11, 64 (Table 6), 68 (Figure 27).
288 Annex 108, Colombia Coca Cultivation Survey 2008, p. 48 (Figure

289
290 Annex 152.
Annex 103, Colombia Coca Cultivation Survey 2003, pp. 12, 15.
291 Annex 108, Colombia Coca Cultivation Survey 2008, p. 13.
1184.34. In the year 2000, in a 10-kilometre strip along the border

there were 20,731 hectares of coca plants. Five years later, in

2005, following the application of the PECIG program, that

figure was reduced to only 4,281 hectares, a 79% decrease.

However, during 2006, following th e first suspension of aerial

sprayings in the 10-kilometre strip at the request of Ecuador, the
cultivated areas increased by 72%, a marked setback in the

progress achieved with the sprayings carried out in the previous

five years.292

(3) A LTERNATIVES TO AERIAL SPRAYING

4.35. The particular conditions of the growing of illicit crops

in Colombia, including the fact that large areas are devoted to

their cultivation, that the crops are situated in remote locations

and associated security problems, make aerial spraying the most

effective eradication method. However, the Colombian
Government has endeavoured to include other methods to

supplement the fight. To that effect, two programs have been

implemented in those regions where appropriate conditions

exist: manual eradication and alternative development or crop

substitution programs.

292 “In a belt of about 10 km width along the Ecuadorian border that
covers about 550,000 hectares, in the departments of Nariño and Putumayo,
7,000 hectares of coca cultivation were found in 2006. This represented an
increase of almost 3,000 hectares (or 72 %) compared 2005.” In Annex 107:
UNODC, “Colombia Coca Cultivation Survey 2006”, June 2007, p. 27.
1194.36. Manual eradication is carried out by workers, using a

variety of methods: pulling out the bushes by hand or using
small shovels; cutting the bushes wi th mechanical tools such as

scythes; or applying herbic ides with handheld spraying

equipment. The eradicators are joined by members of the

Colombian National Police, Army or Navy.

4.37. However, as noted above, following the inception of the

Government’s manual eradication program in certain areas of

the country, illegal armed groups involved in drug trafficking

and drug mafias started to syst ematically employ mine fields
and anti-personnel mines in the midst of their illicit crops in

order to impair these efforts, resulting – in the period between

2006 and 2009 alone – in 39 civilian manual eradicators killed

and 149 wounded, as well as 69 policemen killed and 197
293
wounded during manual eradication tasks.

4.38. Alternative development seeks to generate licit

productive activities for the communities engaged in the
cultivation of illicit crops, so as to allow them to provide a

stable income through the sustainable use of natural

resources. 294

293
294 Annex 72, p. 2; and Annex 73.
Agencia Presidencial para la Acción Social y la Cooperación
Internacional [Presidential Agency for Social Action and International
Cooperation] (ACCION SOCIAL), Available at:
http://www.accionsocial.gov.co/contenido/contenido.aspx?catID=217&conI
D=170; and
1204.39. Nevertheless, such projects ar e not possible in all areas:

the projects planned by the Alternative Development Program
have faced enormous obstacles in their implementation given

the pressure exerted by illegal armed groups on the communities

in question. Further, the Alternative Development Program is

ineffective to tackle the cultivation and processing of illicit
crops that such groups undertake themselves.

4.40. The topographic configurati on of the border region, the
extensive areas cultiv ated with illicit cr ops found there, the

difficulties involved in accessing the area, together with the

threat posed by illegal armed groups, have prevented these

alternative methods from bein g implemented consistently
throughout the zone contiguous to Colombia’s border with

Ecuador.

D. Technical Aspects of Aerial Spraying

4.41. The technical aspects of the aerial spraying program

conducted within Colombian territory may be summarized as

follows.

http://www.accionsocial.gov.co/contenido/contenido.aspx?catID=217&conI
D=965 (last visited 10 March 2010).
121 (1) T HE SPRAYING MIXTURE

4.42. In accordance with the legal provisions in force since

2001, the spray mix used in Colo mbia for the eradication of

coca crops in the PECIG program, consists of 44% glyphosate-

based formulated product, 1% su rfactant based on ethoxylated

linear alcohols and 55% water.

4.43. In 2001, the Ministry of Health of Colombia rendered an

opinion according to which:

“[t]he use of the mixt ure Glyphosate + POEA +
Cosmoflux (1%) would fall into toxicological
category III (Moderately Toxic), without the
addition of such additives posing inadmissible

risks. It should be recalled that pesticides within
that category are admissible, in accordance with
their intended action, for use in household
environments, as is the case of common

pesticides such as Baygon spray, Raid
mosquitoes and flies, Rayol spray, Rodasol,
Cupex for flying insects …” 295

4.44. Colombia has not used products with active ingredients

other than those mentioned above, nor has it included biological

agents in the mix during the implementation of the aerial
296
spraying operations. Suggestions to the contrary by
Ecuador 297and in the Menzie Report 298are without foundation.

295
Annex 44: Toxicological Opinion Nº 0685 regarding the
toxicological classification of the mix Glyphosate + POEA + Cosmo-
296x(1%), Colombian Health Ministry, 8 October 2001.
Annex 42: Communiqué of the Ministry for the Environment of
Colombia to the Public, October 2000: “The Ministry for the Environment
1224.45. Moreover, the composition of the spray mixture has been

a matter of public knowledge si nce 2001. The public character

of the chemical composition of a pesticide is a legal requirement

within the rules of the Andean Community of which Colombia

and Ecuador are Member States. In order for a commercial

product to be registered – and hence for its sale to be authorized

– information concerning its active ingredient, as well as other
ingredients that have any degree of toxicity, must be made

public. 299

4.46. Even before this was required by the rules of the Andean

Community, Colombian law required publicity as to the

ingredients of products the sale or use of which was authorized

within Colombia. Thus, all the products used in the PECIG

program have been issued “sales registrations” by the

Colombian Agricultural Instit ute (ICA), in which all the

did not accept the proposal advanced by the United Nations International
Drug Program (UNDCP), to conduct tests with Fusarium oxysporum
Erythroxylum mycoherbicide, given that it considers that any agent external
to our country’s native ecosystems might pose serious hazards to the
environment and human health”, issued on the occasion of the Meeting of
Vice-ministers of Foreign Affairs of the Andean Community in October
2000.
297 EM, paras. 2.43, 5.27 – 5.34.
298 Menzie Report, EM, Vol. III, Annex 158, p. 6-8.
299 Annex 135: Andean Regulation for the Registration and Control of

Chemical Pesticides for Agricultural Use, Andean Community, Decision 436
of 2000, Arts. 57-59.
123ingredients of the product that have any degree of toxicity

appear. 300

(a) Glyphosate

4.47. Glyphosate is a liquid, clear, viscous herbicide; it is

amber in colour and practically odou rless. For over 30 years, it
301
has been the most widely us ed herbicide in the world.

Glyphosate-based products are regi stered in over 130 countries

and are approved for weed control in relation to over 100
302
crops. Glyphosate is an “herbicide registered for use on many

food and non-food field crops as well as non-crop areas where

300 Annex 39: Roundup SL: Sales Registration Nº 0756 of 11 August
1997; Annex 46: Gly–41: Sales Registration Nº 4294 of 2 July 2002; Annex
33: Cosmo-Flux 411F: Sales Registration Nº 2186 of 19 April 1993. See also

301ex 65, p. 2.
“Based upon EPA reviews of domestic use conditions, glyphosate appears
to be one of the most safely-used pesticides in the U.S.” in Annex 142:
United States Environmental Protection Agency (EPA), Details of the
Consultation for Department of State: Use of Pesticides for Coca and Poppy
Eradication Program in Colombia , August 2002. (The text of this annex
appears in: United States Department of State, Bureau of International
Narcotics and Law Enforcement Affairs, Report on Issues Related to the

Aerial Eradication of Illicit Coca in Colombia , December 2003, p. 2
302rtially included as EM, Vol. III, Annex 143)).
For instance, in the US, glyphosate was re-registered in September
1993 after the EPA reviewed new studies and concluded that the use of
glyphosate-based studies in accordance with label directions “will not pose
unreasonable risks or adverse effects to humans or the environment”. In EM,
Vol. III, Annex 132 (United States Environmental Protection Agency (EPA),
R.E.D. (Re-registration Eligibility Decision) Facts,September 1993). By

1999, the Environmental Protection Agency (EPA) of the United States, one
of the strictest and most widely resp ected environmental agencies in the
world, had authorized the use of 237 glyphosate-based products for more
than 400 uses or applications in the United States. The EPA estimated total
global use of glyphosate to be between 350-360 million pounds of glyphosate
per year.
124 303
total vegetation control is desired.” No other herbicidal active

ingredient is comparable in te rms of the number of approved
304
uses.

4.48. In Colombia, there are 52 glyphosate-based products

publicly available on the market under different commercial
305
names. As the CICAD I study points out, glyphosate “is

widely used in agriculture and for purposes other than
306
eradication of coca and poppy”. Thus it has been used for

sugarcane ripening for over 30 ye ars and as an herbicide in

crops such as coffee, banana, ri ce, cocoa, pasture-land, African

oil palm and citrus fruits. 307

303
304 EM Vol. III, Annex 132.
According to the United Nations Food and Agriculture Organization
(FAO), glyphosate has “a broad spectrum of applications in agriculture,
horticulture, viticulture, forestry orchards, plantation crops, amenities, home
gardening and greenhouses for the control of annual and perennial grasses
and broadleaved weeds. Furthermore it is used for weed control on aquatic

areas, industrial areas, railroad trackand on other non-cultivated areas.
Besides the weed control it is used for root sucker control, for reseeding of
grassland and to facilitate harvest.” In Annex 101: Food and Agriculture
Organization of the United Nations (FAO), “Specifications and Evaluations
for Plant Protection Products. Glyphosate N-(phosphonomethyl)glycine”,
2000/2001, p. 17.
305 See Colombian Agriculture and Livestock Institute, Registros de

Venta de Plaguicidas Químicos de Uso Agrícola [Sales Registrations for
Chemical Pesticides for Agricultural Use], 30 April 2009. Available at:
http://www.ica.gov.co/getdoc/2dae6093-c021-49d1-8b29-
c9dfebce2757/REGISTROS-DE-VENTA--PQA-24-01-09.aspx (last visited
10 March 2010)
306 Annex 116, CICAD I, p. 9.
307 Annex 141: United States Environmental Protection Agency (EPA),
Addendum to memorandum. Subject: Description of Glyphosate Use in the

U.S. as a Basis for Comparison to Glyphosate Use in Colombia for Coca
Eradication, From: Virginia Werling and Timothy Kiely (EPA, BEAD
Biological and Economic Analysis Division) to Jay Ellenberger (EPA,
BEAD Biological and Economic Analysis Division), 21 August 2002, p. 2.
1254.49. In Ecuador itself there are 47 glyphosate-based
308
registered products and glyphosate is widely used both in

relation to agricultural crops as well as for weed control. Of

those 47 products, 24 possess the same glyphosate concentration

as the herbicide used in Colombia’s aerial spraying eradication

program. Herbicides containing glyphosate are used by wildlife

organizations to protect and restore wildlife habitats threatened
by invasive, non-native vegetation in Ecuador. For example,

according to Ecuadorian official sources, Roundup, the main

commercial brand worldwide that contains glyphosate as an

active ingredient, was selected to control invasive weeds in the

unique and particularly sensitiv e Galápagos Islands ecosystem,

which is protected by strict environmental measures. 309

4.50. With regard to the form ulated products used in

Colombia’s illicit corps erad ication program, since 2000 only

two glyphosate-based commercially formulated products have

308
Ministry of Agriculture, Livestock, Aquaculture and Fishing of
Ecuador, Ecuadoran Livestock and Agriculture Sanitary Service (SESA), List
of Pesticides Registered in SESA, October 2002. Available at:
http://www.sica.gov.ec/agro/insumos/plgregecg.htm (last visited 10 March
3090).
Annex 89: “Manual for the Identification and Management of
Weeds on the Galapágos Islands”, Charles Darwin Foundation and
Galápagos National Park Service, 2006, pp. 16-18, 21-29, 32 (charts).
Additionally, “Glyphosate is also used by wildlife organizations to protect
and restore wildlife habitats threatened by invasive non-native vegetation.
Conservation groups have chosen glyphosate formulations because of their
effectiveness against most weeds and because they have very low toxicity to
wildlife.” In Annex 132: Center for Toxicology and Environmental Health,

L.L.C., University of Arkansas for Medical Sciences, GLYPHOSATE
Frequently Asked Questions, 2009, p. 1.
126 310
been used, namely, Roundup SL and GLY41. Roundup SL,

used until 2004, is registered in many countries for agricultural

use and contains 41% glyphosate salt and 59% inert
311
ingredients. The commercial product GLY41, used since

2004, also contains 41% glyph osate salt and 59% inert
312
ingredients.

310
Annexes 39, 46. Also, Saftety Data Sheet for Roundup SL (Annex
133), and Label and Safety Data Sheet for GLY-41(Annex 134). See also
EM, Vol. III, Annex 115 (Label for Roundup SL).
311 According to the United States Environmental Protection Agency
(EPA) “An inert ingredient means any substance (or group of structurally
similar substances if designated by the Agency), other than an active
ingredient, which is intentionally included in a pesticide product. Inert

ingredients play a key role in the effectiveness of a pesticidal product. For
example, inert ingredients may serve as a solvent, allowing the pesticide's
active ingredient to penetrate a plant's outer surface.” In: EPA, Inert
Ingredients Permitted in Pesticide Products, date unknown. Available at:
http://www.epa.gov/opprd001/inerts/lists.html (last visited 10 March 2009).
Also: “Since glyphosate is only effective if absorbed by plant foliage,
glyphosate is combined with a surfactant to facilitate its absorption…

Surfactants are commonly used as wetting agents with herbicides and in
other products such as laundry and dishwashing detergent.” In United States
Environmental Protection Agency (EPA), Details of the Consultation for
Department of State: Use of Pesticides for Coca and Poppy Eradication
Program in Colombia, August 2002. In Annex 142, p. 10 (partially at EM,
Vol. III, Annex 143).
312 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 GLY41. Although that product
contains some POEA as a surfactant, it is present in very small proportions,
and produces no adverse effects on human health. According to Williams et
al, “Glyphosate, AMPA and POEA were not teratogenic [gestational
malformations] or developmentally toxic. Likewise there were no [medium

or long-term] adverse effects in repr oductive tissues from animals treated
with glyphosate, AMPA or POEA in chronic and/or sub-chronic studies. It is
concluded that under present and expected conditions of use, Roundup
herbicide does not pose a health risk to humans”. Annex 125, pp. 117, 160.
127 (b) Cosmoflux

4.51. The addition of adjuvants or wetting agents to herbicides

is a common practice in agriculture. Adjuvants are added to

herbicides in order to reduce evaporation and improve their

efficiency upon spraying, since they add weight to the drops and

render them homogeneous in size. Further, the addition of
adjuvants ensures adhesion of the mix to the surface of the leaf,

reduces dispersion of drops in the air and th e percentage of mix

deposited on the ground, and contri butes to reducing drift. The

fact that the adjuvant increases the effectiveness of the

application of the spray mix – allowing for more precise control

of the targeted plant species – does not in itself increase the

toxicity of the mix.

313
4.52. In the case of Cosmo-Flux 411F, used in Colombia’s

spraying program, the CICAD I study stated that the addition of

the adjuvant Cosmo-Flux to the glyphosate did not change its
toxicological properties to mammals. 314 In 2001, the United

States EPA determined that all of the ingredients in Cosmo-Flux

411F may be used when applied for crops destined for human

consumption given their low toxicological risk. 315The

313 See Annex 138: COSMO-FLUX® 411-F Technical Data Sheet,
Cosmoagro, June 2002.
314
315 Annex 116, CICAD I, p. 78.
Annex 142, p. 23 (partially at EM, Vol. III, Annex 143). Also, “the
Cosmo-Flux 411F added to the glyphosate in Colombia has little or no effect
on the overall toxicity of the formulated product” (Ibid., at p. 39).
128toxicological category to whic h the Cosmo-Flux used in the
316
PECIG program has been allocated is IV, i.e., slightly toxic.

4.53. The Menzie Report annexed to Ecuador’s Memorial

states correctly that “the addition of various surfactants and
additives to enhance the effectiveness of the herbicide

formulation has been the subject of research by the U.S.

Department of Agriculture (USDA) (Collins and Helling

2002)”. 317 However, the research carried out by Collins and

Helling (2002) is mistakenly cite d to suggest that the products

tested may have been used in the Program for the Eradication of

Illicit Crops by Aerial Spraying with Glyphosate (PECIG). In

fact, the USDA tests were ca rried out in greenhouses in

Maryland (United States) and the field studies were carried out

in Hawaii (United States). None of the products tested has ever

been or is currently used in the PECIG program; i.e., they have
not been used in Colombia for illicit crops eradication purposes.

Collins and Helling simply recommended “adding a Colombia-

manufactured agricultural adjuvant with properties similar to

one of the most effective surfactants tested in this research”. 318

Colombia chose the adjuvant Co smoflux 411F, an agricultural

adjuvant manufactured in Colombia, which does not increase

316 Annex 34: Provisional Toxicological Opinion LP-0593-93 (Cosmo-
Flux 411F), Colombian Health Ministry, 30 July 1993; Annex 44.
317 EM Vol. III, Annex 158, Section 3.3, referencing Collins, R.T. &

C.S. Helling. 2002. Surfactant-enhanced control of two Erthroxylum species
by Glyphosate. Weed Technol. 16: 851-859. (The study also appears as EM
318. III, Annex 141.)
EM Vol. III, Annex 141 (Collins & Helling (2002)).
129the toxicity of the mixture but does increase the effectiveness of

the herbicide application from a physical perspective.

4.54. Ecuador states in its Memorial that the glyphosate-based

mix is “adulterated by Colombia” by adding “another chemical,
known as Cosmo-Flux 411F” that “is not sold in the United

States”.319 Cosmo-Flux has been reviewed by the EPA 320 and is

sold in a number of countries, in cluding Ecuador itself. In this

regard, the Ecuadorian Memorial notably omits to mention that

Cosmo-Flux 411F, exactly the same substance as that used in

the PECIG program, is lawfully imported into and used in

Ecuador in accordance with an au thorization of the Ecuadorian

Ministry of Agriculture, Livest ock, Aquaculture and Fishing,

under registration No. MAG 3199076.

4.55. Ecuador also omits to mention that the Cosmo-Flux
411F label it submitted as Annex 113 of its Memorial

corresponds precisely to that a ffixed to the container of a

319 EM, para. 5.20, p. 123. Also: “While the specific spray adjuvant
product identified as that used in Colombia is not sold in the U.S., similar
substances and products are commonly used.” United States Environmental
Protection Agency (EPA), Details of the Consultation for Department of

State: Use of Pesticides for Coca and Poppy Eradication Program in
Colombia, August 2002. In Annex 142, p. 4 (partially at EM, Vol. III, Annex
320).
See para. 4.52 above.
130product made in Colombia in November 2007 and imported into

Ecuador, as the stamps on the label show. 321

4.56. Besides Colombia and Ecuador, Cosmo-Flux 411F is

registered for use in a number of other Latin-American
322
countries, including Panama, Costa Rica and Bolivia, among

others.

(2) P ROCEDURES FOLLOWED IN THE PECIG PROGRAM

4.57. It must be recalled at the outset that the eradication of

illicit crops by aerial spraying in no way constitutes a procedure

followed solely in the Colombian provinces bordering Ecuador;

it has been applied throughout Colombian territory in areas

affected by illicit crops. Although in its current form it began in

2000, aerial spraying has been in use since the early 1990s and

has become one of the most effective methods in Colombia’s

fight against the scourge of drugs.

321 EM, Vol. III, Annex 113. See also CCM, Annex 74: Main
Fertilizers Import and Manufacturing Companies (date unknown), Ministry
of Agriculture, Livestock, Aquaculture and Fishing of Ecuador, pp. 2-3.
322 Registration in Panama, Agriculture and Livestock Development

Ministry, National Directorate of Vegetation Health, Agrochemicals
Division, available at:
http://www.panamatramita.gob.pa/Formularios/2006_8_8_2006_9_3_3.pdf
(last visited 10 March 2010); Registration in Costa Rica, Agriculture and
Livestock Ministry, Phytosanitary Service of the State, available at:
http://www.protecnet.go.cr/insumosys/ConsultarInsumo.asp?cCodigo=4042
&sTipoQry=Plaguicidas (last visited 10 March 2010); Registration in
Bolivia, National Service for Agricultural and Livestock Health and Food
Safety, Vegetation Health National Unit, available at:
http://www.senasag.gov.bo/egp/productossv3.html (last visited 10 March
2010).

1314.58. In Colombia, aerial spraying is carried out following a

detailed procedure, divided into a number of stages consisting of

detection, spraying and control, each of which includes a
323
planning phase.

4.59. The detection procedure consists of the identification,

through the interpretation of satellite imagery processed by the

SIMCI Project and verification by aerial photography, of areas
324
affected by illicit crops. This data is mapped using digital

cartography with the purpose of determining the exclusion zones

and the 100-meter contiguous st rips around them, in which no

spraying takes place. 325 Exclusion zones are applicable to areas

including human settlement, bodi es of water and rivers, and

areas where illicit crop substitution projects are underway. 326

National Natural Parks and indi genous reserves constitute

special management zones. 327

323 Annex 67, p. 1.
324 Ibid., pp.1-2.
325 Annex 30, Arts. 87, 102; Annex 50, Num. 3.2.1, 3.2.2. See also
Annex 67, pp. 1-2.
326
327 Ibid.
For National Natural Parks, the relevant provision reads as follows:
“Taking into account that there is evidence of illicit crops within these zones
[National Natural Parks], which threatens their conservation and
sustainability, the implementation of PECIG is authorized therein, following
the submission to the National Narcotics Council of the environmental and
social characterization of the areas to be sprayed”. Resolution Nº 013 of 27
June 2003 of the National Narcotics Council of Colombia, Article 1, para. 2

(Annex 49). As for indigenous reserves, the relevant provision reads as
follows: “Taking into account that, according to Colombian legislation,
indigenous communities have the right to ancient traditional use of coca
leaves and that on some occasions crops planted in their territories exceed the
1324.60. The data resulting from the detection phase together with

data as to the course of international boundaries are loaded onto

a specialized computer appli cation with which the spraying

planes are fitted, in order to ensure that the operation is carried
328
out with precision.

4.61. The spraying operations are planned on a daily basis and

a coordination briefing with the personnel involved in the
329
spraying tasks takes place. Following an assessment of

whether the minimum requisite security and weather conditions

– including temperature, wind di rection and speed and relative

humidity – are present in the ar eas to be sprayed, the operations

begin. Any alteration to these conditions entails the immediate
330
cancellation of the spraying mission.

4.62. Upon reaching the area to be sprayed, prior to releasing

the spray mix, the spraying aircraft descends to an average flight

ones required for such use, a prior consultation process is carried out with
indigenous communities in order to determine the eradication method to be
applied”. In Annex 70, p. 6.
328
329 Annex 67, p. 7.
During the briefing, the pilots are informed of security conditions,
and are reminded of the requirements set forth in the Environmental
Management Plan, as well as of the aeri al and industrial safety measures to
be observed, among others. Each pilot receives the maps or plots affected by
illicit crops to be eradicated. See An nex 67, pp. 3-5, 2.2.2. Operation
Planning Meeting.
330 “When an aircraft is the object of hostile fire...[t]he flight leader will
cancel the spray mission” and “The mission may be cancelled if during its
course, reliable information is recei ved on possible attempts against the
aircraft of the spray team”. In Annex 67, p. 5.

133altitude of 30 meters, with a maximum operational air speed of
165 miles per hour. 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 pilots.331 With the aid of the computerized system and

with the cockpit screen showing the exact location of the aircraft

with regard to the targeted plots, the pilot activates the release of

the spray mix that exits through the nozzles. The 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
332
hectare.

4.63. The aircraft used for spraying operations are AT-802

planes manufactured by Air Tractor; they are specially designed
to operate with precision dur ing those tasks and possess a

system of tank, nozzles and pumps similar to those used for the

spraying of crops in other part s of the world. The spraying

personnel – pilots – are provide d by DynCorp, Inc., a private

company contracted by the United St ates Department of State.

The pilots are specifically certif ied, trained and experienced.

They are expressly instructed to comply strictly with the
relevant legal provisions concerning the duties of pilots engaged

331 Annex 50, Table Nº 1 (Operational Parameters of the Program for
the Eradication of Illicit Crops by Aerial Spraying).
332 Annex 48, Preliminary Section; and Annex 67, pp. 7-8.
134in aerial spraying of pesticides or herbicides, such as those
333
concerning exclusion zones and contiguous strips.

4.64. 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. 334 A

record is signed by the Base Commander and the personnel

involved in the operation.

4.65. As set out in Chapter 2, the border between Colombia

and Ecuador extends for some 717 km. To the extent that

sprayings have actually taken place in the immediate area
contiguous to the border with Ecuador, they have only been

conducted along two limited segments, as acknowledged by the

Ecuadorian Memorial. 335 The spraying has only taken place in

those areas of Colombian territory where clusters of illicit crops

have been found. Moreover, those sectors of the border with

333 Annex 30: Colombian Decree 1843 of 1991, Article 102, Duties of
pilots. As for the contiguous strips, Article 87 of the same decree provides:
“The application of pesticides in rural areas may not be carried out within 10
meters if land-based and 100 meters for aerial [spraying] as safety strips in

relation to bodies of water or watercourses, main roads, human or animal
334lei, or any other area that requires special protection.”
335 Annex 67, p. 8.
EM, Maps 5, 6 & 7.
135Ecuador that are marked by watercourses are protected by a

100-meter strip on the Colombia n side over which no spraying
takes place.

4.66. Any given plot affected by illicit crops is normally

sprayed once a year. Exceptionally, an area may be sprayed
twice in a year. Spraying can occur at any time of the year as

long as the weather conditions allow.

(3) M INIMIZING DRIFT

4.67. The operation of the PECIG program has always taken

into account the phenomenon of dr ift, inherent to aerial

application of herbicides, and th e factors that determine it have
been carefully considered.

4.68. Spray drift depends essentially on wind speed and

direction, as well as on a number of other atmospheric factors
including temperature, relativ e humidity and atmospheric

stability. It is also dependent on the altitude at which spraying

takes place and the air speed of th e spraying aircraft, as well as
the calibration of the spraying eq uipment, the density of the

spray mix and the initial size of the spray droplets.

4.69. The PECIG, taking into account all these factors, set
minimum and maximum figures in the Environmental

Management Plan for the parameters upon which drift is

136contingent, with the purpose of re ducing it as much as possible.

These parameters are strictly observed by the personnel

involved in spraying operations. 336

4.70. Ecuador’s Memorial claims that “fear of hostile ground-

fire from narcotraffickers prot ecting their illicit crops causes

pilots to fly above the otherwis e mandated altitude”, and relies

on assumptions contained in the Menzie Report as support for
337
that scenario. The Menzie Report asserts that “the spray

missions are often conducted und er highly adverse and violent

conditions” and concludes that “ It is reasonable to expect that

the hostility would be a reason why pilots would fly higher and

faster than they would during normal agricultural crop-spraying.

Under those circumstances, the released spray would be prone to
338
greater drift”. But no spraying operations are authorized on

plots that are assessed as be ing high risk until military

operations to guarantee appropriate security conditions are

carried out; as noted above, if the aircraft comes under fire, the
339
mission is cancelled.

336 Annex 50, Table Nº 1 (Operational Parameters of the Program for
the Eradication of Illicit Crops by Aerial Spraying).
337 EM para. 5.95, p. 162
338 EM, Vol. III, Annex 158, p. 13.
339 Above, para. 4.61 and note 330; and see Annex 67, p. 4.

137 E. Conclusions

4.71. Aerial spraying is one of th e main techniques used by
the Colombian Government in order to combat illicit crops and

consequently, cocaine production throughout the country.

4.72. Starting in the late 1990s, the Colombian Government
decided to enhance its aerial er adication program, within a

precise domestic legal framework. That decision was taken for a

variety of reasons. On the one hand, there had been an
accelerated increase in the total area given over to production of

illicit crops in Colombia, a trend towards growing those crops in

remote areas and an increase in th e size of the plots themselves.

On the other hand, those phenomena were coupled with an
accompanying deterioration of security conditions and a surge in

violence in the areas affected, in particular in the border area

with Ecuador where the most hectares devoted to illicit crops
were to be found.

4.73. Prior to the implementation of the eradication program
by aerial spraying with glyphosat e herbicide, the Colombian

Government requested scientif ic studies. The program only

began once the competent health and environmental authorities

had issued favourable opinions.

4.74. Colombia enacted rules and standards governing the

aerial spraying program in order to ensure that standards relating

138to 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 modification.
These rules were embodied in the Environmental Management

Plan that governs the eradication program up to the present time.

The EMP ensures the adequate management of the

environmental aspects of the program by all agencies involved
in its implementation, as well as verifying the effectiveness of

the aerial spraying operations. Additionally, the program is
340
overseen by a permanent external audit.

4.75. From 2000 to 2005, the areas cultivated with illicit crops

in the Colombian provinces of Nariño and Putumayo, which

previously had the most hectares in the country devoted to coca,

experienced a significant decrea se. Throughout that period,

during which sprayings were ca rried out in the 10-km strip
along the border, that zone alone registered a 79% decrease in

areas affected by illicit crops. By contrast in 2006, when

spraying operations were suspended in that strip, the areas in

which illicit crops were cultivated increased by 72%.

4.76. Colombia has never sought to conceal the composition

of the mix used to eradicate coca crops. Pursuant to the relevant

regulations on the matter, whic h are freely and publicly

340 Appendix 1 to Annex 66.

139available, the composition of the mix used to eradicate coca

crops in Colombia is well-known; the spray mix consists of 44%
formulated glyphosate commercial product, 1% Cosmo-Flux as

adjuvant and 55% water.

4.77. Glyphosate, the active ingredient of the herbicide in the
spray mix, is contained in numerous registered commercial

products which have been widely used worldwide for over three

decades. Those products have been used for multiple purposes,
both agricultural and non-agricu ltural, in over 100 countries,

including, as stated above, ecolo gically sensitive areas such as

the Galápagos Islands of Ecuador. The advantage of glyphosate

isthatitdoesnotcompromi se soil fertility since it has no
residual effects. It has been vetted by international organizations

entrusted with protecting human health and agricultural

productivity, including the WHO and the FAO, as well as by
other specialized agencies such as CICAD (OAS) and the EPA

in the United States. Rigorous scientific studies conducted in

Colombia and elsewhere corrobo rate the findings of those

independent organizations.

4.78. In order for herbicides to achieve their desired results in

certain types of application, it is common practice to add an
adjuvant that, in facilitating the absorption on the targeted plant,

increases the effectiveness of th e active herbicide ingredients.

The adjuvant in the mix used for the eradication of illicit crops
in Colombia is Cosmo-Flux; the addition of that ingredient does

140not alter the formula’s toxicity level, which remains low. The

inclusion of Cosmo-Flux as adjuvant is the only addition to the
formulated glyphosate product, and implies no “alteration”.

4.79. Aerial spraying operations were carried out along only in

certain limited portions of the shared border and took place only
over areas in which illicit crops were present. Moreover,

throughout the entire Colombian territory, according to the work

schedule, any given area is no rmally sprayed only once or,
exceptionally, twice a year.

4.80. The parameters under which the PECIG program is

conducted in Colombia manifests due diligence on the part of
Colombia. Moreover, as will be shown in Chapter 7, the

scientific appraisal on the basi s of which the PECIG program

was approved has been thoroughly vindicated by subsequent
research, as well as by field experience in Colombia.

141142 Chapter 5

THE DISPUTE WITH ECUADOR

5.1. The Memorial of Ecuador provides a one-sided and

incomplete version of the diplomatic history of the present case.

Indeed, as the record shows, from the outset, Colombia’s

attitude toward its neighbour’s perceived concerns was always
cooperative and forthcoming, with a view to preserving bilateral

relations while continuing its struggle to curtail the growth of

illicit crops in the border area. For its part, as will also be

shown, Ecuador’s account is contradicted by both the

contemporaneous factual background, as well as by the acts and
official positions adopted by Ecuador itself.

A. Overview of Diplomatic and Other Exchanges 2000 to
2004

5.2. Shortly after the aerial spra yings began in the area of

Colombian territory adjacent to the border with Ecuador, on 24

July 2000, the Ecuadorian Foreign Ministry – following the
protests of certain “communal groups” in the provinces of

Sucumbíos and Esmeraldas – st ated its concern concerning

future sprayings of coca crops “w ith toxic and/or biological

substances that may cause serious impacts on human health and
the environment, with possible repercussions for Ecuador”. 341

341 EM Vol. II, Annex 36.

1435.3. It is noteworthy that:

(1) Ecuador speculated th at the sprayings would
have “repercussions for Ec uador”, albeit without

providing any substantiation of these concerns;

(2) Ecuador seems to have assumed on the basis of
no evidence that the spra yings would be carried

out “with toxic and/or biological substances”;

and

(3) Ecuador anticipated, albeit again without

providing any support for its assertion, that the

sprayings might “cause serious impacts on

human health and the environment”.

342
5.4. In its Memorial Ecuador refers in a misleading fashion
to a short note verbale sent by Colombia on 29 December 2000

as if it were a refusal on the part of Colombia to hold bilateral

meetings in order to discuss the spraying program. In fact, the

Colombian note refers to and encloses a proposal from the

Government of Panama – not Ecuador – addressed to several

other countries proposing “t o hold a meeting on the
implementation of Plan Colombia”. 343It was the Panamanian

proposal to discuss “the implementation of Plan Colombia” with

342 EM, para. 3.8.
343 See EM, Vol. II, Annex 37.
144a number of other countries that Colombia found inappropriate,

as was clearly stated in the note verbale.

5.5. The Colombian Government had thoroughly explained

the components and scope of its policy in the fight against the

world drug problem on several occasions in different fora,

including at the highest level. Particularly noteworthy are: the

meeting of the Presidents of Colombia and Ecuador on 23

August 2000; 344 the meeting of the Presidents of South America

held on 1 September 2000 at Brasilia; 345 the meeting of Andean

Foreign Ministers convened by the Foreign Minister of Peru and

held at Lima on 16 March 2001; 346 the meeting of Andean

Presidents held in April 2001 at Cartagena, on Colombia’s

initiative;347and the meeting of Presidents of ATPA-beneficiary

344 Speech by the President of the Republic, Andrés Pastrana Arango, at
the dinner hosted for his Ecuadorian counterpart, Gustavo Noboa Bejarano,
Bogotá, 23 August 2000, in Ministerio de Relaciones Exteriores de
Colombia, La Política Exterior de Colombia. 1º de julio – 30 de septiembre

3452000, Fondo Editorial Cancillería de San Carlos, Bogotá, 2000, p. 186.
Speech by the President of the Republic, Andrés Pastrana Arango, at
the work session of the Presidents taking part in the Summit of South
American countries, Brasilia, 1 September 2000, in Ministerio de Relaciones
Exteriores de Colombia, La Política Exterior de Colombia. 1º de julio – 30
de septiembre de 2000 , Fondo Editorial Cancillería de San Carlos, Bogotá,
2000, p. 227.
346 Press Communiqué of the Meeting of Andean Foreign Ministers,
Lima, 16 March 2001, in Ministerio de Relaciones Exteriores de Colombia,

La Política Exterior de Colombia. 1º de enero – 30 de abril de 2001 , Fondo
347torial Cancillería de San Carlos, Bogotá, 2001, p. 195.
“‘Full Support’ of Andean Community to Plan Colombia”, Press
Release by Colombian News Agency (ANCOL), concerning support for Plan
Colombia by the States members of the Andean Community, Cartagena, 18
April 2001, in Ministerio de Relaciones Exteriores de Colombia, La Política
Exterior de Colombia. 1º de enero – 30 de abril de 2001 , Fondo Editorial
Cancillería de San Carlos, Bogotá, 2001, p. 283.

145Andean countries and the President of the United States, held on
348
23 March 2002 at Lima.

5.6. On the other hand, the Colombian Government was

under no obligation to hold meeti ngs with any other State in
relation to the “implementation” of Plan Colombia, a bilateral

agreement concluded with a thir d State – the United States of

America – which, as has been explained, concerned activities to

be carried out by Colombia within its own territory related not

only to the fight against illicit dr ugs, but also to other sensitive

issues such as the country’s s ecurity and its social, economic

and political development.

5.7. In July 2001, Ecuador again expressed its concerns on

the alleged adverse effects that the sprayings might have on

human health and the environm ent in Ecuador, and demanded
that the sprayings take place at a distance of at least 10 km from

the border, in order to preven t the likelihood of their reaching

Ecuadorian territory as a result of the al leged drift supposedly

caused by the winds. 349

348 Press Releases concerning Presidential meeting between Andean
ATPA-beneficiary Presidents and the United States, Lima, 23 March 2002, in
Ministerio de Relaciones Exteriores de Colombia, La Política Exterior de
Colombia. 1º de enero – 30 de abril de 2001 , Fondo Editorial Cancillería de
San Carlos, Bogotá, 2001, p. 187. ATPA refers to the Andean Trade

349ferences Act (USA).
See EM, Vol. II, Annex 41 (Diplomatic Note N° 55416/2001-
GM/SOI/SSN of 2 July 2001, from the Ecuadorian Foreign Minister to the
Colombian Foreign Minister).
1465.8. Ecuador’s demands were unreasonable. For Ecuador to

demand that Colombia suspend an entirely lawful activity such

as the eradication of illicit crops in the fight against drug

trafficking is surprising. This is particularly so given, on the

one hand, that the lawful activity in question was to be carried

out within Colombia’s own territory, under the strict control and
supervision of the relevant national authorities; and on the other,

that the demand was made w ithout any evidentiary basis

whatsoever to support it.

5.9. Nevertheless, in its reply of 14 July 2001 the Colombian

Government explained that the program for the eradication of
illicit crops was conducted unde r technical and controlled

parameters in order to minimize any risk to the environment, or

human and animal health. 350 The note gave details of the spray

mix, a fact which the Ecuadorian Memorial, despite annexing
351
the full text of the note verbale, passes over in silence.

5.10. In the same note, Colombia proposed the holding of a

seminar/workshop with the purpose of informing Ecuadorian

officials of technical aspects rela ting to the eradication of illicit

crops by aerial spraying with glyphosate in Colombia. In that

regard, the note stated as follows:

350 EM, Vol. II, Annex 42 (Diplomatic Note No. DM/AL 25009 from
the Colombian Foreign Minister to the Ecuadorian Foreign Minister, 14 July
2001).
351 EM, Vol II, Annex 42.
147 “Bearing in mind the c oncerns expressed by
Your Excellency, I propose holding a seminar-
workshop in Colombia within the next thirty (30)

days. During this event, the Government of
Colombia will have the opportunity to illustrate
its program of eradication of illicit crops and

provide all the technical information required by
Ecuadorian officials, in view of allaying all
existing concerns and strengthening the spirit of
bilateral cooperation.” 352

5.11. Following the Ecuadorian Government’s acceptance, 353

the proposed seminar was held in Bogotá on 13-15 February
354
2002. During the seminar, representatives of the Colombian

agencies in charge of the im plementation of the eradication

program explained the technical issues concerning spraying

operations, including the composition of the spray mix,

observed variables and technical parameters for aerial
application, as well as aspects relating to the environment and

human health. One of the activities forming part of the seminar

was a site visit to one of the zones s ubjected to the sprayings,

where Ecuadorian officials were able to observe firsthand how

the operations had been carried out.

352 EM, Vol II, Annex 42.
353 Annex 5: Diplomatic Note No. 72523/2001-SG/SSN from the
Ecuadorian Foreign Minister to the Colombian Foreign Minister, 12
September 2001.
354
Annex 7: Note E. 455/90 from the Colombian Ambassador in Quito
to the Colombian Foreign Minister, 26 March 2003, enclosing the non-paper
given to the Ambassador by Ecuadorian authorities at the meeting held on the
same date.
1485.12. In its Memorial, Ecuador attributes to the Colombian

officials who attended the workshop the view that objective and
impartial scientific research studying the effects of the spray

mix on health and the environment was lacking. 355 Ecuador

further contends that the Colombian participants in the seminar

assured their Ecuadorian counterparts that Colombia would

leave an 8 to 10-km safety margin from the San Miguel River,
356
but that this assurance was “quickly broken”.

5.13. These views were allegedly recorded in a “Joint report

from the workshop”, which was filed by Ecuador as Annex 163
of its Memorial. However, the doc ument is not at all a “Joint

report”; it was neither jointly drafted, nor agreed by both parties.

5.14. As is clear from the handwritten annotation on the top

right margin of its first page, the document filed at Annex 163
357
of Ecuador’s Memorial is a draft dated 6 March 2002. This

document is on the letterhead of the Ecuadorian Ministry for the

Environment and bears no indicatio n that it was ever approved
in final form or, for that matter, that it was ever subscribed to –

or, indeed, even seen – by any of the Colombian officials who

had attended the workshop. More over, the general tone of the

document and the use of the firs t person plural when discussing

Ecuador’s position clearly demonstrates that the document was

355
356 EM, para. 3.19.
357 EM, paras. 3.20-3.21.
EM, Vol. III, Annex 163.
149intended to present Ecuador’s position as a result of the seminar.

It cannot in any way be assimilated to a “joint report” of the

proceedings.

5.15. As to the alleged “assurance” given on behalf of

Colombia, the document itself refers to an oral assurance and
adds that a “declaration should be officially requested through

the Foreign Ministry”.

5.16. Ecuador’s allegations are further disproved by the

record. As was made clear in the Note from the Embassy of

Colombia to the Foreign Minister of Ecuador dated 20 February
2002, 358i.e. a few days after the seminar, Colombia continued to

maintain its position with regard to the aerial sprayings. In

particular, this Note stated, in relevant part:

“During the visit made to Colombia by an

Ecuadorian delegation last week, they were able
to verify that sprayings do not have harmful
effects on Ecuadorian territory or in Colombia,
for that matter; the care with which they are
carried out and the products used are a guarantee
359
of their innocuousness.”

5.17. As a result of the seminar held in February 2002,

Colombian and Ecuadorian offi cials recommended to their

governments that they should crea te a Scientific and Technical

Commission to analyze any possi ble effects in the Ecuadorian

358 EM, Vol. II, Annex 43.
359 EM, Vol. II, Annex 43.
150border provinces of the operati ons involving the aerial spraying

of glyphosate carried out over Colombian territory. Thus, as

stated in a 2003 non-paper prepared by the Ecuadorian

Government,

“one of the commitments made during the

Binational Spraying Seminar-Workshop, carried
out from 13 to 15 February 2002 in Bogotá, was
the implementation of an inter-institutional
technical commission in charge of controlling
and monitoring these tasks at national level, in

order to assess the potential impacts to the
environment and the health of the Ecuadorian
communities that live in the border with
Colombia -Carchi, Esmeraldas, and Sucumbíos-,
produced by the spraying plan in Putumayo”. 360

5.18. In pursuance of the proposal to create a bilateral
361
mechanism and in response to Ecuador’s concerns, in

February 2003 during talks held in Quito between the then-

Director of Plan Colombia, Sandra Suárez, and Nina Pacari, the
Foreign Minister of Ecuador at that time, Colombia renewed its

proposal for the creation of a Jo int Scientific and Technical

Commission entrusted with the ta sk of analyzing the issue of

spraying operations carried out in Colombian territory near the

Colombia-Ecuador border. As stated in the non-paper prepared
by the Ecuadorian Government in early 2003:

“…A meeting was held on 26 February 2003,
with the participation of members of the Ministry

360An7. ex
361 EM, Vol. II, Annexes 45 and 46.
151 of Environment, the Minist ry of Foreign Affairs,
and the Government, and the official in charge of
the Plan Colombia Mrs. Sandra Súarez, who
suggested the implementation of a committee,

composed of members from the Ministry of
Environment of both c ountries, in order to
discuss common matters such as the ecological

effects of spraying.
….

In addition, it was suggested that the Binational
Commission, supported by international

organisms, c362d carry out scientific
research…”

The Joint Scientific and Technical Commission was eventually

set up later that year, in September 2003.

5.19. The Commission met on four separate occasions

between October 2003 and August 2004. In the course of these

meetings, the Colombian delegation provided detailed

information on the aerial spraying program to the Ecuadorian
363
delegation. Ecuador’s Memorial acknowledges that

Colombia submitted four reports on the subject in November

2003. However, it goes on to dism iss them as a whole, without
referring to the content of the studies and without any scientific

362An7. ex
363
EM, Vol. II, Annex 50 (Diplomatic Note VRE/DBR No. 40153, of
12 November 2003, from the Colombian Foreign Minister to the Ecuadorian
Foreign Minister); CCM, Annex 9: Note Verbale DBR/CAL unnumbered
from the Colombian Foreign Ministry to the Ecuadorian Embassy in Bogotá,
13 November 2003; EM, Vol. II, Annex 53 (Diplomatic Note DM/DBR Nº
47356 of 15 December 2003, from the Colombian Foreign Minister to the
Ecuadorian Foreign Minister); EM, Vol. II, Annex 54 (Diplomatic Note
DBR/CAL No. 1405 of 14 January 2004, from the Colombian Foreign
Ministry to the Ecuadorian Embassy in Bogotá).
152or technical arguments. The Memo rial simply states that “the
Colombian Foreign Mini ster sent Ecuador four documents of

limited scope”. The delegations decided to perform field visits to

locations situated on both side s of the border in order to

establish the veracity of the complaints regarding alleged

adverse effects of the sprayings.

5.20. While the work of the Joint Scientific and Technical
Commission was on-going, the Fore ign Minister of Ecuador

wrote to the Colombian Embassy in Quito to announce that

Ecuador had received a number of complaints regarding the
364
aerial sprayings from bordering regions and communities.

Ecuador’s letter attached a proposed Memorandum of

Understanding formally suggesting for the second time a 10-

kilometre security or “buffer” no-spraying zone in Colombian
territory, measured inward from the land border. 365 Again, the

Ecuadorian Note did not refer to or produce any evidence

demonstrating that there had indeed been adverse impacts on the

health of the Ecuadorian population or on the flora and fauna in

Ecuadorian territory as a result of the eradication program.

5.21. On 18 September 2003, the Government of Colombia
replied that for national security reasons, which could also have

364
EM, Vol. II, Annex 47 (Diplomatic Note No. 23205/GM of 10 April
2003, from the Ecuadorian Foreign Minister to the Colombian Ambassador
in Quito).
365 EM, Vol. II, Annex 47 (Diplomatic Note No. 23205/GM of 10 April
2003, from the Ecuadorian Foreign Minister to the Colombian Ambassador
in Quito).
153a serious impact on Ecuador’s terr itory, it was not feasible to
declare a no-spray zone in the Colombia-Ecuador border region.

The Note reads in relevant parts as follows:

“To declare a zone of crops free of aerial
sprayings in the common border, would open the
door to the free movement of the drug-trafficking
international networks, gunrunning and chemical

precursors smuggling, in collaboration with the
terrorist groups that operate in the south of
Colombia, and would create a greater risk for the
security of the two nati ons, particularly, for the
366
residents of the border zone.”

5.22. Ecuador contends in its Memorial that at the second

meeting of the Joint Scientific and Technical Commission held

in Quito on 9 February 2004, Colombia agreed to provide

information regarding prior fumi gations, including flight paths,
367
altitude, wind and other parameters, but never did so.
However, Colombia’s contemporary minutes of the meeting of

February 2004, which are attached as Annex 51, and the

correspondence with Ecuador follo wing the meeting, contradict

Ecuador’s allegations. According to these minutes and the

agreed conclusions attached therein as Annex 3, the two

delegations agreed as follows, in relevant part:

“Development of a model that avoids mistakes in
aerial sprayings, instead of referring to a distance

to spray from the borderline. In this regard, it was

366 Annex 8: Note Verbale VRE No. 32759 from the Colombian
Foreign Ministry to the Ecuadorian Embassy in Bogotá, 18 September 2003.
367 EM, para. 3.34.
154 highlighted that it is important to share
information about the technical aspects of

spraying processes. [The] Director of the DNE
[National Narcotics Direct orate] made clear that
for security reasons under no circumstances the
activities schedule would be informed.” 368

5.23. In disregard of this understanding, on 16 February 2004,

the Government of Ecuador requested that, until the procedures

agreed upon during the second Me eting of the Scientific and

Technical Commission were comp leted, aerial sprayings near
369
the border with Ecuador should be suspended.

5.24. Once again, the Colombian Government demonstrated

its willingness to collabor ate: on 23 February 2004, the

Colombian Foreign Ministry replied to Ecuador that, due to the

calendar and priorities set for th e remainder of the year, no

aerial spraying would in fact take place in areas adjacent to the

368 Annex 51: Note Rec’d. No. 29157 from the Director of the National

Narcotics Directorate (Head of the Colombian Delegation to the Joint
Commission) to the Colombian Foreign Minister, 6 April 2004. The letter
read, in Colombia’s translation: “I submit the report of the meeting that I had
the responsibility to preside on behalf of the Colombian Commission, as well
as the conclusions that were drafted in coordination between the
representatives of both countries.” Annex 3 to the Note, entitled
“Conclusions of the Colombia-Ecuador Scientific and Technical
Commission” reads: “The following are the conclusions agreed to between
the Colombian and Ecuadorian delegations of the Colombia-Ecuador
Scientific and Technical Commission.”
369 EM, Vol. II, Annex 56 (Diplomatic Note No. 10181/2004-GM of 16
February 2004, from the Ecuadorian Foreign Minister to the Colombian
Foreign Minister. See also, EM, Vol. II, Annex 58 (Diplomatic Note

15715/2004-GM of 9 March 2004, from the Ecuadorian Foreign Minister to
the Colombian Foreign Minister).
155 370
border. However, the Colombian Foreign Ministry also

observed that if surveillance flights carried out in the zone were

to find new illicit crops, the aerial spraying program in that area

would need to be resumed.

5.25. Through a subsequent diplom atic Note dated 7 April

2004, the Colombian Foreign Ministry explained the

effectiveness of the strategy for the eradication of illicit crops,
which includes aerial spraying anywhere in the national territory

that the Anti-Narcotics Police determined the existence of illicit

crops. 371

5.26. A few days later, in a dipl omatic Note of 14 April 2004,

the Government of Colombia clar ified that it had at no time

subordinated its national policy of eradication of illicit crops,

including by aerial sprayings, to any conclusions of the

Scientific and Technical Commission; still less had it made any

commitment as a result of Ecuador’s pressure for a suspension
372
of spraying in the border area.

370 EM, Vol. II, Annex 57 (Diplomatic Note DM/DBR No. 8092 of 23
February 2004, from the Colombian Foreign Minister to the Ecuadorian
Foreign Minister).
371 Annex 12: Diplomatic Note DPM/CDR Nº 20125 from the
Colombian Acting Foreign Minister to the Ecuadorian Foreign Minister, 7
April 2004.
372 Annex 14: Diplomatic Note VRE/DAA/CAL Nº 18067, from the

Colombian Vice-Minister of Foreign Affairs to the Ecuadorian Vice-Minister
of Foreign Affairs, 14 April 2004.
1565.27. Also on 14 April 2004, Colombia’s Narcotics

Directorate submitted information to the Ecuadorian Scientific

and Technical Commission concerning earlier sprayings in

Colombian municipalities bordering Ecuador and explained the
373
methodology followed in the PECIG. It also reiterated that

the operational parameters of the Program were described in the

Environmental Management Plan, which had been furnished to
374
Ecuador in November 2003, explained the mathematical

formula to calculate drift, and enclosed several reports,

including a progress report by the Colombian National Health

Institute pursuant to Record Nº 7 of the EMP, technical data on

Cosmo-Flux 411F and the Acute Toxicity Study on Laboratory
375
Animals conducted by Inmunopharmos laboratory.

5.28. As a consequence of the work of the Joint Scientific

Technical Commission and during a visit of the Ecuadorian

President to Colombia in March 2004, it was agreed to promote

inspection and scientific assessment visits on both sides of the
376
border. However, it was only possible to carry out fieldwork
in Puerto Mestanza (Ecuador) one day late, on 26 May 2004,

due to the fact that the Ecuadoria n delegation did not turn up at

373
Annex 13: Note SARE-142 from the Director of the National
Narcotics Directorate of Colombia to the President of the Scientific and
Technical Commission of Ecuador, 14 April 2004 (partially translated at EM,
Vol. II, Annex 62).
374An9. ex
375An13.x
376 Annex 52: Report from the Colombian Delegation, Joint Scientific
and Technical Commission on Aerial Sprayings, Orito, Putumayo, 25- 28
May 2004, p. 1. See also, EM, Vol. II, Annex 61.

157the agreed venue on the following day. 377The report of the

Colombian delegation at the time observed that “although the

objective of the visit was to inspect in situ the alleged effects on

both human health and the environment on Ecuadorian territory

caused by aerial sprayings with glyphosate, in the Ecuadorian

delegation there were no official s from the Health Ministry and

378
the Ministry of Agriculture.”

5.29. In subsequent correspondence addressed to Ecuador, the

Government of Colombia expressed its regret as to the refusal of

the Ecuadorian Scientific Commission to take part in seven of

the eight site visits scheduled to take place in the region of the

border between Colombia and Ecuador and which had been
379
agreed through dipl omatic exchanges. Colombia also stated

that it was not willing to pursue that bilateral mechanism if the
380
agreed agenda was not respected.

377 “Although at the end of the first day it was agreed with the
Ecuadorian delegation that on the following morning they would be picked
up in the vicinity of the San Miguel International Bridge, in order to conduct

the field visit to the locations in the Putumayo province, the representatives
of the neighbouring country did not show up…. the Colombian delegation
notes for the record, that it fulfilled its share of the commitment, which could
not be completed due to the no-show of the delegation of Ecuador, despite
the efforts of the Energetic and Ways Special Battalion 9 of Orito [Putumayo
province], that had deployed military units to cover all the locations …”.
(Annex 52, pp. 4-5).
378An52e.x
379
Annex 15: Diplomatic Note VR/ST/CVICE Nº 27776 from the
Colombian Vice-Minister of Foreign Affairs to the Ecuadorian Ambassador
in Bogotá, 9 June 2004; Annex 16: Diplomatic Note VRE/DAA/CAL Nº
28511 from the Colombian Vice-Mini ster of Foreign Affairs to the
Ecuadorian Vice-Minister of Foreign Affairs, 17 June 2004.
380An15e.x
1585.30. Despite the failure of the site visits by the Joint
Scientific Technical Commission, water samples were taken in

Ecuador in April and May 2004 in the province of Sucumbíos in

an area allegedly affected by spraying with glyphosate carried

out in Colombia. Those samples were taken by an Ecuadorian
group, the Ecuadorian Comm ission on Atomic Energy

(Comisión Ecuatoriana de Energía Atómica), which had formed

part of the Ecuadorian Delegation, in order to determine the
possible presence of glyphosate in the rivers of Sucumbíos.

5.31. The results of the analyses of the samples were

published by the Foreign Ministry of Ecuador in Press Bulletin
N° 388 of 25 June 2004 entitled: “No glyphosate residues exist

in the waters of the rivers of Sucumbíos province” . The Press

Bulletin read as follows:

“The Foreign Ministry reported that the
Ecuadorian Commission on Atomic Energy
(CEEA) that takes part in the Ecuadorian
Scientific and Technical Commission (CCTE),
created to analyze the possible effects of

sprayings with glyphosate in the northern
Ecuador-Colombia border between Ecuador and
Colombia, travelled to the border area during the
months of April and May, in order to analyze the
possible presence of that component in the waters
of the rivers of Sucumbíos.

Samples were taken from the rivers Conejo, San
Francisco, San Miguel, La Punta, Zancudo,
Mataje, wherein no glyphosate residues were
found. The report presents favourable results

159 which have assuaged the population that inhabits
that sector.”81

5.32. In July 2004, the Ecuadorian Commission on Atomic

Energy again visited the border ar ea, this time focusing on the
western zone , in the province of Esmeraldas. Again the

Commission carried out thorough studies and again took water

samples that were taken back to Quito for analysis.

5.33. The last meeting of the first Joint Scientific and

Technical Commission, which incl uded a seminar organised by

Colombia with the purpose of explaining various aspects

relating to the PECIG program to Ecuadorian officials, was held

on 2 August 2004. In a letter addressed to the Ecuadorian
Foreign Minister, the Director of the National Narcotics

Directorate of Colombia regist ered his satisfaction with the

work of the Commission in the following terms:

“[T]he existing differences were settled well, in
particular, that relating to the drift of the Program
for the Eradication of Illicit Crops – PECIG,

issue concerning which it was agreed that
conceptual differences and differences regarding
mathematical formulae would never be settled on
a blackboard; that this case in particular could be
resolved by the presence of members of the

Ecuadorian Scientific and Technical Commission
in a validation process to which they would be

381 Annex 80.

160 invited, with the purpose of verifying PECIG’s
real drift on the field.”382

Moreover, the Director stated:

“The activity of the Scientific and Technical
Committee was finished to the extent that it was
not deemed necessary to set a date for a new
383
meeting.”

5.34. At the last meeting of th e Commission “the Ecuadorian

delegation at the Scientific and Technical Commission

abandoned the request of setting a 10 km strip along the border

between Colombia and Ecuador where the PECIG would not be
carried out.” 384

5.35. At the same meeting, the re presentative of Colombia

reiterated that, “should sprayings along the border continue, the

technical conditions necessary to prevent the spray from

reaching Ecuadorian territory shall be guaranteed”. 385 In fact, at

this meeting Colombia offered to notify Ecuador when aerial

spraying operations were taking place in the border area so that

samples could be taken and analys es carried out. As noted in

the Minutes of the meeting:

“The Colombian Party shall notify, by the fastest

means, at the moment that such sprayings are
being conducted along the border area, so that the

382 Annex 53: Note Nº SARE-321 from the Director of the National
Narcotics Directorate to the Colombian Foreign Minister, 11 August 2004.
383Ibid.
384
385Ibid.
EM, Vol II. Annex 64.
161 Ecuadorian Commission may take samples and
conduct the respective analyses in a timely
manner.” 386

Colombia also informed its Ecua dorian counterparts that aerial

sprayings for the year had ended in May 2004 and would not be
387
resumed until the following year, unless deemed necessary.

5.36. Shortly after the last meeting of the joint Commission,
the Ecuadorian Government publicly announced its decision to

abandon its position of requesting Colombia to refrain from

spraying in an area of 10 km of Colombian territory parallel to

the border with Ecuador, and stat ed that it would be sufficient

simply to adopt certain preventive measures. The most

important daily newspaper in Ecuador – El Comercio, published

in Quito – ran a front page article on 4 August 2004 transcribing

statements from the Vice-minister of Foreign Affairs of

Ecuador, entitled “Ecuador acce pts sprayings to continue”,
excerpts of which read as follows:

“Thus, the Ecuadorian Foreign Ministry
abandoned the thesis that the Government of
Bogotá was to refrain from carrying out

sprayings with glyphosate in a strip of 10-km
from the border line. The agreement was reached
yesterday, during the cl osing of the fourth
meeting of the Technical Scientific Commission
388
in Quito.’”

386
387 EM, Vol II. Annex 64.
388Ibid.
Annex 139: Press item: “Ecuador accepts sprayings to continue”, El
Comercio (Ecuadorian newspaper), Quito, 4 August 2004.
1625.37. The analyses of the samples collected in July 2004 had

in the meantime been performed by Ecuadorian scientists. The

results were published by the Fore ign Ministry of Ecuador in
Press Bulletin N° 480 dated 26 August 2004, under the title “ No

glyphosate residues were found in Esmeraldas, border with

Colombia”. The Press Bulletin stated, in particular, as follows:

“The Foreign Ministry presented the report of the
Analysis of Glyphosate Residues performed on
water samples collected in the Province of

Esmeraldas (San Lorenzo – Zona Mataje), by
technical personnel of the Ecuadorian
Commission on Atomic Energy, on 14 July 2004.

These studies were carr ied out as follow-up to
the works that the Ecuadorian Scientific and
Technical Commission is currently undertaking,
as implementation of the Minutes signed at the
IV Meeting of the Scientific and Technical

Commissions of Ecuador and Colombia. And
they were carried out by the scientists of the
CEEA using, for that purpose, the EPA 547
procedure, high-tech equipment and the
analytical standards provided by the company
Riedel de Haën, obtaining as a result, the non-

existence of glypho389e re sidues in any of the
analyzed samples.”

5.38. On 15 October 2004, during a meeting held in

Esmeraldas in Ecuador, the Presidents of Colombia and Ecuador
issued a joint declaration in which, inter alia, Ecuador expressed

389 Annex 81 (emphasis added).

163itself satisfied with the explan ations and documents received

from Colombia on the possible effects of the PECIG program:

“9. The Presidents of both countries expressed
they were pleased with the presentations by
experts in the workshops that have taken place in

that regard, and with th e delivery of studies
conducted in Colombia on the possible effects of
the sprayings with glyphosate on illicit crops.” 390

5.39. The Colombian Ministry of Foreign Affairs

subsequently informed Ecuador through diplomatic channels on

4 November 2004, as soon as it itself was informed by the

Colombian National Police, 391that the sprayings had resumed

two days earlier. The relevant diplomatic Note stated:

“The Ministry of Foreign Affairs of the Republic
of Colombia presents its compliments to the
Honourable Embassy of the Republic of Ecuador

and informs that the Anti-narcotics Police has
notified us of the resumption of the aerial
sprayings with glyphosate in the border zone that
392
will extend until late December.”

390 Annex 17, p. 2.
391 The Note of the National Police read as follows: “In accordance
with what was agreed on [at the meeti ng of the] Scientific and Technical

Commission between Colombia and Ecuador, held on 2 August 2004 in
Quito, I kindly ask your Office to inform the Ecuadorian Government that
the National Police – Anti-Narcotics Direction, will carry out from now and
until the end of December spraying operations in the border area.” (emphasis
added) Annex 55: Note Nº 001727 from the General Director of the National
Police of Colombia to the Colombian Foreign Minister, 2 November 2004.
392 Diplomatic Note N DPM/CDR 65881 from the Colombian Foreign
Minister to the Ecuadorian Foreign Minister, 4 November 2004 (EM, Vol. II,
Annex 65), following the Note Nº 001727 from the National Police General
Directorate – Anti-Narcotics Directorate to the Colombian Foreign Minister,
2 November 2004 (CCM, Annex 55).

1645.40. Ecuador complains in its Memorial that this statement

did not make clear “whether th e aerial sprayings had already
393
recommenced.” However, there is no ambiguity in the

language of this letter and it is incorrect to argue, as Ecuador

does, that Colombia’s “promise was quickly broken when, just
three months later, Colombia conducted sprayings in the area

without appropriate notice to Ecuador” 394or that Colombia did

not fulfil its promise to inform Ecuador “by the fastest means, at

the moment that such sprayings are being co nducted in the

border area”. 395 Such notice was given as soon as was

reasonably possible, i.e. immedi ately after the National Police

informed the Foreign Ministry that the aerial sprayings had been

resumed.

5.41. On 8 November 2004, the Government of Ecuador

replied to the Note of 4 November 2004, requesting that
technical measures be taken into account in order to ensure that

the drift caused by the sprayi ngs did not reach Ecuadorian

territory.96 That communication show s that Ecuador had well

understood the meaning of Colombia’s communication.

5.42. Throughout the work of the Joint Commission,

Colombia displayed a cooperative approach and sought to meet

Ecuador’s concerns, to the extent that this was reasonably

393
394 EM, para. 3.45.
395 EM, para. 3.3.
396 EM, para. 3.45.
EM, Vol. II, Annex 66.
165possible. For instance, on thr ee separate occasions during the

course of November and December 2003 and January 2004,

Colombia provided Ecuador with technical studies and other

information in a continuing effo rt to assist Ecuador in its
397
requests.

5.43. Ecuador’s Memorial curs orily dismisses the study

entitled “Environmental Risk of the Herbicide Glyphosate” 398

which was sent to Ecuador on 14 January 2004 as a “wholly

inadequate assessment of the risks of the aerial spraying
399
program”. However, that report had been specially

commissioned by the Anti-Narcotics Directorate of the National

Police of Colombia and, as evidenced by the detailed analysis

contained therein, it did not corr oborate Ecuador’s catastrophic

assessment of the adverse impact of the principal chemical used

in the sprayings.

5.44. Colombia’s cooperative at titude continued in the

subsequent months of 2004. As noted above, no aerial sprayings

were carried out until early November 2004 in the areas adjacent

to the common border with Ecuador, although Colombia

397 EM, Vol. II, Annex 50 (Diplomatic Note VRE/DBR No. 40153
from the Colombian Foreign Minister to the Ecuadorian Foreign Minister, 12
November 2003); CCM, Annex 9; EM, Vol. II, Annex 53 (Diplomatic Note
DM/DBR Nº 47356 from the Colombian Foreign Minister to the Ecuadorian
Foreign Minister, 15 December 2003); EM, Vol. II, Annex 54 (Diplomatic

Note DBR/CAL No. 1405, from the Colombian Foreign Ministry to the
398adorian Embassy in Bogotá, 14 January 2004).
399 EM, Vol. II, Annex 101.
EM, para. 3.33.
166reserved the right to resume the sprayings if its aerial

surveillance revealed the presence of any new illicit crops in the
400
border area. The Joint Scientific and Technical Commission

continued to meet, and on 14 April 2004 Colombia sent Ecuador

further information regarding the aerial sprayings in response to

a new request sent by the Ecuadorian Foreign Minister on 31
401
March 2004. At the fourth meeting of the Joint Commission

held in August 2004, Colombia renewed its commitment that
“should sprayings along the border continue, the technical

conditions necessary to prevent the spraying from reaching

Ecuadorian territory shall be guaranteed”. 402

B. Ecuador Confirmed in December 2004 that There Were
No Adverse Effects Due to the Sprayings

5.45. Ecuador also alleges in its Memorial that, following the

resumption of the sprayings in the border areas in November

2004, “several weeks afterwards , representatives of Ecuador’s

Foreign Ministry visited the communities of El Afilador and
Santa Marianita in northern Sucumbíos in an effort to collect

evidence of the sprayings’ impact. They were unable to do

so.”403 The inference appears to be that failure to collect any

evidence was in some way the fault of Colombia. Since

Ecuador avoids mentioning the factual background in this

400 EM, Vol. II, Annex 57.
401An13.x
402
403 EM, Vol. II. Annex 64.
EM, para. 3.47 (emphasis added).
167respect, it is important to set out in some detail what actually

happened at the time.

5.46. In December 2004, the Foreign Minister of Ecuador

commissioned a high-level dele gation under the Director for

Amazonian and Regional Affairs, to visit the border area of the
Province of Sucumbíos, in order to verify reports of alleged

adverse effects caused by the sprayings, as well as to collect soil

and water samples to check fo r glyphosate residues. Also
forming part of the delegation we re a member of the cabinet of

the Vice-Minister of Foreign Affairs, a colonel from the

Ecuadorian National Police who wa s the head of the Liaison

Office of the Government Ministry at the Foreign Ministry, and
scientists from the Ministry for the Environment and the

Ecuadorian Commission on Atomic Energy. The delegation was

accompanied by representatives of the print media, radio and
television.

5.47. The relevant Press Bulletin issued by the Ecuadorian

Foreign Ministry under the title: “Commission of the Foreign
Ministry finds no evidence of sprayings following verification

in the entire area where sprayings were claimed to have taken

place”, recorded that, during the second part of the field visits
conducted in late December 2004, the Ecuadorian Delegation

collected plant, soil and water samples for analysis:

“The Special Commission appointed by the
Minister of Foreign Affairs to verify the

168 existence of sprayings, concluded the second

stage of a thorough tour of the area where,
according to recent complaints, Colombian
sprayings would have taken place.

The visit conducted on Wednesday, 29
December, this time incl uded the sites of El
Afilador, Santa Marianita and a site facing the
Colombian locality of Azul, sites located on the
banks of the San Miguel River…

...

Engineer Santiago Salazar of the Ministry for the
Environment and Dr. Ramiro Castro, Head of the
Laboratory of the Ecuadorian Commission on
Atomic Energy took samples of plants, soil and

water with the purpose of subjecting them, along
with the evidence collected last week, to
chemical and biological analyses which will be
made public in the next few days.” 404

5.48. It might also be noted th at the Ecuadorian Memorial

completely omits to mention that the results of the first part of

the Delegation’s visit to the ar ea, conducted between 23 and 24

December 2004, and reflected in a Memorandum delivered to
Ecuador’s Foreign Minister, c onfirmed that no flights or

sprayings had taken place over Ecuadorian territory. The

internal Memorandum of the Foreign Ministry of Ecuador of 24

December 2004 reported as follows:

“Thursday, 23 December.

404 Annex 85: Press Bulletin N° 732 of the Ecuadorian Foreign
Ministry, “Commission of the Foreign Ministry finds no evidence of
sprayings following verificationsthe entire area where sprayings were
claimed to have been conducted”, 30 December 2004.
169 In the city of La go Agrio…. having been
informed that in the constant patrols of this
Special Forces Group and Jungle Battalion ‘56
Tungurahua’, there have been no reports of

violations to the Ecuadorian airspace nor have
they had any knowledge of sprayings on the
Ecuador-Colombia border in the past weeks.

Likewise, we were inform ed that the flux of
Colombian and Ecuadorian nationals has been
normal; the acting Governor and the Police
Quartermaster stated that due to the requirement

of the ‘Judicial Record’ [certificate] violence
statistics have dropped, without their having
received any complaints of sprayings or
displacement of Ecuadorians…” 405

5.49. However, the Ecuadorian population did complain to the
delegates about other issues. As recorded in this report:

“….they stated their concern due to the lack of
basic services evidenced in the Province and
requested the Foreign Mini ster to convey to the
Presidency of the Republic, their needs in

[matters of] roads, health, education, 406iculture
and support for the National Police.”

5.50. During the visit to Lago Ag rio, a number of persons

stated that the sprayings had caused damages in Puerto Nuevo,

Puerto Mestanza and their surroundings. However, they
407
provided no evidence of the supposed damage. In an attempt
to verify these complaints, the Delegation decided to visit the

405 Annex 83: Memorandum of the Foreign Ministry of Ecuador, 24
December 2004.
406 Ibid.
407 Ibid.
170locations in question, together with the civil and military

authorities of the Province. Once on site they learned from the

local civilian and military authorities in the area that the
complaints lacked any basis. As recorded in the Memorandum

produced by the Ecuadorian Foreign Ministry dated 24

December 2004:

“Puerto Nuevo: In the afternoon we went to
Puerto Nuevo… where we interviewed the only
civilian authority present, Mr. Lenin López,

Secretary of the Committee for Improvements,
who stated that there was a population of 800,
most of which were Colombian, that there was a
calm environment, that the incoming and
outgoing flux of people was normal, that they are
aware of the existence of sprayings in Colombia,
without having ever seen any aircraft in the zone

or any displacement of Ecuadorians further into
the Province.
[…]

Friday, 24 December. Puerto Mestanza.

In the early morning the Delegation went to
Puerto Mestanza…

We received the testimonies of 2 inhabitants,
who stated that activities were being carried out
as usual, despite the fact that they have heard
rumours of imminent sprayings in the zone. One
of the people interviewed even stated that on
Tuesday the 21 stat noon, he had seen 2 small

aircraft and 5 helicopters in the distance, on the
Colombian side…
…Confirming the information provided by

Colonel José Pastor, Commander of the Special
Forces Group ‘24 Rayo’, Colonel Robert Arauz,
Commander of the Jungle Battalion ‘56

171 Tungurahua’, stated over the telephone to the

Head of the Delegation that the Armed Forces
conduct permanent patrols all over the areas
where there have been claims of alleged
sprayings, without having recorded any
incidents.”08

5.51. During its visit, the Delegation also met with officials of

the United Nations High Commissioner for Refugees (UNHCR)

assigned to the area, one of whom stated his surprise at reports

of alleged sprayings, and declared that they had had no news of

any such sprayings in the sector. Again, the Memorandum of the
Ecuadorian Foreign Ministry of 24 December 2004 states:

“Meeting with UNHCR Delegates.
In the evening, the Delegation met with Mr.

Oscar Butragueño, representative of the United
Nations High Commissioner for Refugees
(UNHCR) in Sucumbíos, who mentioned he had
received a visit by de legates of FORCCOFES
[Federación de Organizaciones Campesinas del

Cordón Fronterizo Ecuatoriano de Sucumbíos ]
last Tuesday, the 21st, during which they had
reportedly informed him of the existence of 900
displaced Ecuadorians, purportedly as a result of
the fear caused by the violence present on the

Colombian side of the border. The UNHCR
official stated that this was the first complaint he
had received on the matter and clarified that he
had not received any news – including UNHCR
in Colombia – concerning sprayings in the sector

and that, moreover, out of all the Colombians
applying for refugee status, only 0.5% have

408An83.x
172 claimed tha409s the reason for coming to this
country...”

5.52. The result of the Delegation’ s first visit and inquiries

was made public through Press Bulletin N° 721 of the Foreign

Minister of Ecuador of 24 December 2004. The conclusion was

categorical:

“The Special Commission appointed by the
Minister of Foreign Af fairs… was able to

physically ascertain, on the basis of testimony
from the population and local authorities that no
spraying whatsoever took place affecting the
Ecuadorian territory in that sector...

…The Commission determined, on the other

hand, that in none of the visited zones had there
been any population displa cement, as a result of
the alleged resumption of sprayings, situation
that was confirmed by the UNCHR

Representative410 Sucumbíos, Oscar
Butragueño.”

5.53. On 29 December 2004, the Delegation delivered a

further Memorandum to the Fo reign Minister of Ecuador,

Patricio Zuquilanda, reporting on the second part of the
411
investigation carried out in the border area. The report stated
that the situation was calm, daily activities were being carried

out normally, the population was in good health and there were

no incidents concerning animals or agriculture. The

409
410An83.x
Annex 82: Press Bulletin Nº 721 of the Ecuadorian Foreign
411istry, 24 December 2004.
An84.x
173Memorandum referred to the visit of the joint Commission on

23 and 24 December 2004, their colle ction of samples and the
evidence that no planes had b een witnessed over Ecuadorian

territory. As noted in the Memorandum of 29 December 2004:

“Pursuant to your instructions and the
announcement by the Foreign Ministry upon the

conclusion of the verification mission that took
place on 23 and 24 December, on this date the
technical commission visited the villages facing
the Colombian sites of El Azul and El Afiladero,
as well as Monterrey, in the sector of Santa
Marianita, with the purpose of fulfilling the
mandate of gathering te stimony and collecting

soil and water samples relating to the alleged
resumption of sprayings in Colombia and their
repercussions in Ecuadorian territory.

On the first verification site, facing El Azul in
Colombia, Dr. Ramiro Castro, of the CEEA, took
a water sample from the San Miguel River, a few
meters off the place where the helicopter landed;
for its part, the technical commission interviewed
two locals – of Colombian nationality – who
inhabited the farm at the visited site. The
interviewees stated that the previous week –

without specifying what day – they had sighted
between six and eight helicopters and two small
aircraft leaving a trail in Colombian territory.
They also indicated that they had no incidents to
report, were carrying out their activities as usual
and that both their animals and their crops were
in good condition.

[…]

On the third site of verification, Monterrey, Santa
Marianita Sector, on the banks of the San Miguel
River, a local, the owner of the visited farm –
also of Colombian nationality – stated to the

174 technical commission that he had not seen or
heard anything out of the ordinary with respect to
the purported sprayings in Colombia. He

indicated that he is in good health and that there
were no incidents with regard to his animals –
cattle, farm birds or his cocoa plants.” 412

5.54. The Ecuadorian Delegation summarized its findings as

follows:

“The technical commissi on has verified in two

stages, as announced to the public, the main sites
named in the complaints gathered by the press
concerning the resumption of sprayings in
Colombia and their purported repercussions in

Ecuadorian territory.

The second stage, completed today, has allowed
us to confirm the conclusion that was arrived at
during last week’s visit: There is an environment
of calm; daily activitie s are conducted normally;

the locals who were interviewed are in good
health; animals and crops are in order. There has
been no violation of the Ecuadorian airspace, nor
has there been any displacement of persons.” 413

5.55. The final part of the Report is clear:

“In any case, even if it was to be accepted that
Colombia had resumed aerial sprayings, the truth
is that the observations conducted and the

gathered testimony do not afford evidence 414any
repercussions in Ecuadorian territory.”

412
413 Annex 84 (emphasis added).
414 Ibid.
Ibid. (Emphasis added).
1755.56. On 30 December 2004, the Ecuadorian Foreign Ministry

issued a further Press Bulleti n, N° 732, entitled: “Commission
of Foreign Ministry finds no evidence of sprayings after

carrying out verifications in the entire area where sprayings

were reported to have b een conducted”. This document

supplemented the information contained in the previous bulletin,
described the latest reconnaissa nce activities carried out and

added the following clarifications:

“The Special Commission appointed by the
Minister of Foreign Affairs to verify the
existence of sprayings, concluded the second

stage of a thorough tour of the area where,
according to recent complaints, Colombian
sprayings would have taken place.

The visit conducted on Wednesday, 29
December, this time incl uded the sites of El
Afilador, Santa Marianita and a site facing the
Colombian locality of Azul, sites located on the
banks of the San Miguel River, where ecologist
organizations had claimed that there had
purportedly been ‘intense sprayings that even
caused the displacement of local residents.’

Out of the physical corrobor ation carried out last
week as well as on this opportunity, the
Commission ascertained that in the entire area

there is an environment of calm and no
displacement of population has been evidenced.
In its report, the Commission observes that the
citizens interviewed in the zone are in good

health, no alterations were observed in plants or
animals, as well as that, according to locals’
accounts, there has been no violation to the

176 national airspace, nor have any vestiges of
415
glyphosate reached Ecuadorian territory.”

5.57. Following the analysis of the soil and water samples

collected in December 2004 by re presentatives of Ecuador, the
Foreign Ministry issued Press Bulletin Nº 027 of 24 January

2005, which reported as follows:

“The Ministry of Foreign Affairs, just as it had
announced days ago, th e reports of the
glyphosate Residual Analysis made on soil and
water samples collected in the Sucumbíos
Province during the two special missions sent

following the claims submitted by farmers living
in the border area and members of Acción
Ecológica, regarding the possible resuming of
aerial sprayings on illicit crops in Colombia and

its alleged repercussions on Ecuadorian territory.
These reports were carried out based on nine
samples collected from the sectors of Tapi,

Puerto Mestanza, La Balastrera, Azul Chiquito,
and the Ecuadorian area facing the Colombian
sector of Afilador.

The result obtained by scientists of the
Ecuadorian Atomic Energy Commission, who
are also members of the Ecuadorian Scientific

and Technical Commission, from the analyses
made is that no glyphos ate residues were found,
the same as on the two previous occasions when
similar claims arose.” 416

415 Annex 85 (emphasis added).
416 Annex 86: Press Bulletin Nº 027 of the Ecuadorian Foreign
Ministry, 24 January 2005. (Emphasis added).
1775.58. In its Memorial, Ecuador omits to mention the findings

of the missions. Instead it suggests that, when the

representatives of Ecuador’s Fo reign Ministry “visited the

[border] communities of El Afilador and Santa Marianita in
northern Sucumbíos in an effort to collect evidence of the

sprayings’ impact”, “[t]hey were unable to do so”. 417 Asthe

record shows, the reason the representatives of the Foreign

Ministry were “unable” to collect evidence of the impact of the
spraying is that there was no such impact. The results of the

analyses conducted at the tim e on the ground by the joint

Commission, combined with Ecuador’s own contemporary

official accounts, provide a str ong contrast with the version of
the facts provided by Ecuador in its Memorial.

5.59. It is also noteworthy that the witness statements gathered

by Ecuador for these proceedings portray an entirely different

scenario than that provided by th e testimonies of “the citizens
interviewed”, collected in the field by the Delegation of the

Ecuadorian Foreign Ministry in 2004.

5.60. A review of some illustrative examples drawn from the
witness statements filed with Ecuador’s Memorial against the

background described above is revealing:

“…I remember particularly a period in 2004 and

early 2005 during which there were many
patients complaining of headache, vomiting and

417 EM, para. 3.47.

178 skin problems, which coincided with border
sprayings…”. 418

“…Finally, in 2004, I had to leave the
community to work in the city of Puerto Nuevo,

because after the sprayings, there was nothing to
harvest in Yana Amarum, and life had become
too hard…”. 419

“…I remember clearly a spraying approximately
four years ago that extremely affected the cattle.
Shortly after the planes came, the cattle died. I

have several neighbors whose cows we420
pregnant and had miscarriages…”

5.61. These statements cannot be reconciled with the

declarations made by the joint Commission in the relevant years

and with Ecuador’s own official reports.

5.62. It should also be noted that the analyses carried out in

July 2004 in the Esmeraldas province, neighbouring the

Colombian province of Nariño, confirmed that no adverse

effects due to the sprayings had been experienced. Thus, there
can be no connection between the alleged adverse effects on

human health and the aerial spraying , at least up until that date.

Sprayings in the border areas ended in December 2005 and no

418 EM, Vol. IV, Annex 188: Declaration of Dino Juan Sánchez
Quishpe, 15 Jan. 2009 (emphasis added).
419
EM, Vol. IV, Annex 212: Declaration of Witness 28, 17 Feb. 2009
420phasis added).
EM, Vol. IV, Annex 198: Declaration of Witness 10, 16 Jan. 2009
(emphasis added).
179further sprayings took place in the relevant ar eas of Nariño
421
province until 17 December 2006.

5.63. The same applies to Sucumbíos province, neighbouring

the Colombian border region of Putumayo, where research and

investigations carried out by th e Ecuadorian authorities shortly

after the sprayings conducted in December 2004 confirmed the

absence of any adverse effects in Ecuador. No further sprayings
took place in this area until 24 September 2005 and they were

suspended thereafter on 11 December 2005. 422

C. The Position of the Parties as from 2005 to 2008

5.64. On 25 July 2005, nearly a y ear after the Ecuadorian

Government had abandoned its position of requesting Colombia

to refrain from spraying in a 10-km area within its territory
423
parallel to the border, the new Government of Ecuador again

changed its position during a meeting between the Foreign

Ministers of the two States, held in Quito, and reverted to

requesting the establishment of a no-spray zone adjacent to the
border. 424 The Colombian Foreign Minister explained the

Government’s policy and maintained that the sprayings were not

harmful.

421An67.x
422Ibid.
423
424 See above para. 5.36.
EM, Vol. IV, Annex 178.
1805.65. In a further meeting at ministerial level, this time held in

Bogotá in August 2005, 425Ecuador reiterated its request for the

establishment of a 10-km no-spray zone, invoking the

precautionary principle. 426 For its part, Colombia stated that it

was not possible to accede to that request. 427 The Colombian

authorities however invited Ecuador to send observers to

observe directly how the aerial spraying was being carried out;

Ecuador did not accept the invitation. 428

5.66. The Colombian Minister of Foreign Affairs expressed

her willingness to undertake stri ct monitoring of the sprayings,

including with the participation of Ecuadorian observers, as well

as to receive and investigate in situ any complaint concerning

alleged damage to human health and/or the environment due to

the effect of aerial sprayings. 429 Again Ecuador did not accept

that offer.430 However, during the cour se of the same meeting,

the Colombian Minister of Fore ign Affairs accepted a proposal

agreed to in a prior joint meeti ng of experts and officials, to

resort to a specialized internat ional organization in order to

conduct studies to determine any possible effects of the use of

425 Annex 87: Press Bulletin Nº 419 of the Ecuadorian Foreign

426istry, 2 September 2005.
Annex 56: Aide-Mémoire “Aerial Spraying Issue with Ecuador”,
Ministry of Foreign Affairs of Colombia, Division of Multilateral Political
Affairs, Sub-division for Drug Affairs, September 2005, pp. 6-7.
427 Annex 87.
428 Annex 56, pp. 6-7, 8.
429 Ibid., pp. 6-7, 8.
430 Ibid., pp. 6-7, 8.

181glyphosate in aerial sprayings, th e conclusions of which would

be submitted to both Governments. 431

5.67. Given that the differences in this regard were having a

growing negative impact on bilateral relations, the Colombian

Vice-Minister of Foreign Affair s announced in a meeting held

with his Ecuadorian counterpart on 17 November 2005 that the

Colombian Government would temporarily suspend the

sprayings in those areas locate d less than 10 km from the

common border. 432 In a meeting between the Foreign Ministers

of both countries held in December 2005, Colombia confirmed

the decision “to suspend tempor arily aerial sprayings in areas

bordering Ecuador as of January 2006”. 433 In fact, the sprayings

were suspended earlier: on 11 December 2005 in Putumayo, and

27 December 2005 in Nariño. 434

5.68. Colombia’s gesture of goodwill is all the more

significant in light of the fact that it was adopted in spite of the

fact that a number of intern ational agencies, including CICAD

and the Inter-American Commission on Human Rights, 435 had

seen no reason to recommend a suspension of the sprayings

431
432 Annex 87.
Annex 88: Press Bulletin Nº 593 of the Ecuadorian Foreign
Ministry, 18 November 2005.
433 EM, Vol. II, Annex 72 (Joint Communiqué, Ecuador-Colombia
Meeting of Foreign Ministers, Quito, 7 December 2005).
434An67.x
435 See below, paras. 5.69-5.70.

182conducted on Colombian territory in the border area with

Ecuador.

5.69. In particular, Ecuador makes no mention in its Memorial

of the fact that the Inter-Amer ican Commission had received a
request for precautionary measures submitted by the

Ombudsman of Ecuador on 18 August 2005, 436 pursuant to

Article 25(1) of the Inter-American Commission’s Rules of

Procedure. Article 25(1) provides that:

“In serious and urgent cases, and whenever
necessary according to the information available,
the Commission may, on its own initiative or at
the request of a party, request that the State

concerned adopt precautionary measures 437 to
prevent irreparable harm to persons.”

5.70. In response to that reque st, the Inter-American

Commission’s Executive Secretary informed the Colombian
rd
Foreign Minister on 18 November 2005 that during its 123
Session, the Inter-American Co mmission had reviewed the

436 Annex 18: Note Nº DSF40.1/3.1.3-4-00421 from the Executive
Secretary of the Inter-Am erican Commission on Human Rights (IACHR) to
the Colombian Foreign Minister, 12 September 2005, Ref. 187-05
“Transboundary effects of the sprayings in Colombia”, referring to the
request for precautionary measures, addressed by the Ombudsman of
Ecuador to the IACHR of the Organizatio n of American States (OAS). The
requested measures were intended to establish mechanisms of surveillance,
control and monitoring, in order to ensure that sprayed chemicals do not
enter Ecuadorian territory.
437
Rules of Procedure of the Inter-American Commission on Human
Rights, Article 25 (emphasis added). Available at:
http://www.cidh.oas.org/Basicos/English/Basic18.RulesOfProcedureIACHR.
htm (last visited 10 March 2010)
183Ecuadorian request and Colombia’s reply. The Note stated that

“in that regard, after considering all the available information on

this question... it was d ecided that for the moment there was no

basis for invoking the mechanism of provisional measures
438
provided for in Article 25 of its Rules”.

5.71. The sprayings in the border area were not resumed until

almost a year later, when they were resumed for a period of

slightly over a month. The sprayings took place from December

2006, at which time the resumption was duly explained and
439
communicated to the Ecuadorian Government. They ceased

in January 2007. No spraying in the border area has taken place
440
since.

5.72. Following the Colombian Government’s acceptance in

August 2005 of resort to a specialized international
441
organization the President of Ecuador at the time, Alfredo
th
Palacio, during his sp eech before the 60 Ordinary Session of

the United Nations General Assembly, referred to the “aerial
spraying of glyphosate as herbicide to elimin ate illicit crops in

areas neighbouring the Ecuador-Colombia border” and

requested that the United Nations system “promote a

comprehensive, reliable and credible study on the actual impact

438 Annex 20: 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 November 2005 (emphasis added).
439 See below, paras. 5.81-5.85.
440An67.x
441 See above, para. 5.66.

184 442
of this spraying.” In a Note addressed to the United Nations
Secretary-General, the Ecuadoria n Government formalized that

request and stated that “the Permanent Mission of Ecuador to

the United Nations is willing to contribute with the Secretary-

General to coordinate the studies that the Ecuadorian
443
Government requests”.

5.73. On 29 November 2005, the United Nations replied to
Ecuador’s request for a study on the impact of the aerial

fumigations as follows:

“It has been agreed to send to Ecuador a
technical mission of the United Nations system,
with a preliminary character... [whose] objective
will be to explore the viability of the requested
444
study”.

5.74. On 7 December 2005, during a bilateral meeting, the

Colombian Foreign Minister agreed to analyze the results of the
study requested by Ecuador from the United Nations and to

assess the adoption of any measures that may be necessary. The

Joint Communiqué issued at the conclusion of the meeting

provided:

“20. Bearing in mind that both Governments

have not reached an agreement on the
innocuousness of the effects of the glyphosate
herbicide and its coadyuvant [ sic] on health and
the environment, the Government of Colombia

442 EM, Vol. II, Annex 68, at p. 2.
443 EM, Vol. II, Annex 69.
444 EM, Vol. II, Annex 71.
185 has duly noted the request made by the
Ecuadorian Government to the United Nations
for a prospective study on this issue and has
agreed to participate in the definition of the terms

of reference of the study. Colombia further
agreed to review the results of the study and
evaluate the adoption of relevant measures.” 445

5.75. On 20 January 2006, the Government of Ecuador

notified the Colombian Government of the commencement of

the United Nations technical mission:

“The Ministry of Foreign Affairs… is honoured
to inform you, to the ends established in

paragraph 20 of the Joint Communiqué of the
Ministers of Foreign Af fairs of Ecuador and
Colombia of 7 December 2005, that the office of
the Representative of the United Nations

Development Programme in Ecuador has notified
this Foreign Ministry that a UN Mission will visit
the country from 13 February to begin working
on a prospective study of the effects of aerial

sprayings with glyphosate and its a446vants on
the northern border of Ecuador.”

5.76. It should be noted that this communication was a simple

notification to Colombia of the visit of the UN delegation to

Ecuador and not an invitation to participate in its work, which in
any event would have been inappr opriate. It is thus not clear

what Ecuador means when it states in the Memorial that

“Colombia chose not to participate” in this visit. 447

445 EM, Vol. II, Annex 72. (Emphasis added)
446 EM, Vol. II, Annex 73. (Emphasis added)
447 See EM, para. 3.56.
1865.77. The Terms of Reference of the UN Mission referred to

the Joint Communiqué and took note of its precise terms, insofar

as they recalled that what Ecuador had requested, and Colombia

agreed to, was that the studies to be pr oposed would be of a

prospective nature. In relevant part, the Terms of Reference

provided:

“On 7 December 2005, the Ministers of External
Relations of both Colombia and Ecuador issued a

joint press release based on a meeting that same
day in which they discussed a number of issues
including the aforementioned fumigations. In

this joint communication, the Government of
Colombia took due note of the GoE’s request to
carry out a prospective study on the impact of
these fumigations.” 448

5.78. Between 13 and 24 February 2006, the “Preliminary

Mission to propose studies on the im pact of the aerial sprayings

and complementary actions in the northern border of Ecuador”

held meetings with public an d private organizations in
Ecuador. 449 In that regard, the Missi on recommended that five

studies be conducted: 450

448
Annex 21: United Nations, Terms of Reference of the “Scoping
Mission for a Detailed Assessment Study on the Potential Impact to Human
Health, the Environment and Agriculture that Occurs in Ecuador’s Territory
as a Result of Glyphosate Fumigations Done by the Government of
Colombia in the Border Area with Ecuador”, January 2006. (Emphasis
added) (Enclosure to Note Nº E-104/032 from the Colombian Ambassador in
Quito to the Colombian Foreign Minister, 27 January 2006).
449 Annex 106, p. 12.
450
It is noteworthy that the Spanish original of EM Annex 28 states
that “options are presented regarding five studies… in health
187 (1) “Retrospectie vpeidemio logical study of morbid-
mortality trends in communities exposed and non

exposed to aerial sprayings”;

(2) “Experimental toxicological studies of acute and

sub-acute effects of the mix used in aerial
sprayings”;

(3) “Study to assess the po ssible impact of aerial

sprayings on water biota and soil in exposed and

non exposed areas in the northern border”;

(4) “a Study to assess the pathogen population

dynamics in soils exposed and non exposed to

aerial sprayings on the northern border” and

(5) “Study on retrospective assessment of the

behaviour of agricultural production in sprayed
451
and non sprayed areas”.

5.79. Taking into account the fact that no evidence had been

produced of any damage caused in Ecuador by the Colombian

spraying program, as well as the short half-life and limited
452
effects of the spray chemicals in the environment, the

(epidemiological retrospective study…)” [“ se presentan opciones sobre
cinco estudios… en la salud (estudio epidemiológico de cohorte
retrospectivo…)”]. However, in its English translation, Ecuador omitted the
word “retrospective”. See: EM, Vol II, Annex 28, p. 5.
451 Annex 106, p. 33.
452 The principal component of the spray mixture, glyphosate, adsorbs
rapidly to soils, and dissipates in amatter of a few weeks; the added
188Colombian Government considered that only studies with regard

to future sprayings were releva nt. It accordingly objected that

the character of the proposed studies was not in conformity with

what had been agreed to by the Foreign Ministers in December
453
2005.

5.80. Ecuador’s Memorial claims that “the Colombian Foreign

Minister’s pledge to suspend aerial fumigations in January 2006
was empty” because “[h]istorically, most of Colombia’s aerial

fumigations along the border had taken place during the final

months of one year and the beginning of the next” and therefore

no further sprayings would have been planned until the end of

2006. 454 The truth is otherwise. First of all, sprayings in any

given area can be carried out at any time of year. For instance,

in 2005, for the Nariño province, the schedule foresaw sprayings

in the months of January, February, March, April, November

and December; In relation to Putumayo, the relevant months
455
were those from July to October.

5.81. More importantly, any suspension of the spraying could

not be accepted lightly as they would represent a serious set-

components also biodegrade rapidly. See below, paras. 7.84-7.88; also, paras.
7.73-7.75, 7.79-7.82, 7.90, 7.92 and 7.94.
453 Annex 26: Diplomatic Note from the Colombian Foreign Ministry
to the Ecuadorian Embassy in Bogotá, 20 December 2006.
454 EM, para. 3.54.
455 The months of operation according to the spraying schedule refer to

the entire territory of each province, and not solely to spraying in those areas
adjoining the border.
189back for Colombia’s fight against drugs. The first suspension of

the sprayings in the 10-km strip parallel to the border as of

December 2005, resulted in a substantial growth of illicit crops

in that strip. In the period during which spraying was
suspended, there was a 72% increase of illicit crops detected

within the 10-km no-spray area. 456This was compounded by an

increasing amount of terrorist and illicit activities perpetrated by

illegal armed groups in the region. The Colombian Government
was compelled to resume aerial spraying operations in that area.

5.82. The difficulties of reconciling these national security

considerations with Ecuador’s demands are evident in a speech
delivered in October 2006 by the Colombian President, who

stated in particular:

“Almost a year ago, for considerations of

fraternity, solidarity, sisterhood with the
Ecuadorian Nation and its Government, we
agreed to suspend the spra yings in a zone of 10
kilometres from the border line, inland in
Colombian territory. The result is very

worrisome.
And I have to state this concern to my fellow
citizens today, and I intend to convey it to my

good friend Alfredo Pala cio, President of the
sister Ecuadorian republic.
Terrorism has abused this decision and has

strengthened the planting in those 10 kilometres.
Today, there are more than 8 [ sic] thousand coca
hectares in the 10 kilometre zone of the border

456 See above, para. 4.34 and see Annex 107, Colombia Coca
Cultivation Survey 2006, p. 27.

190 with Ecuador. It is advancing without measure,
unstoppable. It is a cha llenge to the democratic
world. They are growing it with all of the
technology, with fertilizer s, it would seem that

what is there is a zone of impunity. They grow it
as if it were highly competitive crops in the
lawful sector of the agricultural economy.

What happens with that coca? It fuels
terrorism…” 457

5.83. The Government of Colombia shared its concern with

the Ecuadorian Government at th e highest level. The President

of Colombia called the President of Ecuador on 5 December

2006 and the Colombian Defence Minister did the same with

regard to his Ecuadorian counterpa rt shortly thereafter. In that
regard, the Colombian Minister of Defence stated:

“We explained to the Ec uadorian authorities,
President Uribe spoke to President Palacio, I
spoke to the Minister of Defence of Ecuador, and
we explained to them why we are taking that

step. We cannot allow the production of coca
plants to keep growing, or the proliferation of
processing laboratories, or the presence of the
guerrillas and the increas e in violence in that
458
zone of the country”.

457
Speech by the President of Colombia, Álvaro Uribe Vélez, during
the Consejo Comunal de Gobierno (Communal Government Council) No. 8,
Puerto Asís – Putumayo, 7 October 2006. Available at: At:
http://www.presidencia.gov.co/prensa_new/discursos/discursos2006/octubr…
cc_puertoasis.htm (last visited 10 March 2010)
458 Annex 148: Press item: “Defence Minister welcomes the new Super
Tucano airplanes of the Air Force”, 14 December 2006.
1915.84. The Ministers of Foreign A ffairs and National Defence

of Colombia officially reiterat ed that decision on 12 December

2006 in the following terms:

“...taking into account that it has been a year

since this temporary susp ension, and that it has
resulted in a substantial increase of illicit crops in
that strip, through which narco-terrorism is

financed, the Government of Colombia has
adopted the decision to resume the spraying tasks
in that zone”. 459

5.85. Ecuador claims in its Memorial that Colombia ‘violated’

its commitment made in December 2005 “to suspend further

sprayings altogether”. 460 However, the decision taken by

Colombia and communicated to Ecuador was to “ temporarily

suspend spraying”. 461 This is expressly recognized in paragraph

3.53 of Ecuador’s Memorial. Colombia always reserved the

right to resume spraying in th e border areas if circumstances

required it.

5.86. Two days later, Ecuador protested against Colombia’s

decision and again requested the suspension of the aerial
462
sprayings in the 10-km zone. Colombia responded with two

Notes sent on 20 and 21 December 2006 by the Foreign

459 Annex 58: Communiqué of the Ministers of Foreign Affairs and
National Defence of Colombia, 12 December 2006 (emphasis added).
460 EM, para. 3.59 (emphasis added).
461
462 See above, para. 5.67.
EM, Vol. II, Annex 76
192Minister and President of Colombia, respectively. 463 Those

Notes are ignored by Ecuador in its Memorial.

5.87. The note sent by the Colombian Minister of Foreign

Affairs, provided in relation to the resumption of the sprayings:

“As expressed in the Press Communiqué of the
Ministries of Foreig n Affairs and National

Defence on 12 December 2006, the Government
of Colombia decided to resume aerial spraying
tasks within 10 kilometres from the borderline
shared by the two countries, due to the significant
increase of illicit crops in this area, which are

used to promote narco- terrorism activities,
posing a great risk to people and democracy in
Colombia. In this conn ection, the Ministry of
Foreign Affairs kindly reasserts to the

Honourable Government of Ecuador the
invitation to understand this measure that the
Colombian Government was forced to take for
obvious reasons of national security, and in
compliance with the obligations our country has

in the fight against the world drug problem, a
scourge with devastati ng effects, not only in
Colombia but in the entire region.” 464

5.88. In his Note to the Ecuadorian President, the Colombian

President explained that the reasons justifying the resumption of

the sprayings included:

“The alarming proliferati on of illicit crops as a
result of that decision [sc. the temporary

suspension of aerial spra yings in a 10-km strip

463 Annex 26; Annex 27: Note from the President of Colombia to the
President of Ecuador, 21 December 2006.
464An26.x
193 inside Colombian territory adjacent to the
border], as well as the unfortunate consequences
that those illicit crops have in the fight against
the world drug problem and narco-terrorism”. 465

5.89. Soon thereafter, at a bila teral meeting between the

President of Colombia, Álvaro Uribe, and the then recently

elected President of Ecuador, Rafael Correa, held on 10 January

2007, the Parties agreed to establ ish a second Scientific and

Technical Commission. That Co mmission was entrusted with a

task similar to that of the first Joint Scientific and Technical

Commission, i.e. to determine, through the means of in situ
fieldwork involving sample-tak ing and verification, whether

glyphosate used in the eradication of illicit crops in Colombian

territory had any impact on Ecuadorian territory. 466

5.90. In a further gesture of good will towards Ecuador and

with a view to promoting a nd improving bilateral relations,
notwithstanding the risks and difficulties with which Colombia

was confronted, Colombia again decided to suspend the aerial

sprayings in areas within the 10-km strip of Colombian territory

running parallel to the borde r. The suspension of aerial

sprayings has been continuously maintained ever since.

5.91. The decision to suspend the fumigations in a 10-km
corridor parallel to the border with Ecuador was confirmed by

465An27.x
466 EM, p. 88, para. 3.65.
194Colombia in a press release issu ed by the Minist ry of Foreign

Affairs on 8 February 2007. The press release reads as follows:

“The Minister of Foreign Affairs of Colombia,
María Consuelo Araújo, announced that

Colombia ceased yesterday the aerial sprayings
carried out in the [Provinces] of Nariño and
Putumayo – bordering Ecuador. Likewise, she
confirmed the beginning of manual eradication
467
works on 12 February…”

5.92. On 9 February 2007, the G overnment of Colombia

addressed a Note to the Ecuadorian Government referring to the
suspension of aerial sprayings on the border:

“I wish to inform that on this date the aforesaid
spraying program was concluded and next week,
the manual eradication program, to which over
468
2,200 people will be devoted, begins.”

5.93. In its Memorial, Ecuador claims that during a meeting

between the Foreign Ministers in May 2007, the Colombian

Foreign Minister “concluded by communicating his
government’s position that it would not suspend aerial spraying

operations in the border area”. 469 However, Ecuador omits to

mention that, at that time spra ying near the border had already

been suspended by the Colombian Government some three

months earlier and that, as reca lled above, that suspension had

467
Annex 60: Press Release of the Colombian Foreign Ministry, 8
468ruary 2007.
Annex 28: Diplomatic Note DM/VRE Nº 6454 from the Colombian
469eign Minister to the Ecuadorian Foreign Minister, 9 February 2007.
EM, para. 3.72 (emphasis added).
195officially been notified by a diplomatic Note from the
470
Colombian Foreign Minister.

5.94. The first meeting of the new Scientific and Technical

Commission took place in April 2007. At the meeting the two
delegations agreed to exchange , through diplomatic channels,

documents in support of their respective views. 471 During the

meeting, Colombia renewed its willingness to notify Ecuador if

sprayings were scheduled to resume again in the area, so as to

allow Ecuadorian officials to verify the lack of any effects in

Ecuadorian territory.

5.95. Ecuador’s Memorial contends – on the basis of an Aide-

Mémoire prepared by Ecuado r’s Scientific-Technical

Commission at the time – that at the first meeting of the new

Commission, Colombia refused to provide Ecuador with the
chemical formulation used in the spray mixture and that it

rejected Ecuador’s proposal that the Commission determine the

implications of the precautionary principle for the spraying

program. Ecuador also alleges that Colombia never sent to

Ecuador the environmental im pact study concerning the
472
sprayings. However, the position then expressed by the

Colombian Government was far from being unreasonable.

470
471An28.x
472 EM, Vol. II, Annex 80.
EM, paras. 3.68-3.69.
1965.96. First, it is not at all clear from the Aide-Mémoire of the

meeting that the Ecuadorian delegation requested the

composition of the formula, but only that it stated that an “issue

to consider was the chemical composition used for the

eradication of illicit crops”. The spray mix was a matter of

public knowledge, and had been communicated to Ecuador as
473
early as 2001; the Environmental Management Plan of the

spraying program had also been furnished to Ecuador in
474
2003. Further, the position taken by the Colombian delegation

that “it should first be prove d that glyphosate crosses into

Ecuador” before the precautionary principle could be invoked

was perfectly reasonable.

5.97. The report of the Ecuadoria n Scientific and Technical

Commission of April 2007 describes the composition of the
475
spray mix based on Colombian official sources. Having

asserted that “about Cosmofl ux, nothing is known”, the report

provides a lengthy descrip tion of the product and its
476
components. However, the Report lacks scientific rigor and
even goes as far as to question the validity of the scientific

studies conducted by international organizations. 477

473 See para 6.3-6.19.
474 Annex 9, para. 5.27 above; also, para. 6.25 and note 524 below.
475 EM, Vol. III, Annex 157, at pp. 23-24.
476 EM, Vol. III, Annex 157, at p. 30.
477 EM, Vol. III, Annex 157, p. 35 (not translated by Ecuador).

1975.98. The Joint Commission, according to its terms of

reference agreed upon by the G overnments of Colombia and
Ecuador, had the mandate of carry ing out work of a strictly

scientific and technical nature without the power to commit the

Governments in actions or projec ts of any kind. However, the

Ecuadorian Commission strayed from that role and displayed a
markedly political stance in trying to impose the notion that the

sprayings caused damage in Ec uadorian territory, rather than

undertaking the scientific works (f ield visits, studies) that the
Joint Commission had been entrusted with.

5.99. The new Joint Commission met for the second – and last

– time on 9 July 2007. Without attempting to en ter into the
slightest debate on the substa ntive issues, the Ecuadorian

delegation claimed at the outse t that they had encountered

certain “scientific disagreements” and demanded that in order to
continue the dialogue, the Colombian representatives should

accept the Ecuadorian position from the outset that:

“- Glyphosate is not harmless to health and the
environment;

- the problem is not limited to glyphosate and
drift;

- the precautionary principle must be applied, as
there are well-sustaine d scientific doubts and
technical disagreements;

- while the scientific dialogue is underway, aerial
sprayings should be suspended within a 10-km
strip from the border line;

198 - if joint studies are carried out, their purpose

should be to assess the dam478 and establish
compensation measures.”

5.100. The Colombian delegation replied that it could not share

the views of the Ecuadorian delegation on the matter and

proposed that a joint methodology of analysis of risk assessment
be developed, complemented by the corresponding fieldwork

that would allow the Commission to fulfil the task entrusted to

it, i.e., to determine whether sprayed glyphosate could reach and

impact upon Ecuadorian territory due to drift. That proposal was

flatly rejected by Ecuador in the following terms:

“The Ecuador Scientific Commission argued that
it could only continue the dialogue based on the

recognition of the scientific reports of the
damages and a comprehensive approach that
warrant the immediate application of the
precautionary principle.” 479

5.101. On 12 July 2007, a communiqué from the Ministry of

Foreign Affairs of Colombia declared that:

“The programs for the eradication of illicit crops,

and particularly eradicati on by aerial sprayings,
have been conducted by Colombia in full
exercise of its sovereignty, exclusively within the
national territory.

The Government of Colombia reiterates the
importance of collaboration between
neighbouring countries in the fight against

terrorism and the world drug problem, and within

478 EM, Vol. II, Annex 83.
479 EM, Vol. II, Annex 83
199 the framework of that effort, the consolidation of

security in Colombia, that represents security for
the region.

Since the month of February of the current year,
no aerial sprayings are be ing carried out in the
vicinity of the border w ith Ecuador and instead,
the manual eradication of illicit crops is being

undertaken, ef480ts to which over 1,100 people
are devoted.”

5.102. Notwithstanding Colombia’s straightforward

representations, on 16 July 2007 the Ecuadorian Minister of
Foreign Affairs declared at a press conference that she regretted

the outcome of the meetings of the Scientific and Technical

Commissions, that she consider ed the diplomatic means of

resolution to be exhausted and announced her Government’s

decision to bring a case against Colombia before the

International Court of Justice.

5.103. On 18 July 2007 Colombia reiterated its willingness to

continue with the mechanism of the Commission, although

insisting on the performance of the field work that both

commissions had deemed necessary to carry out following the
481
first meeting of the Commission on 10 April 2007. Ecuador,
in its reply of 27 July 2007, asserted that the task of the Joint

Scientific Commission had ended a nd considered that “the path

480 Annex 62: Communiqué of the Colombian Foreign Ministry, 12
July 2007.
481 EM, Vol. II, Annex 85.
200 482
of dialogue” had been exhausted. In addition, it demanded

that the Colombian Government “…formalize a written

commitment regarding the defi nitive suspension of aerial

sprayings, and thus cause the payment of indemnities to

compensate the harmful effects that the sprayings have had on
the health and the environment [to] become viable.” 483

5.104. During the course of 2007, the Government of Colombia

had repeatedly expressed, through its highest officials, its

willingness to receive and examine the complaints of

Ecuadorian citizens and to indemnify any actual damages which

were shown to have occurred as a result of the aerial sprayings

carried out in Colombian territory. For example, at the regional

summit on drugs held in March 2007, the President of Colombia

publicly stated that:

“We have offered –and I so reiterate at this
Summit, before its distinguished participants-

that if any citizen of a sister nation shows that the
Colombian fumigation has caused him harm, the
Government of Colombia will immediately
indemnify him, as it does with its own
484
Colombian nationals.”

482 EM, Vol. II, Annex 86.
483 EM, Vol. II, Annex 86.
484 See, Speech of Colombian Pres ident Alvaro Uribe at the Santo

Domingo Regional Summit on Drugs, Security and Cooperation, 16 March
2007. Available at:
http://www.presidencia.gov.co/prensa_new/sne/2007/marzo/16/09162007.ht
m (last visited 10 March 2010).
2015.105. Despite that offer, to date, not a single Ecuadorian
citizen residing in Ecuador has requested the Colombian

authorities, whether directly or through the Ecuadorian

Government, to provide any co mpensation for actual adverse

effects suffered by reason of th e aerial spraying conducted in
485
Colombian territory.

5.106. In February 2008, in a Diplom atic Note sent to his
Ecuadorian counterpart, the Colombian Foreign Minister stated:

“In this regard, please allow me to again express
the interest of the National Government in
attending to the complaints of Ecuadorian

citizens, with the purpose of paying
indemnification for real and ascertainable
damages, and through the most expedient
mechanism possible, for what the corresponding
legal analyses are being done…” 486

5.107. Ecuador however rejected out right Colombia’s offer to
consider the payment of comp ensation on a case-by-case basis

on the ground that, inter alia , Colombia had not accepted

Ecuador’s demand for an ag reement between the two

governments providing for the definitive suspension of the

aerial sprayings in the areas within Colombian territory adjacent

to the border:

485
It is true that substantial damages have been claimed in court
proceedings in the United States agains t the US company that is contracted
by the US State Department to assist in the spraying operations. Colombia is
not a party to those proceedingsSee above, para. 1.35.
486 EM, Vol. II, Annex 87.
202 “In particular, the [Colombian] note does not

accept the Ecuadorian Government’s demand
that the Government of Colombia sign a formal
agreement and of mandatory undertaking to stop,
definitively and permanently, the aerial sprayings
within 10 kilometres from the border between

Ecuador and Colombia. The note does not accept
in satisfactory terms either, the demand from the
Ecuadorian Government that the Government of
Colombia should indemnify it.” 487

5.108. More than one year after Co lombia suspended the aerial

sprayings on the border area, a nd in spite of that suspension,

Ecuador instituted the present proceedings against Colombia

before the Court.

D. Conclusions

5.109. The account provided in Ecuador’s Memorial of the

diplomatic exchanges, of the f acts and of the actions of the
Colombian Government is inaccurate and misleading. The

documents and other evidence collated in this Chapter

establishes the following:

(1) Ecuador began to speculate on the adverse

impacts the program might have on its territory

without evidence of any kind, and did so even

before the PECIG program was implemented in
the Colombian provinces bordering Ecuador.

487 EM, Vol. II, Annex 88 (Diplomatic Note N°
14087/GM/GVMRE/SSNRF/2008 from the Ministry of Foreign Affairs of
Ecuador to the Embassy of Argentina in Bogotá, 24 March 2008).
203 (2) Colombia was always prepared to pursue

discussions with Ecua dor and cooperated in

addressing Ecuador’s concerns including, inter

alia, by participating in bilateral meetings at
different levels, organizing workshops and

seminars, establishing scientific and technical

commissions, conducti ng site visits, and

supplying Ecuador with technical documents,
scientific studies. Colombia even provided

technical assistance for the establishment of a

public health surveillance system on intoxication
488
by pesticides which Ecuador lacked.

(3) The make-up of the approved spray mix was

never unknown. It was widely divulged by the
competent authorities of Colombia and published

in the press and the Official Journal of Colombia

in both 2001 and 2003. Details of the

composition were also provided in the course of

the diplomatic exchanges and documents
furnished by Colombia to Ecuador, and further

explained, together with the relevant specific

procedures followed in the Program, during

bilateral training and information sessions held
by the entities responsib le for carrying out the

Program.

488 See above, paras. 2.37-2.38 on the Colombia-Ecuador agreement for
a public health surveillance system.

204(4) In November 2003, Colombia sent to the

Ecuadorian Scientific and Technical Commission
the PECIG program’s Environmental

Management Plan, a document which pursuant to

the applicable regime in force in Colombia, was

equivalent to an environmental impact
assessment.

(5) With a view to identifying any possible effects of

the spraying program, the Colombian

Government resorted to a specialized body of the

OAS, CICAD, to request an environmental and
human health scientific assessment of the aerial

spray program. The Ecuadorian Government

chose not to take part in the second phase of the
CICAD study, despite havi ng been invited to do

so by both CICAD and Colombia.

(6) Without any scientific assessment or technical

basis, starting in July 2001 Ecuador began to

request Colombia to refrain from spraying in a
10-km zone parallel to their common border.

Although the Ecuadorian Government at one

point desisted in its requests for the establishment

of a 10-km buffer zone, the Government of
Colombia suspended the sprayings from

December 2005, as a gesture of goodwill towards

Ecuador. Given the renewed increase of illicit

205 crops in the affected areas during the period in

which the sprayings were suspended, Colombia
had no choice but resume the operations for a

brief period, between December 2006 and

January 2007. From the latter date, again as a

goodwill gesture towards Ecuador’s new
administration, the aerial sprayings in the border

region were suspended once more, a situation

that continues to date.

(7) On 18 November 2005, the Inter-American

Commission on Human Rights, in answer to an
Ecuadorian request for precautionary measures in

that regard, found no reason to recommend the

suspension of the sprayings conducted on
Colombian territory in the border area with

Ecuador. Moreover, nothi ng in the scientific

studies and evidence deriving from specialized

organizations, such as CICAD, indicates that any
suspension of the sprayings was called for.

(8) The protests and concer ns voiced in Ecuador

with regard to the alleged effects of the sprayings

were investigated by th e Ecuadorian authorities

themselves, who came to the conclusion that they
were unfounded. Indeed, the Ecuadorian

authorities themselves verified in situ that the

alleged effects had not occurred. It was the

206 Ecuadorian authorities, in cluding high officials

from the Ministries of Foreign Affairs and the
Environment, together with the local authorities

and commanders of the military units assigned to

the border, who were ab le to verify, through

gathered testimony and studies, that up until
December 2004, there had been no damage to

human health, the environment, wildlife,

domestic animals, water or soils in Ecuador as a
result of the spraying.

(9) As to the years subsequent to 2004, following the
verification by Ecuadorian authorities that there

had been no damage in Ecuadorian territory due

to the spraying oper ations conducted in
Colombia, the only time periods in which

sprayings took place in the border zone with

Ecuador have been between 14 January and 26

December 2005, and between 17 December 2006
and 14 January 2007 as regards those areas in

Colombia contiguous to the border of the

Colombian province of Nariño where sprayings
were actually conduct ed; and between 26

September and 10 December 2005, and again

between 11 December 2006 and 21 January 2007

as regards those areas in Colombia contiguous to
the border of the Colombian province of

207 Putumayo. None of the allegations of damage or

of spraying flights co ntained in the witness
statements submitted by Ecuador corresponds to

those time periods. No damages could have

occurred since.

5.110. In the light of a situation where no scientifically rigorous

evidence has been adduced of any harmful effects of the aerial

sprayings in Ecuador or on its population and the environment,
Colombia’s willingness to suspend the sprayings in the border

area, in spite of the inevitable security risks associated with

other eradication methods, such as manual eradication, is

noteworthy.

5.111. In spite of Colombia’s efforts, following the suspension

of aerial spraying in the border area from 2007 onwards, and
while the proceedings relating to the second Joint Scientific and

Technical Commission were on-going, the Ecuadorian

Government suddenly announced its intention to bring a case

against Colombia before the Court, thus unilaterally putting an
end to the dialogue between the Parties.

208 Chapter 6

CERTAIN DISPUTED QUESTIONS OF FACT

A. Introduction

6.1. As the Court will have obser ved, the present dispute

largely concerns questions of fact – whether alleged harm
suffered by Ecuadorian i nhabitants in the border region, and/or

the environment of Ecuador, can be shown actually to have

occurred, and whether such harm could have been caused by the
aerial spraying. There are also many particular disputed

questions of fact: for example, whether Colombia ever promised

to terminate, as distinct from suspending, spraying missions in a

10-km zone on its side of the bor der. Many of these factual
issues have already been disposed of in Chapters 4 and 5. In

this Chapter, three particular i ssues of fact are dealt with in

greater detail. They are:

• Colombia’s alleged failure to disclose the ingredients

of the spray mixture;

• Colombia’s alleged failure to deliver an

environmental impact assessment of impacts of
spraying on Ecuador;

• Colombia’s alleged failure to notify spraying
missions;

These claims will be dealt with in turn.
209 B. Alleged Failure to Disclose the Formula

6.2. Ecuador’s Memorial repeatedly asserts that the mix used

in the sprayings in the zones close to the border is unknown,
implying or insinuating that its composition has been kept secret

by the Colombian authorities. 489 Those assertions lack any

foundation in fact. As stated above, 490 in the operations

conducted in the areas near th e border with Ecuador between

2000 and 2007 the Colombian authorities used exactly the same

mix and spraying procedures as were applied in the rest of
Colombia’s territory.

6.3. In fact the composition and characteristics of the spray

mixture used in the PEGIC program have been widely known

and divulged since 2001 in a va riety of publicly available
documents issued by the competent Colombian authorities.

Colombia has at no point sought to conceal or keep secret the

formula of the sprayed mix, nor could it have done so in light is

its own laws and regulations governing the use of herbicides.

6.4. Further, during the periods in which the sprayings took
place near the border, the Ecuadorian authorities carried out

water analyses on a number of occasions, arriving at the

conclusion that there were no residues of either glyphosate or its

489 See e.g. EM, paras. 1.14, 1.17, 2.7, 2.38-2.39, 2.42-2.43, 5.2, 5.5,
5.19, 5.21-5.22, 5.27, 6.51, 8.40, 8.58, 10.11-10.12 . See also Menzie
Report, Section 3 which (no doubt on instructions) is similarly confused.
490 See above, para. 4.3, and para. 7.43 below.
210 491
metabolite AMPA. There was no mystery about what they

were looking for.

(1) P UBLICLY AVAILABLE C OLOMBIAN DOCUMENTS

6.5. In fact the composition of the spraying mixture was

officially published in 2001, as were subsequent modifications.

6.6. The Colombian Environment Ministry, in Resolution Nº

1065 of 20 November 2001, which is included among the

annexes to Ecuador’s Memorial 492and is freely available on the

Internet, accepted the dosage recommended by the Colombian

Agriculture and Livestock Institute ( Instituto Colombiano

Agropecuario (ICA)) of “8 litres of mix per hectare (Roundup

480 SL + Cosmo-Flux 411)”. 493 AsotiChpt,4

Roundup is a proprietary herbicide whose active ingredient is
494
glyphosate; it is available over the counter in many countries.

Cosmo-Flux 411 is an adjuvant, licensed for use in Colombia
495
and surrounding countries.

6.7. The Ministry based its decision on the opinion issued by
496
the Health Ministry in October 2001 and on the field

491 See above paras. 5.30, 5.31, 5.32 and 5.37; see also paras. 5.56-5.57.
492 EM, Vol. II, Annex 15.
493 EM, Vol. II, Annex 15. (Resolution N° 1065 of 2001, Technical
Considerations, 1 Follow-up to Article 9 of the Resolution 341 of 2000,
Tenth Article, second para.)
494 See above, paras. 4.47, 4.50.
495
496 See above, paras. 4.51, 4.56.
An44.x
211assessments of the agronomical efficiency of the mix performed
497
by ICA.

6.8. The Resolution also stipulated that the mix and approved

dosage were to be maintained and that any change in the
formula would require prior approval by the ICA. Article Tenth

of the Whereas part of the Resolution reads as follows:

“In the event that DNE [National Narcotics
Directorate] intends to use higher doses than
those foreseen here, it shall carry out the relevant

assessments pursuant to the procedure and
protocols approved by ICA to that effect, and
obtain prior approval from that Institute… In the
event that DNE intends to use mixtures with

Glyphosate as the active in gredient, different to
those assessed by ICA and the Health Ministry,
DNE shall carry out the relevant assessments

pursuant to the procedure and protocols approved 498
by ICA and the Health Ministry to that effect.”

6.9. In January 2003, following a favourable technical
opinion rendered by the ICA issued on the basis of differential

analyses of dose efficacy submitted by the National Narcotics

Directorate, 499the Ministry for the Environment authorized the

497 EM, Vol. II, Annex 15 (Resolution N° 1065 of 2001, Technical
Considerations, 1 Follow-up to Article 9 of the Resolution 341 of 2000,
Tenth Article, second para.)
498 EM, Vol. II, Annex 15 (Resolution N° 1065 of 2001, Technical

Considerations, 1 Follow-up to Article 9 of the Resolution 341 of 2000,
499th Article, fourth and fifth paras.).
Annex 47: Note Nº 00500 from Assistant Manager for Agricultural
Protection and Regulation of the Colombian Agriculture and Livestock
212National Narcotics Directorate to use a higher dose per hectare

of glyphosate in the spray mix (an increase to 10.4 litres/ha). 500

Resolution N° 0099 of 2003 was published in the Official
501
Journal of Colombia on 21 February 2003. In relevant part, it

provides:

“Pursuant to the provisions of Resolution 1065 of
2001, the Ministry for the Environment accepts
the opinion issued by the Instituto Colombiano
Agropecuario, ICA, for the provisional increase

of the dose of the commercial formula of the
glyphosate herbicide to 10.4 litres/ha, the mix
being Roundup 480 SL (10.4 litres/ha) + Cosmo-

Flux 411 (0.25 litres) + water (13 litres), for the
eradication of coca crops, in the framework of
the Program for the eradication of illicit crops

with glypho502e [PECIG ] in the national
territory.”

6.10. A number of websites, including some to which

reference is made in Ecuador’s Me morial and in particular that

of the U.S. Department of State, contain detailed information on

the composition of the mix used in aerial spraying to eradicate

illicit crops in Colombia. 503

Institute to the Director of the Natio nal Narcotics Direct orate, 28 January

5003.
Annex 48: Resolution Nº 099 of 31 January 2003 of the Ministry for
501 Environment of Colombia.
502 Ibid.
Annex 48, Preliminary Section.
503 United States Department of St ate, 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,
September 2002 (EM, Vol. III, Annex 144). See also, United States
Environmental Protection Agency (EPA), Details of the Consultation for

213 (2) C OMMUNICATIONS BY C OLOMBIA TO ECUADOR (2000-

2004)

6.11. Quite apart from the fact that information as to the

composition of the spray mix was, and remains, widely

available, the Colombian authorities in fact explained the
composition of the mix and the modalities of its use in detail to

the Ecuadorian authorities on a number of occasions.

6.12. For example, in Octobe r 2000, a Meeting of Vice-

Ministers of Foreign Affairs of the Andean Community was

held in Caracas. The agenda included an item called “Early
Warning Mechanism for the dete ction of biological control

agents in the eradication of illicit crops, that affect Andean

ecosystems”. During the meeting, the Colombian Vice-Minister

of Foreign Affairs stated:

“The Government of Colo mbia has stated that it
does not approve under any circumstance,

experimenting with mycoherbicides exogenous
to our ecosystems and that might affect the
environmental balance and health of the
population. In partic ular, it has rejected

experimentation with fusarium oxysporum. It is
important to clarify that504 mycoherbicide is
currently being studied.”

Department of State: Use of Pesticides for Coca and Poppy Eradication
Program in Colombia, August 2002. In Annex 142, p. 13 (partially at EM,
Vol. III, Annex 143).
504 Annex 136: Report of the Fifth Meeting of Vice-Ministers of
Foreign Affairs of the Andean Community, Caracas, 16-17 October 2000.
2146.13. For his part, at the same meeting the Vice-Minister of

Foreign Affairs of Ecuador stated that:

“…complete and absolute assurances were
received from the authorities of the Colombian
Government to the effect that no use or
experimenting with the fungus fusarium

oxysporum or any other type of mycoherbicide
will be undertaken for the programs for the
eradication of illicit crops. Thus, the
Government of Ecuador deems it unnecessary to
bring this issue to th e next meeting of the
505
Council”.

6.14. The Ministry for the Environment of Colombia
subsequently reiterated the deci sion not to use any biological

agents:

“The Ministry for the Environment, as the
highest environmental authority in the country, in
relation to the possible experimenting, use or
application of the fungus Fusarium oxysporum,

as a mechanism for the eradication of illicit crops
in the national territory, reiterates the following:
The Ministry for the Environment DID NOT

accept the proposal advanced by the United
Nations International Drug Program (UNDCP),
to conduct tests with Fusarium oxysporum
Erythroxylum mycoherbicide, given that it

considers that any agent external to our country’s
native ecosystems might pose serious hazards to
the environment and human health.” 506

505 Ibid.
506An42.x
215This position of the Colombian Government remains

unchanged; the use of biological agents for the control of illicit
crops has never been contemplated.

6.15. A diplomatic Note sent by the Colombian Foreign

Ministry on 14 July 2001 explained each of the ingredients of
the mix and included information on their toxicological

classification. It stated:

“Furthermore, Colombia uses products which
have been demonstrated to have no harmful
effects and it has based its illicit crops eradication

program on reliable and consistent international
scientific studies.
Indeed, the chemical substances used against

illicit crops are the same used by both countries
on their plantations of bananas and flowers for
export... For a better illustration of the matter, we
have been guided by two studies prepared by
NAS /Plan Colombia – PECI and by the
Counternarcotics Direct orate of the National
Police of Colombia. I would like to stress the
following points contained in these studies:

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 company Monsanto Inc.
Its toxicological category is No. IV, which
means that it is at the bottom of the universally

accepted toxicity scale.
[...]

Glyphosate, POEA and the herbicide Roundup
have been tested in numerous sub-chronic,

216 chronic, reproductive, and developmental studies,
including tests to determine adverse effects on
the nervous system, and the conclusion is that

there is no evidence of neurotoxicity in any of
these studies. Therefore, it is evident that no
neuropathies or alterations of the nervous system

or alterations of the foetal nervous system have
been observed.

The surfactant Cosmo-Flux 411F is a coadjuvant
commonly used in commercial farming, added to
many herbicides used in many cash crops,
including rice, corn, sorghum, soy, etc.

Dioxin is 100 times below WHO and FAO

standards, so there is no reason for c507ern
regarding human or animal health.”

6.16. In July 2001 Colombia also proposed the holding of a
seminar/workshop with the purpose of informing Ecuadorian

officials of technical aspects relating to the eradication of illicit

crops by aerial spraying with glyphosate in Colombia. During

the seminar, which was held in Bogotá in February 2002, the

composition of the spray mix was explained in detail to the
508
Ecuadorian delegation.

6.17. In November 2003, in response to an Ecuadorian
509
request, the Colombian Government submitted to the Foreign
510
Minister of Ecuador a toxicity study of the spray mix entitled:

“Toxicity studies in laboratory animals – Inmunopharmos Ltd.”

507
508 EM, Vol. II, Annex 42.
509 See above, paras. 5.11-5.16.
510 EM, Vol. II, Annex 52.
EM, Vol. II, Annex 50.
217This study expressly specified the components of the mix as

follows: “Glyphosate 44% + Cosmoflux 1% + Water 55%”. 511

6.18. In 2004, in reply to a further request from the chairman

of Ecuador’s Scientific and Technical Commission, 512

Colombia’s National Narcotics Directorate again submitted

precise and detailed information on the spray mix. The 2004

Note specified as follows:

“Concentrations used for the spraying of coca
crops:

Mix of glyphosate, wa ter and adjuvant 23,65
litres/ha (10,4 litres of glyphosate per hectare).
2
Spraying of 1,04 millilitres/m .

Maximum concentration of 480 mg/litre.
2
Every m sprayed receives 0,499 ml of active
ingredient.

Concentration below critical values and way far
from LD-50 [lethal dose 50] and LC-50 [lethal
513
concentration 50].”

6.19. Together with the 2004 Note, which was addressed to

the Ecuadorian Scientific and Technical Commission, the

Ecuadorian Government was again furnished with the toxicity

511
Annex 128: O. Saavedra, Inmunopharmos Ltda., Toxicity Study on
Laboratory Animals for two concentrations of Glyphosate 44% + Cosmoflux
1% + Water 55%, Bogotá, 15 February 2002, pp. 1, 2, 7, 13, 14, 37, 38, 45,
50, 62, 67, 74, 77, 80, 88.
512 Annex 11: Note N o CMFS-46-2004 from the President of the
Ecuadorian Scientific and Technical Commission to the Acting Director of
the Anti-Narcotics Direction of the Colombian National Police, 18 March
2004.
513An13.x

218study on animals conducted by the Inmunopharmos laboratory,
and the technical data sheet of the adjuvant Cosmo-Flux 411F

used, which specified the chemical features and toxicological

aspects of that product.

(3) C OLOMBIAN E NVIRONMENTAL A UDIT

6.20. Since 1994, Colombia’s National Narcotics Council has
required that aerial eradication ope rations must be subject to an

external environmental audit in order to ensure that the mix with

which the aerial spraying aircrafts are loaded conforms with the

relevant regulations as to permitted composition and dosage
514
under Colombian law. That requirement was reiterated in a

new Resolution in 2003, as follows:

“The PECIG shall have an external technical
audit […]

The general tasks of the technical audit will be
the following:

- Validate whether the execution of the PECIG’s
operative activities adhered to the procedures and

guidelines set in the PMA515nvironmental
Management Plan) [...].”

6.21. The external environmental auditors and the ICA carry

out random quality control opera tions on the spray mix through

the collection of samples that are sent for analysis to the
National Laboratory of Agricultural Products ( Laboratorio

514 Annex 37; Annex 41.
515 Annex 49: Resolution Nº 013 of 27 June 2003, from the National
Narcotics Council of Colombia (Article 5).
219 516
Nacional de Insumos Agrícolas (LANIA)). In all the analyses

performed, the concentration of the active ingredient,

glyphosate, has been found to be 480 milligrams/litre, just as

has been repeatedly reported and officially indicated to
517
Ecuador.

C. Alleged Failure to Deliver an Environmental Impact

Assessment

6.22. Ecuador alleges in the Memorial that it requested the

Colombian Government to provi de an “Environmental Impact

Assessment conducted prior to sprayings of Glyphosate” and

reiterates repeatedly that Colombia never furnished it with any
518
such study.

6.23. In particular, Ecuador claims that during a meeting of the

Joint Scientific and Techni cal Commission in October 2003,

“the Colombian delegation agreed to provide the information

requested”, i.e. “existing environmental impact studies
pertaining to the fumigations”. 519

6.24. A review of the records of the meeting drawn up by the

Colombian delegation shows th at no such commitment was

made. Nor could it have been made at the time since Colombia

516 See Annex 65, p. 3.
517 Annex 71: Spray Mix Quality Control, National Laboratory for
Agricultural Use Products (LANIA), Colombian Agriculture and Livestock
Institute, 18 February 2009; Annex 65, p. 3.
518
519 EM, paras. 3.1, 3.2, 3.31 to 3.34.
EM, para. 3.28.
220had only developed an Environmental Management Plan, on the

basis of the experience gained by prior experimental spraying
programs and studies, and not an “Environmental Impact

Assessment conducted prior to sprayings of Glyphosate” as

requested by Ecuador. Nevert heless, Colombia offered to

provide Ecuador with the studies at its disposal.

6.25. In that regard, Ecuador’s Memorial fails to mention that,

in November 2003, pursuant to th e commitment undertaken at

the meeting of the Joint Scientific and Technical Commission in

October 2003, Colombia sent to the Ecuadorian Scientific and
Technical Commission, among ot her technical documents and

scientific studies, the Environmental Management Plan. 520

6.26. However, Ecuador continued to insist on an

“Environmental Impact Asse ssment conducted prior to
521
sprayings of Glyphosate”. The PECIG’s Environmental

Management Plan was elaborat ed in 2001, as provided for in

Colombian law in force at the time. According to the
transitional regime which was then in force, the Environmental

Management Plan was equivalent to an environmental impact

520
521An9. ex
See e.g. EM, Vol. II, Annex 51 (Diplomatic Note 75204/2003-GM,
sent from the Ministry of Foreign Affairs of Ecuador to the Ministry of
Foreign Affairs of Colombia); CCM, Annex 10: Note No. 4-2-336/03 from
the Ecuadorian Foreign Ministry to the Colombian Foreign Ministry, 10
December 2003; EM, Vol. II, Annex 61 (Diplomatic Note 20434/2003-GM,
from the Ecuadorian Foreign Ministry to the Colombian Foreign Ministry, 31
March 2004).
221assessment. 522 Having been approved and adopted by

administrative resolution, the Environmental Management Plan

is widely known and publicly accessible. 523

6.27. Ecuador also omits to mention that at the second meeting

of the Joint Scientific and Technical Commission, the

Colombian Delegation explained in detail the environmental

impact of the PECIG, as acknowledged in a note sent by the

Ecuadorian Foreign Minister to his Colombian counterpart,

which states as follows:

“The CCTE [Ecuadorian Scientific and Technical

Commission] again requested the report on
environmental impact that the CCTC [Colombian
Scientific and Technical Commission] possesses.

This report has been considered essential since
the beginning of the research and was presented
during the 2 nd Meeting.” 524

6.28. Colombia presented the available research at that

meeting. The signed Minutes of the meeting recorded a

commitment on the part of Colombia to exchange “documents

522 See paras. 4.10-4.14 above.
523 EM, Vol. II, Annex 15; CCM, Annex 50.
524
EM, Vol. II, Annex 55 (Diplomatic Note 4820/2004-GM from the
Ministry of Foreign Affairs of Ecuador to the Ministry of Foreign Affairs of
Colombia, 10 February 2004), emphasis added. Ecuador was also aware of
the nature, content, scope, objectives and procedures of the Environmental
Management Plan since, after its re quest to the Inter-American Commission
for Human Rights for the indication of precautionary measures, the
Colombian Government was required to reply to the request, and detailed
information on the Plan was included therein (see Annex 19: Diplomatic
Note Nº DDH. 58003 from the Colombian Foreign Ministry to the Executive
Secretary of the Inter-American Commission on Human Rights, 18
September 2005).

222with the respective scientific-t echnical basis that support their
525
positions”. In no way do those minutes reflect any

commitment by Colombia to deliver an “environmental impact

study” of the kind demanded by Ecuador. Ecuador’s assertions

that, “[t]o the surprise of the Ecuadorian delegation, the

Colombian delegation stated that the Colombian Ministry of the

Environment had conducted an environmental impact study…
526
[t]he study was never provided to Ecuador” are therefore
incorrect.

6.29. Quite apart from providing the PECIG program’s

Environmental Management Plan, and consistently showing its

willingness to address Ecuador’s demands and queries, the

Colombian Government on a number of occasions submitted

information and studies relating to the various technical aspects

of the spraying operations conducted in Colombia. The

information and studies were provided to assorted agencies and

departments of the Ecuadorian Government, including members
527
and the President of its National Congress.

525
Minutes of the I Meeting of the 2nd Ecuador-Colombia Scientific
and Technical Commission, 10 April 2007 (EM, Vol. II, Annex 80).
526 EM, para. 3.69.
527 Annex 1: Note Nº E-067 from the Colombian Ambassador in Quito
to the Presidential Adviser for Coexistence, National Security, and Fight
against Crime, 18 January 2001. Annex 2: Note No E-934 from the
Colombian Ambassador in Quito to the Ecuadorian Defence Minister, 16
August 2001; Annex 6: Note No E-1313 from the Colombian Ambassador in
Quito to the Ecuadorian Minister of Agriculture, 15 November 2001; Annex

3: Note No. E-931 from the Colombian Ambassador in Quito to Valerio
Greffa Uquiña, Ecuadorian Congressman, 20 August 2001; Annex 4: Note
2236.30. Legal issues associated with the alleged failure to
528
conduct an EIA are discussed in Chapter 8 below.

D. Alleged Failure to Notify Ecuador of Spraying

Missions

6.31. It is a key feature of the aer ial spraying campaign that it

was conducted across the entirety of Colombia, wherever illicit

crops were detected. It was not focused on border areas but on
529
Colombian territory. It was not necessary to notify any other

country, whether or not neighbouring Colombia, of the precise
timetable of operations being undertaken in Colombia.

6.32. Moreover there were very good reasons not to do so. At

a meeting of the Joint Scientific and Technical Commission in

February 2004, the Colombian dele gation made clear that “for

security reasons informati on under no circumstances the
530
activities schedule would be informed.” Nevertheless,

subsequently, during the last meeting of the first Joint

Commission, held in August 2004, the Colombian Government,

in a gesture of goodwill, agreed to notify “by the fastest means,

at the moment that such sprayings are being conducted along the
border area, so that the Ecuadorian Commission may take

No E-962 from the Colombian Ambassador in Quito to the President of

528ional Congress of Ecuador, 23 August 2001.
529 See below, paras. 8.65-8.91.
530 See above, para. 4.57.
Annex 51, Annex 3 to the Letter.
224samples and conduct the respective analyses in a timely

manner.” 531

6.33. The Government of Colombia did in fact notify Ecuador

of the fact that sprayings were being carried ou t in November

2004, in accordance with the undertaking given at that

meeting. 532 Shortly thereafter, Ecuadorian members of the Joint

Scientific and Technical Commission collected water samples in

Ecuadorian territory near the bor der to carry out analyses, the

results of which were described above. 533

6.34. Furthermore, the suspension of the sprayings in the 10-

km strip along the border with Ecuador from December 2005,

the brief resumption between December 2006 and January 2007,

and the suspension from February 2007 to date, were all duly

notified to Ecuador. 534 It is accordingly not true, as the

Memorial asserts, that “not once has Ecuador received advance

notice of aerial sprayings along its borders.” 535Astoadvance

notifications of individual missions, Colombia was under no

obligation, having regard to the evident security concerns, to

give such notification. It may be noted, however, that Ecuador

531
See above, para. 5.35; EM, Vol. II, Annex 64.
532 See above, para. 5.39.
533 See above, paras.5.30-5.31, 5.37.
534 See above, paras. 5.39, 5.67, 5.71.
535 EM, para. 3.2.

225was invited to observe indivi dual missions, and did so in
February 2002; 536but declined to do so subsequently. 537

E. Conclusions

6.35. Ecuador’s allegations that Colombia failed to disclose

the composition of the spraying mix are unfounded.

6.36. The composition of the mix used for the aerial sprayings

has been publicly availabl e since 2001. The Ecuadorian

Government had been informed all along and was therefore

fully aware of the composition of the spray mix used to

eradicate illicit crops in Colombia.

6.37. Not only was the composition of the spray mixture
publicly available and officially communicated to Ecuador, it

was in fact known and appreciate d by Ecuadorian authorities.

In 2005, the Ecuadorian Scientific and Technical Commission

(CCTE), submitted a technical report on the study

commissioned by the Inter-American Drug Abuse Control

Commission (CICAD I), in which it stated that: “The toxicity

data for the active ingredient glyphosate is obtained from the
literate [ sic] and from acute intoxication tests of laboratory

536 See above, para. 5.11.
537 See above, para. 5.35, 5.65.
226animals undertaken with a mixture of glyphosate and
538
Cosmoflux such as the one used in the spraying program”.

6.38. As to Ecuador’s contention that Colombia never

provided an Environmental Impact Assessment, Colombia did

supply Ecuador, upon its request, in November 2003, with the

PECIG program’s Environmental Management Plan, which

pursuant to the applicable legal regime in Colombia was
tantamount to an Environmental Impact Assessment. 539

6.39. In the same spirit of colla boration, and although it was

under no obligation to do so, Colo mbia also notified Ecuador

that aerial sprayings were being conducted in border areas.

Following January 2007, the PECIG program has been

suspended in this region and has not been resumed.

538 EM, Vol. III, Annex 153.
539
See further, paras. 4.10-4.14, 5.27, 5.98, 6.25, note 524 to para.
5.27, and 8.89-8.90.
227228 PART II

SCIENTIFIC AND LEGAL ISSUES

229230 Chapter 7

THE DAMAGE ALLEGED BY ECUADOR

A. Overview

7.1. As will be demonstrated in this Chapter, Colombia’s
PECIG program simply cannot ha ve caused the kind of damage

to the Ecuadorian population, wild life or environment alleged

by Ecuador. 540

7.2. This was confirmed by th e Ecuadorian authorities

themselves, who expressly and pu blicly stated that, up until

December 2004, no damage to human health, the environment

or the wildlife of Ecuador had occurred in the area adjacent to
541
the border with Colombia. Further, in June and August 2004,

the Ecuadorian Foreign Ministry released the results of scientific

analyses carried out by Ecuadorian official s on water samples

taken from rivers in the provinces of Sucumbíos and Esmeraldas

adjacent to the Colombian border. Those analyses showed no
542
evidence of glyphosate residues in the rivers in question,

whereas it is a predicate of Ecuador’s theory of the case that the
543
rivers were polluted in 2004.

540 For Ecuador’s characterization of losses suffered, see above, para.
1.26-1.31.
541 See above, paras. 5.45-5.63.
542 See above, paras. 5.30-5.32, 5.37.
543
Seee.g. EM, paras. 2.11, 2.25, 3.61, 5.48, 6.3, 6.100, 9.60, 9.70,
9.73, 9.74.
2317.3. Thereafter sprayings were suspended over Colombian
territory in the 10-km strip pa rallel to the border between

December 2005 and December 2006. They were only resumed

for a brief period between December 2006 and January 2007

before they were again suspende d pursuant to a decision of the
Colombian Government announced in February 2007. Since

that date, no spraying has take n place over Colombian territory

in the area within 10 km of the border. Accordingly, no damage
could have occurred in Ecuador ian territory as a result of

spraying by Colombia from late January 2007 to the present.

7.4. After the end of 2004, the only periods in which
sprayings took place in any area near the border area with

Ecuador were:

(1) as regards areas contiguous to the border in the

Colombian province of Nariño, between 14

January and 26 December 2005, and between 17
December 2006 and 14 January 2007; and

(2) as regards areas contiguous to the border area in

the Colombian province of Putumayo, between

24 September and 10 December 2005 and

between 11 December 2006 and 21 January
2007.

None of the witness statements submitted by Ecuador relates to

spraying which occurred in the relevant province during those

232periods. Yet Ecuador itself is on record as denying that any

harm had occurred prior to that time. The testimony concerning
alleged damage or spraying on Ecuadorian territory, including in

particular the ex post facto witness statements procured by

Ecuador and submitted together with its Memorial, are contrary

to the evidence gathered by Ecuador itself at the relevant time
and as such have no probative weight.

7.5. But in any event, 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 – including the observance of 100m exclusion

strips along watercourses – no damage could have occurred in
Ecuadorian territory.

7.6. Moreover, even if Ecuador’s position on spray drift were
tenable (which it is not), the ex tent and scope of the alleged

damage referred to by Ecuador in its Memorial is simply not

credible. Nothing remotely like it has occurred in Colombia,

where the sprayings actually took place.

7.7. This Chapter will substantiate these propositions by

reference to the scientific evidence, much of which is ignored or
misrepresented in Ecuador’s Memorial, and also to the affidavits

so far submitted. Colombia has commissioned an expert study

by Dr Stuart Dobson, a toxicologi st of 30 years experience who
chaired the Expert Task Gr oup which finalised the WHO

233 544
assessment of glyphosate, and who is currently Chair of the

Risk Assessment Steering Group of the International
Programme on Chemical Safety (IPCS), a joint activity of the

World Health Organisation (WHO), the United Nations

Environment Programme (UNEP) and the International Labour

Organisation (ILO). 545 The Dobson Report is appended to this

Counter-Memorial.

B. The Factual Prerequisites for Ecuador’s Case

7.8. Ecuador’s case is contingent on proving four separate
sets of facts: (a) that the aerial applications of the spray mixture

used by Colombia in its drug er adication program are damaging

to human health, to other biota a nd to the environment, (b) that

spraying which took place has (primarily by spray drift) resulted

in the spray mixture being deposit ed on Ecuadorian territory in

significant quantities, (c) that Ecuadorian nationals, their

livestock and farms have there by been exposed to the spray

mixture, and (d) that there is a causal link between the sprayings
and the injuries alleged, i.e. that it was the exposure to the spray

mix that caused the harms of which Ecuador complains in these

proceedings.

544
See WHO, Glyphosate. Environmental Health Criteria 159 (World
Health Organization, Geneva, 1994), (excerpts in Annex 96, full text
available at: http://www.inchem.org/documents/ehc/ehc/ehc159.htm (last
visited 10 March 2010). For more recen t published work by Dr Dobson on
glyphosate see J.P. Giesy, S. Dobson S & K.R. Solomon, “Ecotoxicological
risk assessment for Roundup herbicide”, Reviews of Environmental
545tamination and Toxicology167: 35-120 (2000).
See http://www.who.int/ipcs/en/ (last visited 10 March 2010).
2347.9. The burden of proving these facts lies on Ecuador and
Ecuador has failed to discharge that burden. But independently

of any onus, the scientific and other evidence supports

Colombia’s position.

(1) T OXICITY OF THE SPRAY MIXTURE

7.10. With respect to the first point, there is no scientific
evidence of significant risk to hum an or animal health or to the

environment due to exposure to the spray mixture used by

Colombia in its drug eradication program. Glyphosate is one of

the most commonly used pes ticides world-wide. Both
glyphosate and its various formulations have been the object of

numerous investigative studies in order to ascertain whether

they may have adverse effects on humans, animal species and
on the environment. Likewise, it has been shown that the

adjuvant Cosmo-Flux 411F, that accounts for 1% of the spray

mix, does not alter the formulation’s toxicity.

7.11. For instance, the independent scientific study completed

in 2005 by an expert panel composed of Dr Keith R. Solomon

and other scientists at the re quest of the OAS agency CICAD
(CICAD I), concluded that Colombia’s spray eradication

program did not pose significant risks for humans and most

wildlife.

235 “The toxicity of glyphosate has been rigorously
assessed in a number of jurisdictions and in the

published literature. Glyphosate itself has low
toxicity to non-target organisms other than green
plants. It is judged to have low acute and chronic
toxicity, carcinogenic, mutagenic, or a
reproductive toxicant. With respect to humans, is
not considered hazar dous, except for the

possibility of eye and possibly skin irritation
(from which recovery occu rs). The toxicity of
the formulation as used in the eradication
program in Colombia, a mixture of glyphosate
and Cosmo-Flux®, has been characterized in

specific tests conducted in laboratory animals.
The mixture has low toxicity to mammals by all
routes of exposure, although some temporary eye
irritation may occur. By extrapolation, the spray
mixture is also not expected to be toxic to
terrestrial mammals and vertebrates.” 546

7.12. The subsequent studies forming CICAD II, published in

2009 in the Journal of Toxicology and Environmental Health ,

focused on remaining uncertainties identified as subjects for

further study in CICAD I, in partic ular, the issues of spray drift,
the effects on sensitive wildlife such as amphibians and the

effects on humans. Summing up the effect of these studies, the

overall conclusion of CICAD II was as follows:

“Laboratory and field tests on amphibians
showed that Colombian species were of similar

sensitivity to species tested in other locations and
that they were not especially sensitive to
glyphosate formulations. Tests on larvae stages
of amphibians under realis tic conditions showed

546 Annex 116, CICAD I, pp 10-11.

236that toxicity was reduced, most likely because of
the rapid absorption of glyphosate and its
adjuvants to sediment and particulate matter.
Terrestrial stages of frogs showed a range of

sensitivity, but all had LC50 values less than the
application rate used fo r eradication of coca.
Given interception by folia ge, risks to aquatic
and terrestrial stages of frogs from Colombia,
even from direct exposure to aerial eradication
sprays, are judged to be small to negligible. The
study of the large distribution of large diversity

of frog species in Colombia in relation to coca
production and eradication spraying showed that
there were only a few species of frogs potentially
at risk because of their location in southwest
Colombia. As these species are also found in
Ecuador, the likely small risks are to populations
in Colombia, not the species as a whole. A much

greater risk to frogs in Colombia is from the
other pesticides used by the growers of coca (and
poppy) and particularly the deforestation that
precedes the planting of these crops.

In terms of effects on humans, an
epidemiological study did not provide evidence
of effects on reproductive function in terms of
TTP [time to pregnancy]. In a study on potential
genotoxicity that combined epidemiological
surveys with biological monitoring of MN
[micronuclei] in white blood cells, differences in

the baseline frequency we re observed in relation
to region sampled... In some regions the
frequency decreased after spraying but in one, it
did not. These observati ons do not fulfil all the
criteria for causality , suggesting that if
glyphosate spraying had any influence on MN,
this is small and not of biological significance.

Overall, the risks to sensitive wildlife and human
health from the use of glyphosate in the control
of coca (and poppy) production in Colombia are

237 small to negligible, especially when compared to
the risks to wildlife and humans that result from
the entire process of the production of cocaine
(and heroin) in Colombia.” 547

7.13. The fact that glyphosate has relatively low toxicity levels

may be confirmed by reference to Annex I of the Kiev Protocol
on Civil Liability and Compensation for Damage Caused by the

Transboundary Effects of Indus trial Accidents of 21 March

2003, which addresses the question of civil liability for

“hazardous activity”. 548 The Protocol defines “hazardous

activity” in Article 2(2)(f) as: “any activity in which one or

more hazardous substances are present or may be present in

quantities at or in excess of th e threshold quantities listed in

annex I and which is capable of causing transboundary effects
on transboundary waters and their water uses in the event of an

industrial accident”. Annex I lays down precise criteria for

substances as “Very Toxic”, “T oxic” or “Dangerous to the

Environment”. Glyphosate does not meet the criteria for either

“very toxic” or “toxic”. It meet s the criterion for “dangerous to

the environment” for algae (hardly surprising for a herbicide),

547
548 Annex 131-A, CICAD II, at p. 919.
See Protocol on Civil Liability and Compensation for Damage
Caused by the Transboundary Effects of Industrial Accidents on
Transboundary Waters to the 1992 Convention on the Protection and Use of
Transboundary Watercourses and International Lakes and to the 1992
Convention on the Transboundary Effects of Industrial Accidents, Kiev, 21
May 2003, see http://www.unece.org/env/civil-
liability/documents/protocol_e.pdf for the text of the Convention, which has
only 1 ratification and is not yet in force.
238technically meets it for daphnids, but does not meet it for fish,

still less mammals. 549

7.14. Ecuador claims that the spray mix used in the program

of the aerial eradication of illicit crops in Colombia is atypical

because the glyphosate concentratio n differs from that used in
550
agriculture. The underlying basis for that comparison is

flawed, but in any event it is simply incorrect. As to the basis for

the comparison, glyphosate con centrations always vary

contingent upon the species of plant the herbicide is intended to

control and the desired effects. Further, coca eradication is

evidently a non-agricultural use of glyphosate. But in any case,

Colombia’s aerial eradication program for illicit crops results in

the application of doses equivalent to those which are authorised
551
in other countries.

7.15. In an attempt to stress the toxicity of the spray mix,

Ecuador makes much of the information contained in the labels

549 The acute LD50 for rats (oral) for glyphosate is >5000 mg/kg body
weight. The acute dermal LD50 for rats and rabbits are >2000 and >5000
mg/kg bw respectively. For the environmental criteria, 7 out of 17 values for
EC50 in algae are <10 mg/litre. 1 value for Daphnia species is just <10 (9.7)

mg/litre; the other 4 reported are >10 (12.9 to 25.5) mg/litre. All fish LC50
values are >10 mg/litre. Data from Giesy et al. 2000 (e.g., p. 69) and WHO
5504.
551 EM, paras. 5.13, 6.79.
The EPA found that, based on a comparison of the glyphosate use
pattern in Colombia and the use in the U.S., the most equivalent U.S. uses of
glyphosate would be in forestry or rights-of-way. The glyphosate product
used in Colombia (as at 2004) is registered in both the U.S. and Colombia,
and application rates described as used in Colombia were found to be within
the parameters listed on U.S. labels. In Annex 142, pp. 8-9 (partially at EM,
Vol. III, Annex 143), see also Annexes 143, 144, 149, p. 1.

239on containers of glyphosate and Cosmoflux. 552 Such labelling is

not only standard, but mandatory and is entirely “hazard based”,

indicating the potential for harm ra ther than the probability of

harm (risk). It is the same information that all the labels of

chemical pesticides for agricultural use within the Andean
Community have to contain. N eedless to say, this information

is quite similar to that commonly found – and generally required

– to appear on over-the-counter and prescription medicines,

household products, and even processed food products.

(2) S PRAY DRIFT

7.16. Since every care is taken to ensure that spraying occurs

only on Colombian territory, the present case has as a central

issue the question of the extent and effects of spray drift.

7.17. As described in some detail in Chapter 4 above, 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 density of the spray mix and the initial size of

the spray droplets. 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

552 EM, paras. 5.39-5.47.
240contingent, with the purpose of re ducing it as much as possible.
To summarize:

• Target areas are carefully identified in advance.

• The aircraft are equipped with GPS readers and

computerised information about the terrain.

• Aircrew are fully trained and flights are monitored and

recorded.

• The same area is not sprayed more than twice a year.

• An exclusion zone of 100m is observed in relation to all
water bodies, including the boundary rivers.

These parameters are strictly observed by the personnel
553
involved in spraying operations.

7.18. A specific study during the s econd phase of the CICAD

process addressed the issue of drift in further detail.554 The

study assessed the drift potential under extreme conditions of

flight speed (333 km per hour), wind speed (9.3 km per hour,

with wind direction perpendicula r to the flight line) and air

temperature (35º C). The st udy used computer modelling and
wind tunnel experiments to define a safety zone protective of

553
See above, paras. 4.67-4.69, and Annex 50, Table Nº 1, Operational
Parameters of the Program for the Eradication of Illicit Crops by Aerial
554aying.
Annex 131-B, CICAD II: A.J. Hewitt, K.R. Solomon & E.J.P.
Marshall, “Spray Droplet Size, Drift Potential, and Risks for Nontarget
Organisms from Aerially Applied Glyphosate for Coca Control in
Colombia”, in Journal of Toxicology and Environmental Health , Part A,
72:921-929, 2009.
24195% of non-target species downwind of the spray operations; it

did so on the basis of “the worst-case assumption that the

concentration resulted from di rect overspray of a 15-cm-deep

pool with no exposure reduction via adsorption to sediments and

organic matter, and no inter ception by surrounding plants”. 555

Moreover the risk assessment focused on the most vulnerable

species: juvenile frogs.

7.19. The study found that the great majority of the active

ingredient (>90%) was deposit ed within the target area. 556 As

the figure on the following page shows, 557the amount of spray

drift deposition decreases exponentially as the distance from the

swath edge increases:

555 Annex 131-B, CICAD II, p. 923.
556 Hence, the assertion contained in the Menzie Report and reprised by
the Memorial (EM para. 5.97) to the e ffect that “[T]he turbulence created by
high plane speed causes ‘spray droplets to break apart, and these smaller,

lighter droplets have a potential to be carried further by wind currents’”, is
inapplicable to the actual circumstances of the present case. Also disproven
by the CICAD II study are the Memorial’s assertions with regard to the
prevailing high temperatures along the border as allegedly aggravating drift,
by making the “spray more prone to form small droplets that will be carried
away by the wind. These shrunken drop lets are also more concentrated,
increasing their toxicity.” (EM para. 5.89.)
557 Annex 131-B, CICAD II, at p. 926, Figure 5.

242 10000
OV-10
WIND DIRECTION AT-802
1000 ATT-65

100
g/ha)

on ( 10

epositi
D 1

0.1
–50 0 50 100 150 200 250 300

Distance from swath (m)

Dobson, Figure 1, adapted from Hewitt et al. (2009). Deposition rates for
spray drift (g/ha on a log scale) for different aircraft types at representative
flight speeds.

In reading this Figure 5, it is important to note that the vertical

axis is a log scale, depicting orders of magnitude. In other

words the extent of spray drift resulting from the ATT-802, 558

50 metres downwind, is less than 1% of the spray at the target

point (<100/10000 g/ha). The extent of spray drift 100 metres

downwind is 0.1% (10/10000).

7.20. The study concluded that effects of spraying were

negligible beyond 120 meters even for those plants most

sensitive to the spray mixture. 559 It also found that:

“The extensive vegetation of the forest canopy
and environment around the area where the coca

and poppy plants are sprayed in Colombia will

558 AT 802 are the planes currently in use in the aerial spraying

program.
559 Annex131-B, CICAD II, at pp. 923, 925, 929.

243 afford excellent reductions in spray drift potential
by interception of droplets with leaf and other
surfaces (Raupach et al., 2001). This will greatly
reduce the spray drift e xposure risk from the

values reported in this study by 50–90%
(AgDRIFT, 2008).” 560

7.21. The authors concluded that:

th
“Based on modeled drift and 5 centile
concentrations, appropriate no-spray buffer zones
(distance from the end of the spray boom as
recorded electronically ±5%) for protection of

sensitive plants were 50-120 m for coca spray
scenarios... the equivalent buffer zone for
amphibians was 5m. The low toxicity of
glyphosate to humans suggests that these aerial

applicati561 are not a concern for human
health.”

7.22. In terms of long-distance transport the authors concluded

that:

“Long-distance transport of spray drift particles

is small and not an issue for humans or the
environment beyond 50m downwind at the
maximum permitted wind velocity of 9 km/h for
spraying operations. Long-distance movement of

glyphosate is negligible if appropriate no-spray
buffers are used and nonexistent if the wind
direction during spray is away from the area of
concern.” 562

560
561 Annex131-B, CICAD II, at p. 928.
562 Ibid., p. 921.
Ibid., p. 929.
2447.23. For its part, Ecuador’s Memorial baldly asserts that drift

has reached Ecuador, heavily relying on the unfounded

assertions of the Menzie Report in its discussion of topics such

as spray operation parameters, off-target release of spray,
563
ground-based hostilities and localized weather conditions.

7.24. For instance, the Memorial quotes the Menzie Report’s

reference to “agricultural studies conducted in the United States

[that] have ‘shown that pestic ides delivered through aerial

spraying can be transported miles in the drift … spray drift may
564
extend as far as four to ten miles.’” The Report makes that

assertion based on studies carrie d out several decades ago, in

which the substances under study and their method of

application are not comparable to the spray mixture which has
565
been used in the PECIG program since 2000. In any event,

those same studies found that at 300 metres downwind the spray

deposited would be about 1 gram /hectare (g/ha); on a scale of

563 EM, paras. 5.83, 5.86-5.90, 5.93-5.95, 5.97, pp. 157-163. EM, Vol.
III, Annex 158, Section 4.
564 EM, para. 5.83, p. 157.
565 For example, it cites the study carried out by Currier et al. (1982)
that assessed the application of two active ingredients (Cipermetrina,

Triazopho) – without using an adjuvant – for insect control. For several
reasons, these insecticid es are not comparable to the herbicide used in
Colombia to eradicate co ca crops. For exampl e, the droplet size for
insecticides is much smaller. The study by Westra & Schwartz (1989) also
refers to the application of pesticides without the use of adjuvants. As
previously stated, an adjuvant is used in the PECIG precisely to help reduce
drift as much as possible. The study conducted by Fox (1978) is also cited;
he studied terrestrial applications of di fferent herbicides for weed control in
grape crops. Clearly, terrestrial appli cations are not comparable to aerial
spraying because the equipment and procedures used are significantly
different.

245kilometres, measurements would be in micrograms/hectare

(μg/ha). This is far below th e dose which could affect any
organism, animal or plant.

7.25. The Menzie Report asserts th at “the spray missions are

often conducted under highly adve rse and violent conditions”

and concludes that “It is reasonable to expect that the hostility

would be a reason why pilots would fly higher and faster than

they would during normal agricultural crop-spraying. Under

those circumstances, the released spray would be prone to
566
greater drift”. But this is sheer speculation: as stated in
Chapter 4, 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, 567 and

spraying missions are cancelled if the situation changes. 568 In

any event the work of Hewitt et al. (2009), summarised above,

assumed a worst-case scenario of a spray plane flying at 333 km

per hour: even so, predicted deposition of spray downwind was

as shown in the Table.

7.26. Ecuador attaches great importance to the speculative and

hypothetical discussion contai ned in the Menzie Report

concerning “localized [wind] ci rculation patterns [that] can

566
567 EM, Vol. III, Annex 158, p. 13 (emphasis added).
568 See above, paras. 4.61, 4.70 and note 330, Annex 67, pp. 4-5.
Ibid.
246 569
increase the risk of off-target deposition…” The Menzie

Report seeks to infer local circul ation conditions on the basis of

notions of general circulation of the atmosphere. But local

conditions are random in charact er, depending on the time of

year, time of day, etc. To determine a local wind circulation

pattern, it is necessary to have actual data obtained from direct

measurements from both land and air (radio scanning)

meteorological stations, whic h the Menzie Report does not
provide, but which are checked in situ prior to each spray

mission. 570

7.27. Another speculative item in the Menzie Report, also

recalled in the Memorial, is the suggestion of “frequent ‘thermal

inversions’ 571 along the international frontier” as “another

meteorological condition that increases spray drift into

Ecuador”. 572 The Report states that “thermal inversions are

common in the tropics and are expected to be a frequent
573
occurrence in the Ecuador/Colombia border area”. Butfora

thermal inversion to occur, very particular local meteorological

conditions are needed, such as clear sky at night and calm winds

so that the soil cools down faster than the air lo cated above it.

Even when this phenomenon does occur, it is of short duration

because it usually starts to disappear from the very moment the

569 EM, paras. 5.86-5.87.
570 See above, para. 4.68.
571 The thermal inversion phenomenon is defined as the increase in
temperature as height from land increases.
572
573 EM, para. 5.90.
Ibid. (Emphasis added)
247sun comes up. It would have little or no influence on drift

because spraying operations take place after sunrise.

7.28. An accurate method of meas uring drift of any aerial

spraying, in lieu of wind-tunnel or other type of studies, is by

conducting soil and water sample analyses from the areas
surrounding the sprayed areas to measure the residues of the

applied substance. The Menzie Report, on whose assertions the

Ecuadorian Memorial heavily relie s, drew conclusions without

taking any such samples. In contrast, in 2004 Ecuadorian

scientists performed actual soil and water sample analyses and
574
did not find any glyphosate residues.

7.29. The overall conclusion of CI CAD II on spray drift was

as follows:

“In terms of spray drift, new data showed that

drift from eradication spraying is minimal and
that relatively small buffer zones, ranging from 5
to 120 metres, are protective of sensitive aquatic
animals and the target organisms, plants,
respectively.” 575

7.30. In sum, the scientific evidence does not substantiate

Ecuador’s arguments with regard to any alleged damage in its

territory due to drift.

574 See above, paras. 5.30-5.32, 5.37.
575 Annex 131-A, CICAD II, at p. 919.
248 (3) E XPOSURE OF E CUADORIAN RESIDENTS

7.31. Colombia took full account of the availabl e scientific
knowledge available at the time it decided to start the spraying

program, and adopted a precautionary approach. Even in the

absence of any scientific evidence supporting the allegation that

the spray mix causes serious adverse effects to human health

and to animals, out of an abundance of caution, Colombia has

nevertheless implemented strict procedures in relation to the
576
aerial fumigations of illicit crops in its territory. The conduct
of the aerial sprayings is heavily regulated to minimize any risks

of human exposure and so as to reduce the chances of drift, if

any, to a minimum.

7.32. Prior to any spraying opera tion, the areas affected by

illicit crops are mapped out by digital imagery and the exclusion
zones and safety strips where no aerial spraying takes place are

identified. No sprayings occu r over human settlements, bodies

of water or water courses. Such areas are considered to

constitute exclusion zones, around which 100-metre safety strips
577
are also established. In the case of the border with Ecuador

the rivers are protected by a 100-metre safety strip. The

Colombian Government has also taken the precaution of
imposing set minimum and maximum figures for the parameters

of drift (wind speed, aircraft speed, height, temperature). The

576 See above, paras. 4.21-4.29, 4.41-4.70.
577 See, para. 4.59.
249aircraft fly at low altitudes, usually 30 meters, with a maximum
operation speed of 165 miles per hour. They are specially

designed and equipped for precision flying, and the spray mix is

propagated through automatically calibrated nozzles that release

the same amount of mix. 578

7.33. As set out in Chapter 5 of this Counter-Memorial, a

high-level Ecuadorian delegation conducted analyses of soil and
water samples on at least three separate occasions in the year

2004 when sprayings near the border were being conducted. No

residues of glyphosate were found as a result of the tests carried

out by this delegation, who also received reports from the

people residing in the area, incl uding both the local population

and UNCHR officials, stating th at there had been no sprayings
579
affecting Ecuadorian territory at that time. Thus, these
findings confirm that no drift of spray mixture has occurred over

Ecuadorian territory –or if there was any, that it was

insignificant.

(4) E XISTENCE OF A CAUSAL LINK

7.34. Crucially, Ecuador fails to produce any evidence of a
causal link between exposure to the spray mixture, and the

various illnesses and injuries complained of.

578 See above, paras. 4.62-4.63.
579 See above, paras. 5.45-5.63.
2507.35. Representative of this is the witness statement of Dr.

Dino Juan Sánchez Quishpe, 580an Ecuadorian employee at the

local public hospital in Lago Agrio, prepared for these

proceedings. Dr. Sánchez records flu-like symptoms and skin

problems “which coincided w ith border sprayings”. But

although at the time both of th e events (2004-5) and of his
statement (2009), Dr. Sánchez had access to the hospital in Lago

Agrio (the only one in the region), he produces no medical

evidence of any causal link between spraying and these

symptoms. The two may have coincided, but coincidence is not

enough. Moreover Dr. Sánchez ha s nothing to say about more

serious illnesses or deaths.

7.36. While not producing any direct evidence or scientific

studies supporting its allegations of harms to health caused by

the aerial sprayings, Ecuador’s Me morial depicts the very poor

living conditions of the farm ers and indigenous people who

reside in the areas situated along the Ecuador-Colombia
border. 581

7.37. The fact that these groups of Ecuadorians live in

precarious hygienic conditions a nd only have limited access to

medical facilities is highly relevant for present purposes. In the
light of the situation on the gr ound, and in the absence of any

medical and scientific studies ca rried out at the relevant times

580 EM, Vol. IV, Annex 188.
581 EM, paras. 2.17-2.27.
251which link the alleged injuries with the sprayings in Colombia,

it is impossible to establish whet her 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 conditions.

Alternatively they may be the result of the much more toxic

chemicals used in the cultivation and processing of coca in those
areas. As noted in the UNODC 2009 Coca Cultivation Survey

for Ecuador, there were laboratories in the regions of

Esmeraldas and Sucumbíos that were apparently destroyed by

the Ecuadorian Government during 2007-2008. 582

7.38. The Ecuadorian Memorial 583 uses the maps included in

the annual survey reports of the SIMCI Program to purportedly

show, by way of inference, how close the Ecuadorian localities

are from the locations where illicit crops targeted by the PECIG

582
UNODC, Ecuador – Coca Crop Monitoring, September 2009.
Available at:
http://www.unodc.org/documents/crop-
monitoring/ecu_repo.pdf?bcsi_scan_8896DBBFDB1B0269=0&bcsi_scan_fil
ename=ecu_repo.pdf (last visited 10 March 2010)
583 EM, paras. 6.23, 6.30, 6.36: “As the Court can see, official UNODC
data shows how close to Salinas Colombia conducted sprayings in 2006. The
map also shows how little (if any) drift is required for the toxic herbicide to
reach into and directly affect Ecuadori an communities” (para. 6.23); “… the
Court will see just how little drift is required for the herbicide to reach
Ecuadorian communities on the border” (para. 6.30); “As is clear, the yellow
spray areas actually overlap much of the Mataje River which constitutes the

international border in the area and has historically been the chief source of
fresh water for area residents.” (para. 6.36).
252program are found, or that Colombia has sprayed over border

rivers.

7.39. According to Ecuador’s arguments, minimal (“if any”)

drift would be requir ed for the spray mix to reach Ecuadorian

territory. However, according to the operator of the SIMCI

Program,

“…the maps printed in the reports should not be

used for calculating areas or distances because
that is not what they were created for and do not
possess the adequate cartographic precision; they
work as illustrations. Likewise, attempts to

intersect areas of different kinds should be made
on the basis of detailed information and not on
the basis of the illustrations in the report.”584

7.40. Moreover, the information concerning sprayed areas

shown on the maps should “be used only as a reference to the

areas linked to the [spraying] process, but not to make

measurements or to locate a particular area in a detailed way.” 585

On the basis of the informa tion provided by the UN Office on

Drugs and Crime –UNODC, agency in charge of conducting the

584
Annex 111: Diplomatic Note from the United Nations Office on
Drugs and Crime (UNODC) Representative to Colombia to the Colombian
Vice-minister for Foreign Multilateral Affairs, 20 January 2010, p. 2.
Emphasis added. In reply to Diplomatic Note N° 57491 from the Colombian
Vice-Minister for Foreign Multilateral Affairs to the United Nations Office
on Drugs and Crime (UNODC) Repres entative to Colombia, 19 October
5859 (Annex 110).
Annex 111, p. 2.
253annual surveys of coca crops, the assertions of the Ecuadorian

Memorial are unfounded.

(5) C ONCLUSION

7.41. For these reasons, the fundamental theory of Ecuador’s

case fails. Ecuador has demonstr ated neither actual risk, nor
exposure, nor causation. On th e contrary, as will now be

demonstrated, the available scien tific evidence leads to a quite

different conclusion.

C. The Scientific Evidence Concerning Glyphosate-
Based Sprays

7.42. Scientific evidence supports the position of the

Colombian Government as to the negligible effects upon human

health, wildlife, soil and water of the spraying conducted as part

of the PECIG program. In the present section, the conclusions
of the studies containing such evidence are presented. The

views expressed by organizations or individuals with no

expertise in the relevant di sciplines of human health,

epidemiology, medicine, public health, or eco-toxicology are not
considered as scientific evidence.

7.43. It bears emphasising again th at in the aerial sprayings

carried out up until 2007 over Colombian territory situated close
to the border with Ecuador, ex actly the same modalities, mix

and procedures as were applied in the rest of the Colombian

254territory were used. At no time was there any selective variation

as concerns spraying near the border. 586

(1) A LLEGED EFFECTS ON HUMAN HEALTH

7.44. There is no scientific ev idence showing a significant

impact of the PECIG program on human health. To the contrary,

a number of studies exist concl uding that concerns for human

health are unjustified. 587

7.45. Several competent entities, international organizations

and foreign scientific agencies, among them the Food and

Agriculture Organization (FAO) , the Inter-American Drug

Abuse Control Commission (CICAD), the WHO’s International

Programme on Chemical Safety (IPCS), 588 and the United States

Environmental Protection Agen cy (EPA), have undertaken

586 None of the provisions governing aerial spraying in Colombian

territory make selective reference to the provinces bordering the Republic of
587ador. See e.g., Annex 49.
“Glyphosate has been the subject of hundreds of health, safety, and
environmental studies … it is important to consider the total weight of
evidence from scientific studies provided by regulatory agencies, industry,
universities, governmental agencies, and scientists from around the World.
The U.S Environmental Protection Agency (EPA), Health Canada, European
Commission, U.S. Department of Agricu lture Forest Service, World Health
Organization and other scientists have reviewed this data. Those reviews

applied internationally accepted methods , principles and procedures in
toxicology and have determined that there are no grounds to suggest concern
for human health.” In Annex 132: Center for Toxicology and Environmental
Health, L.L.C., University of Arkansas for Medical Sciences, GLYPHOSATE
Frequently Asked Questions, 2009, pp. 1-2.
588 The International Programme on Chemical Safety (IPCS) is a joint
venture of the United Nations Environment Programme, the International
Labour Organisation, and the World Health Organization.

255studies and issued statements on the innocuous character of one

or both of glyphosate or the sp ray mix used in the PECIG
program.

7.46. In 1994, the Environmental Health Criteria on

Glyphosate issued by the WHO` s International Program on

Chemical Safety alluded to a study conducted in Taiwan in

1991, according to the results of which, “[s]evere effects

occurred only in the cases of in tentional ingestion (80 of the 93
589
reported). Accidental exposures led to only mild effects.”

7.47. Moreover, according to the WHO recommended

classification of pesticides based on the hazards linked to its use

(developed as well by the Inte rnational Program on Chemical

Safety), glyphosate is “unlikel y to present ac ute hazard in
590
normal use”.

7.48. Similarly, the FAO’s 2000-2001 “Specifications and

evaluations for plant prot ection products, Glyphosate ”

concluded that glyphosate exhibits “a low toxicity”, as well as

noting that glyphosate and its metabolite AMPA were found not

589
Annex 96: International Programme on Chemical Safety (IPCS),
International Programme on Chemical Safety, “Environmental Health
590teria (EHC) 159, Glyphosate”, Geneva, 1994, p. 36.
Annex 105: World Health Organization (WHO), “The WHO
Recommended Classification of Pestic ides by Hazard”, 2005, p. 54.
Available at:
http://www.who.int/ipcs/publications/pesticides_hazard_rev_3.pdf (last
visited 10 March 2010).
256to exhibit a mutagenic risk to humans. The FAO further

remarked that “it should be also taken into consideration that

there is no evidence of carcinogenic effects in humans, although

glyphosate products have been in world-wide use for many
591
years”. The Specifications noted that experiments conducted
on rats did not indicate any sp ecific hazards from glyphosate or

AMPA for reproduction or for th e development of embryos or

foetuses.592

7.49. These conclusions were echoed in a scientific study

published in 2000 on Roundup (the principal commercial brand

worldwide containing glyphos ate) and glyphosate more

generally, which stated that: “Reviews on the safety of

glyphosate and Roundup herbicide that have been conducted by

several regulatory agencies and scientific institutions worldwide

have concluded that there is no indication of any human health
concern.” 593

7.50. In 2002, the Office of Prevention, Pesticides and Toxic

Substances of the United States EPA, in a report prepared on the

possible effects on health and the environment of the PECIG

program in Colombia, stated that:

591 Annex 101, p. 24.
592Ibid.
593
Annex 125: G. M. Williams et al., “Safety evaluation and risk
assessment of the herbicide Roundup® and its active ingredient, glyphosate,
for humans” in Regulatory Toxicology and Pharmacology 31:117165, 2000,
p. 117.
257 “As for potential human health effects of the coca
eradication program, there are no risks of concern
for glyphosate, per se, from dermal or inhalation

routes of exposure, since toxicity is very low...
There are no expected toxicological effects of
concern for acute (short-t erm) or chronic (long-

term) dietary exposure through food and 594er
from the coca eradication program.”

Glyphosate is not bio-magnifi ed nor does it move along
595
the food chain.

7.51. With regard to the spraying mixture used in Colombia,

the United States EPA has informed the US State Department

on a number of occasions that th e herbicide mix used in the

sprayings conforms to EPA labe l requirements for comparable

use in the United States. Moreove r, the US Secretary of State

determined and certified, for th e purposes of internal legal

requirements in the US during the continued assessment of the

Plan Colombia supported by th at Government, for the years

2002-2007 that “the herbicide mixture, in the manner it is being

used, does not pose unreasonable risks or adverse effects to
humans or the environment”. 596 Significantly, the EPA also

observed in a 2004 report that: “[d]espite an aggressive search

594 Annex 142, pp. 4-5 (partially reproduced at EM, Vol. III, Annex
143).
595 Annex 116, CICAD I, p. 10.
596 US Department of State, Bureau for International Narcotics and
Law Enforcement Affairs, Memoranda of Justification Concerning the

Secretary of State’s 2002 -2007 Certifications of Conditions Related to the
Aerial Eradication of Illicit Coca in Colombia, Washington D.C., 2002-2007,
Annexes 140 (p. 1), 143 (p. 4), 144 (pp. 3-4), 146 (pp. 1-2), 147 (pp. 1-2),
149 (p. 2).
258for cases, there does not appear to be any evidence that

glyphosate aerial spraying has resulted in any adverse health

effects among the population where this spraying takes
597
place.”

7.52. These conclusions were confirmed by a further

independent scientific st udy conducted in 2005, entitled

“Environmental and Health Assessment of the Aerial Spray

Program for Coca and Poppy Control in Colombia” (CICAD I).

That study was conducted with the collaboration of Dr. Keith R.

Solomon, Dr. Arturo Anadón, Dr. Antonio Luiz Cerdeira, Dr.

John Marshall and Dr. Luz Helena Sanín. The first phase of the

study concluded that both glyphosate and Cosmo-Flux, in the

manner in which they have been used in the eradication program

in Colombia, do not pose any significant risk to human
598
health. The study was republished in 2007 in the peer-
reviewed journal Reviews of Environmental Contamination and

Toxicology. 599

7.53. CICAD I found, inter alia, that:

“[i]n the entire cycl e of coca and poppy
production and eradicati on, human health risks
associated with physical injury during clear-

597 Annexes 144, 146, 147, 149.
598 “The risk assessment concluded that glyphosate and Cosmo-Flux®

as used in the eradication program in Colombia did note present a significant
599k to human health”. In Annex 116, CICAD I, p. 11.
Reviews of Environmental Contam ination & Toxicology 2007;
90:43-125.
259 cutting and burning and the use of pesticides for
protection of the illicit crops were judged to be
more important than those from exposure to
600
glyphosate.”

The study also took note of the pr evailing view in the scientific
community that glyphosate is a substance of low toxicity to

organisms other than green plan ts, and is not a carcinogenic,

mutagenic or reproductive toxicant. 601

7.54. In addition, CICAD I indicated that in chronic and sub-

chronic studies, no adverse effects were found on mammal

reproductive tissues treated with glyphosate or AMPA. Nor did

the results of normal studies with these substances show effects

indicating any endocrine modul ation. It was therefore

concluded that the use of glyphos ate has no adverse effects for

the development, or on the reproductive or endocrine systems of
humans and other mammals. 602 Furthermore, no evidence of

neuro-toxicity due to glyphosate exposure was found. No

neuro-toxicity was observed in numerous studies of acute, sub-

chronic or chronic effects conduc ted on rodents, nor were any

such effects observed in two specific studies conducted on dogs.

Immune-toxicity results suggest that glyphosate does not affect

600 Annex 116, CICAD I, p. 11.
601 Annex 116, CICAD I, p. 10.
602 “Comprehensive reproductive and developmental toxicology studies
carried out in accordance with internationally accepted protocols have

demonstrated that glyphosate is not a developmental or reproductive toxicant
and is not an endocrine disruptor (Williams et al. 2000) (USEPA 1993a)
(World Health Organization International Program on Chemical Safety
1994)”. In Annex 116, CICAD I, p. 52.
260mammalian immune response at realistic exposure
603
concentrations.

7.55. The first phase of the CICAD study was extensively

discussed in both Colombian and in ternational scientific circles,

mainly due to its wide circulation in the academic and scientific

arena. Among the various comments on the first CICAD study,

a critique was prepared by me mbers of Colombia’s National
University (Universidad Nacional de Colombia). 604 In the light

of those comments, CICAD requested the scientific team

responsible for the study to caref ully review the arguments put

forward by the critique and to clarify the technical questions

submitted. The reply of the scientific team begins by explaining

that the objections to the study were mainly due to a

“misunderstanding, lack of knowledge, or possibly a less critical

approach to the subject under di scussion than is desirable”. 605

The reply notes that the Nati onal University team did not

include members with any appare nt expertise in human health,

epidemiology, medicine, public health or ecotoxicology:

“clearly some of the comments in the Critique would not have

been necessary if the team had included expertise in these
606
critical areas”.

603 Annex 116, CICAD I, p. 52.
604 EM, Vol. III, Annex 152.
605 Annex 130: K. Solomon et al., A Response to Tomás León Sicard et

606, CICAD, 2005, p. 1.
Ibid., p. 2.
2617.56. The reply affirms that th e CICAD team members are

independent experts and goes on to address and clarify each of
the issues set forth in the critique, including the field

assessments that were conducte d, and the underlying scientific

basis for each of the conclusions.

7.57. The second phase of the CICAD study (CICAD II),

conducted in 2007, included studies on the potential effects on

human health of the use of the glyphosate formulation and
Cosmo-Flux for the eradication of coca. These were also peer-

reviewed and were published in 2009 in the scientific review

Journal of Toxicology and Environmental Health.

7.58. One of the studies carried out in the framework of the

CICAD process was a cytogeni c bio-monitoring study (i.e.

analyses of blood, tissue, urine, etc., carried out in order to
estimate the genetic effects of exposure to a given external

factor, such as a chemical substance, in the body and in the

cellular components, particularly chromosomes), conducted by a

group of international scientists on people living in five different
regions of Colombia that had be en exposed to the sprayings or

had exposure to glyphosate. It concluded that:

“Overall, these results suggest that genotoxic
damage associated with glyphosate spraying [...]
is small and appears to be transient. [...]

Evidence indicates that the genotoxic risk
potentially associated with exposure to
glyphosate in the areas where the herbicide is

262 applied for eradication of coca and poppy is of
low biological releva nce. [...] Glyphosate

persists in the environment for only a short time
(half-life for biological availability in soil and
sediments is hours, and 1- 3 d in water; Giesy et
al., 2000), is rapidly ex creted by animals and

other vertebrates [...] and chronic effects, if any,
would not be expected.” 607

7.59. Another study looked at possibl e effects of exposure on

human fertility, assessing 2,592 women from 5 different

regions. It concluded that “ae rial spraying of glyphosate was

not consistently associated with delayed time to pregnancy”. 608

7.60. Yet another study involved “a cytogenetic biomonitoring

study... carried out in subjects fr om five Colombian regions,

characterized by different expos ure to glyphosate formulations
609
and other pesticides”. 274 volunteers were involved. The

authors conclude that “[o]vera ll, these results suggest that

genotoxic damage associated with glyphosate spraying... is

607
Annex 131-I, CICAD II: C. Bolognesi, G. Carrasquilla, S. Volpi,
K.R. Solomon, D.C. Cole & E.J.P. Marshall, “Biomonitoring of Genotoxic
Risk in Agricultural Workers from Five Colombian Regions: Association to
Occupational Exposure to Glyphosate”, in Journal of Toxicology and
Environmental Health, Part A 72:986-997, 2009.
608 Annex 131-E, CICAD II: L.H. Sanin, G. Carra squilla, K.R.
Solomon, D.C. Cole & E.J.P. Marshall, “Regional Differences in Time to
Pregnancy among Fertile Women from Five Colombian Regions with
Different Use of Glyphosate”, in Journal of Toxicology and Environmental

609lth, Part A 72:949-960, 2009, at p. 959.
Annex 131-I, CICAD II, at p. 986.
263 610
small and appears to be transient” and of low biological
611
relevance.

7.61. Likewise, Colombian scient ific institutions and

authorities have conducted studi es on possible effects of the
aerial spraying program on human health. In 2001, the Clínica

de Toxicología Uribe Cualla located in Bogotá (a toxicology

clinic), one of the most prestigi ous national centres of this type,

conducted a study to examine the complaints concerning the

alleged adverse effects on human health attributed to aerial

sprayings. It concluded that the health problems observed in the

municipalities under study had prev alence rates similar to those

in the years prior to the start of the sprayings, and that the

problems were also present in municipalities where no aerial
612
spraying had taken place.

7.62. The study also concluded that the symptoms initially

attributed to aerial sprayings originated in multiple sources

including non-chemical sources. In particular, it concluded that

the symptoms in question could be due to the chronic exposure

of the population to the multiple agrochemicals used for

610 Annex 131-I, CICAD II, p. 994.
611 Ibid., p. 995. See also Annex 132, p. 3: “glyphosate does not pose a
risk of heritable (passed from parent to child) or somatic (body cell)

612ations in humans.”
Annex 126: Embassy of the United States of America - Clínica de
Toxicología ‘Uribe Cualla’, Alleged effects of Glyphosate on Human Health ,
Bogotá, December 2001, pp. 52-53.
264growing coca, and that those sy mptoms could also result from
other infectious and allergic causes:

“The illnesses most of ten attributed by the
study’s subjects as secondary to glyphosate
spraying were: gastrointestinal symptoms

(diarrhea, vomit and nausea), skin symptoms
(pruritus or itch, erythema or reddening, vesicles
or blisters, soreness an d sores), eye symptoms
(soreness, reddening, pink eye, pain and
pruritus), respiratory symptoms (dyspnea or

fatigue, cough and croup or rhinorrhea), cephalea
(headache and fever). These symptoms may
originate due to multiple causes, as well as to
exposure to chemical elements.” 613

7.63. In the same year, the Nationa l Narcotics Directorate of
Colombia commissioned a set of six studies in relation the

toxicity of the mix used in the spraying program. 614 Those

studies were carried out by Laboratorio Inmunopharmos Ltda.,

an immunological and pharmaceu tical laboratory located in

Bogotá, specializing in biomedical research for the purposes of

protecting human and animal hea lth and the environment. The

studies, conducted in compliance with good laboratory practices
and according to US EPA guidelines, analyzed the mix of 44%

Glyphosate/1% Cosmo-Flux/55% wa ter. Results from these

studies showed that the effects due to exposure to the spray mix

613
614 Annex 126, p. 54.
The studies conducted by the Laboratorio Inmunopharmos (Annex
128) were: Acute oral toxicity LD50 (EPA Guideline 870-1100), Acute
dermal toxicity LD50 (EPA Guideline 870-1200), Acute inhalation toxicity
LD50 (EPA Guideline 870-1300), Acute eye irritation (EPA Guideline 870-
2400), Acute dermal irritation (EPA Guideline 870-2500), Dermal
sensitization (EPA Guideline 870-2600).
265are slight and not lasting. 615 According to the studies’

conclusions, the mix is to be classified as “slightly toxic”, like

any other agricultural fertilizer or household pesticide.

7.64. The National Health Institute of Colombia, after

collecting and analyzing biologi cal and environmental samples

related to complaints received about alleged adverse effects on

health caused by aerial sprayings as part of the Public Health

Program component of the Environmental Management Plan, 616

found that none of the alleged effects on human health in any of

the claims submitted bore any relation to the aerial sprayings but

were instead due to other causes. 617

The National Health Institute concluded that:

“To date, in relation to the complaints on alleged
adverse effects of the sprayings on human health,

analyses conducted by Colombia’s National
Health Institute showed no relation between the
sprayings and reported symptoms. Instead,

complaints received were related to the endemic 618
health conditions of the regions concerned…”

615An12e8.
616 Under the Program, the General System of Social Security in Health

is responsible for processing claims relating to alleged adverse health effects.
To that end, a procedure has been put in place for de termining whether
exposure to the spray mix may have caused any harmful symptoms. See
Annex 68, pp. 3-4.
617 Annex 68, p. 4.
618 Ibid., and see Annex 69: Report on Complaints submitted to the
Colombian National Health Institute with regard to the Program for the
Eradication of Illicit Crops with Glyphosate Herbicide – PECIG, 2010.

2667.65. In support of its arguments as to the alleged effects of
the spraying, Ecuador refers to the results of an investigation

carried out by the Putumayo Depart ment of Health between 22

December 2000 and 2 February 2001 (“the DASALUD

Report”). 619 While the DASALUD Report notes that some of

the local hospitals reported an increase in the number of

complaints for conditions which were attributed by the patients

to aerial spraying, it does not refer to any medical findings
whatsoever supporting the allegation that those symptoms were

in fact caused by the spraying.

7.66. To the contrary, a passage in the report (which,

significantly, was not included by Ecuador in the translation

provided in Annexes to its Memo rial) casts doubt on the actual

causes of the symptoms complained of by the individuals:

“In the emergency clinical cases for the month of
December 2000 it may be noted that, out of 6
cases examined, 50% attribute their symptoms to

the fumigation, while of the remaining 50%, 2
result from alcoholic intoxication, and one case
from intoxication due to direct manipulation of
chemicals in other circumstances.” 620

That is, in circumstances unrelated to the aerial spraying.

619
620 EM, Vol. II, Annex 90.
Ibid., p. 13 (emphasis added, free translation by Colombia). The
Spanish original reads as follows:“Enlas historias clínicas de atención de
urgencias del mes de diciembre del 2000 se puede constatar que de 6 casos
que consultaron, el 50% atribuye su sintomatología a la fumigación, mientras
que del 50% restante 2 corresponden a intoxicación alcohólica y un caso a
intoxicación por manipulación directa de químicos en otras circunstancias.”
2677.67. In the light of that statement, it is open to serious

question how many of the other cases reported in the
DASALUD Report were in fact cau sed by the aerial sprayings.

Furthermore, direct manipulation of chemicals -including those

used for coca cultivation and pro cessing- is quite different from

exposure to spray mix used in aerial eradication. In the absence

of any medical evaluation of th e illnesses allegedly suffered by

the individuals in question, in light of the dates and times of the

fumigations, there is no way of establishing that the health

problems mentioned in the Report were indeed caused by or

were even related to the sprayings.

7.68. The scientific evidence pr esented above rebuts the

arguments of the Menzie Repor t, on which Ecuador heavily

relies. The Report claims that the reported adverse health
621
effects “are consistent” with the “known risks” and
622
“recognised properties” of the surfactants “believed to be used
623
in the spray mixture.” These assertions are made merely on

the basis of a review of scientific literature and without
conducting any toxicological analysis of the spray mixture, still

less any actual field work.

7.69. As the scientific evidence s hows, cases of accidental

exposure to glyphosate would entail minimal risks to human

621
622 EM, para. 6.44. EM Vol. III, Annex 158, Section 5.
623 EM, para. 6.46. EM Vol. III, Annex 158, Section 5.1.1.1,
EM, para. 6.47. EM Vol. III, Annex 158, Section 5.1.1.1.
268health, generating “primarily eye and skin irritation” symptoms

that are not long lasting. 624 In fact, glyphosate is rapidly

eliminated via urine and feces. Thus, “in one metabolism study

with rats, most of the glyphosate administered (97.5 percent)

was excreted in urine and feces as the parent compound; less

than one percent of the absorbed dose remained in tissues and
625
organs, primarily in bone tissue”.

626
7.70. Moreover, contrary to what the Menzie Report asserts,

glyphosate is a non-v olatile substance and no question of

intoxication due to inhalation of the spray mix arises, since its

permanence in the air is minimal. As the EPA states “the acute

inhalation toxicity study was waived because glyphosate is non-

volatile and because adequate i nhalation studies with end-use
627
products exist showing low toxicity”.

7.71. The overwhelming weight of scientific evidence

supports Colombia’s assertions to the effect that the spray mix

used in the PECIG program ha s negligible effects on human
health. This conclusion is not new: it reflects the general

scientific opinion as of 2000. CICAD II notes that by 2000,

624 EM, Vol. II, Annex 132, p. 4. EM, paras. 5.36, 6.46, 6.47, 6.48.
625 EM, Vol. II, Annex 132, p. 3.
626E4r9a.,
627 EM, Vol. II, Annex 132, p. 2.

269glyphosate and its formulations had already been extensively
628
investigated for potential adverse effects in humans.

(2) A LLEGED EFFECTS ON FAUNA

7.72. Similar conclusions have been drawn as to the effect of

the spray mixture on fauna.

7.73. In 1994, the International Programme on Chemical

Safety (IPCS) study found that in experimental animals,

“technical glyphosate has very low acute [short-term] toxicity
by the oral and dermal administration routes… Long-term

toxicity was also studied in mice and rats. Few effects were

observed and, in almost all cases , at relatively high dose levels

only.” The available studies did not indicate that technical

glyphosate is carcinogenic or capable of causing genetic
629
mutations or alterations to embryos or foetuses.

7.74. Similarly, in 2000, the FAO concluded that:

“on the basis of toxicity data and application
rates for the active subs tance glyphosate, the

risks for birds, mammals, aquatic organisms,
bees, earthworms and micr o-organisms in soil in

628
Annex 131-A, CICAD II, p. 918, citing G.M. Williams et al.,
“Safety evaluation and risk assessmen t of the herbicide Roundup® and its
active ingredient, glyphosate, for humans” in Regulatory Toxicology and
629rmacology, 31:117–165, 2000, pp. 117, 160 (Annex 125).
Annex 96, pp. 5, 6. Also: “EPA would not expect any risk to birds
and mammals, including livestock, based on dietary exposure to the active
ingredient glyphosate.” In Annex 142, p. 5 (partially at EM, Vol. III, Annex
143).
270 observance of correspo nding risk management
630
measures are regarded as slight.”

The FAO study also found that glyphosate and its metabolite

AMPA are not carcinogenic, do not produce genetic mutations

or alterations to embryos or foetuses, and do not impair
631
reproduction.

7.75. In 2005, CICAD I stated that “for the environment, risks

from the use of glyphosate and Cosmo-Flux® to terrestrial
632
animals were judged to be small to negligible”. It added that

“the addition of the Cosmo-Flux® to the glyphosate did not
633
change its toxicological properties to mammals.” Eventhe

effects of direct overspray on farm animals would be minor
634
(temporary eye and skin irritation).

7.76. However, it had been suggested in the literature that

amphibians (especially tadpoles) were more sensitive. It should

be recalled that amphibians are deemed as organisms that attest

to the toxicity, if any, of a give n substance. They are suited for

this role, given their sensitiveness to alterations in their

surroundings and their high reproductive rate. It is scientifically

accepted that toxicity studies conducted on minor species 635 may

630 Annex 101, p. 26.
631
632 Ibid., p. 24
633 Annex 116, CICAD I, p. 11.
Ibid., p. 85.
634 Annex 116, CICAD I, pp. 41-70.
635 Minor species often used include, among others, earthworms, toads,
adult frogs, juvenile frogs, amphibian larvae, bees, rabbits, mice and rats.
271be used for extrapolating their findings to other species and

humans. Hence, the particular issue of amphibians was the
subject of four of the studies in CICAD II:

• Lynch & Arroyo examined which Colombian

frog species might be most at risk as endemic to
636
coca-producing areas. They stress the damage

done to frog habitats by clearing forest areas for

planting: “cultivation of coca is likely to reduce
the resident frog fauna by approximately 90%...

prior to consideratio n of any effects of

glyphosate spraying upon the amphibian

fauna”.637

• Brain & Solomon (2009) compared “hazard
quotients” for amphibians of the spray mixture

compared with major chemicals used for coca

production. The latter “w ere up to 10- to 100-

fold more toxic to frogs that the Glyphos–
638
Cosmo-Flux mixture”.

• Bernal, Solomon & Carrasquilla (2009)

investigated whether the spray mixture was more

lethal to Colombian frogs than to the temperate

region frogs on which most scientific testing had

focused, and concluded that the answer was

636
637 Annex 131-H, CICAD II.
638 Ibid., p. 975.
Annex 131-D, CICAD II, Brain et al. (2009), at p. 937.
272 negative. 639 Glyphosate itself was “essentially

nontoxic to amphibians”, and the addition of

Cosmo-Flux did not change that conclusion. The

surfactant was the source of mild toxicity:

“overall 95% of larval frogs would have LC50s

greater than” that caused by the concentration
640
used in the aerial spraying program.

• In a companion study, the same authors

investigated the impact of the spray mixture on
Colombian frogs under field conditions. 641 They

concluded that “under realistic worst-case

exposure conditions, the mixture of Glyphos and

Cosmo-Flux as used for control of coca in

Colombia exerts a low toxicity to aquatic and

terrestrial stages of anurans and that risks to these

organisms under field conditions are small”. 642

7.77. CICAD II concluded that:

“...when considering the cumulative impacts and
risks of coca production collectively in a

639
Annex 131-F, CICAD II: M.H. Bernal, K.R. Solomon & G.
Carrasquilla, “Toxicity of Formulated Glyphosate (Glyphos) and Cosmo-
Flux to Larval Colombian Frogs 1. Laboratory Acute Toxicity”, in Journal of
640icology and Environmental Health, Part A 72:961-965, 2009 at p. 937.
Ibid., p. 964.
641 Annex 131-G, CICAD II: M.H. Bernal, K.R. Solomon & G.
Carrasquilla, “Toxicity of Formulated Glyphosate (Glyphos) and Cosmo-
Flux to Larval and Juvenile Colombian Frogs 2. Field and Laboratory
Microcosm Acute Toxicity”, in Journal of Toxicology and Environmental
Health, Part A 72:966-973, 2009.
642 Ibid., p. 966.

273 multifactorial context to amphibian populations
in coca growing regions, they are judged to be

greater than those posed by the use of glyphosate
and Cosmo-Flux employed for the spray control
program.” 643

7.78. The matter is considered in de tail by Dobson, in light of

CICAD II’s conclusions. His own view is that to protect 99% of

species of tadpoles in even shallower water depths would

require an increase in the neces sary protective swath to 50

metres: he reached “the overall conclusion that effects as far
away as Ecuador are highly improbable”. 644

7.79. The Ecuadorian Memorial again relies on unfounded

arguments in the Menzie Report 645in an attempt to show alleged

damages to fauna of aerial sprayi ng. For instance, the Report

refers to adverse effects caused by direct ingestion of glyphosate

products in large quantities, 646 a situation that does not occur

with aerial spraying where the amount susceptible of being

deposited on the ground, even in the context of direct overspray

(and a fortiori of drift), is low. Moreover, due to the swift

643 Annex 131-D, CICAD II, Brain et al. (2009), pp. 945-946.
644
Appendix: 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. (2009)
645ex 158, (hereafter referred to as “Appendix – Dobson Report”) p. 21.
E5r8a.,
646 These assertions are based on reports of poison control centres due
to voluntary or accidental ingestion of concentrated glyphosate. The Menzie
Report, p. 19, acknowledges that for such symptoms to occur, an individual
would have to ingest significant quantities of glyphosate. These are simply
impossible to ingest taking into account the actual amount of glyphosate
released per hectare in the course of aerial spraying. See para. 7.80 below.

274degradation and non-persistence of the ingredients of the spray

mix in the environment, there is no question of the ingestion

through drinking water or foodstuffs, since according to EPA

“the chronic dietary risk pos ed by glyphosate food uses is
647
minimal” and “in animals, most gl yphosate is eliminated in
648
urine and feces”.

7.80. Although alleged effects on human health were
discussed in the previous secti on, it is relevant to make the

following precisions at this point: taking into account

glyphosate’s Lethal Dose 50 (LD 50) 649 – 5,000 milligram per

kilogram of bodyweight of the individual) 650– a person would

need to directly sw allow more than 625cc of glyphosate-based

commercial formulated product (i.e., over half a litre or two

cans of soda) or 1.4 litres of the spray mix, in order to ingest the

LD 50 estimated for an individual weighing 60 kg, and that is

assuming the body absorbed 100% of the glyphosate

commercial formulation present in the spray mix. However,

neither the glyphosate-based commercial formulated product nor

the spray mix used in Colombia can be absorbed 100% by the

human body. In fact, studies show that only 30-36% of ingested

647 EM, Vol. III, Annex 132, p. 3.
648Ibid.
649 Lethal Dose 50 is the “statistical estimate of the minimum dose
required to kill fifty percent of a population of laboratory test animals in
controlled conditions.” Annex 135, Annex 1, Glossary, “Mean Lethal Dose,

65050”. See also, para. 7.76 and note 634 above.
Annex 96, p. 29.
275 651
glyphosate is absorbed. This means that only direct ingestion
of at least 1.74 litres of the glyphosat e commercial formulation

as is or 3.95 litres of the spray mix as is by an individual

weighing 60 kg could –in theory- possibly result in death. This

is clearly not possible in the framework of Colombia’s spraying

program since, as recalled a bove, the amounts released in the

PECIG program are 23.65 l/ha of spray mix containing 10.4 l/ha

of glyphosate-based formulated product. Therefore, a 60-kg
individual would have to ingest, in one go, all of the glyphosate

sprayed over 1,673 square meters directly under the spray swath

to achieve this dose.

7.81. It is relevant to recall that Colombia’s spraying program

does not target food crops or pastures. Where coca crops are

sprayed and part of the spray we re to fall on pastures, the
mixture amount would not cause significant effects on livestock,

since a very significant amount of herbicide deposited per

hectare would be required to affect grazing animals to any

degree. According to Dr Dobson:

“It is difficult to take this section of the Menzie

et al. (2009) paper serious ly. It is the perfect
example of the dangers of expressing hazard (the
possibility) without refe rence to exposure and
risk (the probability) of something adverse
652
occurring.”  

651 M. Burger et al, “Exposición al herbicida glifosato: aspectos clínicos
toxicológicos”, Rev Med Uruguay; 20: 202- 207, p. 203. Available at:
http://www.scribd.com/doc/2628324/Glifosato (last visited 10 March 2010)
652 Appendix – Dobson Report, para. 92.
2767.82. The Menzie Report findings of many reported fish
653
kills are unsubstantiated: acco rding to the literature
654
“glyphosate [has] exhibited little chronic toxicity to fish”. In

any event, it should be recalle d that 100-metre no-spray buffer

zones along watercourses or ar ound water bodies are observed

in the PECIG program. Also, results of tests on over 80 water

samples collected in the framework of the PECIG program’s

Environmental Management Plan have shown no presence of
655
glyphosate or its metabolite AMPA.

7.83. Thus, as was clear from the scientific evidence from the

outset, the evidence derived from spraying operations at various

locations throughout Colombia over several years evidences that

the glyphosate-Cosmo-Flux mix does not produce significant

effects on fauna.

(3) A LLEGED EFFECTS ON SOIL AND IMPACTS ON NON -TARGET

CROPS

7.84. According to numerous scient ific studies carried out at

both the national 656 and international levels, glyphosate has no

653 EM, para. 6.104.
654
Annex 124: J.P. Giesy, S. Dobson S & K.R. Solomon,
“Ecotoxicological risk assessment for Roundup herbicide”, Reviews of
655ironmental Contamination and Toxicology 167: 35-120 (2000), p. 74.
Annex 149, p. 2.
656 Analyses of soil samples, collected mainly in the Colombian
provinces corresponding to the Amazon and Andean regions, between 2003
and 2008, were conducted in the framework of the EMP. The analyses show
that the likelihood of glyphosate having a negative or positive incidence on

277long-term effects on soil, as it does not persist in soil. In a

recent field study, the United States Department of Agriculture

found that:

“[I]n two years of field evaluation no significant
effects of glyphosate were detected on soil or
root microbial communities after two in-season

glyphosate applications. Exposure of soils to
glyphosate in a laboratory experiment resulted in
small, short-term changes in the microbial
community and a brief inhibition of microbial
657
activity.”

7.85. Due to these characteristics, the International Program

on Chemical Safety recommended that “the major formulation

Roundup may, for instance, be used in pre-plant treatments for
658
seed bed preparations...” In fact, the sprayed fields are
659
quickly colonized by invading vegetation. Insoilstreated

with glyphosate, any vegetable species can be replanted

immediately following application and the plants will develop

highly acidic soils, such as those in the Amazon region from where most of
the samples were taken, is minimal. Samples of results of soil analyses,
Appendices 2 and 3, to the Environment Ministry Report, Annex 70.
657 United States Department of Agriculture (USDA), Agricultural

Research Service, Maintaining soil resources fo r effective conservation and
herbicide management in mid-south crop production , 2007. Available at:
http://www.reeis.usda.gov/web/crisprojectpages/410786.html (last visited 10
March 2010). Also: “Glyphosate has no residual activity, once adsorbed to
soil it quickly becomes unavailable to plants and no longer has herbicidal
activity… In contrast, some herbicides have month-long or even year-long
residual activity which limits the plants that may be grown following their
use.” In Annex 142, p. 9 (excerpt included at p. 5 in EM, Vol. III, Annex
143).
658 Annex 96, p. 12.
659 Annex 116, CICAD I, pp. 48-49.

278normally. 660 No decrease in produc tivity of crops planted

following application occurs. 661

7.86. In the physiochemical analyses conducted on almost 180

soil samples from sprayed plots in Colombia, no adverse effects

caused by glyphosate have been evidenced in any of the

662
physical, chemical or biological characteristics assessed.

7.87. It is true that glyphosate is a non-selective herbicide: if it

falls on non-targeted plants outside the scope of the eradication

program, they may, depending on the dose, also be affected. In

order to prevent this situati on, all reasonable measures are
663
adopted when spraying as part of the PEGIC program.

Moreover, the Colombian Government has also implemented a

program to compensate for proven incidental damage to lawful
664
crops.

7.88. Glyphosate and AMPA residue s found in soil samples

do not necessarily result from aerial spraying, but may rather be

660
United States Department of Agriculture, “Effects of glyphosate on
soil microbial communities and its mineralization in a Mississippi soil”,
published in the peer-reviewed journal Pest Management Science , in 2007,
Abstract. Available at:
http://www.ars.usda.gov/SP2UserFiles/Place/64022000/Publications/Weaver
/Weaveretal07PMS63.pdf (last visited 10 March 2010)
661
662 Contrary to Ecuador’s assertions in, e.g., EM, paras. 5.54-5.55.
Annex 70, Environment Ministry Report and Appendices 2-3 (soil
sample analyses). See also, Annex 149, p. 2.
663 See above, para. 7.174.
664 See above, para. 4.22 for Colombia’s Complaints Attention
Program.
279due to the chemical treatments us ed by the growers of the illicit

crops themselves. Some results even reported traces of
glyphosate and AMPA prior to spra ying in areas that had never

been sprayed before as part of the PECIG program, evidencing

the use of this very herbicide in coca cultivation for the purpose

of weed control.

(4) A LLEGED EFFECTS ON WATER RESOURCES

7.89. In the context of the PECIG program in Colombia,

bodies of water and watercourses are designated as exclusion

zones: a 100-meter no-spray stri p is left around them. Water

resources in the areas subject to the spraying program are not
targeted.

7.90. But even in case of accidental overspray, as the
Guidelines on drinking-water quality first pu blished by the

WHO in 1998 note: “[t]he low mobility of glyphosate in soil

indicates a minimal potential fo r the contamination of ground

water”. Moreover, glyphosate is chemically stable in water and
is not subject to photochemical degradation. While the WHO

also remarks that glyphosate “can enter surface and subsurface

waters after direct use near aquatic environments or by runoff or
leaching from terrestrial application”, it concludes that “because

of the low toxicity of glyphosate, the health-based value derived

for this herbicide is orders of magnitude higher than the

concentrations normally found in drinking-water. Under usual

280conditions, therefore, the pres ence of glyphosate in drinking-

water does not represent a hazard to human health, and the

establishment of a numerical gu ideline value for glyphosate is

not deemed necessary”. 665 These results were confirmed in the

recent update of the Guidelines. 666

7.91. Colombia’s National Health Institute (Instituto Nacional

de Salud) conducted studies in Putumayo in 2007 of cases of

human consumption of water alle ged to be contaminated with

glyphosate. It was concluded that the symptoms reported and

supposedly attributable to sprayings were in fact due to the

water being contaminated with bacteria. 667 In certain cases, the

reported symptoms have been associated with other highly toxic

pesticides used in the growth and processing of coca.

7.92. According to analyses carried out for the purpose of

assessing glyphosate and AMPA concentrations on surface

waters (CICAD I), in every location and on nearly all occasions,

glyphosate and AMPA residues were below the detection level

665 Annex 98: World Health Organization Geneva, “Guidelines for
Drinking-Water Quality”, 2 nded., Addendum to Volume 1,
Recommendations, 1998, p. 21.
666 World Health Organization Geneva, Guidelines for Drinking-Water
Quality (3rded., incorporating the first and second addenda, Volume 1,

Recommendations, 2008) p. 379. Available at:
http://www.who.int/water_sanitation_health/dwq/fulltext.pdf (last visited 10
667ch 2010).
Annex 64: “Investigation Regard ing Possible Secondary Effects on
Human Health, Allegedly Derived from Glyphosate Spraying in the Rural
Area of Villanueva of the Orito Municipality, Putumayo on 6 August 2007”,
Final Report, National Health Institute of Colombia, June 2008, pp. 14-15
(Conclusions).

281 668
of the analytical method used. That data suggests that “little

or no contamination of surf ace waters with glyphosate at

significant concentrations ha s resulted from the use of

glyphosate in either agricultural or eradication spraying in
669
Colombia”.

7.93. Despite the absence of evidence of harm, there is

continued monitoring to verify that the PECIG program is not

causing adverse effects on the environment in general and water

resources in particular. 670 Attached to the Report of the

Ministry for the Environment is an example of a twice-yearly

Verification Report coveri ng the period September 2008-
671
February 2009. It concluded that:

“- Achieved Field Efficacy was 92,2% and

96,4% for Swath [Efficacy], which indicates that
38.084 hectares of coca crops were effectively
controlled. The figures were above the 7-year
historical average of 89% and it has been the

highest efficacy ever achi eved in the course of
the program for the eradication of illicit crops.

- The accuracy index represented as Off-
Target was 2,2%, an acceptable range within the
parameters set out in the Environmental

Management Plan.

668 This limit is equal to 25 µg/lt.
669 Annex 116, CICAD I, pp. 45-46.
670 Annex 50, Record 5, Num 3.2. See also Annex 66, p. 3; Annex 67,

67114; Annex 70, pp. 4-5, 7.
19th Verification Mission Concerning the Spraying Operations
Conducted Between September 2008 an d February 2009, Technical Report
(July 2009), Appendix 1 to the Environment Ministry Report, Annex 70.
282 - Not a single spray line was found to be
over a [vegetation] cover other than illicit coca
crops. 100% were over plantations targeted for
672
control (Coca).”

Soil and water samples taken at the time are also appended; their
results were negative for glyphosate and AMPA. 673

(5) A LLEGED EFFECTS ON AIR QUALITY

7.94. Although the Ecuadorian Memo rial does no t actually

refer to air pollution as bei ng caused by aerial spraying

operations, some of the anonymous witness statements

submitted by Ecuador do mention it. Such assertions have no

scientific basis since, as confirmed by the United States’

Environmental Protection Agency, the low vapour pressure of
the chemical compound renders it “non-volatile”. 674

(6) E CUADOR S SCIENTIFIC EVIDENCE :THE M ENZIE R EPORT

7.95. In addition to selective citations from a limited number

of published sources, Ecuador produces one additional item, the

Menzie Report. The Exponent team that prepared this Report

consisted of 10 professionals, of whom only one visited the

border zone in Ecuador. That person made field observations

and collected testimonies, wi thout taking any samples or

conducting actual field tests. Furthermore, according to the

672
673 Appendix 1 to the Environment Ministry Report, Annex 70, p. 15.
674 See Appendices 2-5 to the Environment Ministry Report, Annex 70..
EM, Vol. III, Annex 132.
283information provided in the Repor t, none of the team members
has a background or experience in meteorology, physics or the

aerial application of pesticides. 675

7.96. These shortcomings are reflected in the language used in

the Report itself. Its formulations used are highly speculative.

The word “may” is used 67 times; the word “likely” 19 times.

Expressions such as “We believe it is likely that these
conditions would result in...” or “…although we have no

confirmation, it is entirely possible that…” abound. The word

“consistent”, used to try to establish a causal relation between

alleged effects and aerial spraying, is to be found 21 times.

7.97. The Report does refer to literature that is relevant on

certain topics, but either the studies themselves or their
conclusions are presented out of context. For instance, with

regard to the effects on human health, some of the studies

quoted refer to symptoms attributable to dermal exposure, but

the Report does not differentiate the circumstances analyzed.

One study involved tests on volunteers who agreed to have

concentrated glyphosate, among other substances, applied to
676
their skin. In fact, the study found that the herbicide was
“less irritant” than dishwashing detergent, general all purpose

675
EM, Vol III, Annex 158, Section 1.
676 H.I. Maibach, “Irritation, sensitization, photoirritation and
photosensitization assays with a glyphosate herbicide”, Contact Dermatitis,
Vol. 15, 1986, pp. 152-156. (Document deposited with the Registry of the
Court)
284cleaner and even baby shampoo, a conclusion disregarded in the
677
Report. Another study cited by Menzie referred to an episode

of multiple complaints reporting alleged adverse effects of

glyphosate spray in California. The case involved the

accidental exposure of a small num ber of actual applicators to

the concentrated formulation or to the spray solution (i.e. it was

a case of occupational exposure, not of incidental exposure in

the field). The symptoms were very mild and, in any case, the
main conclusion was that large numbers of reports of adverse

effects cannot be construed to indicate the likelihood or severity

of any widespread problem of exposure to glyphosate. 678

7.98. Although the adverse effects of the chemicals used in

coca cultivation and processing on the environment are well

known, the Menzie Report completely omits any mention of

them, exhibiting a lack of balance. An assessment that fails to

factor in those elements as well as other variables present in the
679
area – e.g., felling of the natural forest, inadequate

agricultural practices – is lacking in scientific rigor.

677
D.A. Goldstein, J.F. Acquavella & R.M. Farmer, “An analysis of
glyphosate data from the California Environmental Protection Agency
Pesticide Illness Surveillance Program”, Journal of Toxicology Clinical
Toxicology, Vol. 40, 2002, pp. 885-892. (Document deposited with the
Registry of the Court)
678 D.A. Goldstein, J.F. Acquavella & R.M. Farmer, “An analysis of
glyphosate data from the California Environmental Protection Agency
Pesticide Illness Surveillance Program”, Journal of Toxicology Clinical
Toxicology, Vol. 40, 2002, pp. 885-892. (Document deposited with the
Registry of the Court)
679
It should be recalled that the area has endured the felling of the
natural forest for different purposes , including wood-exploitation activities
2857.99. The Menzie Report states that “the exact mixture used

by GOC [Government of Colombia] is unknown —it has been
680
kept confidential”. However, in the same paragraph it refers

to a document from the Department of State (“U.S. Department
of State. 2002a”) entitled “Report on issues related to the aerial

eradication of illicit coca in Colombia: Chemicals used for the

aerial eradication of illicit coca in Colombia and conditions of

application”. This document, which is Annex 144 to the

Memorial of Ecuador, starts by defining the spray mixture in a
precise and accurate way:

“The aerial spray mixture currently used in the
U.S.-supported program of aerial eradication of
coca in Colombia contains three components:
water, an EPA-registered formulation of the
herbicide glyphosate, a nd a surfactant (Cosmo-

Flux 411F)… These components are mixed
together into a spray mixture in the following
percentages: 55 percent water, 44 percent
glyphosate formulation, and 1 percent Cosmo-
Flux 411F...”

7.100. In addition to its conjectures concerning the components

of the spray mix used in Colombia, the Memorial relies on the

Menzie Report in support of its assertion that surfactants in

general – and those it speculates may be part of the Colombian

spray mix – increase the mix’s toxicity, posing greater risks to

and slash-and-burn practices associated with the establishment of coca as
well as lawful crops.
680 EM Vol. III, Annex 158, Section 3.
286 681
plants, humans and animals. In this regard, research and

studies carried out by Collins & Helling (2002) were cited by

the Menzie Report in support of the conclusion that the addition

of an adjuvant represents a fourfold increase in the mixture’s

toxicity. In fact, what Collins & Helling found was that two of
the tested surfactants “inc reased glyphosate phytotoxicity

fourfold”, i.e., its effect on plan ts. “Phytotoxicity” refers to

growth inhibitors or a substance’s effects on plants, and is quite

different to the notion of “toxicity”, that is the degree to which
a

substance can harm non-target species, i.e. humans or
682
animals.

7.101. With regard to spray drift, the Ecuadorian Memorial
683
relies heavily on the assert ions of the Menzie Report in its

discussion of topics such as spray operation parameters, off-

target release of spray, ground- based hostilities and localized
weather conditions. However, as with the rest of the Report, the

relevant section is essentially theoretical, divorced from any

consideration of the conditions of the border area or actual

681 EM, para. 5.18.
682 “ Phytotoxicity” in The American Heritage® Science Dictionary .
Source location: Houghton Mifflin Company. Available at:
http://dictionary.reference.com/browse/phytotoxicity (last visited 10 March
2010). For further explanations of “phytotoxicity” see:
http://www.agf.gov.bc.ca/pesticides/e_10.htm (last visited 10 March 2010).
“Toxicity” in the United States’ Environmental Protection Agency, Terms of

Environment: Glossary, Abbreviations and Acronyms . Available at:
683p://www.epa.gov/OCEPAterms/tterms.html (last visited 10 March 2010).
EM, paras. 5.83, 5.86, 5.87-5.90, 5.93-5.95, 5.97. The section of
the Menzie Report referred to is at EM, Vol. III, Annex 158, Section 4.
287exposures. It is also lade n with unfounded or misguided

assumptions. 684

(7) A N EXPERT EVALUATION OF THE M ENZIE R EPORT

7.102. Reference has already been made to the evaluation of the
Menzie Report by Dr Stuart Dobson. 685 He comments on the

“major deficiencies in methodol ogy” of the Menzie Report as

follows:

“13. The Menzie et al. (2009) report describes

itself as a hazard assessment. However, it seldom
mentions dose response. Much of the report
emphasises the severity of the hazard (the
possible) without considering whether or not the

effects are probable. It is, therefore, largely
hazard identification rather than hazard
assessment. No attempt is made by the Menzie et
al. report to assess risk.

14. Consideration of hazar d in isolation from
exposure and risk will give a misleading picture
of the real effects of the chemical on either

human health or organisms in the environment.”

7.103. Dr Dobson goes on to review the Report and the sources

on which it relies, disagreeing with most of its conclusions and

qualifying the remainder. This is done as follows:

• spray drift (paras. 22-25);

• effects on human health (paras. 26-32);

684 See above, para.4.70.
685 See above, para. 1.44.
288 • effects on organisms and ecosystems (paras. 33-
35);

• effects on amphibians (paras. 36-80);

• effects on plants and soils (paras. 81-91);

• domestic animals (paras. 92-98);

• fish (paras. 99-100);

• insects and mites (paras. 101-108).

7.104. His specific conclusions have mostly been referred to

under the relevant heads. His general conclusion is as follows:

“127. No environmental effects or effects on crop

plants would be seen more than 50 metres
downwind of the spray swath even taking a
highly precautionary approach. There should,
therefore, be no adverse ecological effects in
Ecuador.” 686

(8) C ONCLUSION FROM THE SCIENTIFIC EVIDENCE

7.105. It may be concluded that the scientific evidence lends no

support to Ecuador’s case. It should be recalled that this case

was avowedly brought because Ecuador had suffered a range of

injuries “that have undone the very fabric of life in the border
region”.687 The fabric of life may well be under threat in the

region but this is because of its poverty, remoteness,

686 Appendix – Dobson Report.
687 EM, para. 6.6, and See above, para. 1.26 for other comparable
Ecuadorian assertions.
289governmental neglect and the presence of the FARC guerrillas.

It has nothing to do with the impacts of aerial spraying.

D. The Categories of Injury Alleged by Ecuador

7.106. Against this background it is proposed to consider in

greater detail the evidence addu ced by Ecuador of injury, under

the various categories.

(1) A LLEGED INJURIES TO INDIVIDUAL LIFE AND HEALTH

7.107. Ecuador contends that the spray mix used in Colombia’s
aerial fumigations program causes serious damage to human

health. Ecuador’s contentions are unsubstantiated and are

contradicted by scientific studi es and by the contemporaneous

documentary evidence.

(a) Ecuado“rindepende nt reports” of harm

7.108. The “independent reports ” cited by Ecuador as

substantiating its claims of harm to the health of Ecuadorian

nationals are the following:

(1) Visits to the area conducted by two Ecuadorian

NGOs in early 2001 and in September 2002 and

by a consortium of Ecuadorian and Colombian
688
NGOs in July 2001;

688 EM, Vol. IV, Annex 162. See also, paras. 7.144-7.146 below.

290 (2) An investigation conducted by the Ecuadorian

Government in July 2003, the findings of which

were essentially based on oral testimonies; 689

(3) An investigation condu cted by an Ecuadorian

Congressman in late 2003; 690

(4) A declaration of an Ecuadorian doctor, prepared
691
on behalf of Ecuador for these proceedings;

(5) The April 2007 report of the Ecuadorian

Scientific Commission; 692

(6) An addendum to the Report of the U.N. Special

Rapporteur on the Rights of Indigenous People,
693
dated 28 December 2006;

(7) A preliminary note dated 4 March 2007 by the

U.N. Special Rapporte ur on the Right to
694
Health;

689 EM, Vol. IV, Annex 166 (per List of Annexes in Vol. I and Annex
cover in Vol. IV, although actual pages of Annex appear mislabelled as
Annex 165). In detailing the results of the analyses carried out following the

Verification Mission, particularly with regard to effects on agriculture, the
document concludes at p. 9: “ In light of the absence of a laboratory result
that confirms or discards whether agricultural damages are caused by the
sprayings within the so-called Plan Colombia, is worth noting that the indicia
reported by peasants are credible enough to deduce that with regard to
agriculture, sprayings are noxious to agricultural crops”. Moreover, it states:
“According to the account of a Colombian peasant who has received the
spraying directly, interviewed during the inspection, it is further deduced that

the pathological symptoms of the crops are similar to those reportedby the
690sants on the Ecuadorian side.” (Emphasis added)
EM, Vol. IV, Annex 166.
691 EM, Vol. IV, Annex 188.
692 EM, Vol. III, Annex 157.
693 EM, Vol. II, Annex 30.
291 (8) A report attributed to the United Nations High
Committee for Refugees (“UNCHR”) of

February 2007; 695

(9) A report of the U.N. Special Rapporteur on the

adverse effects of the illicit movement and
dumping of dangerous products and wastes,

dated 5 May 2007. 696

7.109. Taking these documents in tu rn, the accounts by official

Ecuadorian sources cannot be held out as being “independent”

and their probative weight and import must be assessed

accordingly. Moreover, most of these accounts pre-date the
analyses carried out by the Ec uadorian delegation in 2004 and

2005 and they are contradicted by that contemporaneous

evidence.

7.110. The Report by Congressman López Moreno is

essentially based on oral testimonies, and is as inaccurate as

other documents with equivale nt provenance. It refers
generically to “the deaths of fa mily members due to the impact

of the sprayings” without giving any details whatever. Other

items attributed to the aeria l spraying include “congenital

deformities (though it has not been established which ones)” and

“death of animals due to ingest ion of contaminated water and

694 EM, Vol. II, Annex 31.
695 EM, Vol. II, Annex 29.
696 EM, Vol. II, Annex 32; EM, para. 5.13.
292grass”. The Memorial refers to this as “an official Ecuadorian
697
government investigation”, but it is simply a statement by an

individual Congressman.

7.111. As to the declaration of Dr Dino Juan Sánchez
Quishpe, 698his recollection concer ns the same province,

Sucumbíos, and the same time period in 2004 and early 2005

during which the Ecuadorian o fficial delegation conducted

interviews with the local population in the border areas

allegedly affected by the sprayings. As will be recalled, the

Ecuadorian delegation found that “the citizens interviewed in

the zone are in good health [and] no alterations were observed in
699
plants or animals”. The Ecuadorian delegation also carried

out analyses which showed that the water and soil samples
700
collected in the area contained no glyphosate residues. Dr

Sánchez’ account is inconsiste nt with the contemporaneous
findings of the Ecuadorian delegatio n. He states as follows, in

relevant part:

“I remember in particular a period in 2004 and
early 2005 during which there were many
patients complaining of headache, vomiting and

skin problems, which coincided with border
sprayings. During these periods, patients came to
the hospital to consult with me, privately, about

697 EM, Vol. IV, Annex 167 (per List of Annexes in Vol. I and Annex
cover in Vol. IV, although actual pages of Annex appear mislabelled as

698ex 166); see EM, para. 6.16.
699 EM, Vol. IV, Annex 188.
700An85.x
See above, paras. 5.30-5.32, 5.37.
293 their severe headaches, vertigo and nausea, 701e
also reported diarrhea and skin problems.”

7.112. Ecuador’s official accounts and contemporary statements

by international organisations pa int a different picture. For

example:

• Press Bulletin of the Foreign Minister of Ecuador, N°

480 of 26 August 2004 : “They [the studies] were

carried out by the scientists of CEEA [i.e. the

Comisión Ecuatoriana de Energía Atómica

(Ecuadorian Commission on Atomic Energy)] using,

for that purpose, the EPA 547 procedure, high-tech

equipment and the analytic al standards provided for

by the company Riedel de Haën, obtaining as a result
the non-existence of glyphos ate residues in any of

the analyzed samples.” 702

• Press Bulletin of the Foreign Minister of Ecuador, N°

721 of 24 December 2004: “The special Commission

appointed by the Minister of Foreign Affairs [...] was
able to physically ascertain, on the basis of testimony

from the population and local authorities that no

spraying whatsoever t ook place affecting the

Ecuadorian territory in that sector [i.e.

Sucumbíos]”. 703

701
702 EM, Vol. IV, Annex 188.
703An81.x
Annex 82.
294 • Declaration by the represen tative of the U.N. High

Commissioner for Refugees (UNHCR) in Sucumbíos

to the Delegation on 23 December 2004 , in response
to allegations that th ere were 900 Ecuadorian

refugees due to violence on the Colombian side of

the boundary: “The UNCHR official remarked that

this was the first denunciation he received on the

subject and specified that he knew of no sprayings in

the sector – including from UNCHR in Colombia –

and that, in addition, of the Colombians seeking

refuge, only 0,5% affirmed coming to the country for
704
this reason.”

• Report of the Ecuadorian Delegation to the Foreign

Minister of Ecuador of 24 December 2004 ,

concerning a visit to Lago Agrio: “no violations of

the Ecuadorian air space have been reported nor is

there any knowledge of sprayings in the Ecuadorian-
Colombian boundary during the last weeks. [...]

Similarly we were told that the movement of

Colombian and Ecuadorian people has been

normal.” 705 Similar reports were received by the

Delegation during their visi ts to Puerto Nuevo and
706
Puerto Mestanza.

704
Reported in Foreign Ministry of Ecuador, Memorandum of 24
705ember 2004 (Annex 83).
706 Ibid.
Ibid.
295 • Report of the Ecuadorian Delegation to the Foreign

Minister of Ecuador of 29 December 2004 , referring

to the visit and collection of water and soil samples

carried out by the joint Commission in the
Ecuadorian villages facing the Colombian locations

of Azul, el Alfiladero a nd Monterrey in the Santa

Marianita sector:

“The interviewees stated that the previous
week [...] they had sighted between six

and eight helicopters and two small
aircraft leaving a trail in Colombian
territory. They also indicated that they
had no incidents to report, were carrying
out their activities as usual and that both
their animals and their crops were in good

condition. [...] In any case, even if it was
to be accepted that Colombia had
resumed aerial sprayings, the truth is that
the observations conducted and the
gathered testimony do not afford evidence

of any repercussions in Ecuadorian
territory. It was of particular utility that
important media participated in the
mission. They could verify on the ground
the reality of the facts and [will be able
to] reach, we hope, the same conclusion
707
as this report.”
• Press Bulletin of the Foreign Minister of Ecuador, N°

732 of 30 December 2004 : “[T]he citizens

interviewed in the zone are in good health, no

alterations were observed in plants or animals, as

707 Annex 84.

296 well as that, according to locals’ accounts there has

been no violation to the national airspace, nor have
any vestiges of glyphosate reached Ecuadorian

territory.”08

• Press Bulletin of the Forei gn Ministry, Nº 027 of 24

January 2005: “The result obtained by scientists of

the Ecuadorian Atomic Energy Commission, who are

also members of the Ecuadorian Scientific and
Technical Commission, from the analyses made is

that no glyphosate residues were found, the same as

on the two previous occasions when similar claims

arose.”709

7.113. Thus the situation on the ground in the Colombia-

Ecuador border at relevant times, as recognized by the

Ecuadorian authorities themselves , was the following: up until
January 2005 no glyphosate residues had been found in any of

the soil and water samples analyzed and there had been no

officially-recorded damage to hum an health, flora or fauna due

to the aerial sprayings in Ecuador’s regions adjacent to the

border with Colombia. Colombia informed Ecuador when it

resumed the sprayings in December 2006. In spite of this

notice, Ecuador was still unable to obtain any water or soil

708
709An85.x
An86.x
297samples resulting in a positive identification of glyphosate

residues.

7.114. Ecuador also relies on a report of the Ecuadorian

Scientific Commission of 27 April 2007 in support for its

assertions that the local popul ations suffered physical and
710
psychological damage. However, the scientific standing and

general reliability of this document are highly questionable. It
bases all of its purported findi ngs on erroneous assumptions as

to the composition of the mix sprayed by Colombia, 711and it

contains few references to field tests or medical analyses of the

individuals allegedly affected. 712 Such findings cannot be held

to evidence any alleged impacts of the Colombian spraying

program.

7.115. The report’s true political motivations can be seen from

the statements given to the press on the same day the report was

received by the Ecuadorian Mini ster of Foreign Affairs.

According to Press Bulletin N° 510 of the Ecuadorian Foreign

710
711 EM, para. 6.18 and Vol. III, Annex 157.
See e.g., EM, Vol. III, Annex 157, p. 21, where the conclusions are
based on the analysis of “a chemical package consisting of at least
glyphosate, POEA and Cosmo Flux 411 F, possibly made hazardous with a
microbiological additive ( Fusarium oxysporum).” (Emphasis added). See
also, ibid., pp. 29-31, where unsubstantiated and speculative assertions are
made as to the components of the sp ray mix, and their characteristics or
effects.
712instrance, ibid., p. 53, to describe what they term as the
“exposure pattern”, the Report of the Ecuadorian Scientific Commission
relies on interviews with only four individuals in Mataje, held on a single

day, where reported symptoms were mostly mild, and on other studies (i.e.,
Maldonado, 2002; Ordoñez, 2002), details of which are not provided.
298Ministry, the then Minister of Foreign Affairs, María Fernanda

Espinosa, stated: “We have also implemented a legal way,

preparing a case that Ecuador will bring against Colombia

before the International Tribunal at The Hague... [the Minister]

vetted the report, after stating that the government’s policy has
been supplemented ‘by the scientific way’. [...] President Correa

reiterates that repor t is basis for compensations sought by

Ecuador for those affected.” 713

7.116. Given its avowed purpose, this report, issued long after

Colombia had suspended the aerial sprayings in the 10-km

corridor parallel to the border wi th Ecuador in February 2007,

cannot be considered an objectiv e, still less an independent
714
scientific study.

7.117. Ecuador repeatedly relies on the position taken by the
then U.N. Special Rapporteur on the Rights of Indigenous

People in 2006. 715 Mr Stavenhagen, a distinguished Mexican

sociologist, per Ecuador’s invitation, visited that country for 10

days in April-May 2006, at a time when aerial spraying within a

10 km band on the Colombian side of the border was suspended.

He recounts what he was told but he conducted no independent

examination of the position. He does however emphasise the

713 Annex 150: Press item: “Ecuador will sue Colombia over sprayings
in the border”, Revista Caminos, 2 July 2007. See also, EM, Vol. III, Annex

714.
715 Annex 60.
EM, Vol. II, Annex 30, para. 28; the Report is cited at EM, paras.
1.15, 2.3, 2.18, 2.50, 3.61, 5.48, 6.107, 9.23, 9.62, 9.73.
299poverty of indigenous communities in the northern border

region, the lack of basic services , and the destructive effect of
716
illegal logging.

7.118. With respect to the preliminary note by the U.N. Special
Rapporteur on the Right to H ealth of 4 March 2007, and the

report of an “international observer mission” of April 2007,

neither of these documents constitutes – and, indeed, neither

purports to be – a scie ntific assessment of the effects of the

fumigations. This is expres sly recognized by the Special

Rapporteur on the Right to Heal th when he states: “[T]he

mission did not take samples or conduct laboratory tests,

because it was not a scientific mission.” It is also significant

that the Special Rapporteur did not take a final stance on the

matter but concluded as follows:

“The Special Rapporteur will carefully consider
all information received before taking a final

stance regarding the issu e of aerial spraying and
the right to the highest attainable standard of
health and before submitting his report to the
Human Rights Council on the issue.” 717

In the event the Special Rapporteur never did submit a

concluded report on the issue. 718

716 EM, Vol. II, Annex 30, paras. 27, 30, 35.
717 EM, Vol. II, Annex 31, para. 26.
718 According to the website of the Office of the High Commissioner

for Human Rights, there has been no further report on this subject by the
Special Rapporteurs. Paul Hunt was replaced by Anand Grover as from 1
August 2008 and the latter’s reports (A/HRC/11/12 and A/64/272) do not
deal with the matter of the sprayings.
3007.119. Ecuador’s Memorial also refers in passing to the report

of what it terms an “international observer mission” which

reported damage to human health in the area along the border in
early 2007. 719 This document was filed as Annex 29 of

Ecuador’s Memorial. It is not a reproduction of the original, but

a re-typed version of the first two pages of a document, without

the cover page or any indication as to its source. This document
is listed in Volume II of the Memorial as having been authored

by “UNHCR et al.” While it is impossible to ascertain the

actual provenance of this document, it is clear from its contents

that it does not concern an “i nternational observer mission” and
that it is not a UNHCR report.

7.120. According to the introductory paragraph of this

document (which was not transl ated by Ecuador), this mission

was carried out by the social conc erns ministry of the Catholic
parish of San Miguel of Sucu mbíos and of the diocese of

Mocoa-Sibundoy-Putumayo. A f ootnote indicates that the

mission was accompanied by representatives of other

organizations, including offi cials from the UNCHR in
Putumayo and Sucumbíos. UNCHR played no substantial role

in connection with this mission, and appears to have had no part

in drafting of the document. In the circumstances, Ecuador’s

characterization of this report as an “international observer

719 EM, para. 6.17, referring to Vol. II, Annex 29.

301mission” and as having been co-authored by UNHCR would

appear to be both incorrect and misleading.

7.121. In any event, judging from the very short excerpts of the
document filed by Ecuador, this report contains no reliable

information as to possible damage to human health as the

document appears to be only based on hearsay and not on

medical examinations or other sc ientific analysis. References
are merely made to anonymous reports of “symptoms very

similar to those of the flu” allegedly suffered by Colombian and

Ecuadorian children in generic areas at unspecified times, i.e. “a

few days after the fumigation passed”. Perhaps the vagueness
of the terms of this document is explained by the fact that the

mission took place at a time – from 29 January to 2 February

2007 – when the sprayings had al ready been suspended in the

relevant areas. The evidentiary weight of this document for the

purposes for which Ecuador seeks to rely upon it is nil.

7.122. The Memorial also relies on a report of Okechukwu

Ibeanu, the U.N Special Rapporteur on adverse effects on the

illicit movement and dumping of dangerous products and waste
of 5 May 2007 to support Ecuador’s allegation that: “Reports

indicate that spray mixture used by Colombia for drug

interdiction contain much more glyphosate than is typically used
720
in agricultural applications”. However, Ibeanu’s report does

720 EM, para. 5.13, citing EM Vol. II, Annex 32, para. 20.

302not refer to the specific mix us ed by Colombia in its aerial

sprayings. The report contains no reference to Colombia and its
use of herbicides to destroy drug crops. Ecuador’s reliance on

this report is misplaced.

7.123. Finally, in its Memorial 721 Ecuador relies on the

recommendations made to the National Narcotics Council 722 by

the Colombian Office of the Ombudsman following a site visit
723
conducted in Putumayo from 15 to 25 January 2001 and on a

mission carried out in April 2008 in the same region by a team
724
of Colombian NGOs investigating the human rights situation.
Neither of these documents provid es any technical or scientific

assessment of the health situati on in the relevant areas, nor do

they provide any valid findings as to the actual effects of the

fumigations on human, animal h ealth or the environment.

Rather, they recount third-party after-the-fact accounts which at

most warn of the alleged risks for human health, fauna and the

environment of the sprayed mixture.

(b) Ecuadow r’tnststements

7.124. Ecuador purports to rely on a series of witness

statements from Colombian and Ecuadorian nationals residing
in the border area, concerning alleged damages caused by the

sprayings.

721
722 EM, paras. 5.106-5.107.
723 EM, Vol. II, Annex 92.
724 EM, Vol. II, Annex 91.
EM, Vol. IV, Annex 171, and see EM, para. 5.109.
3037.125. All the statements of individuals filed by Ecuador are

anonymous. 725 Ecuador does not explain how the witnesses

were selected among the entire population settled along the

border area, or the circumstances in which the statements were

rendered, particularly in the case of witnesses supposedly
residing in Colombian territory who were selected by

Ecuadorian officials to render their statements before official

Colombian notaries public.

7.126. A striking feature of those statements is the lack of any

medical evidence or other subs tantiation in support of the
allegations of adverse impacts s upposedly caused by the aerial

sprayings. Moreover the witnes s statements are studiously

vague as to the timing of the sp rayings and locations where the

sprayings allegedly took place, making it impossible to check

the assertions against the documented record of spray missions.

Such statements can hardly be verified or falsified – unless, as
with members of the Mestanza family, it is possible to show that

at relevant times they lived somewhere else and that their
726
affidavits were concocted.

7.127. On several occasions, the Court has expressed
reservations as to th e evidentiary value of witness testimony of

725 See para. 1.38, note 100 above.
726 See above, para. 1.37.
304the kind produced by Ecuador in these proceedings. For

instance, in the Nicaragua v. Honduras case, the Court said:

“The Court notes, as to that latter category of
evidence, that witness statements produced in the
form of affidavits should be treated with caution.
In assessing such affidavits the Court must take
into account a number of factors. These would

include whether they were made by State
officials or by private pe rsons not interested in
the outcome of the proceedings and whether a
particular affidavit attests to the existence of facts
or represents only an opinion as regards certain
events. The Court notes that in some cases
evidence which is contemporaneous with the

period concerned may be of special value.
Affidavits sworn later by a State official for
purposes of litigation as to earlier facts will carry
less weight than affidavits sworn at the time
when the relevant facts occurred. In other
circumstances, where there would have been no
reason for private persons to offer testimony

earlier, affidavits prepared even for the purposes
of litigation will be scrutinized by the Court both
to see whether what has been testified to has been
influenced by those taking the deposition and for
the utility of what is said. Thus, the Court will
not find it inappropriate as such to receive
affidavits produced for the purposes of a

litigation if they attest to personal knowledge of
facts by a particular indi vidual. The Court will
also take into account a witness’s capacity to
attest to certain facts, for example, a statement of
a competent governmental official with regard to

305 the boundary lines may have greater weight727an
sworn statements of a private person.”

In the present case, the witness statements are given by persons
interested in the outcome, long after the event, and were

prepared for purposes of litig ation. Unless independently

corroborated, they are entitled to no weight, notably insofar as

they purport to express any opinion as to causation.

7.128. Moreover, the witness statements submitted by Ecuador

in the present case are not only uncorroborated by any
contemporaneous evidence, they are also contradicted by other

contemporaneous evidence emanating from Ecuadorian sources.

To the extent that a time fram e can be identified, the vast

majority of the witness statemen ts concern spraying episodes in

the years 2000-2002, i.e., prior to the Ecuadorian official reports

of June, August and December 2004 and January 2005,

announcing that, through field vis its and laboratory analyses
performed in Ecuadorian terr itory by that country’s own

authorities, no effects were eviden ced in Ecuador as a result of

the aerial spraying in Colombia. 728 This makes it even less

likely that the sprayings could have caused the extreme effects

of which some of Ecuador’s witnesses complain.

727 Case concerning Territorial and Maritime Dispute between
Nicaragua and Honduras in the Caribbean Sea (Nicaragua v. Honduras) ,
Judgment of 8 October 2007, para. 244.
728 See above, paras. 5.30-5.32, 5.37.
3067.129. The witness statements submitted by Ecuador are also

inconsistent with statements and communiqués issued by the

Ecuadorian authorities themselves at the relevant time when the

complaints arose. Such statements – rendered by

representatives of the Ecuadoria n Government with respect to

the relevant boundary region in Ecuadorian territory – are in
direct contradiction with the position adopted by Ecuador in

these proceedings. 729 Similar statements have been held by the

Court to amount to admissions agai nst the interests of the party

that made them. As observed by the Court in Nicaragua v.

United States of America:

“The Court takes the view that statements of this
kind, emanating from high-ranking official

political figures, sometimes indeed of the highest
rank, are of particular pr obative value when they
acknowledge facts or conduc t unfavorable to the
State represented by the person who made them.
They may be construed as a form of
730
admission.”

7.130. Ecuador’s witness statements carry even less probative
weight due to the fact that they were prepared in 2009, i.e., years

after the alleged facts occurred, and were prepared for the

purposes of this case. As noted above, in Nicaragua v.

Honduras the Court, in examining affidavits rendered by State

officials for purposes of litigation, noted that:

729 See above, paras. 5.31-5.32, 5.37-5.38, 5.45-5.61, 7.1-7.6, 7.112-
7.113.
730 Military and Paramilitary Activities in and against Nicaragua
(Nicaragua v. United States of America), Judgment, I.C.J. Reports 1p.6,
41, para. 64
307 “Affidavits sworn later by a State official for
purposes of litigation as to earlier facts will carry
less weight than affidavits sworn at the time
when the relevant facts occurred. In other

circumstances, where there would have been no
reason for private persons to offer testimony
earlier, affidavits prepared even for the purposes
of litigation will be scrutinized by the Court both
to see whether what has been testified to has been

influenced by those taking th731eposition and for
the utility of what is said.”

7.131. For these reasons, Ecuador’s witness statements should

not be accorded any probative value. Moreover, a more detailed

analysis of these statements s hows that they are riddled with

inconsistencies.

7.132. The witnesses residing in Sucumbíos province are

Witnesses 2, 3, 5, 9, 20, 22 and 23. Witness 2 732 states that he is

domiciled in the city of Nuev a Loja, like all other witnesses

from this area. He testifies that he lived from agriculture and

that he suffered twice on account of the sprayings. The witness

provides only an approximate te mporal reference for aerial
sprayings with respect to the first instance of drifts from

sprayings which are said to have occurred “about seven or eight

years ago”, i.e. in about 2001 or 2002, given that the statement

was made on 16 January 2009. No time frame is given for the

731 Territorial and Maritime Dispute between Nicaragua and Honduras
in the Caribbean Sea (Nicaragua v. Honduras), I.C.J. Judgment of 8 October
2007, para. 244.
732 EM, Vol. IV, Annex 190.
308second instance of spraying, which is laconically described as:

“when the sprayings returned”. The symptoms which the

witness testifies he experien ced – bumps all over the body,

headaches and dizziness – are sa id to have appeared “a few

weeks after the planes came by spraying”. In fact the temporary

symptoms of exposure to the sp ray mixture, e.g., eye irritation,
appear shortly after exposure, and not “a few weeks after”; 733

they disappear within a day or so.

7.133. The account of Witness 2 is vague in the extreme, refers

to no specific time-frame and does not attempt to show any

direct link of causality between the specific alleged sprayings
and the harm allegedly suffere d. Furthermore, no medical

reports are filed in support of the testimony as to the symptoms

allegedly suffered and there is no evidence that medical

assistance was ever sought.

734
7.134. The testimony of Witness 3 is similarly unhelpful.
Again, it only provides a vague te mporal reference (“seven or

eight years ago”) for the time wh en the witness apparently saw

planes fly over the San Miguel river and turn around to go back

to Colombia, allegedly causing the chemicals to drift onto

Ecuadorian territory. The symptoms of which complaint is
made – which are likewise said to have appeared two weeks

after the sprayings – are also sim ilar. As in the case of Witness

733 See e.g., Annex 116, CICAD I, pp. 10-11, 52-53, 76.
734 EM, para. 6.27; EM, Vol. IV, Annex 191.
3092, no medical reports are appende d and no mention is made of
any medical consultation having be en sought at the time or, for

that matter, at any time. No attempt is made to substantiate a

causal link between the alleged spraying and the harm allegedly

suffered.

7.135. Witness 5 alleges in 2009 that 8 years earlier, in 2001

(no further specification is provi ded) Colombian planes entered
Ecuadorian territory over Salinas and that the sprayings caused

“an epidemic”. 735 He also contends th at, since these sprayings,

his land “has remained affected and has low productivity”. This

statement is in direct contradi ction with the te sts conducted by

Ecuadorian authorities upon water and soil samples in the

general region, which showed no traces of glyphosate.

7.136. The same criticisms apply to the testimony of Witness 9

who states that “the soil is damaged and no longer produces as

before the sprayings”. Witness 9 also claims that the local

schools were forced to close temp orarily due to the fact that a

number of children were sick for two weeks due to the
736
sprayings. According to this witne ss the schools were closed

apparently following a decision taken by the teachers, without
the intervention or involvement of the medical authorities.

735 EM, Vol. IV, Annex 193.
736 Ibid., Vol. IV, Annex 197.
3107.137. Witnesses 20 and 23 recoun t similar experiences:

sprayings allegedly occurred dire ctly over Ecuadorian territory
or spray mixture drifted with the wind, the occurrence of

symptoms and damage to plants and animals that continued for

a period of time. 737 They however claim that in their case the

symptoms were immediate. Witness 22 has a very vague

recollection of the timing of th e sprayings, “since the year two-

thousand and two, or thereabouts” and claims that for years
738
nothing was harvested due to the sprayings. All of this has

nothing to do with the characteri stics of glyphosate, which is

eliminated from soils within, at most, a few weeks, and which
allows almost immediate replanting. 739

7.138. It is to be emphasized that absolutely no evidence is

provided to support the very se rious accusations made by these

witnesses, particularly in light of the fact that, as explained in

Chapter 4, the components of th e spray mix are scientifically

classified as non bio-accumulating, i.e., as having minimal and

non-lasting effects. 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 s ought medical assistance, whether

for themselves or for their children.

737
738 EM, Vol. IV, Annexes 206 and 209.
739 Ibid., Vol. IV, Annex 209.
See Annex 116, CICAD I, pp. 22-23, 47, 78.
3117.139. The picture is very similar for the witnesses residing in

the Esmeraldas province around th e town of Mataje: witnesses
740
30, 32, 33, 34, 36, 37, 38 and 39. To the extent that a time

frame can be ascertained, the testimonies concern spraying
episodes dating back to the years between 2000 to 2004.

Witness 32 in particular refers to three episodes in increasingly

uncertain terms: the first episode in 2000, the second “a few

years later” and the third “I th ink it was in 2004”. Witness 31
(whose statement is not referred to at all in the Memorial) does

not provide any date at all for the sprayings.

7.140. In some of these cases, the illnesses alleged were
supposedly reported to health centres or nurses and medical

assistance was apparently received. However, no

documentation is provided in this regard, and in the absence of

any medical reports and specific dates that can be matched to

actual spraying episodes, it is impossible to establish what
caused the symptoms reported or whether there is any

connection with the aerial sprayings.

7.141. The same comments apply to the statements of the other
witnesses annexed to Ecuador’s Counter-Memorial, including

the statements of Colombian nationals: to the extent that a

temporal framework is provided at all, it is done in a vague and

imprecise manner, no contempor ary medical records or other

740 EM, Vol. IV, Annexes 214, 216, 217, 218, 219, 220, 221 and 222.

312scientific evidence are referred to or produced, and no attempt is

made to substantiate the connection between specific instances

of spraying and the injuries of which complaint is made.

7.142. In the absence of any direct studies or medical records

supporting the allegations contai ned in the testimony of the

various witnesses, Ecuador is forced to look elsewhere for

substantiation. In th is regard, it relies heavily on a 2001 field
study report which is wholly unrelated to the witness statements

filed with Ecuador’s Memorial. 741 That report – prepared in

2001 by a Commission composed of representatives of

Ecuadorian social organizati ons and NGOs – mentions
complaints of alleged advers e impact of the fumigations

provided orally to the member s of the Commission by some

residents in the border areas in the provinces of Esmeraldas,

Carchi and Sucumbíos.

7.143. The complaints recorded are very similar to those made

by the witnesses whose statements have been put forward by

Ecuador and they suffer from the same flaws and shortcomings:

they are not corroborated by any medical or scientific account or
evidence, they refer to no specifi c instances of spraying, they

provide no details as to the localities where the sprayings

allegedly occurred and they offer no evidence of a causal link –

741 EM, para. 6.14; Vol. IV, Annex 162.

313or indeed any connection – between the sprayings and the

symptoms allegedly experienced by the local population.

7.144. Similarly, the interviews of the hospital staff at Mataje,

in the Esmeraldas region, conducted in 2001 by the Commission

and cited in the Report, show that no connection between the
sprayings and the symptoms complained of was ever

established.742 This is particularly significant since the

interviews concern the same re gion and were carried out in

2001, and therefore correspond to one of the time periods when

Ecuador’s witnesses claim that they suffered health problems

due to spraying.

7.145. Some of the doctors interviewed stressed the lack of

infrastructure and medical personne l in the area which made it

impossible to ascertain what had caused the ailments suffered by

the populations of the border areas. Furthermore, the doctors

interviewed also emphasized that no investigations of the
symptoms reported had been conducted on the ground. The

interviews, in Ecuador’s English translation, 743 speak for

themselves:

“We have no reports of people being poisoned by

the fumigations; we do not know what happened
in Mataje. We received reports from volunteer
collaborators indicating the presence of ARI
(Acute Respiratory Infec tion), but they have not

742 See paras. 7.144-7.146, above.
743 EM, Vol. IV, Annex 162.
314told us if it is on account of the sprayings. An
investigation should be carried out”. (Dr. Dueñas,
Epidemiologist)

“In October, we had a large number of children
with conjunctivitis, skin lesions, respiratory
problems and diarrhea. This was attributed to the
insecticides from the palm plantations, but we do
not have anything available here to work with or

investigate. We have only three attending
doctors and we had to br ing in the rural doctors
to assist at the hospital because they were not
being fully utilized in th eir locations. There are
brigades that go on visits , but they did not go
there.” (Dr. Merino)

“In the case of an epidemiological emergency,
the auxiliaries notify us of changes in health
circumstances, and we go take samples. The unit
provides attention based on its means; it is an

acute care hospital that cannot keep patients for
more than two days. When we learned of what
had happened, we went, but we could not take
samples. We reported it, but that was as far as it
went, no one came to investigate, not even the
MAG, which is in charge of monitoring the palm
plantations. The Red Cross came and took
samples, and also inves tigators from television

channels.” (Dr. Dueñas)
“Our epidemiological monitoring system does
not allow us to follow up on all of this. We have

a laboratory for basic analysis and in Mataje we
do not have the means to replace the light bulb of
the microscope, which burned out months ago.
Moreover, there is a serious lack of personnel.
The doctors who reside here are from here, but
those who come here have gone to Colombia, to
Tumaco, where they earn US$1,200 a month.”

(Dr. Merino)

315 “Rounds are done every two months in order to
provide medical attention to the population, but
they are not very effective. Normally, there is a
doctor (for vaccinations), a dental surgeon (only

extractions), and an obstetrician and assistant, but
our primary health care rates are very low”. (Dr.
Dueñas)

7.146. A number of important conc lusions can be drawn from
these interviews:

• The medical staff working in hospitals located in the
Esmeraldas region in 2001 st ated that they had “no

reports of people being poisoned by the

fumigations”;

• Moreover, while they had reports of acute respiratory

problems, the doctors acknowledged that they had no

reason to believe that the health problems allegedly
occurring in the area at the time were due to the

sprayings and in fact attributed some of the

symptoms to the effects of insecticides used in
spraying palm plantations;

• No samples were taken due to lack of resources and
no investigations were carried out at the time;

• Basic health care in the border areas was inadequate;
there was a lack of personnel and very limited

means;

316 • There is every reason to believe that the same

situation – shortage of human and financial resources

and proper medical care, lack of adequate

investigations – applied in subsequent years and still

exists today.

7.147. Whether for these or other reasons, Ecuador does not

seem to credit its own witnesses’ allegations concerning human

deaths. As discussed in Chap ter 8, it does not rely on the

alleged deaths as constituting a violation of right to life of

individuals in the border regions, and only refers to the deaths as

part of its claim in relation to a lleged violations of the right to

health. The reason seems clear. First, as shown above, the

scientific evidence demonstrates that Colombia’s aerial

eradication program poses no signi ficant risks to human health.

Moreover in the section of the Memorial dealing with “The
Harm to People”, 744 references to deaths in the border region are

curiously absent: the only commen t is that “there are a number

of reports of deaths among young children following early spray

events in particular”, 745 and the section as a whole confines

itself to alleging “serious harm to human health”, 746an

unwontedly restrained phrase if indeed there were as many
747
deaths as the witness statements would suggest. The truth is

744 EM, paras. 6.7-6.53.
745
746 EM, para. 6.50.
747 EM, para. 6.7.
See above, para. 1.28. Elsewhere Ecuador is less restrained, but in a
way which demonstrates its lack of knowledge of the facts: e.g. EM, para.
317that, as opposed to the very de tailed figures it provides for
748
alleged events of fish and farm animal deaths, Ecuador does
not know the numbers or the causes of any of the deaths alleged

by its own witnesses, a gap which it tries to fill by promising to

provide detailed information at the quantum phase, once

Colombia’s responsibility is definitively established. 749 But in

the context of responsibility fo r incidental transboundary harm,

injury is the very essence of the wrongful act, and proof of

injury (that is, of damage caused by the Respondent State)
750
cannot be postponed in this way.

7.148. In the circumstances, given the poor livi ng and health

conditions of the Ecuadorian population in the Colombia-

Ecuador border areas, coupled wi th the problems relating to

medical care and the lack of any investigation or other scientific

assessment of the reported complaints, Ecuador has fallen far

short of discharging the burden upon it of establishing a causal

link between the aerial fumiga tions and injuries allegedly

suffered by Ecuadorian residents. Ecuador’s complaints of
serious injury to health are not substantiated.

10.30 (“There have been deaths of numerous young children connected to the
spraying episodes”). The adjective “n umerous” is telling: Ecuador does not
know how many deaths its witnesses allege to have occurred as a result of
748ay drift.
749 EM Vol. I, para. 6.84, referring to Vol. II, Annex 165.
750 See e.g., EM, paras. 10.3, 10.4, 10.20.
See above, paras. 7.8-7.9; also, further below, paras. 8.37, 8.49,
8.122 (5).
318 (2) A NIMALS

7.149. The same considerations apply, mutatis mutandis, to
Ecuador’s claims of illnesses and death of animals, both

domestic and wild. Even direct overspray could not produce the

kind of injuries asserted, ma ny years later, by Ecuador’s

witnesses.

7.150. In support of this claim that the spray mixture can cause

significant risks to domestic animals, Ecuador cites the
following US warning label for Roundup: “DOMESTIC

ANIMALS… ingestion of this product or large amounts of

freshly sprayed vegetation may result in temporary
751
gastrointestinal i rritation (vomiting, diar rhea, colic, etc.).”

This worst-case scenario, appropriate no doubt for a label, bears

no relationship to Ecuador’s claims . All the label predicts is
“temporary gastrointestinal irri tation”. By contrast, what

Ecuador asserts (at least so far as animals are concerned) is

death.

7.151. Ecuador relies on a report prepared by the NGO Acción

Ecologica of October 2002 for data of animal deaths for

Sucumbíos alone and limited to the year 2001 as a precise
11,828, including 188 cows and 43 horses. 752 On what basis

188 cows and 43 horses in Sucumbíos in 2001 came to eat the

751 EM, para. 5.57.
752 EM, para. 6.80, relying on an NGO report (Vol. IV, Annex 165).
319necessary vast quantities of fres hly sprayed vegetation is quite

unclear and is not explained. What is however crystal clear

from the 2002 report is that the data it contains is not actual data

of animal deaths but rather corr esponds to the claims made by

campesinos from different Ecuadorian communities to the

Office of Human Rights of the Ombudsman of Lago Agrio in
753
2001. The report acknowledges that the data was never

independently verified , since it states that: “no government
official has traveled to the areas to check, in the field, the

damage reported in this claim.” 754 The allegations of the

campesinos remain wholly unsubstantiated.

7.152. It is instructive to make a tally of animal victims from

the numbered witness statements. The claims of injuries and

death range from livestock and fa rm animals such as pigs and

chickens to wild fish, birds and monkeys, which are said to have
755
disappeared as a result of the sprayings near the border.

Specific numbers are not provided, but it is apparent that the

statements wish to convey the image of large numbers of

animals killed or seriously in jured – “a heavy toll on animals,
756
both domestic and wild”.

753 EM, Vol. IV, Annex 165, p. 3.
754Ibid.
755
756 EM, paras. 6.88-6.105.
EM, para. 6.82.
3207.153. Illustrative examples include the following: “I found

several dead chickens near the tree”; 757“[I]n recent years there

has been a decline in some species such as the monkeys and

guatuzas, … before there were many of them and now there are

hardly any”; 758 “I had forty chickens and nearly all of them

died” 759; “[W]e used to see a lot of monkeys and parrots…

760
Now, one seldom sees a monkey or a parrot”. Someofthe

witnesses testify that fish – bot h wild and farm-raised – were

particularly affected by the spra yings and state that they saw
761
dead fish floating on the surface of the San Miguel River.

“The wild birds have disappear ed, and they can no longer be
762
heard singing in the mornings or the evenings”.

7.154. A complaint written by Mr. Victor Mestanza to the

Ecuadorian Ministry for the Environment on 14 October 2002 is

more specific as to the number of animals allegedly affected.

Mr. Mestanza alleges that he su ffered a huge financial loss due

to the fumigations and states th at he lost 30,000 fish in 2000,

60,000 fish in January 2002 and 400 ducks and 80,000 fish in
763
September 2002. However, quite aside from the fact that the

scientific evidence shows that the mixture used for the sprayings

757 Witness 4, EM, Vol. IV, Annex 192.
758 Witness 3, EM, Vol. IV, Annex 191.
759
760 Witness 9, EM, Vol. IV, Annex 197.
761 Witness 2, EM, Vol. IV, Annex 190.
EM, paras. 6.93-6.94 and 6.99-6.100.
762 EM, para. 6.87, citing Vol. IV, Annex 166 (President of the
community of Monterrey).
763 EM, Vol. IV, Annex 237.
321causes no significant effects on fa una in general and little
764
chronic toxicity to fish in particular and the vast number of

fish claimed for, Mr. Mestanza’s testimony is questionable.

7.155. Victor Mestanza is one of the individual plaintiffs in the
Dyncorp case before the U.S. courts and a member of the same

Mestanza family whose personal injury claims were found to be

fabricated, with the result that three of them were dismissed

with prejudice from the case. 765 In spite of the fact that in the

questionnaire submitted in the U.S. proceedings Mr. Victor

Mestanza claimed that he incurred over $600,000 in damages to

his property, he has produced no business or farming records to
766
support his claim.

7.156. With respect to the other testimonies reporting harms to

animals as a result of the aerial sprayings, there is no attempt, as
was the case with alleged injuries to human health, to establish a

causal link between the incident s reported and any specific

sprayings and no independent verification of these allegations is

made. The statements consist of mere assertions uncorroborated

by any contemporaneous scientific or documentary evidence.

764 See above, paras. 7.72-7.83.
765
766 See EM, Vol. IV, Annexes 153, 154. See above, para. 1.37.
See Annex 156: Defendants’ Motion for Sanctions Against the
Arias/Quinteros Plaintiffs for Violatio ns of Discovery Orders, 26 January
2010, p. 24.
3227.157. Ecuador further relies on th e Menzie Report’s statement

that “[s]praying with glyphosat e-based herbicides may also
reduce the local food supply for domesticated animals, which

may lead to decreased body condition and performance in

livestock and other animals.” 767 Itmaybenotedwhatthe

Menzie Report does not say, viz., that the spray mix can cause

the deaths of cows and horses. But even its more modest

conclusion is directly contradi cted by the scientific studies

concerning the effects of the spray mixture on fauna.

7.158. As noted by Dr Dobson, empirical studies on human
volunteers show that skin ex posure to the concentrated

formulation of glyphosate cause at most irritation and are “less

irritant than a standard liquid dish washing detergent and a

general all purpose cleaner.” 768 Dr Dobson’s conclusion is that:

“The likely effects in domestic animals following exposure to

spray drift are the same as those seen in humans: transient eye

irritation and respiratory discom fort. I do not accept that these

mild to moderate effects would impinge on growth or
reproductive performance of these animals.” 769

(3) L AWFUL CROPS

7.159. An illustration of the confused claims made with respect

to crops in Ecuador is the statement of Witness 1, who claimed:

767
768 EM, Vol. III, Annex 158, Executive Summary.
769 Appendix – Dobson Report, p. 24.
Ibid., p. 25.
323 “Before the sprayings began in our area, I used to

sell a lot of coffee. I had sixteen (16) hectares
planted with coffee, and each hectare produced
sixty (60) to ei ghty (80) quintals of the product

annually. Now, I can barely harvest e770t quintals
of coffee per hectare each year.”

7.160. Other witnesses make similar claims. They allege that

entire plantations were destroye d, trees and plants died and

pastures were lost. 771 They also contend that the devastation

continued months or years after the sprayings. 772 Mr. Mestanza

claims that the sprayings are “practically liquidating my project

and my finances, and, as a result the source of employment for

many people in the area who work at my farm.” 773 ButMr.

Mestanza has so far produced no business or farming records to

support his very extensive claim for property damage in the U.S.

proceedings against Dyncorp. 774

7.161. As demonstrated above, spray drift could not produce

the damage alleged by Ecuador and its witnesses. Neither could

direct overspray. Land sprayed with glyphosate is able to be

replanted within days, or at most a few weeks, since the

pesticide rapidly breaks down and does not bioaccumulate. 775

The obvious reason why a harves t might drop by 80% or more

770
771 EM, para. 6.63; Vol. IV, Annex 189.
772 EM, Vol. IV, Witness 21, Annex 207 and Witness 8, Annex 196.
EM, Vol. IV, Witness 37, Annex 220 and Witness 32, Annex 216.
773 EM, para. 6.68; Vol. IV, Annex 237.
774 See Annex 156, p. 24.
775 See e.g. Annex 116, CICAD I, pp. 20-23, for the short half-life of
glyphosate in tropical soils.

324over some years is exhaus tion of the soil, a common
phenomenon in tropical soils and one of the factors underlying

“slash and burn” agriculture.

7.162. Ecuador relies on the re port prepared by the NGO

Acción Ecologica in October 2002 for data concerning the
776
alleged damages to agricultural crops. The data reported by

Ecuador for instance refers to 1,215 hectares of coffee, 785
hectares of grass and 185 hect ares of bananas allegedly

damaged due to the sprayings in 2001. However, as in the case

of alleged harm to animals, these figures do not concern verified

losses, but correspond to the clai ms filed by some local farmers

with the Office of the Huma n Rights Ombudsman in Lago

Agrio, claims that had not been officially checked on the
777
ground.

7.163. Ecuador also refers to the observations made by the

International Federation of Human Rights, several NGOs and

the Defensoría Nacional del Pueblo during a visit to Sucumbíos

in 2005 as stating, in relevant pa rt, that “All those living along

the border agree that the sprayings have weakened the soil
778
quality and its production capacity.” This is another example
of Ecuador’s selective interp retation of the documentary

evidence: the same document, in a passage from the conclusions

776 EM, para. 6.56, relying on Annex 165, p. 3 (Vol. IV).
777 EM, Vol. IV, Annex 165, p. 3.
778 EM, Vol. IV, Annex 169, p. 13.
325not translated by Ecuador, while maintaining that there had been

a change in the quality and its production capacity of the soil,

acknowledges that: “there is no agreement about the harmful

effects of the components of the sprayings announced by
779
Colombia.” With respect to alleged damages to health, the

document also admits as follows:

“From 2003 until today we have detected some
40 patients a year for intoxications due to

pesticides, but the cause, (domestic use or
fumigations) has not been established.” 780

The same reasoning a fortiori must apply to plants and crops:

any alleged damage could very well have been caused by other

pesticides and by contamination from other sources, including,

notably, the agrochemicals used for growing coca.

(4) N ATIVE FLORA AND THE ENVIRONMENT

7.164. As shown above, scientific studies have determined that

glyphosate has no long-lasting effect on soils. 781 Moreover, any

effects are negligible beyond 50 metres downwind and there is

no effect at all if the wind direc tion is away from the area of

concern. 782 In the circumstances, there could have been no

significant effects for non-targ eted crops, flora and the

environment of Ecuador as a result of accidental drift.

779 EM, Vol. IV, Annex 169, p. 23.
780 Ibid., pp. 8-9.
781
782 See above, paras. 7.84-7.88.
Annex 131-B, CICAD II, at p. 929.
3267.165. Ecuador repeatedly alle ges that glyphosate is “a
powerful herbicide specifically designed to kill all plants upon

contact (even in very small quantities)”, 783and asserts that they

kill “virtually every plant they touch”. 784 Ecuador argues that

the fact that the spray mixture may have been released at a

distance from the plant is immaterial and contends – on the basis

of the warning written on the Roundup Pro label – that “even

minute quantities of this pr oduct [Roundup] can cause severe
damage or destruction to crops, plants.” 785 This is incorrect and

misleading. It confuses hazard (the potential to cause harm) with

risk (the probability that harm actually occurs). Classification

and labelling worldwide is hazard-based and takes no account of

risk.

7.166. Of course glyphosate is an he rbicide and as such it is

supposed to kill plants; however, whether it kills a particular

plant or not depends on the administered dose (application rate).

Glyphosate kills after being absorbed into and transferred

through the plant and requires expo sed green parts of the plant

to be available so that it can be absorbed and transferred. At

high application rates, there is sufficient herbicide to kill the
plant whereas at lower applicat ion rates, photosynthesis is

inhibited and growth is reduced. When applied at lower rates,

glyphosate is a plant-growth regulator; for example, it is used in

783 EM, para. 1.17.
784 EM, para. 5.73.
785 EM, para. 6.79.
327 786
small quantities to ripen sugar cane. Moreover, glyphosate

has no lasting effect on soil, and plants that are planted
787
following application will grow normally.

7.167. If the implication is that gl yphosate is so potent that a

single drop drifting across the border into Ecuador would kill a

plant, this is an absurd allegati on. The application rate falls off

very rapidly with distance downw ind of the spray swath and no

sub-lethal, let alone lethal, effect s on plants are expected more
788
than 50 metres from the spray swath. Onlyaverysmall

proportion of the spray drifts any great distance and this

proportion represents very small droplets – the spray application

is set up to maximise the droplet sizestoensuremostofthe

applied herbicide drops immediat ely. These fine droplets that
drift represent a tiny proportion of the applied dose and a very

low application rate.

7.168. Ecuador argues that the fact that the spray mixture may

have been released at a distance from the plant is immaterial and

contends – on the basis of th e warning written on the Roundup

786
See, for instance: http://w ww.washingtonpost.com/wp-
srv/liveonline/01/world/major_011201.htm (last visited 10 March 2010).
787 United States Department of Agriculture, “Effects of glyphosate on
soil microbial communities and its mineralization in a Mississippi soil”,
published in the peer-reviewed journal Pest Management Science , in 2007,
Abstract. Available at:
http://www.ars.usda.gov/SP2UserFiles/Place/64022000/Publications/Weaver
/Weaveretal07PMS63.pdf (last visited 10 March 2010)
788
See Figure 1 of the Appendix – Dobson Report, and the
accompanying analysis.
328Pro label – that “even minute quantities of this product

[Roundup] can cause severe damage or destruction to crops,
plants.”789 This is incorrect for several reasons.

7.169. First of all, the Roundup Pro warning concerns that

particular product, which is not used in Colombia’s program for

the eradication of illicit crops. Furthermore, it is the application

rate of the herbicide that dete rmines whether plants will be
790
killed or not. The application rate fa lls off very rapidly with

distance downwind of the spray swath and no sub-lethal, let

alone lethal, effects on plants are expected more than 50 metres
from the spray swath. 791 Only a very small proportion of the

spray drifts any great distance and this proportion represents

very small droplets – the spray application is set up to maximise

the droplet sizes to ensure most of the applied herbicide drops

immediately. These fine droplet s that drift represent a tiny

proportion of the applied dose and a very low application rate.

th
7.170. The 19 Verification Mission (2 009) reported an off-
target average rate of 2.23%, w ith lower figures for Putumayo

(1.04%) and Nariño (1.09%):

“Drift was estimated with the Off-Target (OT)
indicator, that calcula tes the percentage of
vegetation cover other than coca that was

789
790 EM, para. 6.79.
791 See Appendix – Dobson Report, Section 6.2, paras. 81-91.
See Figure 1 of the Appendix – Dobson Report, and the
accompanying analysis.
329 affected with regard to the coca plot subject to
control. The average obtained in the verification

mission was of 2.23%. This indicator allows for
concluding that for every sprayed coca hectare,
0,022 hectares of a different [vegetation] cover
were affected. This range falls within what is
considered as acceptable in Record [1] of the
Environmental Management Plan that foresees a
792
range of 5.”
Virtually all of this overspray would have been deposited within

tens of metres of the intended spray area.

7.171. Colombia has also adopted a rigorous methodology in
the implementation of the PECIG program, precisely in order to

avoid drift and minimize the risks to health and the

environment. Colombia has produced as Annex 67 of this

Counter-Memorial, a report by th e Anti-Narcotics Direction of
the Colombian National Police (“DIRAN”) which explains the

carefully monitored process governing the spraying operations.

7.172. The first step consists in th e identification of the illicit

crops through satellite imagery. Once the clusters of coca crops
are identified, the sprayings ta ke place under strict observance

of safety strips and special management areas and the process is

carried out in separate stage and in the respect of detailed

guidelines. All spray aircraft are equipped with a satellite
monitoring system which gua rantees the accuracy of the

operations and ensures that the sp rayings are carried out on the

792 Appendix 1 to the Environment Ministry Report, Annex 70, p. 9.

330areas targeted. Wind conditions are constantly monitored by the

aircraft and if they are not with in the parameters allowed, the
793
mission is annulled or postponed. Each operation is recorded

in detail, including the place, hour , number of hectares sprayed,

spraying locations, etc. The am ount of products used in the

operation (glyphosate and adjuvant ) is also stated in the
794
operational records.

7.173. The operational parameters of the aerial spraying

program were outlined in Chapter 4. They result from the

various field tests conducted by the Colombian Agriculture and

Livestock Institute (“ICA”) in order to ensure the effectiveness

of the Program while preserving the environment. 795 ICAalso

performs chemical analyses of both the glyphosate commercial

formulation and the mix used in the sprayings in order to assess

the proper concentration of pure ac tive ingredient used therein.

Routine unannounced visits are c onducted to the operation base
796
and samples of the mix are taken to be analyzed.

7.174. The collateral effects, if any, of the sprayings on

vegetation and lawful crops adjacen t to the illicit crops sprayed

are verified through a specific procedure which involves the

cooperation of several specialized agencies. As part of this

793 Annex 67, p. 10.
794 See above, para. 4.64.
795 Annex 67, p. 7. See also, EM, Vol. II, Annex 15, Tenth Article and
Part No. 1 (Program for Management of the Fumigation, second para.); CCM

796ex 50, Table 1, Operational Parameters.
Annex 65, p. 3; and see above, para. 4.24.
331process, DIRAN also reviews the complaints filed by

individuals for alleged damages to lawful crops during the aerial
797
spraying operations.

7.175. The results of analyses conducted on Colombian
territory by the Colombian Ministry for the Environment, the

agency charged with supervising the implementation of the

Environmental Management Plan (“EMP”), confirm that there

have been no effects on soil and water samples taken from areas

of Colombian territory that were directly sprayed. As stated by
the Ministry:

“The Ministry for the Environment, Housing and
Territorial Development... as supervisor of the
implementation of the Environmental
Management Plan of the Program for the
Eradication of Illicit Crops by aerial spraying

with Glyphosate Herbicide – PECIG, verifies and
controls that the mixture used in the eradication
of illicit crops, Gly-41, Cosmo-Flux 411F, and
water, complies with the technical specifications
stipulated under Reso lution 099 of 2003, which

relies upon the field studi es carried out by the
Colombian Agriculture and Livestock Institute –
ICA , and that the PECIG program has followed
the guidelines of the environmental regulations in
force on the matter and has been subjected to

adequate and continuous environmental controls.
The results of environmental monitoring
activities conducted as part of the EMP –
including the laboratory analyses of the water
and soil samples collected in monitored areas –,

have shown that the implementation of the

797An67.x
332 PECIG to date does not pose a risk to the
environment or to human health, including that
of the persons involved in the environmental
monitoring who, on account of their tasks, are

occasionally directly exposed to the spray
mixture. None of them has reported adverse
effects on their health as a result of such
798
exposure.”

7.176. For these reasons, Ecuador’s contentions of extensive

harm to crops and native plan ts due to spray drift onto its
territory have no basis. As seen above, stri ctly monitored

procedures to preserve the environment and minimize the risks

of drift beyond the illicit crops sprayed are used, buffer zones in

relation to all water bodies are respected and routine analyses on

water and soil samples taken from areas directly targeted by the
799
sprayings are conducted. The results of these analyses and

environmental controls show no risk for human health or the

environment. In the circumstances, Ecuador’s allegations are
groundless.

(5) T HE “SPECIAL HARM ” TO THE INDIGENOUS COMMUNITIES

7.177. Ecuador also creates a separate category of harms,

relating to its complaint that the aerial sprayings severely

disrupted the cultural well-being and natural environment of the

indigenous people living in bo rder regions of Ecuador.

798
799An70.x
Examples of soil and water analyses are at Appendices 2-5 of the
Colombian Environment Ministry Report, Annex 70.
333According to Ecuador, these communities have suffered “special

harm” due to the impact of th e sprayings on nature and the

environment, on which they rely for their cultural traditions and
800
religious beliefs. Ecuador has submitted in support of its

contentions nine witness statem ents of indigenous people and
three “independent reports”.

7.178. Starting with the witnesses, none of them provides any

dates for the sprayings, with the exception of witnesses 11 and

17.801 Witness 11 refers to a spraying episode that took place

“at the beginning of the year two thousand and one” and

suggests that more sprayings took place in subsequent years, but

gives no indication as to the time, merely asserting: “When the
802
planes returned, the disease returned.” This witness testifies

that her daughter was directly sprayed (“My baby was with me

at the farm when the planes came and the liquid fell on her”)
and attributes the death of th e baby, which occurred on 25

September 2001, months after the alleged sprayings. Witness

11 also states that another one of her daughters died two years

later, on 10 September 2003, “during a period of spraying”. 803

7.179. Witness 17 vaguely refers to an episode of spraying that

occurred: “about six or seven years ago” and states that some of

the alleged effects of the spra yings – skin rash, vomiting and

800
801 EM, paras. 6.106-6.113.
802 EM, Vol. IV, Annexes 199 and 203.
803 EM, Vol. IV, Annex 199, lines 8-9.
Ibid., lines 17-18.
334fever – appeared some time later (days or weeks) and lasted
804
several months. Heevenstatesthathesufferstothisdayof
severe headaches and dizziness. 805 It is highly unlikely that this

condition is an effect of the sprayings. As noted above, to the

extent that exposure to the mixture may cause temporary

symptoms, such as eye or sk in irritation, these appear

immediately after exposure and disappear within a day or so. 806

7.180. All nine indigenous people who have provided witness

statements on behalf of Ecuador testify that, since the time when

the sprayings started in Colombia, their traditional way of life
has been severely disrupted, entire families have migrated and

those who were left behind lived in fear, that the crops and the

plants that the tribes use for f ood and religious rituals have died

and entire animals species have disappeared. The statements

provided by these witnesses descri be an environmental disaster

of epic proportions, but fail to provide any temporal framework

or to establish any causal link between the damages complained

of and the sprayings. None of these statements is accompanied
by any contemporaneous corroborating evidence whatever.

Moreover, given the vagueness of the witness statements, there

is no way of verifying these accounts.

804
805 EM, Vol. IV, Annex 203, lines 2-9.
806 Ibid., lines 5-6.
See above, para. 7.127.
3357.181. The “independent reports” to which Ecuador refers for

its allegations of injuries on the indigenous peoples also fall

short of substantiating Ecuador’s claims since all they do is

recount the complaints of the local populations. They are not

corroborated by any independent evidence; those conducting the

inquiries were not scientists, and no testing was done by them.
In fact, some of the reports cited by Ecuador in support of its

allegations clearly state that th ere is no agreement as to the

harmful effects of glyphosate and that there is no established

cause for the illnesses allegedly suffered by the local peoples.

7.182. For instance, the NGOs that visited Sucumbíos in 2005 –
whose observations have been discussed above in connection

with alleged damages to crops – observed that there was no

agreement as to the harmful effects of the sprayings and that the

cause of intoxications (domestic use or fumigations) reported

from 2003 to 2005 as being due to pesticides had not been
807
established. This document also acknowledges, in a passage
that Ecuador omitted to translate:

“It is said that the illnesses reported in Chical are
caused by glyphosate but so far no studies of any
kind have been carried out.” 808

7.183. Another important aspect of the factual context that

emerges from these reports is the extreme poverty of these

indigenous communities living on the Colombia-Ecuador border

807 See para. 7.160.
808 EM, Vol. IV, Annex 169, p. 6.
336and the fact that they are aba ndoned to their own resources by

the Ecuadorian Government. A number of the documents filed

by Ecuador with its Memorial – including those that Ecuador

cites in support of its claim of “special harm” to the native

peoples – point to this deplorable state of affairs.

7.184. For instance, a November 2005 report by two NGOs

which visited the province of Esme raldas in the Summer of that

year stated as follows, again in a paragraph which has not been

translated by Ecuador:

“State attention in these areas is deficient, and
there are no real proj ects or programs for
sustainable development. The infrastructure is
obsolete, the people do not have any basic health

service, potable water, sewage system,
telephones, medical assi stance, nutrition or
education, that would allow them to have an
adequate level of life. The communication routes
leading to these territories are in a very bad state,

due to which, access t809hese areas is either made
on foot or by canoe.”

7.185. Similarly, the 2006 report of the U.N. Special

Rapporteur on the Rights of I ndigenous People on his visit to
Ecuador, which is cited extensively in the Memorial, lists oil

exploration and development as one of the main factors causing

the degradation of the envir onment and living conditions of

809 Ecolex and Aida Environmental Report on the Impacts in Ecuador
of the Fumigations under Plan Colombia, EM, Vol. IV, Annex 170, p. 3.
(Excerpt not included in Annex 170; Colombia’s translation)
337these indigenous communities. 810 The report notes that some

tribes “complain that oil activities have polluted their rivers and

affected the health of the region’s indigenous peoples.” 811

7.186. In the light of this generally recognized state of affairs,

the “special harm” allegedly suffered by the indigenous people
of the border areas, cannot be attr ibuted to Colombia’s aerial

sprayings. Given the vague a nd unsubstantiated nature of

Ecuador’s witness statements and the reports on which Ecuador

relies, as opposed to the scie ntific evidence and the rigorous

methodology adopted by Colombia in the organization and

management of the aerial sprayings program, it can be
concluded that Ecuador’s alle gations are not only unsupported

by any evidence, but are contradicted by the evidence on the

record.

E. Conclusions

7.187. As shown in this chapter, Ecuador has failed to prove its

allegations of harms due to Colombia’s aerial sprayings in its

territory for the following reasons:

• Ecuador has demonstrated neither actual risk, nor

exposure, nor harm, nor causation.

• The available scientific evidence lends no support
to Ecuador’s case. There is no scientific evidence

810 EM, Vol. II, Annex 30, paras. 18-25.
811 Ibid., para. 20.
338 of serious threats to huma n and animal health or

the environment due to exposure to the spraying
mixture used by Colombia in its PECIG program.

On the contrary, scientific studies support

Colombia’s position that the PECIG program

poses no significant risks to human or animal
health and the environment. This is confirmed by

the continuous scientific monitoring of sprayed

areas in Colombia, conducted by the Colombian
authorities with international tech nical support.

The results revealed no evidence of glyphosate or

glyphosate residues in wa ter and soil samples

collected in directly-sprayed areas.

• To the extent that “the fabric of life in the border

region” is under threat as Ecuador alleges, this is
because of the poverty, remoteness and

governmental neglect that characterize this region,

coupled with the presence of the FARC guerrillas.
It has nothing to do with any alleged effects of

Colombia’s aerial sprayings.

• Colombia has adopted a cautious stance and has

taken all reasonable precautions in order to ensure

that the sprayings take place only in its own
territory and that spray drift beyond the targeted

areas is avoided. Strict protocols are followed to

that effect before, durin g and after the operations

339 and no sprayings take place over human

settlements, bodies of water or water courses
which are considered exclusion zones;

• Analyses conducted by the Ecuadorian authorities
themselves in 2004 on samples taken from rivers

in the provinces adjacent to the Colombia-Ecuador

border confirmed that no evidence of glyphosate

was present in Ecuadorian water and soil;

• Ecuador has failed to prove that the sprayings were

the cause of the injuries alleged.

7.188. Ecuador’s case is so extreme that its credibility is

undermined. Even assuming – quod non – that spray drift into
Ecuador’s territory occurred during Colombia’s aerial sprayings,

the quantities would have been minimal and any damage would

have been negligible even for the most sensitive plants, let alone

for humans and animals. It is simply not plausible, both in light
of the facts and the scientific evidence that Ecuadorian

nationals, flora and fauna suffered the kind of damages and

injuries alleged by Ecuador.

340 Chapter 8

ECUADOR’S CLAIMS FOR TRANSBOUNDARY
INJURY OR HARM

A. Introduction

8.1. In its Memorial, Ecuador accuses Colombia of “violation
812
of a broad array of fundame ntal international norms”. For

once, Ecuador does not exaggerate, at least insofar as it refers to
a “broad array”. Indeed, an armoury of treaties and other

instruments is thrown at Colombia, with little reference to

whether Colombia or Ecuador are parties, or even eligible to

become parties, to them, and without regard to the clauses of

these treaties which define th eir own application. New and
aspirational instruments (such as the UN Declaration on the

Rights of Indigenous Peoples) are combined, higgledy-piggledy,

with established propositions such as that of the Trail Smelter
813
arbitration or the Court’s dictum in the Advisory Opinion on

Legality of the Threat or Use of Nuclear Weapons that:

“The existence of the general obligation of States
to ensure that activities within their jurisdiction

and control respect the environment of other
States or of areas beyon d national control is now

812 EM, para. 6.134. Cf. EM, para. 6.6 (“the harms occasioned by
Colombia’s fumigations constitute an integrated, mutually reinforcing whole
that have undone the very fabric of life in the border region”).
813 Trail Smelter (United States of America v. Canada), Award of 11
March 1941, 3 RIAA 1905, p. 1965.
341 part of the corpus of international law relating to
814
the environment.”

8.2. Ecuador overstates the law almost as much as it does the

facts – and unnecessarily so, since if the facts were as alleged in
the Memorial, then the law, or that part of it concerned with the

causing of transboundary harm, would largely follow. If the

aerial spraying program had killed scores of people in Ecuador

and thousands of farm animals, la id waste to old-growth forests
815
and destroyed “thousands of hectares of valuable crops”, then

any threshold in relation to the causing of transboundary harm
would be exceeded. The heavy invocation of a wide range of

human rights and indigenous right s would be unnecessary. On

the other hand, if the facts and the science are as Colombia has

shown – that, at most, there ma y have been some discernible

drift of spray downwind from Colombian spraying operations,
but if so it was harmless about 50 metres from the spray zone –

then the case is quite different and most if not all of the norms

relied on by Ecuador are irrelevant.

8.3. Colombia will nonetheless address Ecuador’s legal

arguments as such, and without re gard to the realities of the
aerial spraying operations as demons trated in Chapters 4 and 7.

It is proposed to do so under the following rubrics:

• breach of Ecuador’s sovereignty;

814 Advisory Opinion, ICJ Reports 1996, pp. 241-242, para. 29.
815 EM, para. 10.9.
342 • transboundary harm and the failure to cooperate;

• breaches of human rights;

• breaches of indigenous rights.

8.4. This Chapter will deal with the former category of

obligations, i.e., those concerning alleged breach of sovereignty

and alleged transboundary harm. Chapter 9 will deal with the
superadded elements Ecuador seeks to bring to the dispute based

upon the rights or claimed rights of people or groups within

Ecuador.

8.5. Before turning to these questions, however, it is

necessary to say something about the applicable law.

B. The Applicable Law

(1) T REATY AND CUSTOM IN INTERNATIONAL
ENVIRONMENTAL LAW

8.6. An initial point to make is that the major modality for
the development of international environmental law in modern

times has been by treaty, bilatera l and especially multilateral.

Such customary international law rules as exist in relation to the

environment are of a general and residual character. The point
was made by the Court in its Advisory Opinion on Legality of

the Threat or Use of Nuclear Weapons , when it referred to the

“general obligation of States to ensure that activities within their

343jurisdiction and control respect the environment of other
816
States”.

8.7. Ecuador proceeds on the apparent assumption that

customary international law relating to the environment

develops at the pace of the fastest, most “advanced”, treaty.

Moreover it blithely cites as authority treaties which are not now

in force and may never enter into force, and regional treaties (for
example, those sponsored by the UN Economic Commission for

Europe) which Colombia and Ecuador are ineligible to join even

if they wanted to. It may be recalled that regional agreements or

practices cannot become embodied in customary rules unless

they are shown to be accepted by each State concerned; 817 a

fortiori they cannot be exported to other regions of the world

which may (as in the present case) face different problems and

have a different level of soci o-economic development. The

environmental problems of Europe are not the same as those of
818
South America.

816
Legality of the Threat or Use of Nuclear Weapons, Advisory
817nion, ICJ Reports 1996, pp. 241-242, para 29 (emphasis added).
818 Asylum (Colombia v. Peru), ICJ Reports 1950, pp. 276-278.
As would be the case, for instan ce, of Andean Community Decision
505 of 2001 (Andean Cooperation Plan for the Control of Illegal Drugs and
Related Offences). Indeed, the Program of Action, within the foreseen
reinforcement of nationalstrategies (S ection I), includes a section on the
technical eradication of illegal crops (Sub-section B.), para. 1 of which reads
as follows:
“1. Build up national capacities to implement programs for the technical
eradication of illegal crops by hand or bair, in the countries that need it,

keeping in mind the environmental sta ndards established by the competent
authorities.” (Emphasis added)
3448.8. In its treatment of the applicable law, Ecuador relies on

the following treaties to which Colombia and/or Ecuador are not

parties (the list is not exhaustive):

• Convention on Long-range Transboundary Air

Pollution, Geneva, 13 November 1979; 819

• United Nations Convention on the Law of the Sea,

Montego Bay, 10 December 1982; 820

• Convention on Environmental Impact Assessment in

a Transboundary Context, Espoo, Finland, 25
821
February 1991;

• Convention on the Protection and Use of

Transboundary Watercourse s and International

Lakes, Helsinki, 17 March 1992; 822

• UN Convention on the La w of Non-Navigational

Uses of International Wa tercourses, New York, 21

May 1997; 823

Thus, in 2001, the Andean countries -Ecuador included- were in agreement
with having aerial eradication in the countries that needed it, and went no
further than to recommend that such countries "keep in mind" whatever
environmental standards were set by their own domestic authorities.
819
1302 UNTS 217 (a UN ECE Convention); neither Ecuador nor
820ombia eligible to be parties.
821 1833 UNTS 3; Colombia not a party.
1989 UNTS 309 (a UN ECE Convention); neither Ecuador nor
Colombia eligible to be parties.
822 1936 UNTS 269 (a UN ECE Convention); neither Ecuador nor
Colombia eligible to be parties.
823 A/51/869 (not yet in force); neither Ecuador nor Colombia a party.

345 • Convention on Access to Information, Public

Participation in Decision-Making and Access to
Justice in Environmental Matters, Aarhus, 25 June

1998. 824

8.9. On the other hand Ecuador fails to have proper regard

for the treaties to which Ecuador is a party, notably the 1988

Narcotics Convention, and it a ssumes without demonstration

that the standards for Colombia’s eradication program are to be
derived from non-binding texts (not ably the ILC Draft Articles

on Prevention of Transboundary Harm from Hazardous

Activities of 2001), rather than th e relevant treaty between the

Parties. It is necessary to say something in turn about each of

these points.

(2) T HE SIGNIFICANCE OF THE 1988N ARCOTICS C ONVENTION

8.10. Ecuador relies on the 1988 Convention for

jurisdiction,25but it systematically ignores the implications of

the Convention for the presen t case. The purpose of the

Convention is stated in Article 2(1) as follows:

“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

824 2161 UNTS 447 (a UN ECE Convention); neither Ecuador nor
Colombia eligible to be parties.
825 EM, paras. 4.15-4.21.
346 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.”
The emphasis is on more effective measures of enforcement,

building on the base of the earlier treaties of 1961 and 1971.

8.11. The 1988 Convention deals expressly with eradication

programs in Article 14(1)-(3). These provide:

“1. Any measures taken pursuant to this

Convention by Parties sha ll not be less stringent
than the provisions applicable to the eradication
of illicit cultivation of plants containing narcotic
and psychotropic substances and to the
elimination of illicit demand for narcotic drugs
and psychotropic substances under the provisions
of the 1961 Convention, the 1961 Convention as

amended and the 1971 Convention.
2. Each Party shall take appropriate
measures to prevent illic it cultivation of and to
eradicate plants containing narcotic or

psychotropic substances, such as opium poppy,
coca bush and cannabis plants, cultivated illicitly
in its territory. The measures adopted shall
respect fundamental human rights and shall take
due account of traditional licit uses, where there
is historic evidence of such use, as well as the
protection of the environment.

3. a) The Parties may co-operate to
increase the effectiveness of eradication
efforts. Such co-operation may, inter alia,
include support, when appropriate, for

integrated rural development leading to

347 economically viable alternatives to illicit

cultivation. Factors such as access to
markets, the availability of resources and
prevailing socio-economic conditions
should be taken into account before such

rural development programmes are
implemented. The Parties may agree on
any other appropriate measures of co-
operation.

b) The Parties shall also facilitate the
exchange of scientific and technical
information and the conduct of research

concerning eradication.
c) Whenever they have common

frontiers, the Parties shall seek to co-
operate in eradication programmes in
their respective areas along those
frontiers.”

8.12. Ecuador reads Article 14(2) as incorporating by

reference the corpus of internat ional law rules, including treaty

rules, concerning human rights and the environment,
826
irrespective of whether Colombia is a party to those treaties.

But there are fundamental probl ems with such a reading of
Article 14(2).

8.13. First, the 1988 Convention is neither a human rights

treaty nor an environmental tr eaty, still less one for the
827
protection of indigenous rights. Rather its concern is that

826 Cf. EM, para. 9.11 (“imported directly into the present dispute”).
See also EM, paras. 8.71-8.79.
827 For the Convention’s emphasis on effectiveness, see e.g. preambular
paras. 4, 6, 7, 15, 16, Arts. 2(1), 3(6), 5(1), 7, 9(1), 14(3)(a), 24.
348measures taken should be efficient and adapted to reducing the

growing traffic in illicit drugs. Under 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.” (emphasis added)

Under 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 Convention.”
Neither under Article 24, nor under the 1961 and 1971

Conventions whose effect is pr eserved by Article 25, is there

any provision even arguably in corporating human rights,

indigenous rights or environmental obligations. 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 25. Colombia relies on both

Article 24 and 25. In its opinion the aerial spraying program is

“desirable or necessary for th e prevention or suppression of

illicit traffic”. Furthermore it has a right to conduct the program
on its territory pursuant to the 1961 Convention (to which both

States are parties).828 By contrast Ecuador’s interpretation of

828 Art. 26(2) of the 1961 Convention provides, inter alia , that the
parties “shall destroy the coca bushes if illegally cultivated”: see Single
Convention on Narcotic Drugs, New York, 30 March 1961: 976 UNTS 105.
349Article 14(2) is incoherent because it ignores Articles 24 and 25

of the 1988 Convention.

8.14. Furthermore there is no refe rence to human rights (still

less any incorporation of huma n rights obligations) in other
provisions of the 1988 Convention where such rights are more

obviously relevant than they are in the context of Article 14(2).

This is true, for example of Ar ticles 3 (offences and sanctions),

5 (forfeiture), 6 (extradition) a nd 7 (mutual legal assistance).

When the drafters of the Convention wished to take into account
human rights considerations, th ey did so expressly, not by

reference (e.g. in Articles 6(6) (discriminatory prosecution),

7(18) (certain protections of w itnesses) and 17(5) (boarding at

sea)). In general the Conve ntion displays a pronounced

deference to domestic law and jurisdiction, even in relation to
those matters which might directly implicate human rights. 829

8.15. Ecuador misrepresents the actual language of Article

14(2) by disregarding its syntax. Ecuador argues that:

“it seems clear that the phrase ‘respect for
fundamental human rights ... as well as
protection of the environment’ used in Article

14(2) is intended to incorporate the relevant
requirements of internat ional environmental law
[...] and with respect to human rights [...]”.

829 For the 1988 Convention’s deference to domestic law and
jurisdiction, see e.g. Arts. 2(3), 3(1)(c), 3(2), 3(9), 3(11), 4(3), 5(5)(a),
7(5)(c), 9(1)(c).
350The relevant part of Article 14(2) reads:

“The measures adopted shall respect fundamental
human rights and shall take due account of
traditional licit uses, where there is historic
evidence of such use, as well as the protection of

the environment.”
Ecuador’s ellipsis between th e references to “fundamental

human rights” and “protection of the environment” does

violence to the actual language of the provision. On normal

syntactical principles, the verb “respect” qualifies only

“fundamental human rights”, while “protection of the

environment” is governed by the words “take due account of”,
which also applies to “traditional licit uses”. Further, the words

“as well as” connect “protection of the environment” to

“traditional licit uses”, rather th an to “respect for fundamental

human rights”. The correct shorte ning of Article 14(2) should

read “respect human rights a nd take due account of ... the
830
protection of the environment”.

8.16. Ecuador further argues that the Court should adopt an

“evolutionary interpretation” of Article 14(2) of the 1988 UN

Narcotics Convention, so as to take account of intervening

developments in international human rights and environmental
law in the period since its adoption. 831 But “evolutionary

interpretation” is not a principle which permits modification of

830 The same formulation is reflected in the French and Spanish texts of
Art. 14(2).
831 EM, paras. 8.75-8.77.
351the essential effect of a provision; at most, it allows account to

be taken of changes in the law, especially in the case of treaties
of long standing. The 1988 Tr eaty represents current

international policy on cooperatio n in the fight against drugs,

and it expressly requires eradication programs, subject to certain

safeguards. Ecuador exaggerates the extent and meaning of the

safeguards and completely igno res the point of the provision,

which is the effective eradicati on of illicit crops, a process in

which, pursuant to Article 14(3)(d), it should itself have

cooperated.

8.17. Ecuador also refers to the declaration made by Colombia

on ratifying the 1988 Convention, 832which it deems significant

as “reflecting Colombia’s recognition of the importance given to

the protection of the environmen t and the rights of indigenous
833
communities in the context of drug control”. Colombia’s

concern in making that declaration was, inter alia, to maintain a

balance between criminalisation of coca cultivation and “a

policy of alternative development, taking into account the rights
of the indigenous communities in volved and the protection of

the environment”. As demonstrated, Colombia does indeed

maintain a policy of alternative development, and a concern for

the protection of the environment. 834

832
833 EM, para. 4.17.
834 EM, para. 4.17.
See above, paras. 4.8-4.14, 4.35, 4.38-4.39.
3528.18. A final reason why Ecuador ’s incorporationist

interpretation of Article 14(2) cannot work is that it gives no

guidance at all as to which treaties are incorporated by

reference. Given that this is a multilateral treaty, they cannot be

confined to treaties to which both Ecuador and Colombia are

parties (or were parties at some point in time). Yet it cannot be
supposed that the parties to the 1988 Convention, by a side-wind

in Article 14(2), committed themselves to a whole range of new

treaty obligations in the fi eld of human rights and the

environment. The natural interpretation of Article 14(2) is that

general respect for human rights and the environment is called

for in the context of the require d eradication measures. Again,
there is no basis for suggesting that Colombia or its officials did

not comply with that obligation.

(3) T HE 2001 ILC DRAFT ARTICLES ON PREVENTION OF
TRANSBOUNDARY HARM FROM HAZARDOUS ACTIVITIES

8.19. In its Memorial, Ecuador relies heavily on the ILC Draft

Articles on Prevention of Transboundary Harm from Hazardous

Activities of 2001 (hereafter the ILC Prevention Draft
835
Articles). Particular issues concerning the ILC Draft Articles
will be dealt with in Section D of this Chapter as they arise. But

two general points should be made here, concerning first the

835 See generally EM, Ch. VIII. The ILC’s Prevention Articles are
relied on in paras. 8.5, 8.6, 8.11, 8.14, 8.16, 8.18, 8.19, 8.25, 8.26, 8.27, 8.31,
8.38, 8.39, 8.43, 8.44, 8.47, 8.56, 8.67, 8.68.
353reception of the Articles, and second, their application in

principle in this case.

8.20. The ILC Prevention Draf t Articles – and their

counterpart text, the Draft Prin ciples on the Allocation of Loss

in the Case of Transboundary Harm rising out of Hazardous
836
Activities of 2006 – are the product of a long and difficult

evolution in the ILC, as is well-known. The 2006 Articles do
not even purport to be anythi ng other than an exercise in

progressive development. 837 But there are at least four

indications that much the same is true of the ILC Draft

Prevention Articles.

8.21. The first indication is that in the ILC’s debates in 2001,

few speakers expressed the view that the Articles were reflective

of general international law. 838 Responding to criticisms of the

text the Special Rapporteur (P.R. Rao) noted that “[m]any States

had also indicated that non-disc rimination, information to the

public and other aspects of emergency preparedness were

elements of progressive development, rather than current

836 Report of the ILC on the Work of its 58th Session, A/61/10 (2006),
p, 106 ff.
837 See, in particular, the General Commentary to the Draft Principles,
Report of the ILC on the Work of its 58th Session, A/61/10 (2006), pp. 110-
114, in particular at 113 (para. 11): “ As the draft principles are general and
residuary in character they are cast as a non-binding declaration of draft
principles.”
838 th
See 2675 Meeting, 11 May 2001; ILC Yearbook 2001, vol. I, p. 58
et seq.
354practice in international law.” 839 Professor Brownlie, with

characteristic pungency, stated:

“the draft was not about transboundary harm, but
about the management of ri sk. It was in effect a
new subject, and one that had proved difficult to

deal with. As many members who had refrained
from taking the floor at the current meeting were
aware, the draft was creative, and in certain
respects, indeed, radical.” 840

8.22. Second, the ILC’s intention wa s that the Articles should

form the basis for the elaboration of a convention, 841 and they

are accordingly drafted in treaty form. In those circumstances,

the wording used is naturally imperative, but this does not mean

that any particular provision re presents codification or rather

constitutes an exercise in progressive development.

8.23. The third indication relates to the lukewarm reception

accorded to the Draft Articles in the debate in the General

Assembly’s Sixth Committee in 2001. In the debate on the

Draft Articles, 30 States made comments, representing the views
842
of a total of 34 States. No consensus position emerged, other

839
Ibid., 66 (para. 55).
840 Ibid., para. 58.
841 Report of the ILC on the Work of its 53rdSession, Yearbook of the
ILC, Vol. II(2), p. 145 (para. 94).
842 The statement by Sweden (A/C.6/5 6/SR.17, paras. 18-21) was also
made on behalf of the other Nordic countries (Denmark, Finland, Iceland and
Norway).

355than to await results of the ILC’s work on liability. 843 A few

delegations (Netherlands, Mexico, Australia) expressed the view

that the draft Articles “broadly reflected customary international

law”, 844 but most reserved thei r position. Some States,

including Colombia, limited themselves to expressing the view

that there exists an obligation of due diligence to prevent or

845
minimize transboundary harm.

8.24. The fourth indication is the even more lukewarm debate

in the Sixth Committee in 2007. In 2006, the General Assembly

by Resolution 61/36 took note of th e ILC’s draft principles on
846
allocation of loss, and resolved to include the topic of

“Consideration of prevention of transboundary harm from

hazardous activities and allocation of loss in the case of such

harm” on the agenda of the next session. In 2007, both ILC

texts were discussed: 14 States made statements, representing a
847
total of 20 States. There was little support for a general

843 A/C.6/56/SR.11, A/C.6/56/SR.12, A/C.6/56/SR.13,

A/C.6/56/SR.14, A/C.6/56/SR.15, A/C.6/56/SR.16, A/C.6/56/SR.17,
A/C.6/56/SR.18, A/C.6/56/SR.22, A/C.6/56/SR.23, A/C.6/56/SR.24. A
decision on further action was postponed: Resolution 56/82, A/RES/56/82
(the general Resolution on the Work of the ILC at its 53ession), para. 3.
844 See A/C.6/56/SR.16, para. 101 (Netherlands).
845 A/C.6/56/SR.16, para. 42 (“with regard to the fulfilment of the
obligation of due diligence, which governed the principle of prevention,
special consideration should be paid to the socioeconomic development of

the parties, the scientific and technological facilities available and the
practical realities of the context in which activities liable to cause
transboundary harm were carried out”).
846 A/RES/61/36, para. 2.
847 The statement made by New Zealand was made on behalf of the
CANZ countries (Australia, and Canada): A/C.6/62/SR.12, paras. 12-15;
356convention on either subject. China described the work of ILC

on transboundary harm as “good examples of the progressive
848
development of the relevant rules of international law”. The

United States said that “both texts went beyond the current state

of international law and practice. Both were designed to

encourage national and internati onal action in specific contexts
849
rather than form the basis of a global treaty.” Although Japan

expressly stated that “the oblig ation of preven tion per se had
become a part of customary international law”, it also stated that

Articles 6 and 12 as to prior au thorization and exchange of

information could not be said to be a codification of custom. 850

In the result the General Assembly commended the articles “to

the attention of Governments, without prejudice to any future

action”, and postponed further debate until 2010. 851

8.25. As regards the debates in both 2001 and 2007, it should

be borne in mind that the Sixth Committee was concerned with

the question whether the Article s should form the basis of a

treaty. The occasional statements expressing the view that

particular provisions should be strengthened, cannot, in the

absence of a clear statement to that effect, be read as an

Norway spoke on behalf of the other Nordic countries (Denmark, Finland,
Iceland and Sweden): Ibid., para. 16.
848 Ibid., para. 21.
849 Ibid., para. 29.
850 Ibid., para. 36.
851 General Assembly Resolution 62/68, A/RES/62/68 (“Consideration

of prevention of transboundary harm from hazardous activities and allocation
of loss in the case of such harm”), paras. 3 and 6.
357expression of opinion as to the content of customary

international law or the extent to which the Articles accurately
codify it. Rather they may be taken as relating to the need to

ensure the effectiveness of any eventual treaty.

8.26. The fifth indication is the discrepancy between the
formulations in the ILC Draft Articles of 2001 and the Court’s

own formulation of the position in Threat or Use of Nuclear

Weapons, cited in paragraph 8.1 above. The Court did not
articulate a category of “hazardou s activity” but confined itself

to affirming a general obligation to “respect the environment of

other States”.

8.27. Against this background there is no warrant for Ecuador

to assume that the ILC’s Prevention Draft Articles of 2001

reflect customary international law en tant que tel. But the
Articles are important in helping to show the limits of the law:

in light of the Court’s form ulation, provisions of the ILC

Articles purporting to impose specific obligations can be

assumed not to reflect custom. A fortiori, Ecuadorian arguments
which would take the position still further must be rejected.

8.28. But even if, for the sake of argument, the ILC Articles
were taken to reflect the existing state of international law, they

would not apply to the aerial spraying program. Both the 2001

Prevention Articles and the Draft principles on allocation of loss

358 852
of 2006 are limited to “hazardous activities”. This is further

qualified by Article 2(a), which defines “Risk of causing
significant transboundary harm” to include “risks taking the

form of a high probability of causing significant transboundary

harm and a low probability of causing disastrous transboundary

harm;”. 853 It must be stressed that the ILC Articles do not

proceed by deeming specific activi ties or classes of activities to

be “hazardous”; the idea of a list of activities was expressly
854
rejected. Therefore the question is whether the particular

activity, with its specific characteristics, meets one of the two

criteria – “a high probabi lity of causing significant
transboundary harm” or “a low probability of causing disastrous

transboundary harm”. In the present case, for the reasons stated

in Chapter 7, the spraying program meets neither of these

criteria. There is no possibility of aerial spraying in Colombia

causing “disastrous transboundary harm” in Ecuador, so the

second criterion does not apply. Nor is there even “a high

probability of causing significant transboundary harm” in

Ecuador. The consequence is th at, whatever their status, the
ILC Prevention Draft Articles do not apply in the present case.

8.29. What does apply, after the two major treaties to which

Colombia is a party have been taken into account, is the general

852
See Commentary to Art. 1, para. (2): ILC Yearbook. 2001, Vol. II
853 149.
854 Ibid., 151-152, and see Commentary, paras. (3), (4).
See Commentary to Art. 1, para. (3): ILC Yearbook. 2001, Vol. II
(2) 149.
359obligation of respect which the Court endorsed in Legality of the

Threat or Use of Nuclear Weapons, cited already. Colombia

certainly complied with that ge neral obligation, as will be
further demonstrated in this Chapter.

8.30. In the interests of responsiveness, Colombia will

however deal, in Section D belo w, with Ecuador’s arguments
based on the ILC’s Prevention Draft Articles. Nothing said in

that discussion should be take n as an acceptance that the

Prevention Draft Articles apply in terms of the present dispute.

C. Ecuador’s Claim of Breach of Sovereignty

8.31. First, however, it is conveni ent to deal with Ecuador’s

arguments of breach of sovereignty. These are elaborated in
Chapter VII of the Memorial. Essentially there are three

complaints: (a) aerial spraying which results in the deposit of

any detectable amount of the spray mixture on Ecuador’s
territory is a breach of its sove reignty; (b) aerial spraying which

causes harm to Ecuador, its citizens or its environment is a

breach of its sovereignty; (c) unauthorised overflight is a breach

of its sovereignty. The first tw o points are connected and will
be dealt with together.

(1) T RANSBOUNDARY EFFECTS AS A PER SE BREACH

8.32. In its Memorial, Ecuador argues that the “deposit of

toxic herbicides on the territory of Ecuador (as well as their

360dispersion in Ecuador’s airspace) in quantities that are

significant and harmful” is a violation of its sovereignty. 855 But

– subject to questions about the threshold of significant harm –

this aspect of the case is already covered by the obligation under

general international law to exercise all due diligence not to

cause to another State signifi cant, proven, transboundary
harm. 856 To describe the causing of such harm as a breach of

sovereignty does nothing to further the analysis.

8.33. Ecuador relies in this context on Article 2(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-intervention in the domestic affairs of
other States.”

But the references in Article 2 to the principles of “sovereign

equality and territorial inte grity of States” and of “non-

intervention in the domestic affa irs of other States” do not have

the effect of imposing any additional substantive obligations on

the States Parties to the 1988 C onvention. Rather, the ordinary

meaning of the formulation used (“carry out their obligations
[…] in a manner consistent with”) indicates that reference is

made to those principles in order to define the way in which the

855 EM, para. 7.1.
856 See Section D below.
361substantive obligations under the Convention are to be carried

out, and in fixing the outer limits of those obligations.

8.34. However, Ecuador also seems to put its case in another
way. It argues that:

“Colombia’s actions have violated Ecuador’s
right to determine for itself what acts may take
place within its territory, and in particular
Ecuador’s right to determine the level and nature
of any harmful pollution to which its territory,
people and natural resources will be exposed.

The spraying and drift of herbicides onto the
territory and natural resources of Ecuador further
violate Ecuador’s permanent sovereignty over its
natural and biological resources.” 857

8.35. This appears to be a claim that no detectable or

measurable trace of the spray mix, no matter how small, should

be allowed to enter Ecuador’s waters, airspace or land territory.
Such a claim would indeed be based on sovereignty. Indeed it

would be a hermetic kind of s overeignty, the sovereignty of a

King Canute, one which claims 100% national control over the

environment within a State’s territory. On this view, Colombia

must refrain from the exercise of its own sovereignty on its own
territory (elimination of illicit co ca plantings) in order to ensure

the absolute freedom of Ecuador ian territory from any trace of

Colombian activity, whether or not any significant harm is

caused. But it has long since been pointed out that where more

857 EM, para. 7.2.

362than one sovereignty is engage d, reconciliation of conflicting

interests is to be achieved, in general terms, by reference to
considerations of reasonableness and proportionality. 858 This is

the balance that the law relating to transboundary harm seeks to

strike, in the context of alle ged transboundary air pollution as

much as of transboundary rivers. To use the term “sovereignty”

to strike another balance, more favourable to the downwind or

downstream State, is to beg the question.

8.36. Ecuador cites the Australian Memorial in the Nuclear

Tests cases as authority for the propos ition that a State is free to
decide that no amount of any toxic substance should be

deposited on its territory, however non-toxic the amount

deposited may be. 859 And it implies that, through the

combination of the provisional measures order and its final

disposition of the two cases, the Court showed sympathy
860
towards the “absolute exclusion” view.

8.37. In fact what the Court said at the provisional measures
phase of both cases was that:

“Whereas for the purpose of the present
proceedings it suffices to observe that the
information submitted to the Court, including

Reports of the United Nations Scientific

858
Cf. the approach of the Tribunal in the Lac Lanoux case, (1957) 12
859IAA 285, 315-316.
EM, para. 7.9 (“independent right to determine what acts shall take
860ce within its territory”).
EM, para. 7.10.
363 Committee on the Effects of Atomic Radiation
between 1958 and 1972, does not exclude the
possibility that damage to Australia [New
Zealand] might be shown to be caused by the

deposit on Australian [New Zealand] territory of
radio-active fall-out resulting from such tests and
to be irreparable;” 861

This does not imply that damage need not be shown in order to

establish responsibility in the context of transboundary effects.

8.38. Nor do the Court’s subsequent decisions in these cases

add anything on this point. 862

8.39. In any event the Nuclear Test cases concerned long-

distance radioactive fallout from atmospheric nuclear testing, a

very different proposition than incidental and marginal spray
drift from lawful activities in th e 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 u npreventable) are initially

difficult to detect and expensive to treat. In all these respects,

glyphosate and Cosmoflux are differ ent: they have a half-life
measured in days, they do not bioaccumulate, they present no

disposal problem, they have minimal toxicity to humans and

861
Australia v France, Interim Meas ures of Protection, ICJ Reports
1973, p. 105, para. 29; to the same e ffect, New Zealand v. France, Interim
Measures of Protection, ICJ Reports 1973, p. 141, para. 30.
862 Cf. EM, para 7.10, citing New Zealand v. France, Judgment, ICJ
Reports 1973, p. 457.
364animals in the quantities used, and the minor irritation that direct

exposure to spray may cause is obvious (principally mild

irritation in the eyes or skin), temporary and easily treated.

8.40. For these and other reasons, there is no duty based on the

principle of sovereignty not to cause any change at all in the

chemical composition of the waters or atmosphere of a

neighbouring State, and to the extent Ecuador suggests
863
otherwise its position should be rejected.

(2) O THER ALLEGED BREACHES OF SOVEREIGNTY

8.41. Finally under this rubric, Ecuador refers to “instances of

direct overflight into the airspace of Ecuador”, to which it has
864
never consented.

8.42. In the absence of a treaty or other consensual

arrangement, or when specific circumstances recognized under

international law are not present, overflight of a State’s territory

would breach the sovereignty of the State which is overflown. 865

863 To similar effect see ILC Prevention Articles, commentary to Art. 2,
para. (5):

“The ecological unity of the planet does not correspond to political
boundaries. In carrying out lawful activities within their own
territories, States have impacts on each other. These mutual impacts,
so long as they have not reached the level of ‘significant’, are
considered tolerable.”
In ILC Yearbook 2001 Vol II Part 2, 152.
864 EM, para 7.17.
865 Cf. Convention on International Civil Aviation , Chicago, 7
December 1944 (as subsequently amended), ICAO Doc. 7300/9, Arts. 3-5.

365A fortiori so would unauthorised aerial spraying actually
conducted on another State’s territory. Colombia does not claim

otherwise.

8.43. It should be stressed that, ev en in relation to the period

before 2004, Ecuador does not give any particularised examples

of overflight, still less does it subs tantiate that any actual harm

occurred as a result. As a result of enhanced technology
introduced in 2004, as well as gr eater experience in conducting

the program, the possibility of occasional error – if it ever

occurred – in this regard has greatly diminished.

8.44. It may be observed that Ecuadorian agents seem to have

entered Colombian territory, inter alia, with a view to taking

statements from the Colombia n witnesses included in its
Memorial. 866 Colombia did not consent to this. The process of

taking evidence on foreign te rritory is a breach of

sovereignty. 867

D. Ecuador’s Claim concerning Non-Prevention of
Transboundary Harm

8.45. The exposition of Ecuador’s claims as regards

transboundary harm is contained in Chapter VIII of the

866
See also the Ecuadorian admissionsthat its officials have entered
Colombian territory to seek evidence in relation to the sprayings on at least
two occasions: EM, paras. 6.14 and 6.15.
867 In re Westinghouse Electric Corporation [1978] A.C. 547, 616
(Lord Wilberforce); 639-640 (Lord Diplock).
366Memorial; these claims form the heart of the case. Ecuador’s

claims in that regard are grouped under essentially three

headings, namely:

(1) breach of the obligation to prevent transboundary

harm;

(2) breach of obligations of cooperation in relation to

management of transboundary effects of spraying

(including failure adequately to assess potential

impacts of spraying; failure to inform and

consult; and failure to cooperate in management
of transboundary risks).

(3) failure to cooperate in relation to respect for the
protection of the environment, in breach of

Article 14(2) of the 1988 Narcotics Convention.

(1) P REVENTION OF TRANSBOUNDARY HARM

(a) Overview of Ecuador’s arguments

8.46. Section I of Chapter 8 sets out Ecuador’s arguments in

relation to the obligation to prevent transboundary harm. The

obligation to prevent transboundary harm is argued to exist not

only as a matter of general intern ational law, but is also argued

to be “imported directly into these proceedings” by Article 14(2)
of the 1988 Narcotics Convention. 868 This interpretation of the

868 EM, para. 8.1; see also Ibid., para. 8.2.
367effect of Article 14(2) has already been discussed and

rejected.869

8.47. Ecuador argues that Colombia has breached the posited

obligation to prevent transboundary harm by:

• causing or failing to prevent spraying causing

significant harm to persons, property, natural

resources and the environment in Ecuador;

• failing to take precautionary measures to prevent and

control the alleged harmful effects of spraying on the

health, livelihood, private and family life and

property of persons within Ecuador;

• failing to take precautionary measures to prevent and

control the alleged harmful effects of spraying on the

environment and natural resources within the
870
territory of Ecuador.

It may be noted that the second and third categories of alleged

breach in fact concern the same breach, but in relation to

different alleged harms; the reason would appear to be in order

to allow the recycling of the claims as regards human rights
arguments also under the heading of transboundary harm.

8.48. Ecuador seeks to preempt any argument of

countervailing benefit, or that Colombia’s obligations under the

869 See above, paras. 8.11-8.18.
870 EM, para. 8.9.
3681988 Narcotics Convention justify its alleged failure to take

precautionary measures. It argues that spraying has not resulted

in an overall reduction of co ca production, and that other

effective means of eradication, including manual eradication are
available. It is accordingly argued that the harm to Ecuador

outweighs any benefit to Colo mbia and that the lack of

proportionality violates the e quitable balance of interests
871
required by Article 10 of the Draft Articles on Prevention.

(b) Colombia’s response: preliminary remarks

8.49. Colombia’s principal res ponse to all of this is
straightforward. The spraying program on Colombian territory

did not cause the harm attributed to it by Ecuador, or anything

remotely like it. Indeed, it has not been shown to have caused

any harm at all to Ecuador or its nationals. Moreover, Colombia
took reasonable precautions, monitored the program throughout,

modified it in light of further information (e.g. in 2004), and

sponsored further scientific work on its impacts. In light of the

substantial body of knowledge about glyphosate as a Class III

substance (i.e. mildly toxic) al ready existing in 2000, Colombia
was entitled to proceed with the spraying program in 2000.

Subsequent studies, including in particular CICAD I and

CICAD II confirm that Colombia was and remains entitled to

continue its spraying program. The factual and scientific
evidence underlying this response was dealt with

871 See EM, para. 8.31.

369comprehensively in Chapter 7 above. The following remarks

are made without prejudice to that basic contention.

(c) The due diligence obligation of prevention

8.50. As already stated, Colombia accepts the Court’s

conclusion in the Nuclear Weapons Advisory Opinion that the

general obligation of States “to ensure that activities within their

jurisdiction and control respect th e environment of other States
or of areas beyond national control is now part of the corpus of

international law relating to the environment.” 872 But this is an

obligation of means and not of re sult; it is an obligation of due

diligence, not a form of strict liability.

8.51. This point was expressly accepted by the ILC in

adopting the Draft Articles on Prevention of Transboundary

Harm for Hazardous Activities on 2001. It is reflected in Article

3 of that text which provides simply that:

“The State of origin shall take all appropriate

measures to prevent significant transboundary
harm or at873y event to minimize the risk
thereof.”

The commentary notes that:

“The obligation of the State of origin to take
preventive or minimization measures is one of

due diligence. It is the conduct of the State of
origin that will determine whether the State has

872 Legality of the Threat or Use of Nuclear Weapons, Advisory
Opinion, ICJ Reports 1996, pp. 241-242, para 29, relied on in EM, para. 8.3.
873 ILC Yearbook 2001 Vol II Part 2, 153.
370 complied with its obligation under the present

articles. The duty of due diligence involved,
however, is not intended to guarantee that
significant harm be totally prevented, if it is not
possible to do so. In that eventuality, the State of

origin is required … to exert its best possible
efforts to minimize the risk. In this sense, it does
not guarantee that the harm would not occur.” 874

8.52. As to the standard of due diligence, the commentary

concludes:

“The standard of due diligence against which the
conduct of the State of origin should be

examined is that which is generally considered to
be appropriate and proportional to the degree of
risk of transboundary harm in the particular
instance. [...] Issues such as the size of the

operation; its location, special climate conditions,
materials used in the activity, and whether the
conclusions drawn from th e application of these
factors in a specific case are reasonable, are
among the factors to be considered in

determining the due diligence requirement in
each instance. […] [D]ue diligence in ensuring
safety requires a State to keep abreast of
technological changes and scientific
developments.” 875

8.53. By contrast, Ecuador, ha ving referred to the
Commentary to Article 3 of the Draft Articles on Prevention as

874 ILC Yearbook 2001 Vol II Part 2, 154, commentary to Art. 3, para.
(7).
875 ILC Yearbook 2001 Vol II Part 2, 154, commentary to Art. 3, para.
(11).
371 876
to the standard of due diligence, seeks to transform it into a

virtual guarantee. Representative is the following passage:

“it is clear that a very high standard of care is
called for when inherently hazardous activities

such as aerial spraying of toxic herbicides are
undertaken. The only appr opriate standard of
care in the circumstances of the present case is
one that eliminates all risk of transboundary
877
pollution caused by overflight or drift.”
For convenience this will be re ferred to as the “absolute

prevention” thesis. It calls for a number of comments.

8.54. First, this passage confuses overflight and drift.

Overflight has nothing to do with situations of transboundary

harm, as has been seen; nor is the obligation not to fly without

permission into the airspace of another State a matter of due
878
diligence.

8.55. As to drift, Ecuador articulat es its “absolute prevention”

thesis for the category of activity “aerial spraying of toxic

herbicides”, and it does so in the face of st rong scientific

evidence both as to toxicity and drift, which was summarised in

the previous chapter. But as the ILC commentary makes clear,

the standard is one of reasonableness in the particular

876
EM, para. 8.26 citing ILC Yearbook 2001 Vol II(2), 154,
877mentary to Art. 3, para. (11).
EM, para. 8.27 (emphasis added). See also EM, para. 8.29:
“Colombia has a duty to make certain that transboundary pollution cannot
878 does not occur” (emphasis added).
See above, para.8.42.
372circumstances. Further, there is no basis for the suggestion that

aerial spraying in Colombia was an inherently hazardous

activity so far as concerns persons or the environment across the

border in Ecuador (this was not the case even within Colombia).
The information available – regu larly updated – indicated no or

none insignificant risk of deposition of the spray mixture in

Ecuador.

8.56. Ecuador seeks to support its “absolute prevention” thesis

by reference to the observ ation of the Court in Gabčíkovo-

Nagymaros Project that:

“in the field of environmental protection,
vigilance and prevention are required on account

of the often irreversible character of damage to
the environment and of the limitations inherent in
the very mechanism of reparation of this type of
damage.” 879

But “vigilance and prevention” do not entail the elimination of

all risk whatever. Moreover the spray mix does not cause
“irreversible damage” to the envi ronment, certainly it does not

do so in Colombia (where the spraying actually occurs and

where it is carefully monitore d); it is biodegradable and

disappears as harmless metabolites in the soil within weeks.
However, the Court’s dictum in Nuclear Weapons , a year

earlier, is phrased solely in terms of an obligation to ensure

“respect” for the environment of other States, and the word was

879EM, para. 8.27, citing Gabč íkovo-Nagymaros Project
(Hungary/Slovakia), Judgment, ICJ Reports 1997, p. 78, para. 140.

373no doubt carefully chosen. It is clear that the Court did not

intend to establish a standard in which all risk of harm was
excluded.

8.57. Ecuador’s reliance on the precautionary principle does

not take things further. The precautionary principle does not

constitute as such an internat ional obligation; it is usually

formulated by international tribunals in adjectival terms, and as

an “approach” rather than a “pri nciple”. It may be seen as

providing guidance as to how States should conduct themselves
880
in matters concerning sustainable development. There is no
reason to think that it modifies the substantive law as concerns

transboundary harm. The authorities relied upon by Ecuador as

supporting the contrary view eith er concerned the interpretation

of specific treaty-based obliga tions of prevention, or of EU

legislation in environmental matters which imposes specific

obligations of prevention.

8.58. Ecuador refers to Arti cle 14(1)(d) of the 1992
Convention on Biological Diversity, although it does not as such

appear to argue that spraying violates that provision. 881 Inany

880
Cf. EM, para. 8.18: “The precautionary principle has become one of
the central concepts for organising, influencing and interpreting
881temporary international environmental law and policy” (emphasis added).
See EM, para. 8.25; see also Ibid., para. 8.36, fn. 670. Cf. the later
references to the 1992 Biological Diversity Convention in the context of the
claim as to the alleged failure to asss risk from the spraying, discussed
below, paras. 8.69-8.72: see in particular, EM, paras. 8.51-8.52 cfIbid.,
8.54.
374case, the scope of the obligation imposed by Article 14(1)(d) is

overlooked. It provides:

“1. Each Contracting Party, as far as possible and
as appropriate, shall:

[…]

(d) In the case of imminent or grave danger or
damage, originating unde r its jurisdiction or
control, to biological di versity within the area
under jurisdiction of othe r States or in areas

beyond the limits of nati onal jurisdiction, notify
immediately the potentially affected States of
such danger or damage, as well as initiate action
to prevent or minimize such danger or damage.”

Quite apart from the qualification (“as far as possible and as

appropriate”), the obligation only applies in cases of “imminent
or grave danger or damage”, and it is an obligation “to prevent

or minimize such danger or damage.” It does not support

Ecuador’s “absolute prevention” thesis.

8.59. To summarise, it may be accepted that the State of origin

has an obligation of due diligence, i.e., to take all appropriate

measures to prevent significant ha rm or at least to minimize the

risk of such harm (and Colombia did adopt such measures). But
Colombia is not required to “eliminate [...] all risk of

transboundary pollution caused by [...] drift”. 882

882 Cf. EM, para. 8.27.

375 (d) Colombia complied with the obligation of prevention

8.60. As has been demonstrated in earlier Chapters, Colombia

has exercised due diligence th roughout in establishing and

implementing the aerial spraying program. The following

points are relevant:

(1) The program is conducted on the basis of an

Environmental Management Plan prepared under
883
the relevant Colombian law;

(2) An integral part of the program is the use of

modern equipment and technology, including

GPS and computerised mapping; 884

(3) The composition of the spray mixture is public

and is kept under review (an improved version
885
was adopted in 2004);

(4) There are strict parameters of aircraft height and

speed, wind speed, buffer zones for rivers and
886
human habitation, etc;

(5) There is careful planning of each mission in
887
accordance with these parameters;

(6) Individual plots are spraye d once, at most twice,
888
a year;

883 See above, paras. 4.8-4.14, 4.23-4.27.
884 See above, paras. 4.59-4.64.
885 See above, paras. 4.42-4.56.
886 See above, paras. 4.57-4.60.
887 See above, paras. 4.61-4.64.

376 (7) The program is subject to continuous scrutiny by

the Ministry of the Environment 889 and also to

external audit; 890

(8) There has been continued scientific review of the

program, in particular CICAD I (2005) and II
891
(2009);

(9) The National Health Institute monitors health

effects of the program in Colombia and deals

with complaints; it has not found any evidence of
892
injury to humans arising from the spraying.

8.61. For these reasons, Colombia has from the inception of

the program acted with due di ligence to prevent or minimize

such transboundary harm as could result from aerial spraying of

illicit coca plantations.

(2) O BLIGATIONS OF COOPERATION

8.62. In Chapter VIII, Section II, Ecuador alleges a number of

breaches grouped under the heading of cooperation. In

particular, it is alleged that the obligation of cooperation has
been breached by:

888 See above, para. 4.66.
889 See above, para. 4.26-4.27.
890 See above, para. 4.28-4.29.
891 See above, paras. 3.50-3.58, 4.15-4.19.
892 See above, paras. 7.64, 7.91.

377 (a) Colombia’s failure to assess the potential

transboundary effects of aerial spraying of

herbicides on the territory, people and

environment of Ecuador;

(b) Its failure to ensure that communities within

Ecuador likely to be affected were informed and

consulted; and

(c) Its failure to cooperate with Ecuador in the

control of the transbounda ry risks arising from

aerial spraying, inter alia, by refusing to share

information as to the chemicals in use and their

likely effects on public health and the
environment. 893

8.63. In that regard, it observes that “the obligation of States to

cooperate through notification, consultation a nd negotiation

permeates” the Draft Articles on Prevention and the 1992 Rio

Declaration, and that the obligati on is “clearly articulated in the
Lac Lanoux arbitration as well as in various regional treaties”,

including the 1991 UNECE Convention on Environmental

Impact Assessment in a Transb oundary Context (the Espoo
894
Convention). Reference is made to Article 9 of the Draft

Articles on Prevention as support for an obligation requiring

893 EM, para. 8.40.
894 EM, para. 8.38.
378States “to cooperate in negotiating an equitable balance of

interests”.895

8.64. These alleged breaches falli ng under the umbrella of

“cooperation” are conveniently examined separately, although

the Memorial displays a certain tendency to allow the arguments
to bleed into one another (and, indeed, into Ecuador’s argument

as to the obligation of prevention).

(a) Failure to assess potential transboundary effects

(i) Ecuado crasim

8.65. The first alleged breach relied upon by Ecuador under

the heading of cooperation is the failure to assess the potential

transboundary effects of the aeri al spraying. In this regard,

Ecuador portrays the obligation as one integrally linked to other

alleged obligations to notify, consult and cooperate with

neighbouring States, and to protect communities likely to be
affected, stating that

“Without an environmental impact assessment
(‘EIA’) there can be no m eaningful notification,
consultation and cooperation with neighbouring
States, nor can adequate steps be taken to protect
896
communities likely to be affected.”
Ecuador alleges that Colombia failed properly to assess in

advance the potential impact of the spraying on the territory,

895 EM, para. 8.39.
896 EM, para. 8.41.
379people, natural resources and e nvironment of Ecuador, that it

did not provide Ecuador with information as to the potential

risks associated with spraying and that it did not inform or
897
consult communities in Ecuador likely to be affected.

8.66. It then baldly states that “[i]nternational law provides

that activities likely to cause significant transboundary pollution

or harm must be subject to EIA by the State in which these

activities are to be conducted”, re lying in this regard on Article

7 of the Draft Articles on Prevention. 898

(ii) Colombia r’sponse

8.67. The EIA claim raises two questions: (a) what was

Colombia required to do in 2000, or subsequently, in the matter

of an EIA, and (b) what did it in fact do.

8.68. As to the first question, ther e is no equivalent to the

Espoo Convention in Latin America. Among universal

conventions, Article 206 of the Law of the Sea Convention (to
which Colombia is not a Party) requires “assessment” of

projects which may cause “substantial pollution of or significant

and harmful changes to the marine environment”, but only “as

far as practicable”. 899 This is a relatively high threshold, and

897 EM, para. 8.41.
898
899 EM, para. 8.43.
N. Craik, The International Law of Environmental Impact
Assessment (Cambridge: CUP, 2008) p.88 comments that the indeterminacy
of Art. 206 “leaves States with a broad discretion”.
380even then the requirement is qualified by considerations of

practicality.

8.69. The only treaty provision binding upon Colombia to
which Ecuador refers in this context is Article 14(1) of the 1992

Convention on Biological Diversity, 900 which provides as

follows:

“1. Each Contracting Party, as far as possible
and as appropriate, shall:

(a) Introduce appropriate procedures requiring
environmental impact assessment of its proposed
projects that are likely to have significant adverse

effects on biological dive rsity with a view to
avoiding or minimizing such effects and, where
appropriate, allow for public participation in such
procedures;

[...]
(c) Promote, on the basis of reciprocity,

notification, exchange of information and
consultation on activities under their jurisdiction
or control which are likely to significantly affect
adversely the biological diversity of other States
or areas beyond the limits of national

jurisdiction, by encourag ing the conclusion of
bilateral, regional or multilateral arrangements,
as appropriate;

(d) In the case of imminent or grave danger or
damage, originating unde r its jurisdiction or
control, to biological di versity within the area
under jurisdiction of othe r States or beyond the
limits of national jurisdiction, notify immediately
the potentially affected States of such danger or

900 1760 UNTS 79.

381 damage, as well as initiate action to prevent or
minimize such danger or damage; …” (emphasis
added)

8.70. Apart from referring to Article 14 of the Convention on

Biological Diversity, Ecuador doe s not suggest that Colombia

was under any treaty obligation to conduct an EIA in 1999-2000

in relation to potential tran sboundary impacts of the spray

program on Ecuador. Instead, such an obligation is apparently

to be inferred from the ILC’s Draft Articles on Prevention, the

1992 Rio Declaration, and on provisions of other, non-

applicable treaties (notably th e Espoo Convention); i.e. as a
matter of customary international law.

8.71. Turning first to the Biodiversity Convention, Ecuador is

understated in its reliance on Article 14; as noted above, it is not
901
at all clear that Ecuador even alleges a breach of Article 14.

The reason is clear. The chapeau to paragraph (1) contains the

qualifier “as far as possible and as appropriate”. N. Craik

901 Above, para. 8.58. In the section relating to assessment, Ecuador
limits itself to stating that “Information on the herbicide spraying programme
should also have been made available to Ecuador in accordance with Article
14 of the 1992 Convention on Biological Diversity” and that “interpreted in
accordance with the precautionary principle, and in the circumstances of the
present dispute, compliance with all of these requirements was both ‘possible

and appropriate’, within the terms of the chapeau to Article 14” (EM, para.
8.51). In the conclusions on the claim relating to the alleged failure to assess
the potential transboundary effects of the spraying (EM, para. 8.54), Ecuador
merely asserts that Colombia “should have carried out an EIA first, and
communicated the results to Ecuador, in accordance with the requirements of
international law set out above. In failing to do so it has violated its duty to
assess transboundary risks, has failed to act with due diligence, and has failed
to cooperate with Ecuador as required by international law.” There is no
allegation of breach of the 1992 Convention.
382comments that this qualifier introduces “a degree of discretion

inconsistent with a formal notion of bindingness”. 902 Moreover,

the scope of these provisions is distinctly limited. Article

14(1)(a) envisages “appropriate pr ocedures” only in relation to

projects likely to have “significant adverse effects on biological

diversity”. Article 14(1)(c) prom otes notification, exchange of
information and consultation in re lation to activities “which are

likely to significantly affect adversely the biological diversity of

other States or areas beyond the limits of national jurisdiction”,

as to which it is expressly envisaged that performance will be

achieved through the conclusion of bilateral regional or

multilateral arrangements. Finally, the obligation of notification
under Article 14(1)(d) is limited to “imminent or grave danger

or damage [...] to biological diversity”.

8.72. Ecuador asserts that the relevant obligations have been
903
incorporated into Colombia’s Environmental Code. But the

Memorial is misleading, and doubly so insofar as it
misrepresents not only the terms of Colombia’s Environmental

Code but also the terms of th e Biodiversity Convention. In

relation to the suggestions that the Code provides for “‘prior and

reciprocal communication’ with bordering States regarding

actions taken in one State that may harm the environmental
rights or interests of another State”, the relevant provision of the

Environmental Code relied upon is framed in exhortatory terms

902 Craik, op. cit.
903 EM, para. 8.52.
383(mirroring those of Article 14(1)(c)) of the 1992 Convention): it

requires the Government “to seek to complement” existing

international obligations, or to negotiate further agreements.

Similarly, the suggestion that the Code provides for “‘reciprocal

and permanent’ exchange of information to facilitate the

management of shared resources” is misleading, given that the

Code likewise calls upon the government to seek to complement
904
existing international agreements, or to negotiate others.

8.73. As to the existence of a free-standing obligation to

conduct an EIA, it is a sufficient answer to point out that the

keystone of Ecuador’s argument, Article 7 of the ILC’s Draft
905
Articles, encourages but does not actually require an EIA.

Article 7 reads:

904 The relevant provision of the Code (Art. 10 of Decree 2811/74)
provides in full :

“To prevent or solve environmental problems and to regulate the use
of renewable natural resources shared with borderin g countries, and
without prejudice to the treaties currently in force, the government
shall seek to complement the existing stipulations or to negotiate
others which deal with:

(a) The reciprocation and perm anent exchange of necessary
information for the planning of development and the optimal use of
said resources and elements;

(b) The reciprocation and prior communication of alterations or
environmental imbalances which can arise from works or projected
works of the governments or inhabitants of the respective countries,
far enough in advance that said governments can take the pertinent
actions when they believe their environmental rights and interests
could suffer impairment;”
See EM, Vol. II, Annex 10.
905 For the status of the ILC’s Prev ention Articles, see above, paras.
8.19-8.28.

384 “Assessment of risk

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
906
assessment.”

8.74. What this provision requires is risk assessment: an EIA

is identified as a form of risk assessment, but that is all.

Moreover the commentary stresses that the modalities of

assessment (who should conduct it, its content, etc.) are left to
907
“the domestic laws of the State conducting such assessment”.

This reflects a major difficulty with Ecuador’s customary

international law argument: ther e cannot be an obligation to

conduct an EIA unless internationa l law defines an EIA, which
908
presently it does not do.

8.75. Indeed such a definition would in practice be a

conventional one, laid down by tr eaty. Ecuador’s reliance on

the 1991 Espoo Convention is misp laced; neither Ecuador nor

Colombia are parties to that instrument, which was concluded

under the auspices of the UN Economic Commission for

Europe. 909

906
ILC Yearbook 2001, Vol. II (2), p. 157 (emphasis added).
907 Ibid., 158-9 (paras. (5)-(7)).
908 Craik, op. cit., 123-125 with references to the literature.
909 Cf. Espoo Convention, art. 16: “This Convention shall be open for
signature [...] by States members of the Economic Commission for Europe as
well as States having consultative status with the Economic Commission for
Europe [...]”. An amendment to allow other UN Member States to become

3858.76. Ecuador relies on five decisions of this Court or of the

Law of the Sea Tribunal which have “involved alleged failures

to undertake a transboundary EIA” , not for the actual decisions

themselves (none of which turned on the point) but for instances
of “State practice” supporting its customary law claim. 910 This

raises the question whether arguments before an international

court or tribunal can constitute State practice for this purpose: if

they do, it would only be where the argument was accepted by

the other side in such a way as to manifest opinio juris.

8.77. In any event, the cases are distinguishable from the

present one.

8.78. Request for an Examina tion of the Situation : the
application brought by New Zeal and sought to reactivate the

proceedings in Nuclear Tests (New Zealand v. France) pursuant

to paragraph 63 of the Court’ s 1974 judgment. New Zealand

sought a declaration by the Court that France’s proposed course
of action in undertaking a series of 8 underground nuclear tests

would “constitute a violation of the rights under international

law of New Zealand, as well as of other States”, and further or

in the alternative, that it would be…

parties to the Espoo Convention was adopted in 2001 (ECE/MP.EIA/4, 27
February 2001), but is not yet in force.
910 EM, para. 8.45.
386 “unlawful for France to conduct such nuclear
tests before it has undertaken an Environmental

Impact Assessment according to accepted
international standards. Unless such an
assessment establishes that the tests will not give
rise, directly or indi rectly, to radioactive

contamination of the marine environment the
rights under international law of New Zealand, as
well as the rights of other States, will be
911
violated.”

8.79. The Court disposed of the application on the narrow

basis that New Zealand’s request did not fall within the situation

foreseen by paragraph 63 of th e Court’s 1974 judgment. The

Court accordingly did not reach the merits of New Zealand’s

subsidiary claim for a declarat ion that such tests would be
912
unlawful unless preceded by an EIA. The principal ground
for the New Zealand EIA claim was Article 16 of the Noumea

Convention of 1986, which requi res assessment “within the

capabilities of the parties”, and consultation with the public and

other affected States “where appropriate”. 913 In reply France

relied on its margin of appreciation in determining the

appropriateness of an EIA for its resumed testing program. 914

911 Request for an Examination of the Situation in Accordance with
Paragraph 63 of the Court’s Judgment of 20 December 1974 in the Nuclear
Tests (New Zealand v. France) Case, ICJ Reports 1995, p. 291, para. 6.
912 Ibid., pp. 305-306, paras. 62- 63.
913 Ibid., p. 291, para. 6, referring to Convention for the Protection of
the Natural Resources and the Environment of the South Pacific Region,

914mea, 24 November 1986, 26 ILM 38.
CR 95/20, 12 September 1995, and see Craik, Op. cit., 142-143.
3878.80. Gabčíkovo-Nagymaros Project : Ecuador’s

characterisation of the case as one concerning the failure to
carry out an EIA prio r to the construction of the hydroelectric

installations is, to say the least, disingenuous.915 In fact the case

was dealt with on the basis of the general propriety of the

original barrage project, and the Court made no criticism of the

decision to proceed with the or iginal project. The Court’s

observation, relied upon by Ecua dor, that new environmental
916
standards and norms had to be taken into account was made

in a situation in which Artic les 15 and 19 of the 1977 Treaty

imposed continuing obligations upon the parties in relation to
water quality and the protection of nature; as discussed above,

the Court held that those provisions had to be interpreted on an

evolutionary basis, such that new treaty norms relating to the

environment had to be taken into account. 917 The decision is, to

say the least, not authority for any general customary obligation

to undertake an EIA.

8.81. MOX Plant : the claim was brought on the basis of
UNCLOS 1982, alleging breach of various obligations under

UNCLOS relating to the protection of the marine environment;

the prevention and control of po llution, co-operation in relation

to the risk of harms resulting discharges of mixed oxide fuel into

915
916 EM, para. 8.45.
EM, para. 8.45, quoting Gab číkovo-Nagymaros Project
917ngary/Slovakia), Judgment, ICJ Reports 1997, p. 78, para. 140.
Gabčíkovo-Nagymaros Project (Hungary/Slovakia), Judgment, ICJ
Reports 1997, pp. 67-68, paras. 114, and p. 78, para. 140.
388the Irish sea and related moveme nts of radioactive material

through the Irish Sea, and storage of radioactive materials at the

Sellafield site. Ecuador reli es only upon the 2001 provisional

measures decision of ITLOS, rende red prior to constitution of

the Annex VII tribunal, and in particular does not rely on the

decision as to Provisional Measur es embodied in the Order No.
3 of the Annex VII tribunal of 24 June 2003.

8.82. The 2001 ITLOS decision on provisional measures is

based on the relevant provisions of UNCLOS, including Article

206. Given various undertakings given by the United Kingdom

that there would be no addition al marine transportation of
radioactive waste as a result of the commissioning of the plant

and that there would be no impo rt or export of waste until

October 2002, ITLOS rejected the request for provisional
918
measures in the terms sought by Ireland. However, ITLOS

considered that:

“the duty to cooperate is a fundamental principle
in the prevention of pollution of the marine
environment under Part XII of the Convention
919
and general international law.”
On that basis it stipulated provisional measures requiring the

Parties to cooperate, including by exchanging information as to

possible risks for the marine environment arising out of the

commissioning of the MOX plant, monitoring the risks or

918 ITLOS, MOX Plant (Ireland v. United Kingdom) , Provisional
Measures, Order of 3 December 2001, para. 81.
919 Ibid., para. 82.
389effects of the operation of the MOX plant on the Irish Sea, and

devising appropriate measures to prevent pollution of the marine
920
environment arising from operation of the MOX plant. The

decision did not turn on whether or not there had been an EIA

and was reached in the specific context of the obligation of
cooperation in relation to polluti on of the marine environment

under UNCLOS.

8.83. Land Reclamation : the case was brought by Malaysia

against Singapore in relation to land reclamation measures in the

Straits of Johor which were said to violate various provisions of

UNCLOS. As in MOX Plant, provisional measures were sought

from ITLOS prior to constituti on of the Annex VII arbitral

tribunal. ITLOS relied on the passage quoted above from its

decision in MOX Plant as to the fundamental role of cooperation
921
in relation to the prevention of pollution under UNCLOS.
Having noted that Singapore had not undertaken a prior

assessment of the effect of the land reclamation works upon

waters under Malaysian jurisdiction, 922it concluded that “it

cannot be excluded that, in the pa rticular circumstances of this

case, the land reclamation works may have adverse effects on

920 ITLOS, MOX Plant (Ireland v. United Kingdom) , Provisional
Measures, Order of 3 December 2001, paras. 84, 89(1).
921 ITLOS, Land Reclamation by Singapore in and Around the Straits

of Johor (Malaysia v. Singapore) , Provisional Measures, Order, 8 October
2003, para. 92, quoting ITLOS, MOX Plant (Ireland v. United Kingdom),
922visional Measures, Order of 3 December 2001, para. 82.
Ibid., para. 95.
390 923
the marine environment”, and expressed the view that there

had been insufficient cooperation between the Parties prior to
924
the commencement of the proceedings. ITLOS granted

provisional measures, essentia lly requiring th e parties to

cooperate in assessing the e ffects of Singapore’s land
925
reclamation program. The claim before the Annex VII
926
arbitral tribunal was subseque ntly settled by agreement.

Again, although there had in fact been no prior EIA, that was
not the ground for the provisional measures decision, which was

based on the express UNCLOS ob ligation of cooperation in

relation to pollution of the marine environment.

8.84. Pulp Mills: The dispute essentially concerns alleged

violations by Uruguay of the 1975 Statute of the River Uruguay,

resulting from the authorization by Uruguay of the construction

of two paper mills. Issues about environmental assessment

revolved around the adequacy of the assessment carried out and

interpretation of the Statute. Colombia will reserve comment on

the case pending the Court’s judg ment. It is worth recalling

though, that “in the absence of a ny evidence of imminent harm

923 ITLOS, Land Reclamation by Singapore in and Around the Straits
of Johor (Malaysia v. Singapore), Provisional Measures, Order, 8 October
2003, para. 96.
924 Ibid., para. 97
925 Ibid., para. 106(1).
926 See ITLOS, Land Reclamation by Singapore in and Around the

Straits of Johor (Malaysia v. Singapore), Award on Agreed Terms, 1
September 2005.
391to the environment,” 927 the Court refused Argentina’s

application for provisional measures.

8.85. On analysis, only Request for an Examination concerned

a claim of per se violation of customary international law as the

result of a failure to carry out an EIA, and then only in the

alternative to reliance on an express treaty provision; 928 the

Court was able to avoid ruling on that claim for other reasons.

In each of the other cases, the failure to carry out an EIA (or the

failure to carry out an adequate EIA) was relied upon in relation

to breaches of specific treaty obligations concerning actual harm

caused or threatened. As is implicitly admitted later in the

Memorial, 929 in Pulp Mills and MOX Plant , an EIA had in fact

been carried out, and the claim was that the EIA was inadequate.

8.86. As to the Southern Bluefin Tuna cases, some comment

needs to be made about Ecuador’s presentation of what the

Tribunal actually said. 930 Rather than saying that harm “could

not be excluded” (a quotation taken from Land Reclamation 931),

what the Tribunal said was that:

927 Boyle A. et al, International Law and the Environment , 3 rded.,
Oxford Univ. Press, 2009, p. 140.
928
Request for an Examination of the Situation in Accordance with
Paragraph 63 of the Court’s Judgment of 20 December 1974 in the Nuclear
929ts (New Zealand v. France) Case, ICJ Reports 1995, p. 291, para. 6.
EM, para. 8.48.
930 EM, para. 8.48.
931 ITLOS, Land Reclamation by Singapore in and Around the Straits
of Johor (Malaysia v. Singapore), Provisional Measures, Order, 8 October
2003, para. 96.

392 “there is scientific uncertainty regarding
measures to be taken to conserve the stock of

southern bluefin tuna and that there is no
agreement among the partie s as to whether the
conservation measures take n so far have led to
the improvement in the stock of southern bluefin
932
tuna.”

To the extent that the Memorial might suggest that Southern

Bluefin Tuna concerned an EIA, it is misleading.

8.87. Ecuador presents arguments as to the supposed

requirements under general international law as to the content of

an EIA in an attempt to provide some substance to the alleged

free-standing obligation to carry out an EIA. 933 First,tothe

extent that it relies on the Espoo Convention in supporting the

supposed required content of an EI A, it must be stressed again

that this is a regional conventi on, to which neither Ecuador nor

Colombia is party. As to Ecuador’s reliance on the 1987 UNEP

EIA Goals and Principles, they are a non-binding instrument,

drafted by a Working Group of Experts and adopted by the
934
Governing Council of UNEP. They provide guidance but not

more. The ILC, by contrast, was clear that the modalities of any
935
assessment were a matter for the law of the proponent State.

932ITLOS, Southern Bluefin Tuna (New Zealand v. Japan; Australia v.
Japan), Provisional Measures, Order of 27 August 1999, para. 79.
933 EM, para. 8.54, see also paras. 8.43-8.51.
934
935 Decision 14/25 of the Governing Council of UNEP, 17 June 1987.
See above, para. 8.74.
3938.88. In short, there is no free-standing obligation under

customary international law to conduct a transboundary EIA

prior to commencing a project or program. Nor was Colombia

under a treaty obligation to do so. Even assuming that it

imposes some obligation to act, and to the extent that its breach
936
is actually relied upon by Ecuador, Article 14(1) of the
Biodiversity Convention has no ap plication in this case, since

the marginal and localised effect of spray drift (limited, at most,

to a few hundred metres) c ould not possibly have had

“significant adverse effects on biological diversity”.

8.89. At most, Colombia had to conduct an assessment, under
the terms of the applicable Colombian law, as to whether its

spraying program risked causi ng significant transboundary

harm, and if so, what mitigation measures were appropriate.

Moreover, such an assessment was not a one-off exercise; it

involved keeping the situation und er continuous review. In the

ordinary course, any issues that might have arisen from the
initial assessment could be addressed by subsequent inquiry and

consultation.

8.90. What Colombia actually did was described in some
937
detail in Chapters 4 and 6 above. In particular:

936 See above, para. 8.71.
937 See especially above, paras. 4.8-4.14, 4.23-4.30, 6.2-6.39.
394 (1) The PECIG program, though new, was based on

earlier Colombian experience duly authorised

under the Law on the Environment of 1993. 938

(2) Colombia explained its intentions at international

level on numerous occasions in 2000-2001 but

did not consider it necessa ry to consult Ecuador
939
specifically. However, when contacted by

Ecuador, Colombia proposed, first a workshop to

provide detailed information about the program,
940
then a Joint Commission which met 4 times.

(3) The Law on the Environment, the relevant

Colombian law in force in 1999, allowed for, but

did not require, an Environmental Management

Plan for the spraying program. Nevertheless, in

fact such a Plan was dir ected to be produced for

the PECIG program, entailing an assessment of

the situation.941

(4) The EMP is under continuous review and still
942
applies to the Project in amended form.

(5) The spray mixture was duly approved and

published; it was improved in 2004. 943

938
939 See above, paras. 4.8-4.10.
See above, paras. 5.5-5.6.
940 See above, paras. 5.10, 5.18.
941 See above, paras. 4.11-4.14.
942 See above, para. 4.13.
943 See above, paras. 4.45-4.46, 6.2-6.21.

395 (6) The process of assessment is a continuing one,

including scientific review of newly identified
issues: see e.g. CICAD I (2005) and CICAD II

(2009). 944

(7) A careful appraisal is c onducted twice a year by

qualified scientists with full access to
945
information.

8.91. In the circumstances, this entails full compliance with

whatever obligations of assessment existed under customary

international law at relevant times.

(b) Failure to inform and consult potentially affected
communities in Ecuador

8.92. The second alleged breach of the obligation of

cooperation concerns the failure to inform and consult
communities in Ecuador likely to be affected by the spraying.

(i) Ecuado crasim

8.93. In support of the existence of the obligation to consult as

a matter of international environmental law, reliance is placed
on Principle 5 of UNEP’s EIA Goal s and Principles and Article

13 of the Draft ILC Articles on Prevention. 946 The commentary

to draft Article 13 is said to support the argument that this

944
945 See above, paras. 3.5-3.58, 4.15-4.19.
946 See above, para. 4.26.
EM, para. 8.56. Reference is also made to the 1998 Aarhus
Convention.
396obligation also applies as re gards the population of other
States.947 Finally, reference is made to Principle 10 of the 1992

Rio Declaration and the 1998 Aa rhus Convention as supporting

the “right to information and public participation in EIA and

authorisation processes”. 948

8.94. In conclusion, Ecuador alleges that:

“these precedents show that Colombia failed in its
duty to inform and cons ult those likely to be

affected by its aerial spraying activities, violating
both the requirements of customary international
law with respect to environmental impact
assessment and the rights of those Ecuadorians

whose health, private lif949nd property have been
harmed or put at risk.”

(ii) Colombia r’sponse

8.95. Here again, the short answer to Ecuador is that, having

regard to what was and is known about the risks associated with

the components of the spray mix, the risk of spray drift and the

modalities laid down for spraying operations, Colombia had no

reason to consult the Ecuador ian public, and certainly no
international obligation to do so.

8.96. Ecuador’s argument as to the supposed obligation to

inform and consult communities in Ecuador is problematic. Is it

947 EM, para. 8.56.
948 EM, paras. 8.60-8.61.
949 EM, para. 8.62.
397to be supposed that State A is required to conduct what (under

some national procedures) would be a formal public inquiry on
the territory of State B as to th e future exercise of sovereign

authority within its own territory by State A? What would be

the significance of statements by members of the public of State

B welcoming an activity officially opposed by the Government
of State B? What would be th e significance of local opposition

to a project which the Government of State B had officially

approved? Is State A required or even entitled to refer to local
opinion in such cases? Even accepting, arguendo, that this

might be possible by agreement be tween two friendly States on

some matter of common interest, it can hardly be required by

law.

8.97. Heavy reliance is placed on the UNEP Goals and

Principles of Environmental Impact Assessment and the ILC
Draft Articles on Prevention. Bu t on this point these texts

certainly do not reflect customary international law. The UNEP

Goals and Principles are overt guidelines, using the verb

“should” throughout. The ILC’s commentary to draft Article 13
says that:

“This article is inspired by new trends in

international law, in general, and environmental
law, in particular, of se eking to involve, in the
decision-making processes, individuals whose
lives, health, property and environment might be
affected by providing them with a chance to

398 present their views a nd be heard by those
responsible for making the ultimate decisions.” 950

This proclaims Article 13 to be an exercise in “progressive

development”, a fortiori as concerns consultation with the
951
public of the other State in a transboundary situation.

8.98. The texts relied on in the commentary to illustrate these

“new trends”, while they encourage consultation, do not support

the idea of a legal obligation to consult a foreign public. To take
952
just two examples, Principle 10 of the Rio Declaration deals

exclusively with consultation “[a]t the national level” and

below; it does not mention trans boundary issues. Article 6 of
953
the Framework Convention on Climate Change provides that
the Parties, in their role as pr omoters of the Convention, shall

“facilitate at the na tional and, as appropria te, subregional and

regional levels, and in accord ance with national laws and

regulations, and within their re spective capacities” consultation

on a number of matters including “public participation in

addressing climate change and its effects and developing

adequate responses”. Quite apart from the sequence of provisos

in the chapeau of Article 6, this is very far from the present case.

950 ILC Yearbook 2001 Vol. II(2), 165, para. (3).
951 See also the observa tions of the Special Rapporteur (PS Rao) in the
debate in the ILC in 2001: 2675th Meeting, 11 May 2001; ILC Yearbook
2001, Vol. I, p. 66 (para. 55) (quoted above, para. 8.21).
952
953 Cited in ILC Yearbook 2001 Vol. II(2),, 166 (para. 4).
Cited Ibid. (para. 5).
3998.99. The weakness of Ecuador’s non-consultation claim is
evidenced by its attempt to recy cle its arguments as to human

rights obligations as evidence of practice showing the existence

of an obligation of foreign public consultation as a matter of

customary international law. But the recognition of the right to

information in each of the cases cited was based on

interpretation of the substantiv e provisions of the human rights

instrument in question, some of which are not applicable as
between the parties to the disput e. Moreover none of the cases

involved transboundary harm. In at least some of them (e.g.,

Öneryildiz v. Turkey before the European Court of Human
954
Rights ), the relevant observations were made only in passing,

in circumstances in which the principal concern of the case was

violation of an entirely different substantive right (e.g. the right

to life) which had actually taken place.

8.100. Ecuador concludes that “these precedents show that

Colombia failed in its duty to inform and consult those likely to
955
be affected by its aerial spraying activities”. eh

“precedents” show nothing of the kind. At most they show that,

on the one hand, certain States in Europe have taken a conscious

decision to enter into treaty obligations requiring them to
provide information and to consult with potentially affected

954
ECHR (GC), Judgment of 30 November 2004, cited EM, para.
9.104; it may also be noted that Öneryildiz did not in fact concern
environmental harm as such, but ac tivities potentially hazardous to human
life.
955 EM, para. 8.62.
400individuals in certain circumstances. On the other, they show

that human rights monitoring bodies have interpreted certain

substantive human rights obligations as implying, subsidiarily, a

need for provision of information and consultation with

individuals. But what Ecuador must show is the existence of an

obligation binding on Colombia that consultation with and
provision of information to Ecuadorian nationals is positively

required of it in circumstances such as the present. This it has

failed to do.

(c) Failure to cooperate in the control of transboundary
risks

8.101. Finally, Ecuador puts forward a catch-all claim alleging

failure to cooperate in the cont rol of transboundary risks arising

from the aerial spraying.

(i) Ecuado crasim

8.102. Ecuador asserts that much of the harm alleged could

have been prevented if Colombia had cooperated with Ecuador

“by notifying it of the intende d operation, sharing information

on the chemicals in use and their likely effects on public health
and the environment, and on ways to reduce or eliminate the risk

to Ecuadorian territory and its inhabitants”. 956 It is alleged that

Colombia failed to cooperate in any of these ways. 957The

allegation of failure to provide information or notification to

956 EM, para. 8.63.
957 EM, para. 8.63.
401individuals in Ecuador is recycled as showing a breach, and it is
further alleged that Colombia failed to “consult with Ecuador as

required by international law at an early stage, or in good faith,

or at all.”58

8.103. In support of the existence of a general obligation to

cooperate, it is argued that

“Case law, multilateral and bilateral treaties, the
1988 Narcotics Convention, ILC codifications,

and 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

international law to cooperate in order to control
and minimize the risk of transboundary harm.
They must give each other prior notice of the
activity, provide adequate information about the
substances used, and the risks 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.” 959

However, that submission is immediately followed by a caveat:

“even if quod non notification and consultation in
cases of transboundary ri sk are not independent
customary rules, non-compliance with them is
strong evidence of a failure to act diligently in

protecting oth960States from harm under Rio
Principle 2.”

958 EM, para. 8.63.
959 EM, para. 8.68.
960 EM, para. 8.68.
4028.104. As to breach, it is asserted that “in the course of

Ecuador’s repeated attempts to negotiate a solution Colombia
did not cooperate in good faith”. 961Reference is made to

Colombia’s undertaking to provide advance notification of any

further planned spraying in the border area, and the position

taken by Colombia in the bilateral scientific commission. 962

(ii) Colombia r’sponse

8.105. Yet again, most of what needs to be said on these issues

has been said in Chapters 3-6 above. It has been shown that

Colombia did respond to Ecuador’s initial approach, that it did

provide information including information about the spray mix,

and that it even voluntarily susp ended spraying in the border

region – to its considerable cost. Assuming, arguendo, a

freestanding obligation to cooperate with Ecuador, Colombia in

fact did cooperate – it bei ng understood that there was no
obligation to agree and that Ecuador had no veto over the

conduct of the spraying program on Colombian territory.

8.106. Further, Ecuador’s argument that much of the alleged

damage “could have been avoided or minimised had Colombia
963
cooperated with Ecuador at the outset” presupposes that

Colombia was required to bow to Ecuador’s demands that aerial

spraying should not take place within a buffer zone along the

961
962 EM, para. 8.70.
963 EM, para. 8.71.
EM, para. 8.63.
403border. However, Colombia was not required as a matter of

international law to accede to any such demands in relation to

activities carried out within its own territory and which did not

cause any significant harm within Ecuador.

8.107. In Lac Lanoux, the Tribunal expressly rejected the
suggestion that, as a matter of customary international law or as

a general principle of law France’s proposed project was subject

to any prior agreement with Spain, and held that such a
964
requirement could only be imposed by treaty.

8.108. Ecuador emphasises the Tribunal’s holding that
conflicting interests had to be reconciled by “n egotiation and

mutual concession” and that France was required to give “a

reasonable place to Spain’s inte rests in the solution finally
965
adopted”. In fact, as to the question of negotiation and

mutual concession, the Tribunal simply observed that, even if it

were accepted that there is an obligation prohibiting an upstream
State on an international river from altering the waters in such a

way as seriously to prejudice a downstream State (a question it

did not decide):

“En realité, les Etats ont aujourd’hui parfaitement

conscience de l’importance des intérêts
contradictoires, que met en cause l’utilisation
industrielle des fleuves internationaux, et de la
nécessité de les concilier les uns avec les autres

964 Lac Lanoux (Spain/France), RIAA, vol. XII, p. 281, 308 (1957).
965 EM, para. 8.65.
404 par des concessions mutuelles. La seule voie
pour aboutir à ces compromis d’intérêt est la

conclusion d’accords, s966une base de plus en
plus compréhensive.”
That is a statement recognizi ng the practical necessity of

concluding agreements reconciling the interests of the States

involved rather than a statem ent of an obligation under

international law to do so. The Tribunal continued:

“La pratique internationale reflète la conviction
que les Etats doivent tendr e à conclure de tels
accords; il y aurait ainsi une obligation

d’accepter de bonne foi tous les entretiens et les
contacts qui doivent par une large confrontation
d’intérêts et par une bonne volonté réciproque,
les mettre dans les meilleures conditions pour
conclure des accords.”

That comes closer to an acceptance of an obligation to negotiate,

but it must be read in the light of the Tribunal’s earlier

observations. By way of preface to its discussion of whether
there was a treaty obligation to reach an agreement prior to

commencing the works, the Tribunal rejected the Spanish

suggestion that there existed any rule of international law giving

one State the right of veto ove r works of another on its own
territory on the basis that it might affect the interests of the first

State. In that context it observed that:

“la pratique internationale recourt de préférence à

des solutions moins extrêmes, en se bornant à
obliger les Etats à rechercher, par des tractations
préalables, les termes d’un accord, sans
subordonner à la conclusi on de cet accord

966 Lac Lanoux (Spain/France), RIAA, vol. XII, p. 281, 304 (1957).

405 l’exercice de leurs compétences. On a ainsi parlé,

quoique souvent d’une manière impropre, de
‘l’obligation de négocier un accord.’ En réalité,
les engagements ainsi pris par les Etats prennent
des formes très diverses et ont une portée qui

varie selon la manière dont ils sont définis et
selon les procédures destinées à leur mise en
oeuvre; mais la réalité des obligations ainsi
souscrites ne saurait être contestée et peut être
sanctionnée, par exemple, en cas de rupture

injustifiée des entretiens, de délais anormaux, de
mépris des procédures prévues, de refus
systématiques de prendre en considération les
propositions ou les intérêts adverses, plus

généralement967 cas d’infraction aux règles de la
bonne foi.”

8.109. There is of course a difference between the law of

international rivers and transboundary air pollution: the former
is much more developed, whereas the latter is embryonic – as

the Chernobyl accident showed. In consequence the Economic

Commission for Europe has b een active in pursuing treaty

solutions, including important protocols to earlier texts. But this

treaty practice finds little echo in other regions.

8.110. Ecuador’s complaint about non-cooperation has to be

seen in context. Far from seeking to mitigate risk, from the first

and without any evidence, Ecuador raised questions about the
968
very need for the existence of the spraying program. Asfar

as the record shows, at no stage did it acknowledge or appear to

967 Ibid., 306-307.
968 See above, para. 5.2-5.3.
406take into account the enorm ous social, environmental and

economic problems presented by the illicit drug trade for

Colombia. (It does not do so in the Memorial either.) 969 By

contrast, other States (notably the United States, Colombia’s

partner in the spraying program) and international organisations

(UNODC, OAS-CICAD) have been supportive. 970

8.111. Subsequent developments have included the following:

(1) Colombia pressed for the creation of a Joint

Scientific and Technical Commission on Aerial
971
Spraying and sought to facilitate its work. But

Ecuadorian members failed to attend 7 of 8
972
proposed site visits.

(2) In November 2003, Colombia gave Ecuador

copies of relevant documents, including the

revised EMP. 973

(3) Subsequent Colombian offers for Ecuador to
974
observe spray missions were declined.

(4) So too was an offer to Ecuador, within the

framework of the Commission, to test the impact

969
See above, paras. 5.2-5.44, 5.64-5.108; and cf. EM, paras. 2.54-
9706.
See above, paras. 3.41-3.58.
971 See above, para. 5.18-5.19.
972 See above, para. 5.29.
973 See above, para. 5.19, and for the list of documents see Annex 9.
974 See above, para. 5.66.

407 of aerial spraying from the Colombian side of the
975
border.

(5) The proceedings of the Second Joint Commission

were pre-empted, then aborted, by Ecuador. 976

(6) Ecuador complains that it was not involved from

the beginning as a co-sponsor of the CICAD
977
study. That study covered the whole of

Colombia, and Colombia had no obligation to

involve its neighbour in this way. But it did

inform Ecuador, at the highest level, of the

proposed CICAD study, did invite its

participation, and did separately inform it of the

outcome. 978

(7) In September 2006, Colombia invited Ecuador to
979
participate in CICAD II; CICAD itself had

also extended that invitation in August 2006. 980

Again Ecuador declined. 981

8.112. In the circumstances, Colombia is at least as justified in

claiming non-cooperation as Ecuador. What Ecuador really

sought – and still seeks – is a veto over spraying near the border.

975 See above, para. 5.65.
976 See above, para. 5.94-5.102.
977 See EM, para. 3.36.
978
979 See above, paras. 4.15-4.19.
980 See Annex 24.
See Annex 24, at p. 2, and Annex 25.
981 See Annex 25.
408Quite apart from evidentiary cons iderations, this is not a claim

that can be based on an obligation to cooperate, as the Tribunal
in Lac Lanoux made clear. 982

(3) B REACH OF ARTICLE 14(2) OF THE 1988N ARCOTICS
C ONVENTION

8.113. One of the features of Ecuador’s Memorial is that it
repeatedly relies on assertions about customary international law

when there is an applicable tr eaty between the parties dealing

precisely with the substance of the claim. In relation to

cooperation, the relevant treaty is the 1988 Narcotics

Convention. As noted, the purpo se of the 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”. Most of the provisions of the
Convention deal with modaliti es of cooperation, including

cooperation in eradication of illicit coca production.

8.114. It is significant that in Section II of Chapter VIII, dealing

with Ecuador’s non-cooperation claim, the 1988 Convention is

mentioned only once, a brief and ge neric reference as part of a
983
list of sources. But the explicit le gal basis for cooperation

between the parties was not the inapplicable treaties of the

982
Lac Lanoux (Spain/France), RIAA, vol. XII, p. 281, 308 (1957);
983ve, para. 8.107.
EM, para. 8.68.
409UNECE: it was the 1988 Convention. 984 Ecuador has an

entirely selective approach to the 1988 Convention, relying on it

for jurisdiction and for the in corporation by reference of

environmental and human rights norms, but virtually ignoring it

for all other substantive purposes. This becomes most obvious
in Section III of Chapter VIII of the Memorial, when Ecuador –

having struggled in Section II to construct from disparate

materials an obligation to cooperate focusing only on its

concerns – specifically turns to the 1988 Convention.

8.115. Section III of Chapter VIII is headed “Cooperation and

Respect for Fundamental Human Rights and Protection of the

Environment as Required by the 1988 Narcotics Convention”.

Though a reasonable summary of Ecuador’s case, this is already

a curious reading of the Conven tion, which is treated as if it
were a human rights or environmental law treaty.

8.116. Not merely does Ecuador focus on Article 14 out of

context from the Convention as a whole; it focuses on Article
14(2) out of context from Article 14 as a whole. Significantly,

Article 14 is entitled “Measures to Eradicate Illicit Cultivation

of Narcotic Plants and to Eliminate Illicit Demand for Narcotic

Drugs and Psychotropic Substances”. It provides in full as

follows:

984 See also the Preamble, which emphasises that “eradication of illicit
traffic is a collective responsibility of all States and that, to that end, co-
ordinated action within the framework of international cooperation is
necessary”.
410“1. Any measures taken pursuant to this
Convention by Parties sha ll not be less stringent
than the provisions applicable to the eradication
of illicit cultivation of plants containing narcotic

and psychotropic substances and to the
elimination of illicit demand for narcotic drugs
and psychotropic substances under the provisions
of the 1961 Convention, the 1961 Convention as
amended and the 1971 Convention.

2. Each Party shall take appropriate
measures to prevent illic it cultivation of and to
eradicate plants containing narcotic or
psychotropic substances, such as opium poppy,
coca bush and cannabis plants, cultivated illicitly
in its territory. The measures adopted shall

respect fundamental human rights and shall take
due account of traditional licit uses, where there
is historic evidence of such use, as well as the
protection of the environment.

3. a) The Parties may co-operate to
increase the effectiveness of eradication
efforts. Such co-operation may, inter alia,
include support, when appropriate, for
integrated rural development leading to
economically viable alternatives to illicit
cultivation. Factors such as access to

markets, the availability of resources and
prevailing socio-economic conditions
should be taken into account before such
rural development programmes are
implemented. The Parties may agree on
any other appropriate measures of co-
operation.

b) The Parties shall also facilitate the
exchange of scient ific and technical
information and the conduct of research

concerning eradication.

411 c) Whenever they have common
frontiers, the Parties shall seek to co-
operate in eradication programmes in
their respective areas along those

frontiers.
4. The Parties shall adopt appropriate
measures aimed at eliminating or reducing illicit
demand for narcotic drugs and psychotropic

substances, with a view to reducing human
suffering and eliminating financial incentives for
illicit traffic. These measures may be based, inter
alia, on the recommendations of the United
Nations, specialized agencies of the United
Nations such as the World Health Organization,
and other competent inte rnational organizations,

and on the Comprehensive Multidisciplinary
Outline adopted by the International Conference
on Drug Abuse and Illicit Trafficking, held in
1987, as it pertains to governmental and non-
governmental agencies and private efforts in the
fields of prevention, treatment and rehabilitation.
The Parties may enter into bilateral or

multilateral agreements or arrangements aimed at
eliminating or reducing illicit demand for
narcotic drugs and psychotropic substances.
5. The Parties may also take necessary

measures for early destruction or lawful disposal
of the narcotic drugs, psychotropic substances
and substances in Tabl e I and Table II which
have been seized or confiscated and for the
admissibility as evidence of duly certified
necessary quantities of such substances.”

8.117. The relevant obligation of cooperation here is in Article

14(3)(c): the parties are obliged to “seek to co-operate in

eradication programmes in thei r respective areas along those
frontiers”. In fact, not only does Ecuador not cooperate with the

412PECIG program: it has not been able to prevent its territory

being used by armed groups lin ked to the drug trade, and

provides a major route for getting coca leaf processed and traded
985
out of the region.

8.118. Ecuador accepts that cooperation is required when

carrying out drug eradication in accordance with the 1988

Narcotics Convention, referring in particular to Articles 14(3)(b)
and (c).986 But the bulk of Section III is devoted to its argument

that any breach of obligations under environmental law in the

course of an eradication progr am such as PECIG necessarily

breaches Article 14(2) of the Convention. 987 These arguments

have already been fully discussed. 988

8.119. Article 14(2) refers to “respect [for] fundamental human

rights”, and to taking “ due account of traditional licit uses,

where there is historic evidence of such use, as well as the

protection of the environment”. To the extent that these

stipulations refer to human and indigenous rights, they will be

discussed in Chapter 9. As to the requirement to take “due

account” of the protection of the environment, this is subsumed

under the general international law relating to transboundary

harm; a State which exercises due diligence in accordance with

985 See the findings of the Commission on Transparency and Truth for
the Angostura Case, Quito, December 2009 , as summarised quoted, paras.
2.23-28.
986 EM, para. 8.71.
987
988 EM, para. 8.71 et seq.
See above, paras. 8.10-8.18.
413the international law standard will thereby respect the protection

of the environment.

8.120. Ecuador also alleges that Colombia has breached other

provisions of the 1988 Convention (Articles 14(3)(b) and (c)),

but it does so only briefly. 989 It is noticeable that no reference is

made to breach of any provision other than Article 14(2) in the
990
conclusions of Chapter VIII.

8.121. Obviously, the formulation of neither provision provides

a firm basis for making a credible claim of breach by Colombia.
Article 14(3)(b) speaks merely of facilitating the exchange of

information and technical data, while Article 14(3)(c) uses the

formulation “seek to cooperate”. For the reasons already

given, 991Colombia has complied with its obligations under

these provisions. Whether Ec uador has done so is another

question.

E. Conclusions

8.122. This chapter reaches the following conclusions:

(1) The principal source of applicable law in this

case is the 1988 Narcotics Convention, notably

Article 14, which expressly deals with
eradication of illicit coca crops. Article 14

989
990 EM, para. 8.72.
991 EM, para. 8.84.
See above, paras. 8.113-8.121.
414 simply requires the State concerned to “respect

fundamental human rights and … take due
account of traditional licit uses, where there is

historic evidence of such use, as well as the

protection of the environment”.

(2) The ILC Articles on Prevention do not, as a

general matter, reflect customary international
law, and their reception has been luke-warm at

best. In any event, they do not apply in the

present case, since the threshold criteria in

Article 2(a) are not met.

(3) The principle of sovereignty does not mean that

States are completely immune from
transboundary effects of act ivities carried out on

the territory of other States. The threshold in this

context is a real risk of causing significant harm.

(4) Colombia hasstrictly complied with its

international obligations under the treaties to
which it is a party, notab ly under Article 14 of

the 1988 Narcotics Convention.

(5) Colombia’s aerial spraying program was adopted

and implemented with all due diligence. Further,

it has not been shown to have caused any harm to
Ecuador or its nationals.

415(6) The only treaty obligation of Colombia to

conduct an EIA was contained in Article 14(1) of
the Convention on Biological Diversity. That

provision did not, on any view, require an EIA in

the present case.

(7) At most, Colombia had to conduct an assessment,

under the terms of the applicable Colombian law,
as to whether its spraying program risked causing

harm, and if so, what mitigation measures were

appropriate. Such an assessment involved

keeping the situation u nder continuous review.
Colombia did all of these things.

(8) Having regard to the facts, Colombia had no
reason to consult the Ec uadorian public, and no

international obligation to do so.

(9) In any event, Colombia did not fail to cooperate

with Ecuador in the implementation of the aerial

spraying program. What Ecuador really sought
was a veto over any sprayi ng near the border, to

which it was not and is not entitled.

416 Chapter 9

ECUADOR’S CLAIMS BASED ON HUMAN AND
INDIGENOUS RIGHTS

A. Introduction

9.1. Ecuador, not content with relying on a multitude of

environmental law instruments as a basis of obligation, doubles

up by extensive reliance on human rights treaties (incorporating

the rights to life, health, food,water, a healthy environment,
property, humane treatment, private life, and information) and

on treaties concerning indigenous rights.992 This raises serious

issues of principle as to the relations between established or de

lege ferenda standards of interstate conduct – e.g. with respect

to transboundary air pollution or international watercourses –
and standards concerning individual or peoples’ rights. It must

be stressed again that the people of Ecuador are not the targets

of the aerial spraying program: the target is illicit coca crops in

Colombia. Ecuador’s claim is that spray drifts across the
boundary, causing significant harm in Ecuador. That claim has

not been established as a matter of fact; it is inconsistent with

the scientific evidence, and it is unsustainable under the

applicable treaties – the 1988 Narcotics Convention and the

992 EM, Chapter IX.

4171992 Biodiversity Convention (to the extent that there is a claim

of violation of that latter treaty) – or under customary

international law. That being so, the key point is that a bad

environmental claim does not beco me any better by being re-

presented as a human rights or indigenous rights claim. The

factual underpinning for the claim is still the same, with exactly
the same defects and lacunae.

9.2. It is proposed to illustrate this fundamental point in

further detail under each rubric , starting with the alleged

breaches of human rights.

B. Alleged Breaches of Human Rights

9.3. For the reasons already given, and despite the vagueness

and generality of the language used, it will be assumed here that

Ecuador is only claiming on beha lf of Ecuadorian nationals
993
whose alleged injuries occurred in Ecuador.

(1) R ELATIONSHIP OF HUMAN RIGHTS AND ENVIRONMENTAL
CLAIMS

9.4. At the outset, Ecuador asserts that the aerial spraying

violates obligations in “three di stinct but interrelated areas of

international law”: protection of the environment; human rights,
994
and the protection of indigenous peoples. It goes on to assert

that the “relationship between these three distinct areas of

993 See above, paras. 1.14-1.25.
994 EM, para. 9.1.
418international law lies at the heart of this case”: the

interrelationship is said to arise from the fact that the aerial

spraying has “significantly harmful consequences for the health

and well-being of the people, natural resources and environment
in the affected areas of Ecuador”. 995

9.5. In principle it is of course true that the same conduct can

breach international obligations in different fields. But there are
two important qualifications.

9.6. First, each breach has to be independently established: it

has to be proved as a matter of fact. The conclusion that
particular conduct may have breached the obligation of

prevention of transboundary harm does not entail there has been

a breach of the human rights of those allegedly affected.

9.7. Secondly, however, the conclusion that conduct is lawful
under general international la w has implications for its

lawfulness as a matter of inte rnational human rights Respect

should be paid to the principl e of speciality identified by the

Court in a related context in the Advisory Opinion on Legality
of the Threat or Use of Nuclear Weapons. There it was alleged

that the threat or use of nucl ear weapons in any circumstances

(even in self defence) would violate the human right to life. The

Court stated that the content of that human right could not be

995 EM, para. 9.2.

419determined independently of the most relevant applicable law,

the law relating to the means a nd methods of war. The Court

observed:

“that the protection of the International Covenant

of Civil and Political Rights does not cease in
times of war, except by operation of Article 4 of
the Covenant whereby certain provisions may be
derogated from in a time of national emergency.

Respect for the right to life is not, however, such
a provision. In principle, the right not arbitrarily
to be deprived of one’s life applies also in
hostilities. The test of what is an arbitrary
deprivation of life, however, then falls to be

determined by the applicable lex specialis ,
namely, the law applicable in armed conflict
which is designed to regulate the conduct of
hostilities. Thus whether a particular loss of life,

through the use of a certain weapon in warfare, is
to be considered an arbitrary deprivation of life
contrary to Article 6 of the Covenant, can only be
decided by reference to the law applicable in
armed conflict and not deduced from the terms of
996
the Covenant itself.”
The same is true here, mutatis mutandis. 997 A new set of norms

and balances concerning transboundary harm is not to be

“deduced from the terms of the Covenant itself”, when

international law already lays down the relevant standard.

9.8. International human rights law has a distinct function

from international environmen tal law, and overlap between

996 Legality of the Threat or Use of Nuclear Weapons, Advisory
Opinion, I.C.J. Reports 1996, p. 240, para. 25.
997 On the relation between nuclear weapons and the environment cf.
Ibid., 243 (para. 33).
420them is incidental, not integral . It is not the function of

international human rights law to protect the environment as

such, though in certain serious cas es harm to the environment

may impair the human right to lif e or property. However, in

such circumstances, and given th e facts of the present case,
Ecuador’s assertion that the breaches alleged by it constitute “an

integrated, mutually reinforcing whole”998is a mere construct.

(2) A PPLICABILITY OF HUMAN RIGHTS OBLIGATIONS TO

TRANSBOUNDARY CLAIMS

9.9. Ecuador relies on a great variety of international human

rights instruments in this part of its case. These include:

• the International Covenant on Civil and Political

Rights (“ICCPR”);

• the International Covena nt on Economic, Social

and Cultural Rights (“ICESCR”);

• the American Convention on Human Rights
(“ACHR”);

• the 1989 Convention on the Rights of the Child
(“CRC”);

• the Convention on the Elim ination of All Forms

of Discrimination Against Women (“CEDAW”);

• the 1988 Additional Protocol to the ACHR in the

Area of Economic, Social and Cultural Rights

998 EM, para. 6.6.
421 (the Protocol of San Salvador) (“the ACHR

ESCR Additional Protocol”); and

• ILO Convention No. 169 Concerning Indigenous

and Tribal Peoples in Independent Countries

(“ILO Convention No. 169”);

9.10. Ecuador accepts that the human rights treaties it relies

upon “generally require a State Pa rty to guarantee the relevant
rights and freedoms for every person within its territory”. 999 But

it asserts that those treaties:

“are not, however, limited in their territorial
scope; they impose obligations which are
violated when Colombia authorises actions in its
own territory that have consequences across the

boundary, particularly where – as in the present
case – Colombia and Ecuador are part of the
shared legal space to which these instruments
apply.” 1000

9.11. Ecuador’s reference to the fact that Ecuador and

Colombia “are part of the shared legal space to which these
instruments apply” recalls the comment of the European Court

as to the “ espace juridique” of the European Convention in its

decision on admissibility in Bankovi ć.1001 Quite apart from

whether that remark formed part of the ratio of the Court’s

999
1000 EM, para. 9.9.
1001 EM, para. 9.9.
Banković v. Belgium and 16 Other Contracting States (App. no.
52207/99), Decision on Admissibility of 12 December 2001;
ECHR 2001-XII [GC], para. 80.
422decision, 1002it is an open question whether the Banković

principle applies to the human rights treaties to which Colombia

and Ecuador are actually parties.

9.12. But in any case, Banković was not a case of alleged

transboundary harm caused in State B by a lawful activity
carried out entirely on the territory of State A. The question of

the applicability of the various human rights treaties has to be

considered in light of the fact s of each case: here, the claim

concerns incidental transboundary harm allegedly caused by

aerial spraying in Colombia. It is to be emphasised that the

present case does not concern alle gations of extra-territorial
State action: the actions of Colombia are intra-territorial, and it

is only the alleged effects of those actions which are extra-

territorial. Thus the first quest ion is whether the various human

rights treaties apply to this case – whether human rights drift

across the boundary rivers with the (alleged) residual spray.

9.13. Put more generally, the question is whether a new

international law of transboundary harm is secreted in the

interstices of the ICCPR and ot her human rights instruments.

1002 As to which, see the various subsequent cases concerning State
action outside the territory of the Contracting States: see Issa, Omer,
Ibrahim, Murty Khan, Muran and Omer v. Turkey (App. no. 31821/96),
judgment of 16 November 2004 (Turkish action in Iraq); Pad and Others v.
Turkey (App. No. 60167/00), decision on admissibility of 28 June 2007
(Turkish action in Iran); andAl-Saadoon and Mufhdi v. United Kingdom
(App. No. 61498/08), decision on admissibility of 30 June 2009 (United
Kingdom action in Iraq).
423That raises an issue of principl e. Colombia’s aerial spraying

program is not directed at individuals or groups in Ecuador. If

they are affected by it (quod non), this is merely incidental to its

actual and intended target, the cultivation of illicit hard drugs in
Colombia. That program may or may not be unlawful under the

international law relating to tr ansboundary harm, its prevention

and mitigation. If it is unlawful there is no need to refer to

human rights or the rights of indigenous peoples. If (as
demonstrated in Chapter 8) it is internationally lawful, then it

does not have to survive independent scrutiny under the

international law of human or indigenous rights. As an external

actor acting on its own territory in the public interest, Colombia
cannot be presumed to have obligations to individuals or groups

within Ecuador which exceed its obligations to Ecuador itself.

9.14. Ecuador wants to have it bo th ways. It wants to

represent the people (including the indigenous people) of
Ecuador before this Court. But it wants those people (including

the municipalities, which are organs of the State) to be treated as

individuals with independent rights which it can invoke on their

behalf, and which they can simultaneously invoke before
foreign courts.1003 As the Court implied in the Advisory Opinion

on Legality of the Threat or Use of Nuclear Weapons , the

international law of human rights does not provide a ground for

1003 For the Dyncorp litigation (in which international law provides,
supposedly, the cause of action) see above, paragraph 1.35.

424completely rewriting the law of interstate relations. 1004 Thatis

true of the jus ad bellum and the jus in bello . It is equally true

of those aspects of the law of peace which deal with

transboundary harm occurring inci dentally in the course of

otherwise lawful activities of a State on its own territory. That
law strikes a certain balance betw een the rights and interests of

the two States, and therefore of the national communities they

represent. In respect of conduct not targeted at individuals or

groups within the State but merely incidentally affecting them,

the international law of human ri ghts or indigenous rights does
not supervene. For example, if the international law of the sea

gives State A the right to inte rcept a foreign ship on the high

seas, then that right is not ta ken away because passengers on

board the ship are thereby affected in terms of their human right

to freedom of movement. The balance, having been struck
between the relevant interests of two States and their respective

communities, does not have to be recalibrated by reference to

the potentially disparate interests of individuals within one of

those two States.

9.15. For these reasons, in Colombia’s view the incidental

effects (if any) on individuals or groups in Ecuador of the

internationally lawful conduct of Colombia in its own territory

do not in principle raise issues of the human rights of those

1004 Legality of the Threat or Use of Nuclear Weapons, Advisory
Opinion, I.C.J. Reports 1996 , p. 240, para. 25. Cf., as to the environment,
Ibid., pp. 242-243, paras. 30-33.
425persons or groups. But in any ev ent, the substantive provisions

of the applicable treaties do not purport to apply to the
incidental extra-territorial eff ects of the conduct of a foreign

State acting otherwise lawfully on its own territory. This will

now be demonstrated, using as examples the ICCPR, the

IACHR and the ICESR.

(a) ICCPR

9.16. Both Ecuador and Colomb ia are party without

reservation to the ICCPR and to Optional Protocol 1. Article 2
of the ICCPR, governing the scope of its application, provides:

“Each State Party to the present Covenant
undertakes to respect a nd to ensure to all
individuals within its territory and subject to its
jurisdiction the rights r ecognized in the present
Covenant, without distinction of any kind, such
as race, colour, sex, la nguage, religion, political
or other opinion, national or social origin,

property, birth or other status.”

9.17. Ecuador selectively quotes the Court’s Advisory

Opinion on The Wall in support for the proposition that the

human rights treaties on which it relies are not limited in their

territorial scope. However, the Court’s approach must be read
in its proper context. The Court addressed the interpretation of

Article 2 in considering the a pplicability of the substantive

provisions of the ICCPR (as we ll as of the ICESCR and CRC)
to Israel’s actions in the Occupied Palestine Territories (OPT). It

framed the issue as follows:

426 “whether the two international Covenants and the
[CRC] are applicable only on the territories of

the States parties thereto or whether they are also
applicable outside those territories and, if so, in
what circumstances.” 1005

9.18. The Court observed:

“This provision can be interpreted as covering
only individuals who are both present within a
State’s territory and subject to that State’s

jurisdiction. It can also be construed as covering
both individuals present within a State’s territory
and those outside that territory but subject to that
1006
State’s jurisdiction.”
The Court preferred the latter construction, observing that:

“while the jurisdiction of States is primarily
territorial, it may sometimes be exercised outside

the national territory. Considering the object and
purpose of the International Covenant on Civil
and Political Rights, it would seem natural that,
even when such is the case, States parties to the

Covenant should be bound to comply with its
provisions.” 1007

The Court concluded that

“the [ICCPR] is applicab le in respect of acts

done by a State in the exercis1008 its jurisdiction
outside its own territory.”

1005 Legal Consequences of the Construction of a Wall in the Occupied
Palestinian Territory, I.C.J. Reports 2004,p. 178, para. 107.
1006 Ibid., p. 179, para. 108.
1007
1008 Ibid., p. 179, para. 109.
Ibid., p. 180, para. 111 (emphasis added).
4279.19. In accordance with the Court’s observations in The Wall,

the alleged victim either is on th e territory of the State and thus
within its authority, or else is outside the State’s territory but is

nevertheless subject to the State’s jurisdiction as a result of the

actions of the State’s agents “in th e exercise of its jurisdiction”.

The Court does not seem to have envisaged any other
possibility: in particular, there is no mention of the situation in

which an individual is outside th e State’s territory (and outside

its “jurisdiction”, as that te rm is normally understood), but
nevertheless is injured as a by-product of lawful action taken by

the State within its own territory.

9.20. To summarize, the clear import of the Court’s Opinion
in The Wall is that the two conditions of territory and

jurisdiction in Article 2 are to be interpreted disjunctively: it is

sufficient either that an individu al is on a State’s territory, or
that he is “subject to its jurisdic tion”. Given that individuals in

Ecuador are not “within [Colombia’s] territory”, the crucial

question remains whether they may be said to be “subject to

[Colombia’s] jurisdiction”. Gi ven the Court’s reference in The
Wall to the “primarily territorial” character of jurisdiction, the

words “subject to its jurisdicti on” in Article 2 require some

“exercise of jurisdiction”, in the sense in which that term is
normally understood in public international law, before

applicability of the ICCPR is engaged.

4289.21. Ecuador relies on the deci sions of the Human Rights

Committee in López Burgos and Celiberti, but these are of little
assistance. Both cases again concerned the extra-territorial

actions of State agents and the Committee’s reasoning on its

face has no bearing on the question whether or not an individual

in a State A is able to complain of the effects of the action of
another State, State B, taken in its own territory which may

incidentally have some impact on the exercise of that

individual’s rights. The PECIG pr ogram has been carried out in
Colombian territory.

9.22. The mere fact that actions of Colombia within its own

territory could arguably have some effects on individuals across
the border within Ecuador is not sufficient to render them

“subject to its jurisdiction” for the purposes of Article 2 ICCPR.

9.23. Since the Court’s Advisory Opinion in The Wall , the

Human Rights Committee has expressed its views as to the

scope of Article 2 in General Comment No. 31 . As regards the

applicability of the ICCPR, the Human Rights Committee has
said:

“Article 2 defines the scope of the legal

obligations undertaken by States Parties to the
Covenant. A general oblig ation is imposed on
States Parties to respec t the Covenant rights and

429 to ensure them to all indi viduals in their territory
1009
and subject to their jurisdiction.”
Later on, it observed:

“States Parties are re quired by article 2,
paragraph 1, to respect and to ensure the
Covenant rights to all persons who may be within
their territory and to all persons subject to their
jurisdiction. This means that a State party must

respect and ensure the rights laid down in the
Covenant to anyone within the power or effective
control of that State Pa rty, even if not situated
within the territory of the State Party . As
indicated in General Comment 15 … the

enjoyment of Covenant rights is not limited to
citizens of States Parties but must also be
available to all indivi duals, regardless of
nationality or statelessness, such as asylum
seekers, refugees, migrant workers and other

persons, who may find themselves in the territory
or subject to the jurisdiction of the State Party.
This principle also applies to those within the
power or effective control of the forces of a State
Party acting outside its territory, regardless of the

circumstances in which such power or effective
control was obtained, such as forces constituting
a national contingent of a State Party assigned to
an international pea ce-keeping or peace-
enforcement operation.” 1010

9.24. As in the Committee’s previous decisions in López

Burgos and Celiberti, the emphasis is on whether an individual

is within a State’s “power or effective control” within the

1009 General Comment No. 31; Nature of the General Legal Obligation
Imposed on States Parties to the Covenant, 29 March 2004, CCPR/C/21/Rev.
1/Add.13, para. 3.
1010 Ibid., para. 10.
430context of extended operations such as peace-keeping

operations. There is no indication that the notion of jurisdiction

for these purposes encompasses a situation such as that at issue

in the present case.

9.25. Colombia is not aware that any of the global or regional
international human rights instru ments has been interpreted in

such a way that they would apply to the present circumstances.

The most important decision is probably that of the Grand

Chamber of the European Court of Human Rights in
1011
Banković, which concerned the meaning of the formulation

“within their jurisdiction” in Article 1 of the European
Convention, and which, by extens ion, is illuminating as to the

meaning of the similar notion of “subject to its jurisdiction” in

Article 2 of the ICCPR.

9.26. In Banković, the argument that State action could

“affect” human rights extraterritorially was explicitly rejected
by the European Court of Huma n Rights in circumstances in

which the State action in question (the bombing of a TV station)

was much more obviously targeted on the victims of those

actions than in the present case.

1011 Banković v. Belgium and 16 Other Contracting States(App. no.
52207/99), Decision on Admissibi lity of 12 December 2001; ECHR
2001-XII [GC].
4319.27. The applicants in Banković had argued in the alternative
that the relevant actions were not only the actual dropping of the

bombs which hit the television station, but also the authorisation

of the operation and the giving of the relevant commands (i.e.

that the alleged violations resulted from the extraterritorial
1012
effects of intra-territorial acts). The Grand Chamber appears

not to have regarded that factor as affecting the outcome. It

said:

“the real connection between the applicants and
the respondent States is the impugned act which,
wherever decided, was performed, or had effects,

outside of the territory of those States (‘the extra-
territorial act’). It considers that the essential
question to be examined therefore is whether the
applicants and their deceased relatives were, as a

result of that extra-territorial act, capable of
falling within the jurisdiction of the respondent
States.1013

It could not have made a difference in Banković if the harm had

been caused by a missile fired from the territory of a NATO

member, as distinct from be ing dropped by a bomber flying
from that territory.

9.28. As to the interpretation of Article 1, the Grand Chamber

observed that the words “within their jurisdiction” in Article 1

of the European Convention had to be interpreted in accordance

1012 Banković v. Belgium and 16 Other Contracting States (App. no.
52207/99), Decision on Admissibi lity of 12 December 2001; ECHR
2001-XII [GC], para. 53.
1013 Ibid., para. 54.
432with the normal rules of interpretation set out in the Vienna
1014
Convention. In that regard, it observed:

“As to the ‘ordinary mean ing’ of the relevant
term in Article 1 of the Convention, the Court is

satisfied that, from th e standpoint of public
international law, the jurisdictional competence
of a State is primarily territorial. While
international law does not exclude a State’s

exercise of jurisdiction extra-territorially, the
suggested bases of such jurisdiction (including
nationality, flag, diplomatic and consular
relations, effect, protection, passive personality

and universality) are, as a general rule, defined
and limited by the sovereign territorial rights of
the other relevant States.” 1015

9.29. The Grand Chamber added:

“In keeping with the esse ntially territorial notion
of jurisdiction, the Court has accepted only in
exceptional cases that acts of the Contracting
States performed, or pr oducing effects, outside

their territories can constitute an exercise of
jurisdiction by them within the meaning of
Article 1 of the Convention.” 1016

9.30. Having discussed Drozd and Janousek and the cases

holding that the Convention was applicable to Turkey as regards

the actions of the TRNC in Northern Cyprus, in which

jurisdiction was held to subsists as a result of the “effective

1014
Banković v. Belgium and 16 Other Contracting States (App. no.
52207/99), Decision on Admissibi lity of 12 December 2001; ECHR
1015-XII [GC], para. 56.
1016 Ibid., para. 59.
Ibid., para. 67.
433control” exercised by Turkey over the territory, the Grand
Chamber observed that :

“the case-law of the Cour t demonstrates that its
recognition of the exercise of extra-territorial
jurisdiction by a Contracting State is exceptional:

it has done so when the respondent State, through
the effective control of the relevant territory and
its inhabitants abroad as a consequence of
military occupation or through the consent,
invitation or acquiescence of the Government of

that territory, exercises a ll or some of the public
powers normally to be exercised by that
Government.” 1017

9.31. Turning to the question whethe r the applicants could be

said to be “within the jurisdiction” of the respondent States, the

Grand Chamber rejected the appl icant’s reliance on a modified

“effective control” test according to which the applicability of

the substantive obligations under the Convention could be

“divided and tailored” according to the level of control in fact
exercised by the State. 1018 Of particular interest for present

purposes is the Grand Chamber’s observation that:

“the applicants’ notion of jurisdiction equates the
determination of whethe r an individual falls
within the jurisdiction of a Contracting State with
the question of whether that person can be

considered to be a victim of a violation of rights
guaranteed by the Convention. These are separate
and distinct admissibility conditions, each of

1017 Banković v. Belgium and 16 Other Contracting States (App. no.
52207/99), Decision on Admissibi lity of 12 December 2001; ECHR
2001-XII [GC], para. 71.
1018 Ibid., para. 75.
434 which has to be satisfied in the afore-mentioned
order, before an individual can invoke the

Conventi1019rovisions against a Contracting
State.”

9.32. Those observations are equally apt as regards the

ICCPR; it is necessary first to ascertain whether the ICCPR is

applicable, and whether substantive obligations are thus owed to

individuals, before assessing wh ether particular conduct can be

said to amount to a breach.

9.33. Another relevant admissibility decision of the European
1020
Court of Human Rights is El Mahi and Others v. Denmark .

There an individual and two associations in Morocco brought a

case against Denmark alleging breach of their right to religion
under Article 9, together with the prohibition of discrimination

in Article 14, as a result of the decision of the Danish authorities

not to bring any prosecution in relation to the publication of the

infamous cartoons depicting the prophet Mohammed in the

Jyllands-Posten newspaper. They also alleged breaches of their

right to freedom of expression under Article 10 and of Article

17 (prohibition of abuse of rights).

1019 Banković v. Belgium and 16 Other Contracting States (App. no.
52207/99), Decision on Admissibi lity of 12 December 2001; ECHR
2001-XII [GC], para. 75.
1020
El Mahi and Others v. Denmark (App. no. 58o3/06), Decision on
Admissibility of 11 December 2006, ECHR 2006-XII (5 Section/C).
4359.34. The Court held that the application was inadmissible on

the basis that the applicants were not “within the jurisdiction” of

Denmark within the meaning of Article 1 of the European

Convention. The Court reasoned that the concept of

“jurisdiction” in Article 1:

“must be considered to reflect the term's meaning
in public international law. Thus, from the

standpoint of public international law, the words
‘within their jurisdiction’ in Article 1 of the
Convention must be understood to mean that a
State's jurisdictional competence is primarily
territorial and also that jurisdiction is presumed

to be exercised normally throughout the State's
territory. Only in exceptional circumstances may
the acts of Contracting States performed outside
their territory or which produce effects there

(‘extra-territorial acts’) amount to an exercise by
them of their jurisdiction within the meaning of
Article 1 of the Convention...

Here the applicants are, respectively, a Moroccan
national resident in Morocco and two Moroccan
associations which are based in Morocco and
operate in that country. The Court considers that

there is no jurisdictional link between any of the
applicants and the relevant member State, namely
Denmark, [n]or that they can come within the
jurisdiction of Denm ark on account of any
extra-territorial act. Accordingly, the Court has

no competence to examine the applicants'
substantive complaints unde r the Articles of the
Convention relied upon.” 1021

1021 El Mahi and Others v. Denmark (App. no. 5853/06), Decision on
Admissibility of 11 December 2006, ECHR 2006-XII (5 Section/C), pp. 8-9.
4369.35. Admittedly, the decision is not on all fours with the

present case, given that the breach was as an alleged breach of

positive obligations to protect the right to religion, rather than

direct State action alleged to re sult in a breach. However, the

decision is authority for the propos ition that, if an applicant is
not on the territory of a State (and therefore within its

jurisdiction on that account), and if there is no extraterritorial

act, then the substantive provisions of the European Convention

are simply not engaged, whatever the effects that State action

(or inaction) may have upon the righ ts of an applicant. Similar
reasoning is equally applicable under the ICCPR.

9.36. Reference may also be made to the Court’s 2008 Order

on Provisional Measures in Application of the International

Convention on the Elimination of All Forms of Racial
Discrimination (Georgia v. Russia), in which the Court rejected

Russia’s argument that Article 2 and 5 of CERD were not

applicable to the extraterritori al actions of Ru ssian troops in
1022
South Ossetia and Abkhazia. That provisional conclusion
related to a treaty containing no express limitation on its

applicability and concerned specific provisions phrased in broad

terms. It is not authority in relation to jurisdictional provisions

such as those in Article 1 ICCPR.

1022 Application of the International Convention on the Elimination of
All Forms of Racial Discrimination (G eorgia v. Russia), Request for the
Indication of Provisional Measures, Order of 15 October 2008, para. 109.
437 (b) ACHR

9.37. Both Colombia and Ecuador are parties to the American

Convention on Human Rights without reservation. The English

text of Article 1 of the ACHR provides:

“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
for reasons of race, color, sex, language, religion,
political or other opinion, national or social
origin, economic status, birth, or any other social
condition.”

9.38. The formulation of Article 1 ACHR differs from that of
Article 2 ICCPR (there being no reference to persons “within

the territory” of the States Parties).

9.39. The Inter-American Court has not yet pronounced on the

interpretation of Article 1 in relation to extra-territorial State
action (or intra-territorial State action having extra-territorial

effects). On the other hand, the Court has emphasised the

distinction in Article 1 between the obligation not to violate

human rights (“respect”) and the obligation to “ensure” the fu
ll
and free enjoyment of human rights, 1023 in terms which suggest

that the obligation “to ensure” at least is limited to persons

“subject to their jurisdiction”.

1023 See e.g. Velázquez-Rodríguez v. Honduras, Merits, I-ACtHR, Series
C, No. 4, Judgment of 29 July 1988, paras. 164 ff.

4389.40. By contrast, the Inter-Am erican Commission has had
1024
occasion to deal with the issue. In Saldaño v. Argentina , it

was argued by the applicant, on be half of her son who had been

condemned to death in the Unite d States, that Argentina was
under an obligation to present an inter-State case on his behalf

in respect of alleged violations of his right to a fair trial under

the American Declaration on the Rights and Duties of Man and

the ACHR. The Commission declared the application

inadmissible. Having cited Article 1(1) ACHR, the Commission

observed:

“Accordingly, States Parties have undertaken to

respect and ensure the substantive guarantees
enshrined in the Convention in favour of persons
‘subject to their jurisd iction.’ As implicitly

established by the case law of the Commission
and the Inter-American Court, this protection
must extend to all human beings present within
their national territory, irrespective of their
1025
nationality or status.”
The Commission continued:

1024 Saldaño v. Argentina, I-ACmHR, Report No. 38/99, 11 March 1999.
1025 Ibid., para. 16. The reference to the jurisprudence of the Inter-
American Court is to Advisory Opinion OC-2/82, The Effect of Reservations
on the Entry into Force of the American Convention on Human Rights
(Articles 74 and 75) , I-ACtHR, Series A, No. 2, 24 September 1982, para.
33, where the Inter-American Court observed:
“Viewed in this light and considering that the Convention was designed to
protect the basic rights of individual human beings irrespective of their
nationality, against States of their own nationality or any other State Party,

the Convention must be seen for what in reality it is: a multilateral legal
instrument of framework enabling States to make binding unilateral
commitments not to violate the human rights of individuals within their
jurisdiction.” (emphasis added).
439 “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.”

9.41. Having referred to the decision of the European

Commission on Human Ri ghts in the first Cyprus v. Turkey

inter-State case, in which the European Commission had held

that States Parties to the Eu ropean Convention “are bound to

secure the said rights and fr eedoms to all persons under their

actual authority and responsibility, whet her that authority is
1026
exercised within their own territory or abroad”. The Inter-
American Commission continued:

“This understanding of jurisdiction – and
therefore responsibility for compliance with
international obligations – as a notion linked to

authority and effective control, and not merely to
territorial boundaries, ha s been confirmed and
elaborated on in other cases decided by the
European Commission and Court.” 1027

1026
Saldaño v. Argentina, I-ACmHR, Report No. 38/99, 11 March 1999,
§18, citing Cyprus v. Turkey (App. Nos. 6780/74 and 6950/75) 2 DR 125
10275), para. 8.
Saldaño v. Argentina, I-ACmHR, Report No. 38/99, 11 March 1999,
para 19 (reference omitted). In that re gard, the Inter-American Commission
also referred to, inter alia, the d ecision of the European Court in Loizidou v.
Turkey, Jurisdiction (App. No. 15318/89), Series A, No. 310 [GC] (1995)
and the decisions of the European Commission in X v. United Kingdom (App.
No. No. 7547/76) 12 DR 73 (1977), Bertrand Russell Peace Foundation v.
440The Inter-American Commission di smissed the application as

inadmissible, on the basis that “the claims filed refer to the

alleged violation of the rights of a person who is not subject to
1028
Argentine jurisdiction under the terms of Article 1(1)”.

9.42. The decision of the Inter-American Commission in
Saldaño is strong authority both for the proposition that both

limbs of Article 1(1) apply only to persons “subject to [the]

jurisdiction”, and that such jurisdiction will exist in relation to

extra-territorial action only when the State in question exercises

“authority and effective control”.

9.43. In such circumstances, it is an essential requirement for

applicability of the ACHR that the individuals in question

should have been “subject to the jurisdiction” of Colombia. The

express reliance in Saldaño on the jurisprudence of the

European Court and Europ ean Commission relating to

“authority and effective control” supports that conclusion. The
PECIG program carried out in Colombian territory cannot be

understood as an “exercise of authority or effective control” in

Ecuador.

United Kingdom (App. No. 7597/76) 14 DR 117 (1978); and Mrs W v.
United Kingdom (App. No. 9348/81), 32 DR 190 (1983).
1028 Saldaño v. Argentina, I-ACmHR, Report No. 38/99, 11 March 1999,
para. 23.
441 (c) ICESCR

9.44. Both Colombia and Ecuador are parties to the ICESCR
without reservation.

9.45. The ICESCR contains no expr ess provision concerning

its applicability. The closest it comes is Article 2, which
provides:

“1. Each State Party to the present Covenant

undertakes to take steps, individually and through
international assistance and co-operation,
especially economic and technical, to the
maximum of its available resources, with a view
to achieving progressively the full realization of
the rights recognized in the present Covenant by
all appropriate means, including particularly the

adoption of legislative measures.
2. The States Parties to the present Covenant
undertake to guarantee that the rights enunciated
in the present Covenant will be exercised without

discrimination of any kind as to race, colour, sex,
language, religion, political or other opinion,
national or social origin , property, birth or other
status.

3. Developing countries, with due regard to
human rights and their national economy, may
determine to what extent they would guarantee
the economic rights rec ognized in the present
Covenant to non-nationals.”

9.46. In its Advisory Opinion in The Wall , the Court was

called upon to determine the applic ability of the ICESCR to the

actions of Israel in the OPT. The Court observed:

442 “The International Covenant on Economic,
Social and Cultural Rights contains no provision
on its scope of application. This may be
explicable by the fact that this Covenant

guarantees rights which are essentially
territorial. However, it is not to be excluded that
it applies both to territories over which a State

party has sovereignty and to those over which
that State exercises terr itorial jurisdiction. Thus
Article 14 makes provision for transitional
measures in the case of any State which ‘at the

time of becoming a Party, has not been able to
secure in its metropolitan territory or other
territories under its ju risdiction compulsory
primary education, free of charge’.” 1029

9.47. Having referred to the views of the Committee on

Economic, Social and Cultural Right s to the effect that “the

State party’s obligations under the Covenant apply to all

territories and populations under its effective control” 1030, as

noted above, the Court observed that

“the territories occupied by Israel have for over

37 years been subject to its territorial jurisdiction
as the occupying Power. In the exercise of the
powers available to it on this basis, Israel is
bound by the provisions of the International

Covenant 1031conomic, Social and Cultural
Rights.”

1029
Legal Consequences of the Construction of a Wall in the Occupied
Palestinian Territory, Advisory Opinion, ICJ Reports 2004, p. 181, para. 112
1030hasis added).
1031 Ibid.
Ibid.
4439.48. Thus the ICESR “guarantees rights which are essentially
1032
territorial”. It was only held to apply to Israel’s actions in
the OPT on the basis of Israel’s “territorial jurisdiction” over

those territories as a consequence of its occupation.

(d) Conclusion as to territorial application

9.49. To summarise, the 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 territory.

9.50. In the interests of economy this point will not be
repeated in relation to each of the human rights treaties

discussed below. It is, however, maintained with respect to each

of them.

(3) H UMAN RIGHTS OBLIGATIONS AND ARTICLE 14(2) OF THE
1988 N ARCOTICS C ONVENTION

9.51. This issue has already been discussed in relation to

Ecuador’s environmental claims. 1033 It can therefore be dealt

with here very briefly.

1032
Similar considerations apply to the CRC, the CEDAW and the
ACHR ESCR Additional Protocol, which likewise do not contain any
express provision as to their aplicability, but are likewise “essentially
1033itorial” in their application.
See above, paras. 8.113-8.121.
4449.52. Ecuador’s arguments in this regard consist chiefly of a

cross-reference to the discussion in Chapter VIII as regards

environmental matters. 1034However, perhap s perceiving the

potential weaknesses of its argumen ts as to applicability of the

various human rights instrument s relied upon, it also argues

that:

“Article 14(2) is not limited to fundamental
human rights within the territorial jurisdiction of
the State concerned but applies equally to respect

for the fundamental rights of persons beyond its
borders who are affected by the measures in
question. On that basis Ecuador’s case is that
Colombia has violated not only applicable
provisions of inter alia the 1966 UN Covenants

on Civil and Political Rights and on Economic,
Social and Cultural Rights, the 1969 Inter-
American Convention on Human Rights, and the
1989 ILO Convention No. 169 Concerning

Indigenous and Tribal Pe oples in Independent
Countries, but additionally or alternatively the
1988 Narcotics Convention, Article 14(2).” 1035

9.53. But Ecuador cannot have it both ways. Either the human

rights obligations in question are incorporated by reference, in

which case they must have the same scope ratione loci as they

do in the original treaty, or they are not, in which case Article
14(2) is reduced to a provisi on requiring only “respect” in

general terms for unspecified human rights as a condition of

certain measures to implement the 1988 Convention. Even if

Ecuador’s arguments as to the effects of the Article 14(2) are

1034 EM, para. 9.11.
1035 EM, para. 9.11.
445accepted, those human rights obligations cannot be rendered

applicable merely by incorpor ation by reference in another

treaty, especially where that tr eaty does not as such concern
human rights. If obligations do not arise for Colombia under the

treaties themselves, they cannot rise unde r the 1988 UN

Narcotics Convention.

(4) T HE SCOPE OF HUMAN RIGHTS ALLEGEDLY BREACHED

9.54. In Section II of Chapter IX, Ecuador alleges violations of

a catalogue of human rights, in particular, the rights to life,
health, food, water, healthy environment, property, humane

treatment, private life, and information. As already

demonstrated, the various human rights instruments relied on by

Ecuador are simply not applicable to any extraterritorial effects
of Colombia’s actions on the facts of the present case.

However, in addition to the prim ary response that it has not

been shown by Ecuador that spraying has had any
transboundary effects within Ecua dor’s territory, Colombia has

specific legal and factual answers to each of these individual

allegations, and without prejudice to the general points of

principle made above, the allegations will be dealt with
individually in their turn.

446 (a) The right to life

(i) Ecuado pr’sition

9.55. As regards the alleged violati on(s) of the right to life,

Ecuador relies on Article 6(1) ICCPR, Article 6 CRC and
1036
Article 4 ACHR. It asserts that it is a universal right (relying

also on the African Charter and European Convention), and that
1037
“it is unquestionably part of general international law.

9.56. In attempting to elucidate the content of the right to life,

Ecuador refers to the Inter-American Court’s Street Children
decision 1038 and its reiteration in, inter alia , the decision in

Yakye Axa, 1039 where:

“the Court concluded that lack of access by
indigenous peoples to traditional means of

subsistence, as well as to use and enjoyment of
the natural resources n ecessary to obtain clean
water and to practice traditional medicine to

prevent and cure illnesses, resulted in a breach of
the right to a decent existence, as recognised in
Article 4 of the American Convention.” 1040

9.57. Ecuador concludes that:

1036
1037 EM, para. 9.43.
1038 EM, para. 9.43.
EM, para. 9.44, referring to “ Street Children Case ” , Villagran
Morales et al. v. Guatemala, I-ACtHR, Series C No. 77, Judgment of 19
November 1999, para. 144.
1039 Indigenous Community Yakye Axa v. Paraguay, I-ACtHR, Series C,
No. 125, Judgment of 17 June 2005.
1040 EM, para. 9.47, citing Indigenous Community Yakye Axa v.
Paraguay, I-ACtHR, Series C, No. 125, Judgment of 17 June 2005, paras.
162-168, 176.

447 “Pollution from Colombian spraying of toxic
herbicides poses a well-documented risk to life

and human health. The evidence shows that it has
harmed access to food and clean water, and
denied those living in a ffected border areas of
Ecuador the decent existence to which they are
entitled.”041
and that as a result, Colomb ia has breached the relevant

provision of the ICCPR, ACHR and CRC.

(ii) Colombiar’sponse

9.58. Article 4(1) ACHR provides:

“Every person has the right to have his life

respected. This right sh all be protected by law
and, in general, from the moment of conception.
No one shall be arbitrarily deprived of his life.”

The remainder of Article 4 ACHR is concerned with provisions
relating to capital punishment and the death penalty.

9.59. Article 6(1) ICCPR provides:

“Every human being has the inherent right to life.
This right shall be protected by law. No one shall

be arbitrarily deprived of his life.”
The remainder of Article 6 ICCP R is likewise concerned with

issues relating to the death penalty.

9.60. Article 6 CRC provides:

“1. States Parties recognize that every child has
the inherent right to life.

1041 EM, para. 9.48.

448 2. States Parties shall ensure to the maximum
extent possible th e survival and development of
the child.”

9.61. As noted elsewhere in this Counter-Memorial, it is

striking that, in the section of Chapter 8 devoted to the right to
life, Ecuador does not allege any breach of the right to life as

such (i.e. the causing of deaths of persons), although it does

make reference to a number of allegations of deaths in the

section devoted to the right to health. 1042 In particular, Ecuador

does not as such explicitly e ndorse, adopt or rely upon the

allegations of a number of its witnesses 1043and the allegations

made in a number of the documents contained in the
1044
Annexes, that the spraying has di rectly caused deaths or

serious long-term illnesses.

9.62. Rather, its allegation is of a more diffuse breach of
ancillary obligations which it says are to be read into the

provisions in question, relating to access to food and clean

water, and ensuring the conditions for a “decent existence” and

which are said to protect a right to “a decent life”. Ecuador’s

1042 EM, paras. 9.53, 9.54; and cf. EM, para. 10.9: “Colombia’s
chemical spraying has caused damage and injury to human health, including
illness and death among the people who inhabit the border region”.
1043 See e.g., EM, Vol. IV, Annex 194 (Witness No. 6); Annex 199
(Witness No. 11); Annex 201 (Witness No. 13); Annex 202 (Witness No.
14).
1044
See e.g., EM, Vol. IV, Annex 161, p. 9 (Acción Ecologica); Annex
162 (CONAIE), pp. 12-14, 20; Annex 166, pp. 4, 12, 15, 18 and 20
(Ecuadorian Ministry for the Environment); Annex 166 (Ecuadorian
Congress), pp. 4 and 5; Annex 169 (FIDH et al), p. 12.
449claims are unsupported by evidence; still less can they be

described as “well documented”. Moreover, the ancillary

obligations on which Ecuador he re relies are the positive

obligations of a territorial so vereign, and thus the exclusive
responsibility of Ecuador.

9.63. Further, Ecuador’s allegations of breach of the right to

life (or rather, the alleged righ t to a “decent lif e”) in effect
recycle its claims as to the right to food and the right to water, as

well as its claims as to the rights of indigenous peoples.

9.64. As to the dictum of the Inter-American Court in Yakye
Axa, that pronouncement was made in a case involving a very

different factual situation, viz., action of a State affecting an

indigenous population located on its own territory. That

reasoning cannot simply be extend ed to a situation of alleged

extra-territorial effects of Stat e action carried on within its own
territory. The essentially territorial nature of the Inter-American

Court’s reasoning is apparent from the terms used to define the

scope of its enquiry:

“whether the State generated conditions that
worsened the difficulties of access to a decent life
for the members of th e Yakye Axa Community

and whether, in that context, it took appropri1045
positive measures to fulfill that obligation”.

1045 Indigenous Community Yakye Axa v. Paraguay, I-ACtHR, Series C,
No. 125, Judgment of 17 June 2005, para. 163.

450 (b) The right to health

(i) Ecuado prsition

9.65. As to the right to health, Ecuador relies on Article 12(1)

ICESCR, Article 12 CEDAW, Article 24 CRC and Article 10
1046
ACHR ESCR Additional Protocol. It argues that the right to
health is not limited to health care, relying on General Comment

No. 14 of the Committee on Economic, Social and Cultural

Rights, in which the Committee observed:

“the right to health em braces a wide range of
socio-economic factors that promote conditions
in which people can lead a healthy life, and
extends to the underlying determinants of health,

such as food and nutrition, housing, access to
safe and potable water a nd adequate sanitation,
safe and healthy working conditions, and a
healthy environment.” 1047

9.66. Relying on Article 12(2)(b) ICESCR, it argues that that

provision “imposes a positive obligation on the parties to take

appropriate measures aimed at “t he improvement of all aspects
1048
of environmental…hygiene,” and in that regard again refers

to General Comment No. 14 in which the Committee opined

that that provision includes an obligation of

“prevention and reduction of the population’s
exposure to harmful substances such as radiation
and harmful chemicals or other detrimental

1046 EM, para. 9.49
1047 EM, para. 9.50, citing General Comment No. 14, para. 11.
1048 EM, para. 9.50.
451 environmental conditions that directly or
indirectly impact upon human health”. 1049

9.67. Ecuador next argues that “[v] iolations of the right to

food and the right to water [...] also engage a breach of the

obligation to respect the right to health,” 1050 alleging a violation

of the right to health by “expos ure to toxic herbicides, by the

contamination of drinking water sources, and because of the

destruction of crops that const itute the basis of their nutrition
1051
and medicine.” In that regard, reliance is placed on the
1052
reports of the UN Special Rapporteur on the Right to Health.

(ii) Colombia r’sponse

9.68. Article 12 ICESCR provides:

“1. The States Parties to the present Covenant
recognize the right of ev eryone to the enjoyment

of the highest attainable standard of physical and
mental health.

2. The steps to be taken by the States Parties to
the present Covenant to achieve the full
realization of this right shall include those
necessary for:

(a) The provision for the reduction of the

stillbirth-rate and of infant mortality and for the
healthy development of the child;

(b) The improvement of all aspects of
environmental and industrial hygiene;

1049
1050 EM, para. 9.50, citing General Comment No. 14, para. 15
1051 EM, para. 9.56
1052 EM, para. 9.56.
EM, para. 9.56.
452 (c) The prevention, treat ment and control of
epidemic, endemic, occupational and other
diseases;

(d) The creation of conditions which would
assure to all medical service and medical
attention in the event of sickness.”

9.69. Article 24 CRC provides:

“1. States Parties recognize the right of the child
to the enjoyment of the highest attainable

standard of health and to facilities for the
treatment of illness and rehabilitation of health.
States Parties shall strive to ensure that no child
is deprived of his or he r right of access to such
health care services.

2. States Parties shall pursue full implementation
of this right and, in particular, shall take
appropriate measures:

(a) To diminish infant and child
mortality;
(b) To ensure the pr ovision of necessary

medical assistance and health care to all
children with emphasis on the
development of primary health care;
(c) To combat disease and malnutrition,

including within the framework of
primary health care, through, inter alia,
the application of readily available
technology and through the provision of
adequate nutritious foods and clean
drinking-water, taking into consideration
the dangers and risk s of environmental

pollution;
(d) To ensure appropriate pre-natal and
post-natal health care for mothers;

453 (e) To ensure that all segments of society,
in particular parents and children, are
informed, have access to education and
are supported in the use of basic

knowledge of child health and nutrition,
the advantages of breastfeeding, hygiene
and environmental sanitation and the
prevention of accidents;

(f) To develop preventive health care,
guidance for parents and family planning
education and services.

3. States Parties shall take all effective and
appropriate measures with a view to abolishing
traditional practices prejudicial to the health of
children.

4. States Parties undertake to promote and
encourage international co-operation with a view
to achieving progressively the full realization of
the right recognized in the present article. In this
regard, particular account shall be taken of the
needs of developing countries.”

9.70. Article 12 CEDAW provides

“1. States Parties shal l take all appropriate
measures to eliminate discrimination against

women in the field of health care in order to
ensure, on a basis of equality of men and women,
access to health care services, including those
related to family planning.

2. Notwithstanding the provisions of paragraph I
of this article, States Parties shall ensure to
women appropriate services in connection with
pregnancy, confinement and the post-natal
period, granting free services where necessary, as
well as adequate nutrition during pregnancy and

lactation.”

4549.71. Article 10 ACHR ESCR Additional Protocol provides:

“1. Everyone shall have the right to health,
understood to mean the enjoyment of the highest
level of physical, mental and social well-being.

2. In order to ensure the exercise of the right to
health, the States Parties agree to recognize
health as a public good and, particularly, to adopt
the following measures to ensure that right:

a. Primary health care, that is, essential health
care made available to all individuals and
families in the community;

b. Extension of the benefits of health services to
all individuals subject to the State's jurisdiction;

c. Universal immunization against the principal
infectious diseases;
d. Prevention and treatment of endemic,
occupational and other diseases;

e. Education of the population on the prevention
and treatment of health problems, and

f. Satisfaction of the health needs of the highest
risk groups and of those whose poverty makes
them the most vulnerable.”

9.72. The ICESCR, CRC and the ACHR ESCR Additional

Protocol are all “essentially terr itorial” in their application, and
therefore impose obligations on the territorial State. So much is

clear from the various references to the fact that States “shall

take appropriate measures” (CRC, Art. 24(2)); States shall

“adopt the following measures to ensure that right” (Additional
Protocol, Art. 10(2) ; that “States Parties shall take all

455appropriate measures to eliminate discrimination against women

in the field of health care” (CEDAW, Art. 12(1)), as well as the

catalogue of “steps to be taken by the States Parties” (ICESCR,

Art. 12(2). The measures liste d are all steps which by their
nature are only applicable to th e territorial State within its own

territory.

9.73. As regards the ICESCR, despite Ecuador’s strong
reliance upon it, the same is clear from General Comment No.

14, which clearly envisages the right to health as being primarily

a right opposable to the territorial State. For instance, in the

General Comment, the Committee states variously that

“The Committee recognizes the formidable
structural and other obstacles resulting from
international and other factors beyond the control

of States that impede the full realization of article
12 in many States parties.”
“The right to health in all its forms and at all

levels contains the following interrelated and
essential elements, the precise application of
which will depend on the conditions prevailing in
a particular State party [ ...]”053

9.74. Article 12 CEDAW, is concerned with the elimination of

discrimination in relation to women in the field of health care.

Ecuador presents no colourable basis on which it to base any
claim of discrimination against women in relation to spray drift

from aerial spraying.

1053 General Comment No. 14, paras. 5 and 12.

4569.75. As to the general allegations of symptoms allegedly
suffered by individuals in Ecuador and that “the nature and

extent of these symptoms of toxic spraying indicate Colombian

responsibility for failing to respect the right to health of the

affected populations in Ecuador”, the general point as to
Ecuador’s claims as a whol e obtains: Ecuador has not

demonstrated that those symptoms have in fact occurred – the

only evidence put forward is the statements of a handful of
witnesses, and there are no independently verified expert reports

which demonstrate that the symptoms in question were in fact

suffered by individuals in the border area. Further, Ecuador has

not demonstrated any causal li nk between the sprayings and the
occurrence of the symptoms in que stion, if they did in fact

occur. Rather, the independent scientific evidence shows that

the majority of such symptoms could not have been caused by
the spraying, and certainly not at the chronic levels alleged.

9.76. As mentioned above, it is notable that, although

allegations of deaths caused by spraying are not relied upon by
Ecuador as violations of the ri ght to life of in dividuals, they

surface as regards the right to health. However, they are relied

upon selectively and Ecuador appe ars not to fully endorse the
allegations of the witnesses in question.

9.77. The evidence of these witnesses is of doubtful weight.
To take one example, Witness 11, explicitly relied upon by

457Ecuador, resident in San Francisc o II, alleges that two of her

children died as the result of spraying: one, following spraying

in early 2001, died many months later, in September 2001; the

other died an unspecified peri od after spraying, in September
1054
2003. Ecuador also refers to the statement of Witness 12,

also resident in San Francisco II, who alleges spraying in 2002
and 2004, and notes that four ch ildren died in that small

community – of only 20 families – in 2002, all of them within a

week of the sprayings. 1055 The stories are mutually

incompatible, quite apart from the fact that they do nothing to

establish causation.

9.78. Further, Ecuador’s admission that there is “inadequate

health care” in “these impoverished areas” underlines the lack of

any proof of a causal link between the alleged sp raying and its

alleged consequences. The symptoms described are fully

consistent with other causes in such “impoverished areas”; the

“inadequate health care” and poor infrastructure in those areas,
responsibility for which can only be laid at the door of Ecuador,

is more likely a cause of the symptoms in question.

9.79. Finally, Ecuador attempts to recycle its allegations of

violation of the rights of indigenous peoples, as well as the
alleged violations of the rights to food and water, under the

heading of the right to health. Those allegations are denied for

1054 EM, Vol. IV, Annex 199.
1055 EM, Vol. IV, Annex 200.
458the reasons set out in this chapter; they add nothing to the
alleged violation of the right to health.

(c) The right to food

(i) Ecuado pr’sition

9.80. As regards the right to food, Ecuador relies upon Article

11(2) ICESCR, Article 27 CRC and Article 12 ACHR ESCR
Additional Protocol, alleging that that right is “[c]losely related

to the right to life and the right to health”. 1056 In that regard, it

refers to the ESCR Committee’s General Comment No. 12 on

the right to adequate food, which it states is an “authoritative
1057
interpretation” of Article 11 ICESCR. In this regard, Ecuador

argues:

“General Comment No. 12 indicates that this
right’s core content implies the availability of

food in a quantity and quality sufficient to satisfy
the dietary needs of individuals, free from
adverse substances and acceptable within a given
culture. This right imposes an obligation on
Colombia to respect, protect and fulfil the right to

food, including1058areas outside its
jurisdiction.”

9.81. Ecuador’s allegation of br each, relying on the factual

allegations contained in Chapter VI, is that aerial spraying has

allegedly “degraded and in some cases destroyed the normal

1056 EM, para. 9.57.
1057 EM, para. 9.58.
1058 EM, para. 9.59 (citations omitted)
459 1059
subsistence foodstuffs of local communities in Ecuador”. It

is further alleged that th e spraying has had effects on

domesticated animals, fish in local rivers and wild animals used
1060
as food sources. In this regard, relian ce is placed on official

reports from the Director of Environmental Management for the

Ecuadorian Ministry of Agricult ure and Livestock, the Director

of the National Directorate for the Defence of the Rights of

Indigenous Peoples (DINAPIN), the office of the National
Ombudsman of Ecuador, and representatives of the Provincial

Government of Sucumbíos, alleging “extensive damage to

plantations of maize, plantai n, rice, coffee, cacao, and other

staple foods of the local communities, such as yucca.” 1061 Itis

further stated that there have been reports of loss of animals. 1062

Ecuador alleges that these “extensive losses in Ecuador are

consistent with the damage that has been described on the

Colombian side of the border.” 1063

(ii) Colombia r’sponse

9.82. Article 11 ICESCR provides:

“1. The States Parties to the present Covenant
recognize the right of ev eryone to an adequate

standard of living for himself and his family,
including adequate food, clothing and housing,
and to the continuous improvement of living

1059 EM, para. 9.59.
1060 EM, para. 9.60.
1061 EM, para. 9.61.
1062
1063 EM, para. 9.61.
EM, para. 9.61.
460 conditions. The States Parties will take
appropriate steps to ensure the realization of this
right, recognizing to this effect the essential
importance of international co-operation based

on free consent.
2. The States Parties to the present Covenant,
recognizing the fundamental right of everyone to
be free from hunger, shall take, individually and

through international co-operation, the measures,
including specific pr ogrammes, which are
needed:

(a) To improve methods of production,
conservation and distribution of food by making
full use of technical and scientific knowledge, by
disseminating knowledge of the principles of
nutrition and by developing or reforming
agrarian systems in such a way as to achieve the
most efficient development and utilization of

natural resources;
(b) Taking into account the problems of both
food-importing and food-e xporting countries, to
ensure an equitable dist ribution of world food

supplies in relation to need.”

9.83. Article 27 CRC provides:

“1. States Parties recognize the right of every
child to a standard of living adequate for the
child's physical, mental, spiritual, moral and
social development.

2. The parent(s) or others responsible for the
child have the primary responsibility to secure,
within their abilities an d financial capacities, the
conditions of living necessary for the child's
development.

3. States Parties, in accordance with national
conditions and within their means, shall take

461 appropriate measures to a ssist parents and others
responsible for the child to implement this right
and shall in case of need provide material
assistance and support programmes, particularly

with regard to nutrition, clothing and housing.
4. States Parties shall take all appropriate
measures to secure the recovery of maintenance
for the child from the parents or other persons

having financial responsibility for the child, both
within the State Party and from abroad. In
particular, where the person having financial
responsibility for the ch ild lives in a State
different from that of the child, States Parties
shall promote the accession to international
agreements or the conclusion of such agreements,

as well as the making of other appropriate
arrangements.”

9.84. Article 12 ACHR ESCR Additional Protocol provides:

“1. Everyone has the right to adequate nutrition
which guarantees the possibility of enjoying the
highest level of physical, emotional and
intellectual development.

2. In order to promote the exercise of this right
and eradicate malnutrition, the States Parties
undertake to improve methods of production,
supply and distribution of food, and to this end,
agree to promote greater international

cooperation in support of the relevant national
policies.”

9.85. As with Ecuador’s reliance on the right to health, it bears
emphasising that the obligati ons imposed by the various

instruments in relation to the right to food are essentially

territorial and envisage action by the territorial State in order to

462ensure the rights in question to their population. By Article

11(1) ICESCR the States Parties u ndertake that they “will take
appropriate steps to ensure the re alization of this right”, and by

Article 11(2), in particular, they undertake to “take, individually

and through international co-ope ration, the measures, including

specific programmes, which are needed” in order to “improve
methods of production, conservatio n and distribution of food”

and to “ensure an equitable di stribution of world food supplies

in relation to need”. Those provisions essentially impose
obligations in relation to a State’s own population; to the extent

that they impose wider obligations applicable outside the State’s

own territory, those obligations are ones of international

cooperation.

9.86. Similar considerations apply in relation to Article 12 of

the ACHR Additional Protocol, pu rsuant to which the States
Parties undertake “to improve methods of production, supply

and distribution of food, and to this end, agree to promote

greater international cooperati on in support of the relevant

national policies.”

9.87. The understanding of Article 11 ICESCR set out above

is confirmed by the ESCR Committee’s “authoritative
interpretation” of that provision, contained in General Comment

No. 12, in which the Committee expressed the view that

“Every State is obliged to ensure for everyone
under its jurisdiction access to the minimum

463 essential food which is sufficient, nutritionally
adequate and safe, to ensure their freedom from
hunger.” 1064

9.88. Similarly, the Committee recognized the essentially

territorial nature of State obligations in relation to the right to

food when it observed that

“The most appropriate ways and means of
implementing the right to adequate food will

inevitably va1065ignificantly from one State party
to another.”

9.89. Likewise, Ecuador’s reli ance on the Committee’s

enunciation of the oblig ation to “respect, protect and fulfil” in

General Comment No. 12 is misplaced; the paragraph in

question clearly envisages that it is the territorial State which is
under an obligation to take such steps.

9.90. Similar considerations apply in relation to Article 27

CRC, which, it is to be noted, is concerned not with the right to

food as such, but with the right of every child to an adequate

standard of living; it is notable that Article 27(2) recognizes that

the parent(s) or others responsib le for a child have the primary
obligation, and, pursuant to Articl e 27(3), the obligation of the

State is to “take appropriate m easures” to assist them. Those

obligations can only be incumben t upon the territorial State in

1064 ESCR Committee, General Comment No. 12, para. 14 (emphasis
added).
1065 ESCR Committee, General Comment No. 12, para. 21 (emphasis
added).
464which the child and his or her parents or other responsible

persons are located.

9.91. Further, Ecuador’s suggestion that Colombia is obliged
“to respect, protect and fulfil the right to food, including in areas

outside its jurisdiction” is misleading to the extent that it

suggests that the ESCR Committee so stated in General

Comment No. 12. In fact, although the obligation “to respect,
protect and fulfil” the right to food has been enunciated by the

Committee, it did not express the view that thos e obligations

applied outside a State’s jurisdiction. Indeed, as noted above, it

said exactly the opposite, limiti ng the applicability of the
obligations in that regard to persons “within its jurisdiction”. 1066

As is clear from the accompanying footnotes in the relevant

passage of Ecuador’s Memo rial, the only support which

Ecuador provides for its extraterritorial applicability of Article

12 is the views of the Special Rapporteur, which are in no way
“authoritative”, and go well beyond what the ESCR Committee

has been prepared to endorse.

9.92. As to the evidence relied upon by Ecuador, although it
attempts to give the impression that there are a multitude of

bodies and reports which have pronounced upon the question,

1066 See above, 9.88.

465the only source it cites is a singl e report from the Ecuadorian
1067
Ministry for the Environment.

9.93. As regards reliance on the reports of the UN Special

Rapporteurs on the Right to Food and on the Rights of

Indigenous People, those reports are based on allegations of

individuals in the relevant areas, and have no scientific basis. To

take but one example, the suggestion by the Special Rapporteur
on the Rights of Indigenous Pe ople that the spraying has

resulted in “the diminishing of soil quality” 1068 is clearly

contradicted by the scientific evidence; the products used in

Colombia’s spray mix have negligible effect on the soil, insofar

as they are broken down within a few days. More probable

causes for degradation of the soil and the damage to plants in the

border areas are the practices of “slash and burn” agriculture and

the array of highly toxic chemicals used in the cultivation and
1069
processing of coca crops to produce cocaine.

9.94. Finally, it is notable that Ecuador seeks to make a virtue

out of the extremely fragile situ ation of the population in the

border areas. Conditions of chr onic malnutrition were present
1070
there long before the sprayings began. As acknowledged in

1067 EM, Vol. IV, Annex 166.
1068 EM, para. 9.62.
1069 See above, paras. 3.28-3.39, 7.98, note 678.
1070 EM, para. 9.66, after the quote, citing the report of the Ecuadorian
Scientific Commission: “It is worth recalling that this region was already

characterised, before the sprayings began, as one having ‘a higher level of
malnutrition among the school-age population than in the same population
466the Memorial, “a 2001 Ecuadorian Government study found that

nearly one-third of all the residents in rural areas of Esmeraldas,

Carchi and Sucumbíos, includi ng children, suffered from
1071
chronic malnutrition.” The fact that t hose areas have high

levels of malnutrition is once again a matter for which Ecuador,

and Ecuador alone, bears responsibility. Ecuador cannot seek to
shift the blame for the parlous condition of the population in

areas adjacent to the border by alleging, without any solid proof

or scientific basis, that the s ituation of those individuals is

caused by Colombia’s spraying programme.

(d) The right to water

(i) Ecuado pr’sition

9.95. Ecuador relies on Article 14(2)(h) CEDAW and Article
1072
24(2)(c) CRC as embodying the right to water. In addition,

it notes that in General Comm ent No. 15, the ESCR Committee

stated that Articles 11(1) (right to an adequate standard of
living) and 12(1) (right to an adequate standard of health)

ICESCR embody a right to water, as follows:

“The human right to wate r entitles everyone to
sufficient, safe, acceptable, physically accessible
and affordable water for personal and domestic
uses. An adequate amount of safe water is

over 20 km away from the border.’” (EM, Vol. III, Annex 153: Ecuadorian
Scientific Commission, The Plan Colombia Aerial Spraying System and its
Impacts on the Ecosystem and Health on the Ecuadorian Border (hereinafter
“Ecuadorian Scientific Commission Report”) (April 2007), p. 53.)
1071 EM, para. 2.23.
1072 EM, para. 9.67
467 necessary to prevent de ath from dehydration, to
reduce the risk of water-related disease and to
provide for consumption, cooking, personal and
domestic hygienic requirements.” 1073

9.96. Ecuador alleges that communities in the border region

have no access to running water (i.e., processed and treated tap

water) and that they rely on local rivers “to cook, drink, wash,
bathe or raise domestic animals”. 1074

9.97. By way of conclusion, Ecuador alleges:

“The evidence demonstrates that Colombia is
responsible for failing to protect the right of
access to safe and healthy water. By polluting the
rivers and springs from which the affected

farmers and indigenous peoples in Ecuador draw
their essential live lihoods Colombia has
endangered the health and well-being of the most
vulnerable populations living along the Ecuador-
Colombia border, and si gnificantly interfered

with their rights 1075ater, life, health, property
and private life.”

(ii) Colombia r’sponse

9.98. Article 14(2)(h) CEDAW provides:

“2. States Parties shall take all appropriate
measures to eliminate discrimination against
women in rural areas in order to ensure, on a

basis of equality of me n and women, that they

1073 EM, para. 9.69, citing ESCR Committee, General Comment No. 15,
para. 2.
1074 EM, para. 9.70.
1075 EM, para. 9.74.
468 participate in and benefit from rural development
and, in particular, shall ensure to such women the
right:

[...]

(h) To enjoy adequate living conditions,
particularly in relation to housing, sanitation,
electricity and water supply, transport and
communications.”

9.99. Article 24(2)(c) CRC provides:

“States Parties shall pursue full implementation
of this right and, in particular, shall take
appropriate measures:

[...]

(c) To combat disease and malnutrition,
including within the framework of primary health
care, through, inter alia , the application of
readily available technology and through the
provision of adequate nutritious foods and clean
drinking-water, taking into consideration the
dangers and risks of environmental pollution;”.

9.100. Article 11(1) ICESCR provides:

“The States Parties to the present Covenant

recognize the right of ev eryone to an adequate
standard of living for himself and his family,
including adequate food, clothing and housing,
and to the continuous improvement of living
conditions. The States Parties will take
appropriate steps to ensure the realization of this
right, recognizing to this effect the essential

importance of international co-operation based
on free consent.”
while Article 12(1) provides:

469 “The States Parties to the present Covenant
recognize the right of ev eryone to the enjoyment
of the highest attainable standard of physical and
mental health.”

9.101. As regards the right to water, the core point to be made

is that despite repeated testing by Colombia and even by

Ecuador, the boundary rivers have never been found to be

polluted by glyphosate: in other words, no significant amounts

of the spray mix (in most cases no detectable amounts) have
1076
been found in the relevant rivers.

9.102. The simple answer to this complaint is that, whatever

obligations in the realm of huma n rights or otherwise Colombia

may be under in relation to tr ansboundary waters, there is not

the slightest independent evidence of breach. And this is not

surprising given that the scientific evidence shows that the

ingredients of the spray mix have a short residence time and are
innocuous to humans and animals – a fortiori when further

diluted and dispersed by river waters. 1077

9.103. In relation to Article 14(2)(h) CEDAW, the same point

is to be made as regards Ecuador’s other claims based on

CEDAW, namely that Ecuador has provided absolutely no

evidence of any Colombian di scrimination against women,
whether in Ecuador or in Colombia.

1076 See above, paras. 5.31, 7.24.
1077 See Annex 116, CICAD I, pp. 91-94; Appendix – Dobson Report,
pp. 6-7, 28.
4709.104. As regards Article 24(2)(c) CRC, the specific CRC

obligation relied on requires the provision of infrastructure so as

to provide clean drinking wate r. It is for Ecuador, not

Colombia, to provide such infr astructure: responsibility for
funding basic sanitary facilities and piped water is not to be

transferred to Colombia on the basis of an allegation that it

contributes a minor fraction of the chemical load carried by the

boundary rivers. 1078

9.105. As regards the ICESCR, the Committee’s General

Comment No. 15, in discussing “a ccessibility” as a necessary

component of the right to water, states:

“Water and water facilities and services have to

be accessible to everyone without discriminatio1079
within the jurisdiction of the State party.”

More specifically, as regards “physical accessibility”, the

Committee observed that:

“water, and adequate water facilities and
services, must be within safe physical reach for
1080
all sections of the population.”

1078 Annex 118, CICAD I, Toxicology of Substances used in the
Production and Refining of Cocaine and Heroin: A Tier-Two Hazard
Assessment, p. 19.
1079 ESCR Committee, General Comment No. 15, para. 12(c) (emphasis
added).
1080 ESCR Committee, General Comment No. 15, para. 12(c)(i); see also

para. 12(c)(iii) “Water and water facilities and services must be accessible to
all, including the most vulnerable or marginalized sections of the population,
in law and in fact, without discrimination on any of the prohibited grounds”;
and see para. 14.
4719.106. It is true that General Comment No. 15 goes further,

suggesting that:

“To comply with their international obligations
in relation to the right to water, States parties
have to respect the enj oyment of the right in

other countries. International cooperation
requires States parties to refrain from actions that
interfere, directly or indirectly, with the
enjoyment of the right to water in other countries.
Any activities undertaken within the State party’s
jurisdiction should not deprive another country of

the ability to reali1081he right to water for persons
in its jurisdiction.”

As to this comment, it must be said, first, that there is no
evidence that Colombia is in bre ach of such an obligation, if it

indeed arises under the ICESR, and, secondly, that such general

language cannot be regarded as rewriting the provisions of the

many international watercourse agreements, or for that matter
the 1997 UN Convention, that regulate use of shared

watercourses on a State-to-State basis. No breach by Colombia

of applicable rules governi ng non-navigational uses of

international watercourses has been established.

9.107. As to the supposed eviden ce relied upon by Ecuador as

establishing the breach of a right to water, the principal evidence

is that of individual witnesses, together with the reports of the

Ecuadorian Ministry for the E nvironment, the Commission of

1081 ESCR Committee, General Comment No. 15, para. 31.

472the Ecuadorian Congress and of the Confederation of

Indigenous Nationalities of Ecuador (“CONAIE”).

9.108. The report compiled by the Ecua dorian Ministry for the

Environment in 2003 1082 is based on testimony taken from

residents, with no scientific analysis. The report of the

Ecuadorian Congressional Committee is based on the collection

of testimony taken from residents of the area following a visit in

December 2003 and likewise is not backed up by any scientific
1083
evidence. The report by CONAIE, an advocacy group,

likewise merely produces the hearsay testimony of local
residents.1084

9.109. In the absence of scie ntific studies conducted

immediately after any given alleged instance of spraying,

Ecuador has provided no evidence other than the assertion of

local residents that the sprayi ng is actually the cause of the

problems allegedly experienced. It cannot be excluded that any

pollution of water sources could be due to other causes, in
particular contamination resulting from the use of chemicals in

the cultivation and processing of coca on Ecuadorian

1082
1083 EM, Vol. IV, Annex 166.
1084 EM, Vol. IV, Annex 167.
EM, Vol. IV, Annex 162.
473territory.1085 Similar doubts must be rais ed as to the reports of

the two UN Special Rapporteurs.

(e) The right to a healthy environment

(i) Ecuado pr’sition

9.110. As to Ecuador’s claims in relation to “right to a healthy

environment” (or a “healthy and decent environment” 1086) it

relies on Article 11 of the ACHR ESCR Additional Protocol. 1087

It acknowledges that “[w]hat constitutes a healthy environment

must be determined by reference to the natural, social, economic

and cultural character of the region in question”, 1088 and refers

to the case-law of the Inter-American Court and the Colombian

Constitutional Court. Having acknowledged that the Inter-

American Court has not been ab le to rule directly on the

question given the jurisdictional limitation in Article 19 of the

Additional Protocol, 1089it notes that the Inter-American Court

has considered the scope of the right “in connection with” other

rights, including in Yakye Axa, 1090 where the Court had regard to

Article 11 in determining whether the State had generated

conditions which had “worsened the difficulties of access to a

1085
Annex 118, CICAD I, Toxicology of Substances used in the
Production and Refining of Cocaine and Heroin: A Tier-Two Hazard
1086ssment (2005), pp. 27-28.
EM, para. 9.75.
1087 EM, para. 9.75.
1088 EM, para. 9.76.
1089 EM, para. 9.77.
1090 EM, para. 9.77.

474 1091
decent life for the members of the Yakye Axa Community”,

and concluded that the right to a decent life had been violated.

9.111. Ecuador concludes that:

“protection of the right to a healthy environment
does not amount to restricting the State from
taking any action that impacts the environment.

However, the State, in order to protect human
rights, must exercise due care, take necessary
preventive measures, allow informed
participation of the concerned population, and

provide for ade1092e monitoring
mechanisms”.

9.112. As to breach, Ecuador asserts that none of those

requirements have been complied with by Colombia,

emphasising, inter alia , the connection of the indigenous
1093
peoples in the area to the natural environment.

(ii) Colombia r’sponse

9.113. Article 11 of the ACHR ES CR Additional Protocol

provides:

“1. Everyone shall have the right to live in a
healthy environment and to have access to basic
public services.

1091
EM, para. 9.77, citing Indigenous Community Yakye Axa v.
Paraguay, I-ACtHR, Series C, No. 125, Judgment of 17 June 2005, para.
1092
1093 EM, para. 9.82.
EM, para. 9.83.
475 2. The States Parties shall promote the protection,

preservation, and improvement of the
environment.”

9.114. As regards the interpretati on of Article 11, the basic

point is that that provision is essentially territorial, as is

demonstrated by the reference to “access to basic public

services”: only the territorial state can be obliged to ensure
access to public services, and it is for the territorial state to

ensure that individuals enjoy a healthy environment.

9.115. As to Ecuador’s reliance on the Yakye Axa decision of

the Inter-American Court, two points can be made: first, the
right to a healthy environment was but one of a whole catalogue

of rights referred to by the Court in elucidating its conception of

the right to a “decent life”.1094 Second, as noted above, the case

concerned a very different situation to that at issue in the present

case, given that it involved State actions in relation to an
indigenous community located on its own territory.

9.116. A similar observation may also be made as to the

Saramaka case which also involved St ate action in relation to a

group residing on its own territory. In any event, despite
Ecuador’s attempt to portray th at case as one confirming the

right to a healthy environment, the case principally concerned

violations of the right to prope rty of the applicant indigenous

1094 Indigenous Community Yakye Axa v. Paraguay, I-ACtHR, Series C,
No. 125, Judgment of 17 June 2005, para. 163.
476people under Article 21 ACHR, t ogether with other ancillary

rights, including the right to juridical personality and the right to

judicial protection. Indeed th e Court did not so much as

mention the right to healthy environment contained in Article 11

of the ACHR ESCR Additional Protocol.

9.117. Ecuador relies on the deci sion of the African

Commission in the Ogoniland case, which it treats it as if it

were a decision in relation to th e right to a healthy environment
1095
under Article 11 of the ACHR ESCR Additional Protocol.

In fact it concerned the right of peoples (rather than of

individuals) to a “generally satisfactory environment favorable
to their development” under Article 24 of the Banjul Charter, a

right that has no equivalent in the Inter-American system.

9.118. As to breach, even if it we re accepted that Colombia is

under any obligation as regards the right of individuals resident

in Ecuador to a healthy environment, the simple response is that
Colombia has not committed any of the breaches alleged by

Ecuador; Ecuador has simply not proved that any of the alleged

harms in fact took place. The fact that Ecuador resorts to

exaggeration and hyperbol e, going so far as to allege that

Colombia has destroyed “peoples’ ‘most basic conditions of

1095 Seeinparticular EM, para. 9.81 “With regard to the content of the
right to a healthy environment, th e Commission ruled that [...]” and
“Compliance with the right to a h ealthy environment also includes
obligations related to the right to information and to taking other preventive
measures [...]”, and para. 9.82 “As established by the African Commission,
protection of the right to a healthy environment [...]
477 1096
survival’” does nothing to conceal th e fact that Ecuador has
not even begun to discharge its burden of proof in that regard.

(f) The right to property

(i) Ecuado pr’sition

9.119. As to the right of property, Ec uador refers to Article 21

of the Universal Declaration of Human Rights and Article 21
ACHR. It notes that under the latter provision,

“the right to the use and enjoyment of property
may be limited, but only according to a legal
mandate; a person may only be deprived of his or

her property for reasons of “public utility or
social interest, and in the cases1097 according to
the forms established by law”.

9.120. Ecuador notes that many families near the border “own

little more than a small plot of land, the crops they cultivate and

the few animals they raise.” In that regard, it alleges that:

“The loss of crops and animals occasioned by
Colombia’s aerial spraying of herbicides has

brought irreparable harm to many families,
especially to those located closer to the border.
Over a period of nearly nine years, in many
cases, the soil has not recovered and the

productivity of1098ms has decreased
significantly”

as well as alleging impact on in come resulting in an inability to
repay loans.

1096 EM, para. 9.84.
1097 EM, para. 9.85.
1098 EM, para. 9.86.
4789.121. As to breach, reference is simply made to the alleged

harms set out in Chapter VI of the Memorial. It is alleged that:

“Subsistence crops have been destroyed or

damaged, and domestic animals have died. As a
result, many families have lost a great deal of
their property or livelihood, and many have been
forced to move to other areas.” 1099

This is baldly said to implicate a violation by Colombia of its
1100
obligation to respect the right of property.

(ii) Colombia r’sponse

9.122. Article 21 ACHR provides:

“1. Everyone has the right to the use and
enjoyment of his property. The law may

subordinate such use and enjoyment to the
interest of society.

2. No one shall be deprived of his property
except upon payment of just compensation, for
reasons of public utility or social interest, and in
the cases and according to the forms established

by law.
3. …”

9.123. Again, the key point is that Ecuador has not

substantiated its claims of prope rty damage. It simply makes

vague general claims in that regard without providing any

specification of the damage allege dly suffered. For the indirect

1099 EM, para. 9.87.
1100 Ibid.
479impacts of aerial spraying in Colombia to even begin to raise an

issue under Article 21, the de privation would have to be
substantial, yet the scientific evidence st rongly indicates that

spray drift at a distance cannot have such an effect. 1101

9.124. For the reasons set out in Ch apter 7, Ecuador has failed

to provide any evidence substa ntiating its claims of property

damage. The majority of the statements made by Ecuador’s

witnesses are vague as to the date , or even as to the year in

which the spraying causing damage is alleged to have taken

place. There are a large numbe r of inconsistencies in the
witness evidence.

9.125. Ecuador’s allegations as to the effect of the spray

mixture on soil are contradicted by all available scientific

evidence. There is no evidence that the spray mixture causes

degradation of the soil and thus there is nothing from which the
1102
soil must “recover”. Rather, the scientific evidence is clear

that the spray mix is broken down into harmless compounds
within a short time, and that it does not accumulate in soils or

water. 1103 Tropical soils decline under a regime of “slash and

burn” agriculture for well-understood reasons, which have

nothing to do with Colombia’s aerial spraying in its own

territory.

1101
1102 See above, paras. 7.16-7.30.
1103 EM, para. 9.86.
See above, para. 7.84-7.85.
480 (g) Right to humane treatment

(i) Ecuado pr’sition

9.126. Ecuador’s claim in relation to alleged violation of the
right to humane treatment is based on Article 5 ACHR. 1104

Ecuador alleges that…

“the right to psychol ogical integrity of the
Ecuadorian population in the border region has

been violated because people have been subject
to severe emotional distress caused by the direct
impacts of Colombia’s aerial fumigations on
their lives.”1105

9.127. In that regard, Ecuador alle ges that illnesses and other

harms to health and livelihood have had “a severe psychological
impact on the population”. 1106 Reference is made to certain of

the witnesses who testify as to effects of the spraying upon

them. 1107 Further, Ecuador places reliance on the Report of the

UN Special Rapporteur on the Right to Health who stated that

there is “credible, reliable evidence” that the aerial spraying

damages their mental health” 1108

9.128. Ecuador further argues, th at “the right to humane

treatment, as recognised in Article 5 of the American

Convention, requires protection from other sources of distress”,

1104 EM, para. 9.88.
1105 EM, para. 9.89.
1106 EM, para. 9.89.
1107
1108 EM, paras. 9.89-9.91.
EM, para. 9.92.
481referring to the decision of the Inter-American Court in the

Moiwana case as to the distress caused to indigenous peoples by

being “deprived of the right to continue to live in their
1109
traditional lands”.

(ii) Colombia r’sponse

9.129. Article 5 ACHR provides (so far as relevant):

“1. Every person has the right to have his
physical, mental, and moral integrity respected.

2. No one shall be subjected to torture or to
cruel, inhuman, or de grading punishment or
treatment. All persons deprived of their liberty
shall be treated with respect for the inherent

dignity of the human person.”

9.130. Even assuming that Article 5 ACHR is applicable to the
alleged indirect effects in Ecuador of Colombia’s conduct in law

enforcement on its own territory, a certain degree of severity is

necessary before a violation of Ar ticle 5 will occur, as Ecuador

itself recognizes. Ecuador ha s provided no proof of any

particular psychological harm which any individual has
suffered.

9.131. In that regard, at least some of the witnesses relied upon

by Ecuador are unreliable. Fo r example, the mother who

describes the anguish she allegedly suffered as the result of the

1109 EM, para. 9.93 citing Moiwana Community v. Suriname ,
Preliminary Obections, Merits, Reparatio ns and Costs, I/ACtHR, Series C,
No. 124, Judgment of 15 June 2005, paras. 101-103
482deaths of two of her daughters is none other than Witness 11. As

set out above, given, the inconsistency of Witness 11’s evidence
with that of other witnesses, the Court should harbour strong

reservations as to its accuracy.

9.132. The opinion of the Special Rapporteur on the Right to
Health that there is “credible, reliable evidence” that spraying

damages the mental health of individuals in the border area

would appear to be no more than his account of conversations
with local residents as to distant events. The Special Rapporteur

gives no indication of the precise damage which has supposedly

been caused, nor does it appear th at any efforts were made to

verify the accounts of the residents. The Special Rapporteur is
of course entitled to express his views, but his assessment as to

the quality of the evidence is non-judicial and adds nothing to

what he was told by his informants. In addition it is not
supported by any scientific evidence.

(h) Right to private life

(i) Ecuadopr’sition

9.133. As regards the right to private life, Ecuador alleges that

“the spraying of toxic chemicals on the border
area has severely disrupted the lives of local
communities over many years, to the point that

their lives have been transformed. These
sprayings have interfered with their most

483 intimate aspects of life, affecting their choices
1110
and the way they lead their daily existence.”

It further argues that the spra yings have resulted in the

destruction of means of subsis tence and poor health, and that
1111
this has caused families to abandon their homes. Ecuador

relies in particular on Articl e 17 ICCPR, although noting that

similar obligations are contai ned in Article 11 ACHR and
1112
Article 16 CRC.

9.134. Ecuador alleges that “freedom from arbitrary or unlawful

interference includes having one ’s home and family life free
1113
from significant pollution”, relying on the jurisprudence of

the European Court of Human Rights, in particular the decisions
1114
in López Ostra v. Spain and Guerra v. Italy.

9.135. Ecuador alleges that individuals in the border areas were

periodically exposed to herbicides, with the result that “farmers

and indigenous peoples have found it difficult to work their

fields, to make use of the river, to send their children to school,

to lead their daily lives undist urbed and to enjoy their home

lives in a manner to which they are entitled.” 1115It is further

alleged that families have been forced to leave their homes and

relocate away from the border, with some families being

1110
1111 EM, para. 9.95.
1112 EM, para. 9.95.
EM, para. 9.96.
1113 EM, para. 9.98.
1114 EM, para. 9.98.
1115 EM, para. 9.99.
484separated as a result. 1116 Specific reference is made to the

alleged effects on indigenous peoples, in relation to which the

spraying is said to have caused the abandonment of ancestral

lands, resulting in “an end to enjoyment of homes” and “the
1117
disintegration of families”.

(ii) Colombia r’sponse

9.136. Article 17 ICCPR provides:

“1. No one shall be subjected to arbitrary or
unlawful interference with his privacy, family, or
correspondence, nor to unlawful attacks on his
honour and reputation.

2. Everyone has the right to the protection of
the law against such interference or attacks.”

9.137. Article 11 ACHR provides:

“1. Everyone has the right to have his honor
respected and his dignity recognized.

2. No one may be the object of arbitrary or
abusive interference with his private life, his
family, his home, or his correspondence, or of
unlawful attacks on his honor or reputation.

3. Everyone has the right to the protection of the
law against such interference or attacks.”

9.138. Article 16 CRC provides:

“1. No child shall be subjected to arbitrary or

unlawful interference with his or her privacy,

1116 EM, para. 9.100.
1117 EM, para. 9.101.
485 family, or correspondence, nor to unlawful
attacks on his or her honour and reputation.

2. The child has the right to the protection of the
law against such interference or attacks.”

9.139. Despite the various matters referred to by Ecuador, it
appears only to rely upon the right to private life as regards the

alleged displacement of individuals. 1118

9.140. Even assuming that the right to respect for private life

and home under the various inst ruments is applicable in

principle in the present case, the facts relied upon by Ecuador do

not disclose any breach of that right.

9.141. Ecuador relies on decisions of the European Court of

Human Rights in order to support its argument that aerial

spraying such as that in the present case is capable of giving rise

to a violation of the right to respect for home and private life.

However, the two leading cases, López-Ostra 1119and Guerra, 1120

concerned situations very different from that in the present case;

in both cases, the applicant was re sident on the te rritory of the

respondent State and alleged th at the State had not taken

sufficient steps to protect the applicant against pollution
emanating from third parties, not from the State itself. The cases

thus concerned alleged breach of a positive obligation to take

1118
1119 Cf. EM, para. 9.102.
López Ostra v. Spain (App. No. 16798/90), Judgment of 9
1120mber 1994, Series A, No. 303-C.
Guerra v. Italy (App. No. 14967/89), Judgment of 19 February
1998, Reports 1998-I.
486measures to ensure respect for the right to home and private life,

rather than direct interference by the State.

9.142. Further, even assuming that the approach under Article 8

of the European Convention can be transposed to provisions

which are phrased in very different terms, the European Court

was careful to emphasise that only “ severe environmental

pollution” is capable of interfering with the rights of individuals
1121
under Article 8 of the European Convention. There is no

basis for the claim that occasiona l local spray drift of the spray

mix crosses this threshold: indeed, it would not do so even in the
context of direct ove rspray, given that a particular plot is

sprayed at most twice a year and that the spray is not persistent

in air or soils. 1122 De minimis pollution (even assuming it

occurred) cannot be said to interfere either with enjoyment of

home or private life.

9.143. In the present case, for the same reasons as in relation to

Ecuador’s claims in relation to transboundary harm, Ecuador
has simply not established that aerial spraying in Colombia has

resulted in any significant drift of spray mixture into Ecuador, or

that such drift as may have occurred has resulted in any

1121
See López Ostra v. Spain (App. No. 16798/90), Judgment of 9
December 1994, Series A no. 303-C, para. 51 and Guerra v. Italy (App. No.
14967/89), Judgment of 19 February 1998; Reports 1998-I , para. 60
1122hasis added).
See above, paras. 4.79, 7.84-7.85.
487 1123
harm. A fortiori it cannot meet the standard of “severe
environmental pollution”.

(i) The right to information

(i) Ecuado prsition

9.144. Finally, as regards alleged breach of the right to

information, Ecuador alleges that Colombia…

“has persistently failed to provide adequate
information concerning the aerial sprayings.
Information has been withheld on timings and

locations, and on the chemical composition of the
materials that have been used in the sprayings.
Even now the Government of Ecuador and the
local population in the a ffected areas have not

been informed about the specific composition of
the herbicide compound Colombia has used over
time in the aerial sprayings, the concentration of
the chemicals within the mixture, or the location
1124
or times of the spray campaigns.”

9.145. In support of the supposed right to information, Ecuador
relies on a number of decisions of the European Court of Human

Right, including Öneryildiz v. Turkey and Taskin v. Turkey , as

well as the decision of the African Commission in the

Ogoniland case. 1125As regards the Inter-American system,

Ecuador alleges that it has long been recognized that:

“the failure to enga ge in ‘meaningful

consultation’ with indigenous communities in

1123 See above, para. 7.16-7.30.
1124 EM, para. 9.103.
1125 EM, para. 9.104.
488 connection with activities affecting their
traditional lands will result in a violation of
various human rights” 1126

and that consultation with indigenous peoples and participation

in decision-making are also required by Article 6 of ILO
1127
Convention No. 169 and Article 27 ICCPR.

9.146. As to the alleged breach of the right to information,
Ecuador simply asserts that there was no consultation with local

inhabitants about the spraying programme, and that the failure to

provide even a minimum of information meant that the

population was left entirely uninformed. 1128

(ii) Colombia r’sponse

9.147. Ecuador’s claim as to the s upposed right to information

is not only a reprise of its argument that international

environmental law condemns a failure to provide information; it

is also just another way of putting its claims as to the rights to

life, private and family life under ILO Convention 169 and

“applicable articles” of the ICCPR.

9.148. The foundation upon which Ecuador seeks to build this

supposed right to information in relation to the various

substantive rights is shaky. It relies on decisions of the

European Court and the African Commission, but not to any

1126
1127 EM, para. 9.105.
1128 EM, para. 9.105.
EM, para. 9.106.
489decision by the Inter-American Court of the Human Rights

Committee. Further, the decisions relied on are not on point.

9.149. The observations of the European Court in Öneryildiz

were made in relation to the public’s right to information as part

of procedures for the licensi ng, setting up, operation, security
and supervision of hazardous ac tivities carried out within the

respondent State. 1129 Further, in Öneryildiz, the Court was

concerned with a situation in which death of the victims had

already occurred as the result of an accident in the operation of a

hazardous activity, with the result that the Court was concerned

to examine the positive obligation “to take all appropriate steps
to safeguard life for the purposes of Article 2”; in that regard,

the Court took the view that the positive obligation “entails

above all a primary duty on the State to put in place a legislative

and administrative framework de signed to provide effective
1130
deterrence against threats to the right to life”. It was in that

context that the Court made its observations as to the public’s
right to information

9.150. Similarly, the European Court in Taskin v Turkey was

faced with a situation in which an EIA had in fact been carried

out in relation to a particular activity; the Court emphasised the
importance of public access to su ch information, but it did not

1129 Öneryıldız v. Turkey (App. No. 48939/99), Judgment of 30
November 2004, Reports 2004-XII [GC], para. 90.
1130 Ibid., para. 89.
490formulate the question of access as a freestanding right; rather it
was a factor which should be taken into account in determining

whether the decision-making process as a whole had been

fair.131

9.151. Finally, in the Ogoniland case, the African Commission

was concerned with the right of peoples to “a generally

satisfactory environment favorable to their development”. It is
hardly surprising that the Af rican Commission felt able to

identify a requirement that th e people in question should be

provided with information and consulted. Such an approach is

not transposable to the present case.

9.152. As to the alleged breaches of the right to information,

Ecuador seems to be in two minds as to whether the breach is
one caused by the lack of inform ation provided to Ecuador, or

the lack of information provided to the individuals in the area in

question. 1132 That indecision is telling insofar as it reveals what

appears to be the true motivation behind Ecuador’s claim,

namely to provide an alterna tive vehicle for its own demands

that it be provided with information. Even at a point when there

is no longer any spraying near the border, Ecuador seems to
continue to insist on its own asse rted right (rather than that of

the individuals) to know the co mposition and concentration of

1131 Taskin v. Turkey (App. No. 46117/99), Judgment of 10 November
2004; Reports 2004-X, paras. 118-119.
1132 Cf. EM, para. 9.103.
491the spray mix and the location and times of spraying. Colombia
was under no obligation to provide information to Ecuador with

regards to the timing of flights, in particular taking into account

evident security concerns associ ated with production of illicit

narcotics. Nonetheless, Colombia cooperated with Ecuador and

provided information about the spraying program, as
1133
demonstrated above.

C. Alleged Breach of Indigenous Rights

(1) E CUADOR ’S RELIANCE UPON INDIGENOUS RIGHTS

9.153. In its Memorial, Ecuador places heavy emphasis upon

the rights of indigenous peoples living in Ecuador which, it

says, have been particularly affected by the aerial spraying. For

example it claims

“serious disruption of the traditional way of life
of indigenous communiti es who live, farm and
hunt in the affected areas. Pollution damage has

significantly harmed the natural resources and
environment on which these communities
depend. It has displaced some communities from
their homes, deprived them of traditional

medicines, interfered with their right to use and
enjoy their property, and denied them the right to
enjoy own culture.” 1134

It also endorses descriptions gi ven by a number of witnesses.

For example, according to Witness 40:

1133 See above, paras. 5.9-5.16, 5.27-5.28, 5.35-5.44, 8.105-8.106 and
8.111-8.112.
1134 EM, para. 9.37.
492 “The third time they sprayed was fatal for our
community. …All the plants, big and small,
were destroyed. … I estimate that at least some
thirty species of plants that died were used by us

in the Awá traditional medical treatments. …
We no longer had anything to eat. Our diet
depended on the plants that we sow and those

that are in nature,1135ch have been affected by
the fumigations.”

In short, Colombia’s aerial spraying has been said to produce for
a significant group of indigenous people the “impossibility of

continuing with their traditional lives”. 1136

(2) C OLOMBIA ’S RESPONSE ON THE FACTS

9.154. The short answer to this argument is that the evidence of

differential impact of the spraying on indigenous Ecuadorians is

exiguous. Ecuador’s evidence of sp ecial harm to its indigenous
1137
peoples was reviewed in Chapter 7. It takes the familiar

form of vague, generic assertions relating to events years before,

unaccompanied by medical or other material evidence. On what
basis aerial spraying in Colombia could threaten the lifestyle of

indigenous peoples occupying a su bstantial range of territory

Ecuador does not explain. In no way can it be said to have

made it impossible to continue with “their traditional lives”.

1135
1136 EM, para. 6.117, citing Annexes, Vol. IV, Annex 223.
1137 EM, para. 9.23.
See above, paras. 7.177-7.186.
4939.155. Colombia has conducted aeria l spraying exclusively on

its own territory. It is entitled to do it, as Ecuador accepts. 1138

There is no suggestion that it ha s targeted persons in Ecuador,

still less the indigenous peoples of Ecuador. There is no proof

that the real and urgent problems of those peoples had any
causal relation to aerial spraying.

9.156. This being so, the controversia l subject of the extent of

indigenous rights in internat ional law does not arise.

Nonetheless the following remarks are offered.

(3) A LLEGED BREACH OF INDIGENOUS TREATY RIGHTS

9.157. Ecuador relies on three treaties as a basis for its claim on

behalf of its indigenous peoples. These will be dealt with in

turn.

(a) ICCPR, Article 27

(i) Ecuado p’sition

9.158. First, Ecuador alleges that Colombia is in breach of

Article 27, ICCPR, which provides:

“In those States in whic h ethnic, religious or
linguistic minorities exist, persons belonging to
such minorities shall not be denied the right, in
community with the other members of their

group, to enjoy their own culture, to profess and

1138 E.g., EM, paras. 7.18, 8.36.
494 practice their own religi on, or to use their own
language.” 1139

It goes on to stress the importance of indigenous links with the

land, and of traditional fishing and hunting.

(ii) Colombiaresponse

9.159. A first and obvious point is that in principle the status of
a given minority (indigenous or ot herwise) is relative to the

society in which they live. The primary obligations towards that

people are obligations on the part of the State to which they

belong. The New Zealand Maori are not indigenous peoples in

relation to, say, Australia, nor the Australian Aboriginal peoples
to New Zealand. This is expressly recognised in Article 27 by

the phrase “[i]n those States in which ethnic , religious or

linguistic minorities exist”. It is Ecuador which is responsible

under Article 27 for protecting th e rights and interests of its
indigenous people.

9.160. Secondly, Article 27, which applies to indigenous groups

if and to the extent that they can be regarded as ethnic or
linguistic minorities, is deliberately formulated in negative

terms (“shall not be denied the right”). In fact nothing that

Colombia has done has denied the indigenous peoples of

Ecuador the three enumerated rights.

1139 EM, paras. 9.18-9.23.

4959.161. Thirdly, the rights attributed to minorities by Article 27

are limited ones – insofar as releva nt here, the right to “enjoy

their own culture”. If the pe ople concerned have abandoned

their culture, this is likely to be for a range of reasons, including
the cash economy, indiscriminate logging, the exploitation of

oil, the influx of persons from outside, the introduction of

alcohol and western foods, etc. It is absurd to single out

occasional aerial spraying as the decisive cause of a story of
decline that has been repeated on so many occasions, both

within the region and outside it.

(b) ILO Convention 169

(i) Ecuadopr’sition

9.162. Secondly, Ecuador alleges that Colombia is in breach of

ILO Convention169, Articles 4, 5, 6, 7, 13 and 15, relying on

the same witness statements an d reports. The United Nations
Declaration of the Rights of Indigenous Peoples of 7 September

2007 1140is also relied on, and presented as a restatement of

principles and rights reflected in ILO Convention 169– a

considerable understatement si nce there is much in the
Declaration which is new. In fact the Declaration was explicitly

adopted as “a standard of achieve ment to be pursued in a spirit

1140 UNGA Res 61/295, 13 September 2007 (adopted 143-4:11).

496of partnership and mutual respect”, i.e. as an aspirational
1141
document.

(ii) Colombia r’sponse

9.163. Again, there is an issue as to whose obligations are

engaged. The point is explicit ly addressed in ILO Convention

169, the Convention concerning In digenous and Tribal Peoples

in Independent Countries of 27 June 1989, 1142as it was in its

predecessor Convention, ILO Convention 107. ILO Convention

169 has not been widely ratified (only 20 States parties), but it

has been ratified by both Ecuador and Colombia and is in force

between them.

9.164. Article 1 of ILO Convention 169 defines its scope of

application by contrast with “o ther sections of the national

community”, or by reference to their descent from peoples who

“inhabited the country, or a geogr aphical region to which the

country belongs, at the time of conquest or colonization or the
establishment of present state boundaries”. 1143 In respect of a

1141
It is unnecessary to take any position on whether and to what extent
the provisions of the 2007 Declaration reflect existing international law. For
the view that they do not, see e.g.S. Allen, “The UN Declaration on the
Rights of Indigenous Peoples: Towards a Global Legal Order on Indigenous
Rights?” in A. Halpin & V. Roeben (eds). Theorizing the Global Legal Order
(Hart Publishing, Oxford, 2009), available at SSRN:
http://ssrn.com/abstract=1400665.
1142 1650 UNTS 383.
1143 Art. 1 provides:
1. This Convention applies to:

(a) tribal peoples in independent countries whose social, cultural and
economic conditions distinguish them from other sections of the national
497given indigenous or tribal people in an independent country, it is

the Government of that c ountry which assumes primary

responsibility for the fulfilmen t of the obligations under the

Convention. This is clear, for example, from the key provision
of the Convention, Article 2, which provides:

“1. Governments shall have the responsibility
for developing, with the participation of the
peoples concerned, co-ordinated and systematic
action to protect the rights of these peoples and to
guarantee respect for their integrity.

2. Such action shall include measures for:

(a) ensuring that members of these
peoples benefit on an equal
footing from the rights and
opportunities which national laws
and regulations grant to other

members of the population;
(b) promoting the full realisation of
the social, economic and cultural

rights of these peoples with
respect for their social and cultural
identity, their customs and
traditions and their institutions;

(c) assisting the members of the
peoples concerned to eliminate
socio-economic gaps that may
exist between indigenous and

community, and whose status is regulated wholly or partially by their own
(b) peoples in independent countries who are regarded as indigenous on
account of their descent from the populations which inhabited the country, or
a geographical region to which the count ry belongs, at th e time of conquest
or colonisation or the establishment of present state boundaries and who,
irrespective of their legal status, retain some or all of their own social,
economic, cultural and political institutions.”

498 other members of the national
community, in a manner
compatible with their aspirations
and ways of life.” 1144

The only provision dealing with tr ansboundary issues is Article

32, entitled “Contacts and Co-Operation across Borders”, which

is irrelevant to the present case. 1145

9.165. Indigenous communities in Ecuador have not been

affected by aerial sprayings in Colombia. Their economic and

social conditions are the exclusive responsibility of the
Government of Ecuador.

9.166. This appears from the most relevant of the articles of

ILO Convention 169 cited by Ecuador, Article 7, which

provides:

“3. Governments shall ensure that, whenever

appropriate, studies are carried out, in co-
operation with the peoples concerned, to assess
the social, spiritual, cu ltural 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 activities.

1144
Emphasis added. See also Arts. 6(1)(b), 7(1), 9(1), 14(3), 19, 26,
1145), 29, 31.
Article 32 provides:
“Governments shall take appropriat e measures, including by means of
international agreements, to facilitate contacts and co-operation between
indigenous and tribal peoples across borders, including activities in the
economic, social, cultural, spiritual and environmental fields.”
499 4. Governments shall take measures, in co-
operation with the peoples concerned, to protect
and preserve the environment of the territories
they inhabit.”

Clearly it is the territorial sovereign which has to perform these

functions; only the State in que stion can engage in “planned
development activities”.

9.167. It may be noted that Article 29 of the 2007 UN

Declaration also deals with the environment:

“1. Indigenous peoples have the right to the
conservation and protection of the environment
and the productive capacity of their lands or
territories and resources. States shall establish
and implement assistance programmes for

indigenous peoples for such conservation and
protection, without discrimination.
2. States shall take effective measures to ensure
that no storage or disposal of hazardous materials

shall take place in the lands or territories of
indigenous peoples without their free, prior and
informed consent.
3. States shall also take effective measures to

ensure, as needed, that programmes for
monitoring, maintaining and restoring the health
of indigenous peoples, as developed and
implemented by the peoples affected by such
materials, are duly implemented.”

Again it is a matter for Ecuador to take responsibility for such

matters as assistance programmes for its own indigenous
peoples, the storage of hazardous materials on Ecuadorian

territory and the provision of heal th service programmes there.

500Neither Article 29, nor any other provision of the Declaration,

deals with transboundary harm from another State. In the event,

even the most advanced instrument at the international level in

the field of indigenous rights, the 2007 Declaration, does not

deal with the subject of the present dispute.

(c) ACHR, Article 21

(i) Ecuado prsition

9.168. Thirdly, Ecuador invokes Article 21 of the American

Convention on Human Rights, the property provision, and

emphasises the progressive sta nd taken by the Inter-American
Court in giving effect to it. 1146 It accuses Colombia of having

“fractured these vital anthropolog ical and cultura l relationships

through its chemical fumigations”, 1147thereby impairing the

right of indigenous people to their property, i.e. their traditional

lands.

(ii) Colombia r’sponse

9.169. It is, of course, for Ecuador, acting in conformity with its

human rights obligations, to recognise the land rights of
indigenous groups on its territory. These ri ghts have not been

affected in any way by aerial sprayings in Colombia and for this

reason as well as those already gi ven, there is no factual basis

for alleging a breach of Article 21 ACHR by Colombia.

1146 EM, paras. 9.30-9.36, citing, inter alia , Mayagna (Sumo) Awas
Tingni Community v. Nicaragua, IACHR, 31 August 2001.
1147 EM, para. 9.30.
501 D. Conclusions

9.170. The aerial spraying program in Colombia has not

violated human rights in Ecuador, or the rights of its indigenous

peoples.

9.171. Article 14(2) of the 1988 Convention does not purport to

impose independent obligations as regard fundamental human

rights or indigenous rights. Ra ther, it is concerned with
imposing an obligation upon States to adopt measures to prevent

illicit cultivation of plants co ntaining narcotic or psychotropic

substances, subject to the qualification that such measures
respect or take into account the considerations mentioned.

9.172. Moreover, the obligations de rived from human rights

instruments invoked by Ecuador are the exclusive responsibility
of the states parties to those instruments to individuals falling

within their own jurisdiction. They do not apply to the case of

alleged injury in Ecuador as a consequence of the aerial
spraying program carried out by Colombia within its own

territory.

9.173. What Ecuador’s materials do show is the serious
responsibility attribut able to it due to its abandonment and

neglect of the communities – among them, indigenous

502communities – residing in the borde r area with Colombia in the

provinces of Sucumbíos, Carchi and Esmeraldas. This situation
has deteriorated further due to the oil exploration and

exploitation activities that have long been carried out in these

areas.

503504 Chapter 10

THE REMEDIAL SITUATION

A. Ecuador’s Proleptic Approach to Injury

10.1. In Chapter X of the Memorial, Ecuador repeats its claims

that “Colombia’s actions have caused grave, continuing and
long-lasting harms to Ecuador: to its sovereignty, to its people

and property, including indigenous peoples, and to its

environment.” 1148 It goes on to specif y the remedies sought,

drawing a sharp distinction between what it needs to establish in

order to obtain declaratory relief and what damages it intends

eventually to prove. Relying on the Court’s approach in Armed

Activities on the Territory of the Congo, Ecuador states that it…

“does not propose at this stage of the proceedings
to ‘demonstrate and prove the exact injury that
was suffered’ as a result of specific actions of

Colombia; Ecuador is evaluating all the damages
that have been suffered and will tender specific
and complete evidence on all the harms, together
with a detailed claim for monetary compensation,
1149
in the next phase of these proceedings.”

10.2. There are a number of difficulties with this approach.

1148 EM, para. 10.2.
1149 EM, para. 10.4, citingArmed Activities on the Territory of the
Congo (Democratic Republic of Congo v. Uganda), Judgment, ICJ Reports
2005, p. 93, para. 260.
505• First, it amounts to a confession – albeit a confession

of something which is obvi ous to a reader of the
Memorial – that Ecuador has not yet evaluated the

damages it claims to have suffered. Ecuador covers

up the lack of such an evaluation with generalised

assertions of harm, and by repeating the word “toxic”
119 times. But, as demonstrated above, the gist of

Ecuador’s claim is that Colombia has failed in its

duty of due diligence not to cause significant
transboundary harm; Colombia denies it has failed to

show due diligence, but it also denies that its aerial

spraying has caused tran sboundary harm. This

Colombian claim has very substantial scientific
support. Faced with the scientific evidence, Ecuador

cannot simply postpone to the quantum phase the

proof of a material element of its principal claim.
Given that actual damage is the gist of this ground of

responsibility, if Ecuador cannot prove such damage,

it is entitled to no remedy at all. In particular,

Colombia should not be put to the trouble and
expense of a second phase of the case on the basis

that some harm might have been caused and might

have been linked to aerial spraying. If all Ecuador
can show are these small “mights”, then Ecuador has

failed to make out its case and Colombia is entitled

to a declaration to that e ffect. It would be an odd

quantum phase that in effect held there was no
506 liability on the merits of the principal claim, yet there

is a real risk of this, on Ecuador’s approach to the

matter.

• The point is all the stronger in that, if Ecuador had

been injured as alleged, th is could be very easily

proved. Ecuador would not be reduced to such
claims as the following:

“the extent to which these threats have
become reality remains unknown
precisely because so many of Colombia’s
fumigations have been conducted
adjacent to undisturbed primary

ecosystems. In some cases, there are no
human witnesses to 1150eive the impacts
in those regions.”

The Court is apparently required to intuit harm rather
than having it proved.

• Thirdly, Ecuador does not explain how– other than

through mere assertion in still further unsupported
affidavits – it proposes to “evaluate” transient harms

which allegedly occurred between 6 and 10 years

ago. How to establish causation at such a distance of

time? The effort is likely to be disproportionate to
the result. For example, it may be asked, what

records of injury exist in the Lago Agrio hospital to

which Dr Sánchez did not have access when he

1150 EM, para. 6.105.

507 produced his statement? 1151What tests of water

purity were conducted at re levant times and remain

undisclosed? The Court can infer from Ecuador’s

unwillingness to engage with specific facts of actual
and substantial harm that the answer to such

questions is: none.

10.3. There are also acute difficulties of method, which
Ecuador’s “wait and see” approach conceals. For example, on

any view, the other problems of the region – malnutrition,

poverty, lack of basic infrastruc ture, presence of illegal armed

bands, large-scale use of noxious chemicals in coca processing –
will account for far more of the injuries of which Ecuador

complains than hypothetical and occasional drift of spray from

the Colombian side. Colombia has no responsibility for the

most likely local causes of injury to the population and the

environment. Yet just how these other, much more serious,
elements are to be factored in to a damages calculation Ecuador

does not begin to explain.

10.4. Even more objectionable is the suggestion that the Court
should order guarantees against non-repetition in the absence of

proof of any wrongful act on the part of Colo mbia – i.e. in the

absence of the proof of damage in a case where the principal

claim has damage as a necessary component of the cause of

1151 Cf. above, para. 1.33, note 85.

508 1152
action. The Court has already de monstrated considerable

reticence in the matter of assu rances and guarantees of non-
repetition.1153 The need for caution is even greater in a case

where harm is the gist of the wrong and the Claimant effectively

admits that it has yet to evaluate the harm allegedly caused.

10.5. Further, the claims for compensation, contained in

Chapter X of the Memorial, only attempt to substantiate damage

claimed in relation to certain of the alleged violations of human

rights and the rights of indigenous peoples (right to life and

health; damage to property and livelihood). Ecuador reserves
the right to supplement the heads under which it claims

compensation at a later date. 1154 In effect, then, Colombia has

not yet been told, even at the level of principle, the damages

claim it has to meet.

10.6. In these circumstances, there is in Colombia’s respectful

view no point in yet another general account of the law relating

to full reparation, and no need to swap dicta from cases decided
on their own facts which are quite unlike those of the present

case. For instance, Ecuador relies on the environmental damage

decision issued by the UNCC, which it describes as “the modern

approach”, in disparaging contrast with the outdated Trail

1152
1153 EM, paras. 10.11-10.13.
See e.g. LaGrand (Germany v. United States of America),
Judgment, ICJ Reports 2001, pp. 508-514, paras. 117-127; Avena (Mexico v.
1154ed States of America), Judgment, ICJ Reports 2004, p. 69, para. 150.
EM, para. 10.58. See also para. 9.110.
509Smelter award with its emphasis on direct and proven
damage. 1155 But the analogy between incidental drift of a

glyphosate mixture in the course of a lawful program on the

territory of a State, on the one hand, and the wanton destruction

of oilfields and the causing of deliberate environmental harm

following an invasion in breach of the UN Charter, on the other

hand, is not immediately obvious. Likewise the analogy

between the determination of reparation by this Court after a full
forensic process, on the one hand, and the assessment of damage

by a non-court pursuant to a Secu rity Council resolution laying

down a unilateral process, on the other hand, is not apparent.

10.7. For these reasons, Colombia will address all issues of

legal consequences, including quant ification, if and when they

may arise and in light of the Cour t’s actual findings of fact at
that stage.

B. Ecuador’s Attempt to Restrict Colombia’s
Sovereignty over its Own Territory

10.8. One issue that does require discussion, however, is

Ecuador’s claim that Colombia be ordered not to conduct aerial

spraying operations on Colo mbian territory “near” the
1156
border.

1155
EM, para. 10.18. The UNCC is referred to repeatedly: see EM,
paras. 10.19, 10.33, 10.37, 10.42, 10.44, 10.47, 10.48, 10.52, 10.55.
1156 EM, para. 10.13 (“at, near or acr oss the border with Ecuador”); see
also Submissions, p. 413, para. C (vi) (“on or near any part of its border with
Ecuador”).
51010.9. The diplomatic exchanges between the parties on the

question of a buffer zone we re recounted in Chapter 5. 1157

Whenever the issue was raised, Ecuador sought a 10 kilometre

“no spray” zone along the Colombian side of the border.

Eventually, Colombia acceded to this request, voluntarily and
without prejudice to a resumption of spraying activity as might

be necessary. In consequence, Colombia suspended operations

in that area of the border between December 2005 and

December 2006, and again, since February 2007 up to the

present. In fact, as has been seen, maintaining the 10 km zone

has been expensive in the lives of the manual eradicators and
costly in terms of the success of the struggle against illicit

drugs. 1158 It has led to a significant concentration of illegal

activity in the 10 km zone.

10.10. In its Memorial, Ecuador is surprisingly vague on this

front. It now says:

“To the extent that aerial herbicide spraying
continues, the damage to the territory, population
and environment of Ecuador described in Chapter

VI will continue. Accordingly, Ecuador seeks an
order from the Court that Colombia should fulfil
its international obligation to Ecuador by
refraining from further aerial spraying activities

1157 See above, paras. 5.20, 5.34, 5.36, 5.64-5.71, 5.86-5.92.
1158 See above, paras. 4.4 (and note 46), 4.37 (and note 293), 4.34, 4.75,
5.81-5.84, 5.87, 5.109 (6).
511 that deposit herbicides at, near or across the
1159
border with Ecuador.”

10.11. An initial point is that Colombia does not claim any right
to spray across the border. Its position is that the spraying

activity must occur on Colombian territory, with due diligence

shown to prevent spray drift across the borde r in such a way as

to risk causing significant harm. Furthermore, as described in

Chapter 4, no spraying occurs in a 100m buffer zone from the
boundary rivers, and Colombia does not propose to change that

existing regulation. In Colombia’s view and for the reasons

already given, such a buffer zone is perfectly adequate to

prevent spray drift reaching Ecuadorian waters or territory.

C. Conclusion

10.12. For the reasons already gi ven, however, the issue of

remedies does not arise. It is a sufficient answer to Ecuador’s

claims for the Court to hold that neither has significant harm
been proved, nor has it been proved that any such harm was

caused by aerial spraying. Colombia has exercised due

diligence in the prevention of transboundary harm. That being

so, there can have been no failure to respect human rights or the
rights of indigenous people in Ecuador. Ecuador’s claims

relating to assessment, cooperation, consultation and the

provision of information likewise fail.

1159 EM, para. 10.13 (emphasis added). See also Submissions, para.
C(vi), EM, p.413 (“on or near any part of its border with Ecuador”).
512 SUBMISSIONS

For the reasons set out in this Counter-Memorial,

Colombia requests the Cour t to adjudge and declare
that the claims of Ecuador, as set out in the Memorial

of 28 April 2009, are rejected.

Colombia reserves the right to supplement or amend

the present submissions.

JULIO LONDOÑO PAREDES

Agent of Colombia

The Hague, 29 March 2010

513514 APPENDIX

DR STUART D OBSON,“EVALUATION OF CHEMICALS USED IN
COLOMBIA ’S AERIAL SPRAYINPROGRAM ,AND HAZARDS
PRESENTED TO PEOPLEPLANTS,ANIMALS AND THE

ENVIRONMENT IN ECUADOR ” MENZIE ET A.(2009)
[EM,V OL. IIINNEX 158

515516Birchtree Consultants Ltd.

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. (2009) Annex 158

[email protected]

Birchtree House, Jones Drove, Angle Bridge, Whittlesey, Peterborough
Cambs. PE7 2HW, UK

517Contents

Page

1. Terms of reference and qualifications 2

2. Introduction 3

3. The chemicals used in Colombia’s aerial spraying programme 4

4. Potential for spray drift 5

5. Effects on human health 6

6. Effects on organisms in the environment and ecosystems 8

6.1 Amphibians 8

6.2 Plants and soils 21

6.3 Domestic animals 23

6.4 Fish 24

6.5 Terrestrial arthropods (insects and mites) 26

7. Summary and conclusions 29

8. References 32

1

518 1 Terms of reference and qualifications

1. I am asked to independently evaluate the science of ―Evaluation of chemicals used in

Colombia‘s aerial spraying program, and hazards presented to people, plants, animals and the
environment in Ecuador‖ written by Menzie et al. (2009) and presented as Annex 158 to
Ecuador‘s Memorial before the International Court of Justice.

2. Since the Menzie et al. (2009) report is itself a critique of the CICAD Report
(Solomon et al. 2005), critical evaluation of this second document was also included in the
terms of reference.

3. Scientific papers referred-to in the Menzie et al. (2009) report were obtained and

evaluated.

4. Science published since the Menzie et al. (2009) report, including a collection of
scientific papers comprising the follow-up to the CICAD Report (CICAD2), was also

evaluated.

Qualifications

5. My specific expertise is on the risk assessment of chemicals related to effects on
organisms in the environment and human health. I have had 30+ years authoring, co-

authoring and evaluating chemical risk assessments for a wide range of substances in UK
national and international programmes whilst working for the UK Natural Environment
Research Council. I retired from the Council in 2009 and am currently working as a private

consultant through Birchtree Consultants Ltd., of which I am a Director.

6. I have been a member of UK government Advisory Committees since 1986, most
recently as a member of the Advisory Committee on Hazardous Substances for the
Department of Food, Environment and Rural Affairs (DEFRA) for 7 years until my

retirement last year. I am Chair of the Risk Assessment Steering Group of the International
Programme on Chemical Safety (IPCS) which is a joint activity of the World Health
Organisation (WHO), the United Nations Environment Programme (UNEP) and the

International Labour Organisation (ILO). I also currently chair a scientific advisory group for
UNEP. I am currently a permanent member of the Final Review Board for the IPCS Concise
International Chemical Assessment Document project. I have been a regular member,

rapporteur and occasional chair of Expert Task Groups convened by WHO on chemical risk
assessment for 30 years.

7. Specifically on the substances of relevance to this report:

I chaired the Expert Task Group which finalised the WHO assessment of glyphosate

(WHO, 1994);

I co-authored a review of glyphosate and its formulations published in the scientific
literature (Giesy et al. 2000).

2

518 2 Introduction

8. The determination of effects of chemicals on human health or the environment
requires three steps: hazard assessment, exposure assessment and risk assessment.

9. Hazard assessment involves the identification of toxic endpoints associated with the

chemical, hazard identification, and the establishment of dose-response. Dose response
determines the amount of the chemical associated with any particular toxic endpoint; ideally,
it also establishes ‗no-observed-effect‘, the highest dose at which the particular toxicity is no

longer manifest. Regulation of chemical use requires the testing of the substance at doses
much higher than are likely to be encountered in normal use to establish the type of toxicity
the chemical might exhibit. No-observed-effect establishes whether the effect will actually

occur in use.

10. Exposure assessment looks at the use of the chemical and establishes the amount of
the substance likely to be received by humans or organisms in the environment related to the

specific use pattern of the chemical. Different routes of exposure are considered, ingestion,
inhalation or dermal exposure for humans, concentrations in the environmental media
appropriate for different organisms in the environment.

11. Risk assessment then compares the two. If exposure leads to a dose sufficient to cause

a specific toxic effect, risk is present. Again, ideally, risk is quantified and would state that
exposure is x times greater than or y times lower than the no-observed-effect. Sometimes this

is not possible and qualitative estimates of risk are made: high, medium or low.

12. Hazard assessment, therefore, covers possibilities whilst risk assessment determines
probabilities of adverse effects.

13. The Menzie et al. (2009) report describes itself as a hazard assessment. However, it

seldom mentions dose response. Much of the report emphasises the severity of the hazard
(the possible) without considering whether or not the effects are probable. It is, therefore,
largely hazard identification rather than hazard assessment. No attempt is made by the

Menzie et al. report to assess risk.

14. Consideration of hazard in isolation from exposure and risk will give a misleading
picture of the real effects of the chemical on either human health or organisms in the
environment.

15. Despite its major deficiencies in methodology, the Menzie et al report does raise
issues which need to be addressed:

Has the potential for drift of applied spray been underestimated in the assessments

done for the eradication programme for coca and opium poppies in Colombia?

Is the specific spray used more damaging than formulations used in agriculture?

3

520 Is exposure from spray drift sufficient to cause reported health effects in people across

the border in Ecuador?

Is there an adverse effect on agricultural crops in Ecuador?

Is there an adverse effect on domestic animals including farmed fish?

Do natural vegetation, animals including insects and amphibians, and soils comprising

a sensitive and highly diverse ecosystem in Ecuador suffer damage from spray drift?

These issues will be addressed in this report.

3. The chemicals used in Colombia’s aerial spraying

programme

16. I am somewhat confused by this section of the Menzie et al. (2009) paper which
suggests that the glyphosate spray used in Colombia is a mystery.

17. The spraying in Colombia currently uses a mixture of a commercial formulation of
glyphosate (Glyphos as Gly41) which represents 44% of the spray mixture, Cosmo-Flux
411F which is added as an adjuvant to the tank mix at 1% and the remaining 55% is water

(US Department of State, 2002, Hewitt et al., 2009). The formulation ‗Roundup‘ was
previously used. It is my understanding that no re-formulation
place; commercial formulations have always been used. The only addition is the adjuvant.

18. Of the commercial formulations of glyphosate, 15% is surfactant; the exact surfactant
used varies from formulation to formulation. The Glyphos formulation uses predominantly
POEA as its surfactant, in common with Roundup, the formulation most commonly used in

toxicity testing, though the POEA content of Glyphos/Gly41 is lower than in Roundup. Other
surfactants are used in some formulations in minor amounts. One of these, Silwet L-77 is an
organosilicone surfactant which might be solely responsible for one aspect of the

formulation‘s toxicity, the effects on insects and mites (see Section 6.5); there are also reports
of another organosilicone product ‗Pulse‘ as being present in some glyphosate formulations. I
am informed that Gly41 does not contain organosilicones of this type. Other than the

organosilicones, all of the surfactants present in the formulations have comparable toxicity:
they are mild to moderate skin irritants and more significant eye irritants.

19. All other eco-toxicological effects of any of the commercial formulations are likely to

be represented reasonably by the Roundup formulation. The POEA surfactant present in this
formulation was used in the risk assessment performed by Giesy et al. (2000) because ―the
least POEA LC 50… is at the upper end of the toxicity range for surfactants‖; that is, it is a

worst case for surfactant toxicity. Conclusions based on testing of Roundup will, therefore,
overestimate effects of Glyphos/Gly41.

20. In addition, there is a preservative added. Bradberry et al. (2004) regards the
preservative Proxel (benzisothiazolin-3-one) as responsible for the occasional reports of

photo-contact dermatitis. Menzie et al. (2009) mention formaldehyde as a possible

4

520component of formulations. I can find no basis for this but would assume its addition would

be as a preservative. Although formaldehyde has high hazard and the potential for toxicity, its
presence in low volume as a preservative would not be manifest as risk.

21. Speculation on the presence of: other herbicides, fungal plants pathogens or other
additives, by Menzie et al. (2009) seems unfounded based on published information on the

spray mixtures actually used.

4. Potential for spray drift

22. Hewitt et al. (2009) studied the potential for spray drift of the glyphosate formulation
used in coca control. Droplet size spectra were determined in a wind tunnel. The results were
used as input parameters for the accepted model for determining spray drift along with

conditions appropriate to the local environment in Colombia (temperature, relative humidity,
aircraft type, height and speed etc.). A plot of likely deposition rates (expressed as
glyphosate) was generated (Figure 1).

Figure 1 : Taken from Hewitt et al. (2009). Deposition rates for spray drift (g/ha on a log scale) for different
aircraft types at representative flight speeds.

23. As can be seen from the plot, the dose received on the ground reduces very quickly

with distance. It should be emphasised that the vertical axis of the graph is plotted on a log
scale with each major division a factor of ten lower than its predecessor. At 300 metres from

the edge of the spray swath, received dose at ground level is about 5000 times less than the
peak delivery at the intended area for spraying. Almost all deposition is downwind from the
spray swath with the fall to one five-thousandth of the dose within 20 metres upwind.

24. This result is not incompatible with the observations reported in the studies cited by

Menzie et al (2009) (Currier et al, 1982; Murray & Vaughan, 1970, Robinson & Fox, 1978)
and many more which could be cited. These studies measured spray drift by detecting

5

522 droplets dyed with fluorescent dye on collecting sheets placed at varying distances from the
application. Droplets were detected at distances of 4 to 10 miles. The long tail in the plot in

Figure 1 predicts that a very small proportion of the spray, the finest droplets, will drift for
long distances. However, in assessing effects of this drift, the very much reduced deposition
rate (dose to those on the ground) has to be taken into account.

25. Most of the environmental factors listed by Menzie et al. (2009) are included as
parameters in the model used by Hewitt et al. (2009). Relative humidity was further modelled
by Hewitt et al. who demonstrated that increasing relative humidity from 70% to 90%, more

typical of one area in the spray programme, decreased spray drift. Very local factors
mentioned, such as wind circulation and thermal inversions, could alter the result but, I
believe, would not alter the overall conclusions on risk drawn later. I am not an expert on
spray drift but I consider the authors of the Hewitt et al. (2009) paper to be experts; the

predictions of this model would have to be substantially wrong to alter conclusions drawn in
the risk assessment and I consider this highly unlikely.

5. Effects on human health

26. Irritation of the eyes and skin are indeed the most common and consistent effects
reported by people claiming exposure to glyphosate sprays, not only in Ecuador and

Colombia but also globally. The exposure is to a combination of glyphosate and surfactants
found in commercial formulations plus the extra adjuvant in the Colombian spray
programme.

27. Maibach (1986), cited by Menzie et al. (2009), conducted tests on human volunteers

who had undiluted glyphosate formulation (Roundup) applied to normal and abraded skin. A
general purpose household cleaner, domestic dishwashing detergent and baby shampoo were
tested in parallel. Mild skin irritation was seen with the herbicide formulation described by

Maibach as ―less irritant than a standard liquid dishwashing detergent and general all purpose
cleaner‖; for non-abraded skin, the herbicide was also less irritating than baby shampoo.
Further tests showed no sensitisation, photoirritation, allergic or photoallergic contact

dermatitis. The statement by Menzie et al. (2009) that ―adverse effects from dermal contact
that have been reported for glyphosate formulations including skin irritation and occasional
reports of contact dermatitis‖, referenced as coming from the Maibach paper, is unjustified by

that study.

28. Whilst eye irritation from glyphosate formulations is more significant than skin
irritation, the ―periorbital oedema‖ and other signs and symptoms reported in a sprayer by
Temple & Smith (1992) and cited by Menzie et al. resulted from wiping Roundup

concentrate into the eye. Temple & Smith report that the effects ―settled down over 48 hours
and required no specific treatment‖. Exposure through spray drift would not lead to eye
irritation of this severity.

6

52229. The Goldstein et al. (2002) paper cited by Menzie et al. (2009) examined 815 calls

from the public reporting adverse effects of glyphosate spray in California. The study was
prompted by the unusually high numbers of calls relating to this herbicide compared to other
pesticides. The majority of calls recorded topical irritation of the eye (399 cases), skin (250

cases) upper airway (7 cases) or a combination of these sites (32 cases). Of 187 ‗systemic‘
cases (signs or symptoms elsewhere in the body than those parts directly sprayed), 22 were
regarded as probably or definitively related to glyphosate exposure. Each of these cases is

reviewed in the study. All but one, the firefighter referred-to in Menzie et al., were
applicators exposed either to the concentrated formulation or to the spray solution by accident
or failure of spray equipment. Symptoms were mild including: headache, nausea, sore throat,

burning sensation on the skin. The study concluded that reporting of adverse effects in large
numbers does not reflect either the probability or severity of the outcome. In all cases, even
with inhalation exposure of the spray mix, the respiratory effects were mild.

30. More severe effects are seen following ingestion of glyphosate. Bradberry et al.

(2004) (cited elsewhere in the Menzie et al. paper but not in this section) reviewed cases of
glyphosate formulation exposure from Poison Control Centres and hospitals. Talbot et al.

(1991) had similarly reviewed cases dealt with in hospitals in Taiwan; the study was followed
up by Lee et al (2000). It is the Taiwanese studies which were the basis for Bradberry‘s
conclusions on effective dose. Talbot et al. state that patients ingesting mean volumes of 17 ±

16 ml of concentrate were asymptomatic, those ingesting 58 ± 52 ml showed mild symptoms,
128 ± 114 ml moderate and only those ingesting 185 ± 70 ml showed severe symptoms. The
large errors (the ± values) in these estimates reflect the low numbers of cases investigated and

should not be misinterpreted; the headline value represents the most likely mean with the
errors representing the 95% likelihood limits for that mean. Bradberry et al. describe ―greater
than 85 ml ingestion‖ as ―likely to lead to significant toxicity in adults‖ which is a fair

reflection of the lower error limit on the Talbot et al. value of 185 ml. To receive this dose
orally, a person would have to ingest allf the formulation sprayed over 230 square metres of
ground directly under the spray swath. This is clearly impossible, even for a child where the

dose required would be proportionately smaller. Diarrhoea, even in malnourished
populations, is unlikely to be an outcome from exposure to drifted spray.

31. Menzie et al. (2009) raise POEA, a component of the glyphosate sprays, separately.

The effects described above, and the dosages, are based on Roundup formulation which
contains POEA. The conclusions, therefore, take any POEA exposure into account.

32. The only component of the mixture sprayed for coca control not covered in the
exposures above is Cosmo-Flux 411F. It is a mixture of common surfactants in an isoparaffin

mixture as solvent. It is added to the tank mix at 1%, whilst the glyphosate formulation
represents 44% (Hewitt et al., 2009). An additional 1% of surfactant above the approximately
15% (Giesy et al, 2000) already in the formulation is highly unlikely to affect the conclusions

drawn. Isoparaffin acute toxicity is low with moderate skin irritation and slight eye irritation;
it is unlikely to contribute to adverse effects of glyphosate spraying (Mullin et al., 1990).

7

524 6. Effects on organisms in the environment and ecosystems

33. The thrust of the argument in Menzie et al. (2009) related to environmental effects is:

Species relevant to the area of Colombia and Ecuador are not included in the risk
assessment conducted by Colombia. Local species may be more sensitive to the spray.

The added adjuvant (Cosmo-Flux) increases the toxicity of the spray beyond that of

commercial agricultural formulations tested on organisms as Roundup. The risk
assessment might be flawed as a result.

The drifted spray is sufficient to cause adverse effects on local flora and fauna, both

agriculturally and in the natural environment.

Overall, the risk assessment conducted in Colombia (Solomon et al. 2005) is
insufficiently precautionary.

34. On the first bullet point, this is an oft repeated contention but no evidence has been

presented either by Menzie et al. (2009) or other proponents of the view. The scientific
evidence is to the contrary. A recent systematic study (Maltby et al., 2005) concluded that
there was no statistical difference in species sensitivity distributions between aquatic

organisms originating from Nearctic (cold and temperate North America), Palearctic (Europe,
North West Africa and North Asia) and tropical regions for 16 insecticides. Further studies
on Colombian amphibians conducted since the Menzie et al. report and considered below

confirm that this general conclusion holds specifically for these species. I conclude that there
is no basis for believing that Ecuadorian or Colombian species are more sensitive.

35. The other bullet points will be considered below against each of the species groups
and assemblages raised by Menzie et al. (2009).

6.1 Amphibians

36. How do we establish risk to organisms in the environment from the use of a
chemical?

37. First we establish the sensitivity of a range of species by conducting toxicity tests in

the laboratory. These tests are usually conducted under conditions which maximise exposure
of the organisms and, therefore, likely adverse effects, rather than testing under conditions
which simulate actual exposure in the field. They are worst-case. A sensitivity curve is then

plotted and the most sensitive species identified. If we have enough test results, which is the
case for glyphosate and its formulations, we can then estimate a dose of the chemical (usually
a concentration in water) which does not significantly affect species.

38. We then compare the estimated protective concentration against the concentration we
measure or expect in the environment – the exposure estimate. If the exposure estimate is

8

524greater than the protective concentration then adverse effects are expected. For pesticides, the
exposure estimate will always be greater than the protective concentration for the target
organisms – pesticides are designed to kill target species. It may or may not be greater for

non-target organisms.

39. In the case of aerially applied chemicals, we can then further ask ‗at what distance
from the point of spraying does the exposure estimate fall below the protective
concentration?‘ This gives us a distance in metres from the point of application beyond which

we expect no adverse effects – the effects margin.

40. This process is illustrated graphically in Figure 2.

9

526 Figure 2: Steps in the risk assessment process

10
52641. The Menzie et al. (2009) report criticises the original CICAD report (Solomon et al.

2005). 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 gaps. A second CICAD report has now been published as a series of scientific
papers. I will, therefore, examine the Menzie et al. (2009) criticisms against the newly
published updated risk assessment rather than the original Solomon et al. report.

42. The update includes extra test results on particular species as well as a re-worked risk
calculation.

43. Bernal et al. (2009a) conducted acute (3 day) toxicity tests (for lethality) on tadpoles

of eight species of amphibians found in the region of Colombia where coca control takes
place. These species were exposed to the glyphosate formulation Glyphos with Cosmo-Flux
added, as used in the Colombian coca-control programme. Results were compared to similar

tests conducted on other amphibian tadpoles using the Roundup formulation of glyphosate
(Figure 3); this sensitivity curve for larval amphibians represents step 2 in Figure 1.

Figure 3 : Taken from Bernal et al (2009a). The plot shows sensitivity of Colombian and non-Colombian
amphibian tadpoles to glyphosate formulations. The Colombian species were exposed to the formulation
Glyphos with Cosmo-Flux added as used in coca control. Other species were exposed to the Roundup
formulation. The formulations Glyphos and Roundup are similar.

44. Glyphos and Roundup are similar in terms of glyphosate content and both contain the

POEA surfactant which has been identified as a major contributor to the aquatic toxicity of
glyphosate forumations (Giesy et al., 2000). Colombian species were neither more nor less
sensitive than other species. There is also no indication that addition of Cosmo-Flux, leading

to overall greater concentration of surfactant in the exposure, increases the toxicity of
glyphosate formulations to amphibians. A complementary study exposed terrestrial stages of

frogs, also species local to Colombia (Bernal et al., 2009b). The adult frogs were oversprayed

11
528 with the mixture of Glyphos formulation and Cosmo-Flux surfactant. LC

exposure concentration killing 50% of test frogs) are plotted in Figure 4; this sensitivity curve
represents step 2 in Figure 1 but for adult rather than larval amphibians. Here exposure is

expressed as kilograms of glyphosate per hectare; that is in terms of the application rate.

Figure 4: Taken from Bernal et al 2009b. Relative sensitivity of adult frogs from Colombia to overspraying

with glyphosate formulation (Glyphos) with added Cosmo-Flux surfactant.

45. Menzie et al. (2009) cite several specific studies on amphibian larvae. How do these
particular studies fit with the overall relationship between dose and effect plotted in Figure 3?

46. The Relyea (2005b) study cited is the second of two published studies from this

author (the first is not mentioned). The first study, (Relyea, 2005a) conducted laboratory tests
of the Roundup formulation of glyphosate on the aquatic larvae of 6 species of amphibians

from the Midwestern United States. The LC
glyphosate/litre. These values fit into the mid section of the graph in Figure 2; that is, these
species showed average sensitivity to the glyphosate formulation. The second study (Relyea

2005b), as cited in Menzie et al. (2009), then chose
93%, 94% and 92% mortality to leopard frogs, American toads and gray tree frogs
respectively and applied this to mesocosms (simplified artificial environments which

represent the real environment for experimental purposes). The aim of the study was to see if
the formulation showed similar toxicity in the wild as it did in the laboratory. They found the

96%, 100% and 98% respective mortality cited by Menzie et al (2009), confirming to the
authors that the laboratory tests accurately predicted field mortality. Of more interest in the
Relyea study, though not mentioned by Menzie et al, is a second experiment conducted with

terrestrial amphibians from North America. High mortality, ranging from 68 to 86% was
found in three species directly oversprayed with a Roundup formulation at recommended
field application rate (approximately equal to those used in coca control at 3.7 kg/ha). Only

12
528one dose was used in the experiment, precluding the calculation of an LC , so that the results
50
cannot be plotted in Figure 4. However, mortality rates greater than 50% at exposure doses
approximately equal to the lowest LC 50in Figure 3 suggest that Colombian amphibians are

less rather than more sensitive to the glyphosate formulations than temperate species and that
the Cosmo-Flux formulation is no more toxic to terrestrial amphibians than a standard
agricultural formulation.

47. Smith (2001) claimed that ―relatively low concentrations for a short period of time …
appears to induce high mortality in tadpoles‖ as quoted by Menzie et al. (2009). The doses
given in the original paper, and partially cited by Menzie et al., range from one part per

10,000 (incorrectly quoted by Menzie et al. as 1 part per billion) to one part in ten of the
Kleenaway formulation (0.75% glyphosate); these correspond to doses rising from 7.5
mg/litre (=7,500 µg/litre, the units used in Figure 2) glyphosate to an astonishing 0.75

grammes/litre (=750,000 µg/litre). Since the lowest concentration used (7.5 mg/litre) is close
to the LC 50r the least sensitive amphibian plotted in Figure 2 and approximately 5 times

higher than the predicted concentration in water after application for coca control, it is
certainly not a ―relatively low concentration‖ and it is surprising that any of the exposed
tadpoles survived. The Smith (2001) study does not alter the conclusions to be drawn from

Figure 3. The final sentence in the paragraph on the Smith (2001) study in the Menzie et al.
(2009) paper, although referenced as Smith, actually covers material from the Howe et al.
(2004) study covered in the following paragraph.

48. Howe et al. (2004) studied several amphibian species; their findings are already
plotted in Figure 3 so change no conclusions. Howe et al. looked at a range of species,
different formulations (but not Kleenaway as stated by Menzie et al.) and some individual

components of the formulations. Howe et al.‘s findings confirm that the toxicity of
glyphosate formulations reflect the surfactants rather than the herbicide content. They, and

Menzie et al. (2009), raise an important issue; are tadpoles surviving the test concentration
damaged to the point where their survival longer-term will be affected? Howe et al (2004) did
find sub-lethal effects (tail damage, reduced tail length, rate of development, proportion of

animals reaching metamorphosis, gonadal abnormalities and alteration in a genetic marker for
thyroid hormones (required for metamorphosis in amphibians) in larvae which survived. This
will be considered further below.

49. None of the scientific studies cited by Menzie et al. (2009) throw any doubt on the
range of species sensitivity plotted by Bernal et al. (2009a) and we can regard this as
representative.

50. We now move on to step 3 of the risk assessment (Figure 2) and calculate a
‗protective concentration‘. This, in theory, would be a point at the left hand end of the species
sensitivity plots (Figures 3 & 4) which reflects zero effect on the most sensitive species. From

the straight line graphs of Figures 3 and 4, this appears easy. It is not. Both graphs are plotted
with log scales both horizontally and vertically; the result of this is to straighten the plotted

line. If the vertical scale was not expressed in logs, the curve would be ‗S‘-shaped with the
slope reducing increasingly the further to the left we move. In theory, such curves have no

13
530 defined end; the slope keeps reducing to infinity but never disappears completely. Since the
‗end‘ cannot be defined, it is usual to define a point further along the curve which can be

calculated. This is often the 5% point, or sometimes the 1% point. In terms of protection of
species, these equate to the exposure which protects 95% or 99% of species respectively from
adverse effects. Regulatory systems for chemicals globally accept the 95% point as being

reasonably protective of species in the environment.

51. Using the data plotted in Figure 3, Hewitt et al. (2009) in a follow-up to the Solomon
et al. (2005) risk assessment of the Colombian spraying programme, have calculated the 95%

protection concentration for amphibian larvae at 917 µg/litre in surface waters. This is plotted
as the dark green arrow in Figure 5 below; this represents step 3 of the risk assessment
process in Figure 1. This predicts a 5% or lower chance of there being known or unknown
species not protected at this exposure.

Figure 5: The calculated protective concentration for 95% of species of amphibian larvae
(green arrow) at 917 µg/litre glyphosate (from Hewitt et al. 2009), step 3 in the risk
assessment process from Figure 2.

52. How precautionary is this approach?

53. The protection concentration is based on individuals being killed or adversely
affected. Ecotoxicology usually states its aims as protecting populations of organisms rather
than individual organisms. A stable population in the wild produces, on average, one pair of

offspring which survive to reproduce from each breeding pair. More than this and the
population grows, less than this and the population declines. Most organisms in the wild
produce large numbers of offspring to cope with the large losses from predation and

starvation, the two largest sources of mortality. Killing individuals, even in large numbers,
through human activity such as spraying of pesticides does not necessarily add
mortality in the wild. Both predation and starvation are ‗density dependent‘, that is as

numbers fall both reduce in intensity; the lower the number of individuals in the area, the

14

530lower the likelihood that any individual will either fall prey or starve. If I kill some

individuals by spraying pesticides, the survivors are less likely to fall prey or starve. If I kill
all individuals in a specific area, populations can still recover through immigration from

surrounding areas. Protecting individuals is, therefore, very precautionary for species with
high reproductive rates. Some species have evolved a different strategy; they produce smaller
numbers of offspring but protect and nurture them better to reduce losses. Some amphibian

species in the area of coca spraying use this strategy. Here there will still be some natural
mortality; they also, in a stable population, would expect to lose all but two offspring in a
lifetime of breeding. Protecting individuals is still precautionary but less so than for species

with high reproductive output.

54. Balanced against the inherent precaution of the estimate of the protective
concentration, we must consider:

There are two known species showing greater sensitivity in Figure 3: Xenopus laevis
(from South Africa) and Scinax nasicus (from sub-tropical South America). Whilst
this is not unusual in 95% protection values (which do, obviously, allow for 5% of

species to be more sensitive), it looks less than fully protective. Regulatory systems
sometimes shift the value to the concentration representing the known most sensitive

species as a precaution. Others increase the level of protection to 99% or apply an
arbitrary ‗safety factor‘

The value is calculated from LC 50values (killing 50% of tested animals), rather than

no-effect-concentrations, so individual animals will still be killed at this
concentration. It is not usual to conduct species sensitivity distribution protection
values based on acute toxicity studies (LC s) without correction. Some regulatory
50
systems allow correction of LC 50lues whilst others do not.

Even if we lowered the value to account for the mortality below the 50% level of the

most sensitive species, we still need to cover possible initially sub-lethal effects which
might later adversely affect survival (Howe et al., 2004).

55. Scientifically, the protective concentration derived by Hewitt et al. (2009) should be

adequate to prevent significant ecological effects. Intuitively, this might not appear to be the
case and I will come back to this later.

56. We now move on to step 4 of the risk assessment process (Figure 2) which compares

the protective concentration with estimated exposure concentration.

57. Predicted concentration (from Solomon et al., 2005) is given in Figure 6 (for surface
waters immediately following application of spray, assuming a 15 cm water depth) as a red

arrow; actual application rate is indicated by a similar arrow in Figure 7 for terrestrial
amphibians. The two arrows represent the same application expressed differently for the two
different media (water concentration for aquatic amphibian larvae and application rate in

kg/ha for terrestrial adults).

15
532 Figure 6: Comparison of the protective concentration derived by Hewitt et al. (2009)(Figure

5) with the predicted exposure concentration (from Solomon et al. 2005) for larval
amphibians; step 4 of the risk assessment process in Figure 1.

Figure 7: Comparison of the species sensitivity distribution for adult amphibians (Bernal et
al. 2009, Figure 3) with the predicted exposure concentration.

Since in Figure 6 the predicted exposure concentration is greater than the protective

concentrations (step 5 of the risk assessment process in Figure 2) for larval amphibians and
risk is indicated, we move on to step 6. Adults are less sensitive than larvae with LC
equal to or to the right of the actual application rate (Figure 7); the normal application rate for

coca control at 3.7 kg/ha corresponds approximately to the LC
species tested (Centrolene prosoblepon). Adults will, therefore, be protected to a greater
extent than larvae in the derived effects margin and we do not need further calculations of

risk for adults.

58. How precautionary is the exposure estimation?

16

53259. Exposure has been overestimated throughout the risk assessment; no account has been

taken of mitigating factors such as interception by vegetation or lessened availability to
organisms of the active substances over time. Predicted concentrations in surface waters tend
to overestimate measured concentrations following spraying (see Giesy et al., 2000). Rapid

dissipation of glyphosate itself from water through adsorption to sediment and suspended
particulates is often used to justify a lack of need to assess chronic effects of the substance.
However, the toxicity of formulations of glyphosate is largely due to the surfactants rather

than the glyphosate and it is, therefore, the persistence of the surfactants which needs to be
considered. Whilst surfactants also adsorb to sediments, adsorption is weaker than for
glyphosate. Dissipation is usually rapid for surfactants since they are biodegraded under

aerobic conditions within a few days. Chronic toxicity is also, therefore, unlikely for
surfactants.

60. Risk assessment divides the exposure concentration by the effect concentration. This
multiplies precaution. For example, an over-estimation of exposure by a factor of 10 and an

under-estimation of the protective concentration by a factor of 10, when divided, lead to
precaution by a factor of 100.

61. Overall the approach is highly precautionary.

62. Step 6 of the risk assessment process (Figure 2) converts the protective concentration
into a deposition rate and reads a distance from the spray swath from Figure 1 to establish an
effects margin beyond which no adverse effects are expected.

63. Hewitt et al. (2009) relate their protection value to actual deposition of spray. For

their protective concentration value of 917 µg/litre and based on a water depth of 15 cm, they
estimate that a deposition rate of approximately 1350 g/ha (read from the graph) would lead
to this concentration in water. This deposition rate occurs approximately 5 metres from the

edge of the spray swath. They conclude that all amphibian larvae would be protected outside
this area downwind of the spray.

64. Assumptions about depth of water receiving the spray affect the toxic outcome. The

assumed depth used in the Solomon et al. (2005) and Hewitt et al. (2009) risk assessment has
been criticised by Menzie et al. (2009) as excessive because larval amphibians tend to live in
shallower water than this and are also found in Colombia and Ecuador in very low volume

bromeliad ‗ponds‘. This also needs to be balanced against the general highly precautionary
approach. Scientifically, it is probable that the risk values derived indicate little or no effects
ecologically beyond the effects margin calculated by Solomon et al and Hewitt et al.

65. Their results are summarised in Figure 8 below.

17

534 Figure 8: Summary of the risk assessment calculations and conclusions for larval
amphibians from Solomon et al. (2005) and Hewitt et al. (2009)

18
53466. How can these results be translated into likely effects in the field at the point of the

application of the glyphosate formulation spray and in the adjacent areas?

67. At the point of application of the spray, the area has been cleared of natural vegetation
for the planting of coca. Further, damage has been done to the immediately adjacent natural
forest by the coca production through human activity in general and the application of

pesticides to the coca crop in particular. No further ecological damage is likely in the coca
fields themselves.

68. About half of the tested Colombian species aquatic larvae fall to the left of the red

arrow in Figure 6 with the remainder to the right of it. This suggests significant mortality for
all Colombian species, likely total mortality of the most sensitive species and approaching
50% mortality for the least sensitive species, in the band of natural vegetation immediately

adjacent to the coca fields and still directly over-sprayed at the full dose. The Howe et al.
(2004) study also suggests that some of the immediate survivors would be adversely affected
to the point where their longer-term survival was compromised. Terrestrial adult amphibians

are less sensitive than larvae (Figures 4 and 7). In the worst case, about half of the most
sensitive species would be killed, with mortality reducing for species further to the right of
the red arrow in Figure 7. I would expect no mortality in the least sensitive species.

69. The level of damage falls to insignificance within 5 metres according to the
Solomon/Hewitt risk calculations.

70. The degree of precaution in Solomon et al. (2005) has been criticised as inadequate by
Menzie et al. (2009). This criticism would remain for the Hewitt et al. (2009) update of the

risk assessment since the assumptions remain the same. Menzie et al. (2009) offer no
alternative degree of precaution since they make no attempt at the risk assessment
calculations which are needed to derive it; this is not helpful.

71. How precautionary should we be?

72. There is no definitive answer to this question because, although informed by the
scientific evidence, the answer is not entirely a scientific one.

73. Legislation on chemical regulation globally uses different levels of precaution

depending on the circumstances. Several different levels are possible:

Ideally, no change for any species in the environment should result from the use of
chemicals. This is unworkable if any human activity with chemicals is undertaken and

it has never been incorporated into regulatory law

No adverse change resulting in effects at the level of populations or communities of
organisms. This equates to no structural changes in ecosystems. It is applied, with

some arbitrary added precaution, to the use of industrial chemicals on the ground that
there is no intention of adding the substances to the environment and no benefit
gained by doing so.

19

536 No adverse change resulting in effects at the level of populations or communities of

organisms. This equates to no structural changes in ecosystems. It is applied with no
additional precaution (realistic worst case) for agricultural use of pesticides on the
grounds that the substance is intentionally applied to the environment and there is

some benefit from doing so.

Some adverse change but no change in ecosystem services. This is applied to areas
commercially exploited and seeks to maintain sustainability in the exploitation.

Significant adverse change acceptable. This would be applied to areas such as disease

vector control in extremis (for example displaced human populations from natural
disaster or political upheaval) where substantial benefit to human life is balanced

against environmental damage.

The Solomon/Hewitt approach equates to the third bullet point. The Menzie et al. criticism
would imply either the second or even the first bullet points.

74. Which level of precaution to use is clearly a key decision for the court to make. What

I can do is calculate an effects margin for a higher level of precaution to inform that decision
making.

75. If I take the species sensitivity curve (Figure 3) and apply 99% of species protection
instead of the 95% used above, I generate a protective concentration at 473 µg/litre (Figure

9). To generate the 99% value, I was obliged to read the data points from the Bernal et al.
(2009) curve since the test results are not tabulated by them. To ensure that I read them

correctly, I also re-ran the 95% protection calculation and came up with a value comparable
to theirs at 797 µg/litre (the result will not be absolutely identical to theirs since I used a
slightly different curve-fitting program). I will also divide the derived value by 10 to increase

precaution and account for any sub-lethal effects as postulated by Menzie et al. (2009) and
shown experimentally by Howe et al. (2004). This gives a protective concentration at 47.3
µg/litre. This figure is lower (i.e. more protective)

µg/litre proposed by Menzie et al. (2009) for Colombian amphibians based on their reading
of relative toxicity of formulations.

76. Increasing the protection level in this way leads to an estimate of the effects margin at
18 metres downwind of the spray swath.

77. Menzie et al. also criticised the water depth used by Solomon/Hewitt in their
calculations as being too deep to be representative of water depths favoured by larval
amphibians and pointed to amphibian larvae living in bromeliad ponds. If I reduce water

depth in the calculations to 7.5 cm rather than the 15 cm used by Solomon et al. (2005), the
effects margin is extended to 50 metres from the spray swath. The order of magnitude is still
tens of metres.

78. In other words, increasing protection to levels which regulatory law would consider
over-precautionary and beyond the implied precaution suggested by Menzie et al. (2009) still

20

536does not change the overall conclusion that effects as far away as Ecuador are highly

improbable.

Figure 9: Calculation of the 99% protection value with 50% certainty at 473.3 µg/litre (green arrow) for
amphibian larvae based on data plotted in Figure 2 (using the Burrli-Oz software, ANZECC-ARMCANZ, 2000).

The 95% protection value with 50% certainty is shown as the red arrow for comparison.

79. Does the possibility remain of a highly localised, very sensitive species of amphibian
larvae being eradicated by the spraying? Lynch & Arroyo (2009) investigated this possibility
related to the direct spraying of the coca-growing areas in Colombia using geographical

analysis of the distribution of species compared to the distribution of coca growing and
eradication. They believe that several species of frogs are at risk from the coca growing itself
(habitat destruction and use of a range of pesticides), possibly with a contribution from the

eradication programmes. The chances of similar risk to species in Ecuador exposed only to
drifted spray is much lower and, in my opinion, negligible.

80. The same principles used for amphibians can be applied to other organisms in the

environment.

6.2 Plants and soils

81. The section of Menzie et al. (2009) on direct effects on plants is confusing. They

state, correctly, that the addition of extra surfactant Cosmo-Flux (as an adjuvant in the tank
mix) increases the potency of the glyphosate formulation to coca plants fourfold. They also

state, correctly, that the increased potency is due to surfactant effects on the waxy cuticle of
the coca leaf; glyphosate penetrates to the sensitive inner tissues of the plant via the damaged
cuticle. They also state, correctly, that plants other than coca will, therefore, be more

susceptible to the herbicide spray enhanced with the adjuvant. However, they then imply that

21
538 the increased potency increases the effect of the spray on all plants four-fold, which is not the

case. Plants with cuticular protection comparable to the coca will be killed more effectively
with the enhanced spray; with increased potency of approximately fourfold. Those with
greater protection than coca would show a less than four-fold increase in potency and might

show little or no increased toxicity. Those with no protective coating on the leaves would
show no increased toxicity – it is only possible to die once and the application rate of spray in
terms of the glyphosate itself remains constant for all formulations/adjuvants.

82. The question is, therefore, given a range of plants, some susceptible to the original
glyphosate formulation and some only susceptible to the enhanced formulation, what is the
safe exposure for plants in general?

83. Hewitt et al. (2009) obtained toxicity data on 21 plant species tested against
formulated glyphosate from the USEPA ectotox database. A species sensitivity distribution
was plotted (reproduced in Figure 10).

Figure 10: Taken from Hewitt et al. (2009). The plot shows sensitivity of plant species to glyphosate

formulations.

84. In contrast to the equivalent amphibian data, these tests looked at sub-lethal effects on
the plants measuring growth or development. One value, for Magnoliophyta spp., is an outlier
(top right of the graph) and was not used by the authors to derive the plotted line. With the

exception of this species, all other plants plotted are crop plants, including maize, with little
cuticular thickening or waxing which would be protective against glyphosate formulations.

The outlier is interesting because it suggests that the Magnoliophyta spp. might have
thickened cuticles similar, but not necessarily equivalent, to that of coca. If we apply the four-
fold increase in potency observed with the Cosmo-Flux enhanced spray for coca to the

outlying result, this moves the plotted point to the left and very close to the plotted line. It
still leaves it as the least sensitive species in the distribution.

22
53885. Hewitt et al. (2009) calculated a 95% protection rate (this time expressed as g/ha

application rate for the glyphosate content of the formulation) at 43 g/ha. The criticisms
applied to their calculations for amphibian larvae do not apply here (sub-lethal rather than
lethal endpoints and dilution not a factor) with one possible minor exception. The values are

EC 25 the Effective Concentration causing a reduction to 25% of the control level, in this
case of growth or development. A reduction to 25% is not a reduction to zero. However,
estimating effects on growth or development of plants down to, say, 5% is difficult and the

result will not affect risk calculations significantly.

86. For worst case spray drift (the worst aircraft and nozzle type), plants more than 50
metres away from the spray swath would be unaffected.

87. If there are no sub-lethal effects on plants at this distance, there can be no long-term

effects.

88. The studies cited by Menzie et al. (2009) as supporting long-term effects of
glyphosate refer to direct application of glyphosate to forests to control non-productive

vegetation and are not relevant to likely effects of spray drift.

89. Giesy et al. (2000) reviewed the then available studies on the effects of glyphosate
and the Roundup formulation on soil microorganisms and soil function. Effects covered
included: nitrogen fixation, nitrification, dehydrogenase activity, immobilisation of ammonia,

urea hydrolysis, denitrification, degradation of cellulose, starch and protein, and degradation
of leaf litter. The no-observed-effect concentration in soil for the most sensitive of these tests
corresponded almost exactly with the concentration in soil following Roundup application;

this indicates no risk at realistic soil concentrations.

90. The denitrification reported by Tenuta & Beauchamp (1996), cited by Menzie et al.
(2009) occurred following complete kill of a grass sward with the herbicide. The effect was

linked by the authors to increased nitrate and soil moisture resulting from the death of the
vegetation. The paper reviews other causes of denitrification which include hand hoeing of
weeds. This secondary effect of large amounts of dead plant material in soil is relevant to the

areas directly sprayed for coca control but not to areas outside the 50 metre protective zone.

91. The studies cited by Menzie et al. (2009) on effects on fungi could be relevant to risk.
The application concentrations are within expected deposition rates. The Johal & Rahe

(1988) study placed drops of Roundup directly onto the hypocotyls (early leaf) of beans. The
study was a mechanistic one. The Beyrle et al. (1995) study similarly used realistic
concentrations on orchids. However, these concentrations would only occur close to the spray

swath and fungal effects would not occur beyond 50 metres.

6.3 Domestic animals

92. It is difficult to take this section of the Menzie et al. (2009) paper seriously. It is the

perfect example of the dangers of expressing hazard (the possibility) without reference to
exposure and risk (the probability) of something adverse occurring.

23

540 93. Menzie et al. cite Bradberry et al. (2004) as support for their statement that there are

―even greater concerns about effects on mammals and birds …. related to compounds added
to the tank mix to enhance foliar penetration‖. Bradberry et al. regard the primary toxicity
seen with glyphosate formulations as related to the surfactants present; they discuss only

POEA and certainly say nothing about other potential surfactants or adjuvants to increase
foliar penetration. Bradberry et al. (2004) state that dermal exposure to ready-to-use-
glyphosate formulations (that is the concentrate) ―can cause irritation …. severe skin burns

are very rare‖. Maibach (1986) describes skin irritation following direct application of the
Roundup concentrate to the skin of human volunteers as ―less irritant than a standard liquid
dishwashing detergent and a general all purpose cleaner‖. It is difficult to reconcile these

descriptions of the effects of the concentrated formulation with Menzie et al.‘s picture of
severely irritated domestic animals abrading their own skin, producing ―self-inflicted
wounds‖ and getting secondary infections following exposure to the spray of diluted

formulation some distance from the application site. The Menzie et al. (2009) argument
seems to be that hypothetical additives would have greater irritancy than the surfactants
already present in the commercial formulations. In fact only Cosmo-Flux is added to the

commercial formulations and the components of Cosmo-Flux are no more irritating, and
probably less irritating, than the POEA already there.

94. In contrast to the Menzie et al. statement, POEA has been tested on birds (Giesy et al.,

2000) since it is a component of the Roundup formulation. Bobwhite quail, mallard and zebra
finches fed on a diet containing 5620 or 8064 mg/kg Roundup for 5 days (with a further 3
days of observation) showed no toxicity; the dose is equivalent to 843 and 1210 mg/kg POEA

in the diet. Predicted concentration of glyphosate in food for grazing birds such as mallard
would be 400 mg/kg diet directly under the spray following aerial application of a
formulation (EPPO, 2003); this would be approximately equivalent to 150 mg/kg diet for

POEA. No overt toxicity would, therefore, be expected following spraying for coca control.

95. The Oliviera et al. (2007) paper cited by Menzie et al. also dosed mallard with POEA
as a component of Roundup. They dosed by gavage (directly to the stomach) at doses of 5
and 100 mg/kg body weight (or 0.75 and 15 mg/kg body weight POEA) for 15 days; since no

lower doses were used, a no-effect dose cannot be established for the effects on circulating
testosterone, the epididymus and the testis. These doses are within the range of total daily

intakes likely from contaminated food following spraying with glyphosate. It is unusual to
dose by gavage for this type of experiment and regular dosing for 15 days exceeds likely field
exposure but the effect warrants further study. Whilst these effects could be seen directly

under the spray, they are unlikely to occur from the exposure via spray drift at some distance
away.

96. The introduction of ethylene oxide as a possible contributor to overall toxicity of the
glyphosate formulations here is ludicrous. Ethylene oxide is indeed a highly dangerous

chemical with severe effects on humans at low exposure concentrations. However, it is
present as a contaminant of POEAs (and other detergents) at less than 10 ppm (parts per
million) as stated in the Manufacturer‘s Safety Data Sheet and by WHO (2003). Given that

POEA represents 15% of the formulation, this gives a total of 1.5 mg of ethylene oxide in a 1

24

540litre bottle of the concentrate. Application of the formulation at 3.7 kg/ha as glyphosate
would apply ethylene oxide at no more than 15 mg/ha and deposit it at a maximum of 1.5
µg/m in the area of direct spraying. Assuming all of this ethylene oxide is available in the

breathing zone of humans or domestic animals and is breathed-in within a few breaths,
exposure would be to 1.5 µg/m over a period of minutes. Average concentration of ethylene
3
oxide in uncontaminated outdoor air in Canada is 0.34 µg/m with the highest measured
concentration at 4.9 µg/m ; average concentration in indoor air in houses is 0.17 µg/m with a
3
single maximum value at 4 µg/m . These concentrations are breathed in by Canadian
residents 24 hours a day, 365 days per year. Peak concentrations experienced by workers
showing spontaneous abortion were at 458 mg/m (WHO, 2003) approximately 300,000

times the peak exposure directly under the glyphosate formulation spray and experienced
over a longer period. Ethylene oxide can, with complete confidence, be ruled out as

contributing to the toxicity of the glyphosate sprays.

97. The Campbell & Chapman (2000) book on poisoning in dogs does not give dose.
However, a comparable study on 31 cases of acute poisoning in dogs by Burgat et al. (1998)

make clear that the vomiting and other signs in dogs follows ingestion of the concentrate or
spray tank mixture of the glyphosate formulation. These effects are not relevant to indirect, or

even direct, exposure to the spray in the field. In quoting Bradberry et al. (2004), Menzie et
al. (2009) report reduced weight gain in mammalian species possibly caused by unpalatibility
without completing the quotation; Bradberry et al. say ―since very high dietary concentrations

were used in some of these studies, this effect may have been due to unpalatibility and
reduced calorific intake‖. These very high dietary concentrations (my emphasis) make the

relevance of the findings to field exposure highly unlikely. Similarly, the chronic dietary
exposure to glyphosate formulations referred-to in the product label will not be seen in the
field from spray-drift. Systematic risk assessments of glyphosate formulations have

consistently concluded that there is little or no risk to mammals (WHO, 1994; Williams et al.,
2000).

98. The likely effects in domestic animals following exposure to spray drift are the same
as those seen in humans: transient eye irritation and respiratory discomfort. I do not accept
that these mild to moderate effects would impinge on growth or reproductive performance of

these animals.

6.4 Fish

99. Without going into detail again, the effects on fish of direct spraying over water will
be comparable to, though less severe than, those on amphibians. Fish show lower toxicity to
glyphosate formulations than amphibian larvae. Fish exposed to the spray formulation as

used in Colombia (including the Cosmo-Flux adjuvant) show greater toxicity than to the
formulation alone. However, this toxicity is lower than for the most sensitive amphibian

species included above in the risk assessment. Water depth greater than the 7.5 cm assumed
as worst case amphibians is appropriate for fish since they tend to live in deeper water than
amphibian larvae. The distance of 18 metres associated with no-effect for amphibians (at a

25

542 water depth of 15cm) will be protective for fish. I would expect no effects on fish exposed to
spray drift at distances greater than this.

100. Fish kills could only occur following direct overspraying of aquaculture ponds at the
full application strength. This effect would be either from the direct toxicity of the surfactants
in the spray mix or from indirect effects due to oxygen depletion caused by biodegradation of

dead plant material. The latter, by definition, only occurs where plants have been killed. Dead
plants will only occur within 50 metres of the spray swath which is protective of any adverse
effects on plants (see above). Widespread killing of fish outside the area of direct spraying,

attributed by Menzie et al. (2009) to the Columbian spraying programme, is, therefore, highly
improbable.

6.5 Terrestrial arthropods (insects and mites)

101. Cowles et al (2000), as cited by Menzie et al. (2009) studied the effects of one class of

surfactant on mites, the organosilicones. Their study was conducted in response to previous
studies suggesting direct toxicity of these surfactants to both mites and insects. Its aim was to
determine the relative roles of reduced surface tension caused by the surfactant and other

factors (interaction with other components of the applied formulations and environmental
media such as leaf surface waxes). They found that reduced surface tension was the principle,
if not only, reason for toxicity. Further, they determined the surface tension associated with

lethality. Trisiloxanes, including the Silwet L-77 correctly identified as a component of some
glyphosate formulations by Menzie et al., caused high mortality in mites whilst other
organosilicone surfactants did not. The authors quote Imai et al. (1995) as stating that

trisiloxanes have ―unequalled wetting properties‖ (= reduction in surface tension). The
Goodwin & McBrydie (2000) paper, also cited by Menzie et al., oversprayed and fed
honeybees with 11 different surfactants. Two of these, Pulse and Boost, caused honeybee

deaths at recommended rates of application; both of these surfactants are organosilicones
(Pulse is described on the internet as being a component of Roundup). A tallowamine
surfactant similar to the POEA included in glyphosate formulations did kill bees but only at

doses more than 10 times the recommended application rate. The remaining surfactants were
not toxic to bees. Oral toxicity was seen with the tallowamine surfactant at 0.5% in sugar
syrup; no indication was given as to how realistic this dose was.

102. The Cowles et al. study, therefore, does not support the Menzie et al. statements that

effects on insects (and presumably mites) would be ―fairly unpredictable from laboratory
tests‖ or that ―the addition of special surfactants could also result in increased toxicity to
insects over and above that for Roundup‖. Field effects are predictable from surface tension

reduction induced by particular surfactants and the most potent class of surface-tension
reducers are already included in some commercial formulations of glyphosate but not in
Cosmo-Flux. The apparent quotation or citation ―at sub-lethal surfactant exposures, this can

result in disorientation and altered behaviour, as exposed insects react to the injury to the
exoskeleton and respiratory system‖ ascribed to Goodwin & McBrydie (2000) by Menzie et
al. does not appear in the Goodwin paper and is, presumably, speculation on the part of

Menzie et al. themselves. The Goodwin & McBrydie (2000) results support the view of

26

542Cowles et al. (2000) that the organosilicone surfactants are the most potent insect and mite

killers and also show that the POEA component of glyphosate formulations does not kill
insects.

103. Hislop & Prokopy (1981), cited by Menzie et al. exposed the predatory mite
Amblyseius fallacies to glyphosate using the same technique as Cowles et al. (2000) where

adult mites are stuck onto microscope slides, or placed on leaves, and then dipped into a
solution of the herbicide. Although this is stated by Menzie et al. to be ―glyphosate‖ it is, in
fact, a formulation. The formulation given in Hislop & Prokopy is an emulsifiable

concentrate which seems unlikely for glyphosate. I would assume that, like all glyphosate
formulations of which I am aware, it actually contained surfactants which were responsible
for the effect.

104. How severe would the effects of surfactants on insects and mites be? When would
they occur? What is a safe distance from the spray swath for prevention of the effect?

105. Directly within the spray swath, effects could be severe, even if the particularly potent

surfactants are a small part of the formulation. I would expect significant kills. To kill the
arthropods, the solution of the surfactant needs to wet the surface of the insect or mite in the
same way it would the surface of a leaf. All of the experiments have submerged the arthropod

in the solution. Direct exposure to spray droplets, exposure to droplets remaining after
spraying on leaves or the soil surface would provide such wetting. As droplets reduce in
number and size with distance from the spray swath, exposure becomes less likely and more

arthropods would survive. Deposition has fallen by a factor of 1000 within 50 metres of the
spray swath and falls by a further factor of at least10 within the next 75 metres. Whilst an
exact safe distance cannot be suggested, it seems reasonable to assume that the highly

precautionary safe distance for amphibians at 50 metres would be more than protective for
direct toxicity to arthropods. Since the highly toxic trisiloxanes are probably not present in
Gly41, this conclusion is highly precautionary for the spray applied in Colombia.

106. I was surprised to see Cox (1995) referenced in a serious scientific report. The Journal
of Pesticide Reform is an overtly anti-pesticide website not a scientific journal. Ms Cox is
highly selective in what she takes from the scientific papers she cites. I checked all of the

original papers cited in the Cox (1995) reference.

Brust (1990) tested the relative contributions of direct toxicity to insects (carabid
beetles) and the indirect effects due to loss of plant habitat. They state that there was
no toxic or repellent effect of any herbicide, including glyphosate, when tested in the

laboratory. The herbicides did not have acute or chronic effects on longevity or food
consumption of beetles over a year in greenhouse trials. However, the beetle

population fell following field application; the beetles ―responded to destruction of
plant material‖. Populations returned to normal after 28 days.

Asteraki et al. (1992) ascribed all of the decline in populations of carabid beetles
following Roundup treatment of hedgerows to reduction in the plant community on

which they depended.

27

544 Hassan et al. (1988) conducted laboratory tests on Roundup toxicity to 18 species of

insect pests, predators, spiders or mites. For 13 species, Roundup was classified as
―harmless‖, for 4 species as ―slightly harmful‖ and for 1 species as ―moderately
harmful‖.

Santillo et al. (1989) studied small mammal populations following the use of
glyphosate formulation to clear ground of vegetation; they found significantly reduced
numbers of invertebrates on cleared ground.

Mohamed et al. (1992) found reduced survival of the woodlouse Hemilepistus

reaumuri kept on filter paper dipped in Roundup at 25 mg/litre or 50 mg/litre in the
laboratory. At the lower exposure dose, 90% of males and 60% of females survived;

at the higher dose 50% of both males and females survived. Whilst it is difficult to
equate soil concentrations to those on filter paper (glyphosate is strongly adsorbed
onto soil particles reducing its availability to organisms), the predicted soil

concentration following glyphosate use is 15 mg/kg (Giesy et al., 2000). These seem
high doses.

107. Menzie et al. (2009) then cite further studies but give incomplete coverage of their
conclusions:

Jackson & Pitre (2004) state that ―the Roundup Ready soybean system, including
applications of glyphosate, had no detrimental effects on pest and beneficial insects in
wide-row soybean plantings‖. They also state that ―no significant differences in G.

punctipes numbers between glyphosate-treated and nontreated soybean demonstrated
that glyphosate had no direct effect on G. punctipes densities in 1997 or 1998.
Glyphosate indirectly influenced G. punctipes densities in 3 of 11 weeks. Densities

were correlated with weed cover in the crop, not with glyphosate application.

Guiseppe et al. (2006) is actually a review of the literature on effects of herbicide use
directly and indirectly on ecosystems. They conclude ―indirect effects of glyphosate

herbicide by means of reduced herbaceous plant, shrub and deciduous tree cover, on
insects, birds, small mammals and large mammals can result in short term reduction
in abundance but usually not a decline in species richness or species diversity‖

The study on pine tip moth referenced by Menzie et al. as Guiseppe et al. is by Nowak
et al. (2003), reviewed by Guiseppe et al. The Nowak et al. study is the only one of
more than 30 studies reviewed by Guiseppe et al. which suggests direct toxicity of

glyphosate on arthropods rather than indirect ones.

108. The overwhelming view of researchers in this area is that effects on arthropod
populations are an indirect effect of the use of glyphosate following on from its herbicidal

action. The arthropods using plants as food, refuge, or as sources of prey are likely to decline
following herbicide use; those feeding on dead plant material are likely to increase in
numbers. The literature shows that these generalisations are not universally true; there is

sometimes little or no population decline. These indirect effects can only, by definition, occur

28

544where significant proportions of the plant community have been killed. They are, therefore,

highly unlikely to occur beyond 50 metres of the spray swath as for effects on plants.

7. Summary and conclusions

109. Consideration of hazard (the potential to cause harm) in isolation from exposure and
risk (the probability that the hazard will be realised in practice) will give a misleading picture
of the real effects of chemicals on either human health or the environment. This is certainly

true of the Menzie et al. study.

The spray

110. The constituents of the spray used for coca control in Colombia are clear. A
commercial glyphosate formulation, Glyphos (Gly41), is mixed with an adjuvant, Cosmo-

Flux 411F and water prior to application. The exact nature of the surfactants which form part
of the formulation changed over time as Roundup was replaced with Gly41. However, all of
the surfactants used throughout the spraying programme have similar toxicology and

ecotoxicology. The most common, and most toxic to aquatic organisms, surfactant used is
POEA which is present in the Roundup formulation, the most tested. Roundup test results are
precautionary for other formulations since Gly41 contains less POEA than Roundup.

111. One surfactant present in Roundup but probably not present in Gly41 has significant
ecotoxicity above the general; this has relevance for possible effects on insects and mites.

112. Preservatives may also play a significant role in specific effects on humans.

113. The Cosmo-Flux adjuvant has similar properties to the general surfactants in the

formulation and its addition simply increases the dose of surfactants to humans and
organisms in the environment. Surfactants play the principal role in the environmental
toxicity of formulations of glyphosate and the Colombian spray mix; they are probably key in

symptoms in humans.

Spray drift

114. Droplet distribution has been measured under conditions relevant to the Colombian
environment and spray drift modelled. This indicates that the vast majority of deposition

occurs within 100 to 200 metres downwind of the spray swath. Only minor deposition will
occur further away.

Effects of human health

115. Published analyses globally of reports of exposure to glyphosate formulations and

their spray indicate that short-term eye irritation, minor skin irritation, mild respiratory effects
and sore throat are the most common symptoms of exposure to spray. More serious effects

are restricted to direct exposure to the concentrated formulation or the spray mix being
applied and have only been seen in spray applicators themselves (with the exception of a

29

546 single individual who was oversprayed). It is, therefore, highly unlikely that serious effects

could be related to drifted spray.

Effects on organisms in the environment and ecosystems

116. The scientific evidence shows that tropical species are neither more nor less sensitive
to pesticides than species from temperate or cold areas. Recent testing has confirmed that this

is specifically the case for Colombian amphibians.

117. Toxicity directly under the spray swath will be high for a number of groups of
organisms: amphibian larvae, to a lesser degree amphibian adults and fish, plants, insects and

other arthropods. Populations of the most sensitive organisms could be eliminated directly
under the spray. However, the ecology of coca fields will have already been largely destroyed
by the coca growing; adverse effects are, therefore, restricted to the margins of coca fields

still directly over-sprayed.

118. Risk has been assessed for each group of organisms and ‗safe‘ distances downwind of
the spray swath suggested for each.

119. For amphibian larvae, the most sensitive of the aquatic species tested, the Colombian

risk assessment suggests no adverse effects would be seen more than 5 metres from the spray
swath. This conclusion is justified by the science and the level of precaution is consistent
with those used in chemical regulatory legislation globally.

120. To inform decision making, I have calculated the distance using a more precautionary

approach as suggested by the Menzie et al. (2009) report. Even at a very high level of
precaution, no adverse effects would be seen at a distance downwind of the spray swath of 50

metres.

121. Adult amphibians are less sensitive than the larvae and would also be protected at this
distance.

122. For plants, neither acute not chronic sub-lethal effects would be seen at more than 50

metres from the spray swath. There are no reports of effects on soil function or fungi at soil
concentrations expected directly under the spray. There is, therefore, no risk for soils.

123. Domestic animals would be affected in comparable ways to humans with mild to
moderate transient effects on eye, skin or respiratory system. No significant harm is expected.

124. Fish would be killed by direct overspray. However, they show lower sensitivity to the
glyphosate formulation than amphibians and 18 metres from the spray swath would be
protective for fish using the highly precautionary approach. Indirect effects caused by oxygen

depletion resulting from decaying plant matter would, by definition, only occur if plants were
killed; the 50 metre protection value for plants would also protect fish from indirect effects.

125. The organosilicone surfactant constituent of some commercial formulations is toxic to

insects and mites, and probably to other terrestrial arthropods. This toxicity would occur
directly under the spray. Whilst an exact safe distance cannot be established, I would expect

30

546arthropods to be fully protected beyond 50 metres. Since the formulation currently used in
Colombia (Gly41) does not contain organosilicones of the type causing these effects, this risk

estimate is again highly precautionary.

126. Indirect effects on arthropod populations will certainly occur at doses of the herbicide
spray which kill plants. Beyond 50 metres, where no plants are adversely affected, no effects
are expected.

Environmental effects and effects on crop plants

127. No environmental effects or effects on crop plants would be seen more than 50 metres
downwind of the spray swath even taking a highly precautionary approach. There should,
therefore, be no adverse ecological effects in Ecuador.

1 February 2010

31

548 8. References

ANZECC/ARMCANZ (2000) Australian and New Zealand guidelines for fresh and marine
water quality. National Water Quality Management Strategy, Australian and New Zealand

Environment Conservation Council, Agriculture and Resource Management Council of
Australia and New Zealand, Canberra, Australia
(http:/www.deh.gov.au/water/publications/index.html)

Asteraki EJ, Hanks CB, Clements RO (1992) The impact of the chemical removal of the
hedge-base flora on the community structure of carabid beetles (Col., Carabidae) and spiders
(Araneae) of the field and hedge bottom. Journal of Applied Entomology 113: 398-406

Behrle HF, Smith SE, Peterson RL, Franco CMM (1995) Colonization of Orchis morio
protocorms by a mycorrhizal fungus: effects of nitrogen nutrition and glyphosate in
modifying the responses. Canadian Journal of Botany 73: 1128-1140

Bernal MH, Solomon KR, Carrasquilla G (2009a) Toxicity of formulated glyphosate

(Glyphos) and Cosmo-Flux to larval Colombian frogs 1. Laboratory acute toxicity. Journal of
Toxicology and Environmental Health Part A 72: 961-965

Bernal MH, Solomon KR, Carrasquilla G (2009b) Toxicity of formulated glyphosate
(Glyphos) and Cosmo-Flux to larval Colombian frogs 1. Field and laboratory microcosm
acute toxicity. Journal of Toxicology and Environmental Health Part A 72: 966-973

Bradberry SM, Proudfoot AT, Vale JA (2004) Glyphosate poisoning. Toxicological Reviews
23: 159-167

Brust GE (1990) Direct and indirect effects of four herbicides on the activity of carabid
beetles (coleoptera: Carabidae). Pesticide Science 30: 309-320

Burgat V, Keck G, Guerre P, Bigorre V, Pineau X (1998) Glyphosate toxicosis in domestic
animals: A survey from the data of the Centre National d'Informations Toxicologiques
Veterinaires (CNITV). Veterinary and Human Toxicology 40: 363-368

Campbell A, Chapman M (2000) Handbook of poisoning in dogs and cats. Blackwell
Science, Oxford, UK. 272pp

Cowles RS, Cowles EA, McDermott AM, Ramoutar D (2000) ―Inert‖ formulation ingredients
with activity: Toxicity of trisiloxane surfactant solutions to twospotted spider mites (Acari:
Tetranychidae). Journal of Economic Entomology 93: 180-188

Cox C (1995) Glyphosate fact sheet. Part 1: Toxicology and Part 2: Human exposure and
ecological effects. Journal of Pesticide Reform 15

Currier WW, MacCollom GB, Baumann GL (1982) Drift residues of air-supplied carbaryl in
an orchard environment. Journal of Economic Entomology 75: 1062-1068

32

548EPPO (2003) Environmental risk assessment scheme for plant protection products. EPPO

Bulletin 33. European and Mediterranean Plant Protection Organization, Paris, France

Giesy JP, Dobson S, Solomon KR (2000) Ecotoxicological risk assessment for Roundup
herbicide. Reviews of Environmental Contamination and Toxicology 167: 35-120

Goldstein, Acquavella, R.M. M, Farmer(2002) An analysis of glyphosate data from the

California Environmental Protection Agency Pesticide Illness Surveillance Program. Clinical
Toxicology 40: 885-892

Goodwin RM, McBrydie HM (2000) Effect of surfactants on honey bee survival. New
Zealand Plant Protection 53: 230-234

Guiseppe KFL, Drummond FA, Stubbs C, Woods S (2006) The use of glyphosate herbicides
in managed forest ecosystems and their effects on non-target organisms with particular
reference to ants as bioindicators. Technical Bulletin 192. Maine Agricultural and Forest
Experiment Station, The University of Maine. 51pp

Hassan SA, Bigler F, Bogenschütz H, Boller E, Brun J, Chiverton P, Edwards P, Mansour F,

Naton E, Oomen PA, Overmeer WPJ, Polgar L, Rieckmann W, Samsøe-Petersen L, Stäubli
A, Sterk G, Tavares K, Tuset JJ, Viggiani G, Vivas AG (1988) Results of the fourth joint
pesticide testing programme carried out by the IOBC/WPRS-Working Group "Pesticides and
Beneficial Organisms". Journal of Applied Entomology 105: 321-329

Hewitt AJ, Solomon KR, Marshall EJP (2009) Spray droplet size, drift potential, and risks to

nontarget organisms from aerially applied glyphosate for coca control in Colombia. Journal
of Toxicology and Environmental Health Part A 72: 921-929

Hislop RG, Prokopy RJ (1981) Integrated management of phytophagous mites in
Massachusetts (USA) apple orchards, 2: influence of pesticides on the predator Amblyseius
fallacis (Acarina: Phytoseiidae) under laboratory and field conditions. Protection Ecology 3:

157-172

Howe CM, Berrill M, Pauli BD, Helbing CC, Werry K, Veldhoen N (2004) Toxicity of
glyphosate-based pesticides to four North American frog species. Environmental Toxicology
and Chemistry 23: 1928-1938

Imai T, Tsuchiya S, Fujimori T (1995) Aphicidal effects of Silwet L-77, organosilicone
nonionic surfactant. Applied Entomology and Zoology 30: 380-382

Jackson RE, Pitre HN (2004) Influence of Roundup ready soybean production systems and
glyphosate application on pest and beneficial insects in wide-row soybean. Journal of
Agricultural and Urban Entomology 21: 61-70

Johal GS, Rahe JE (1988) Glyphosate, hypersensitivity and phytoalexin accumulation in the
incompatible bean anthracnose host-parasite interaction. Physiological and Molecular
Pathology 32: 267-281

33

550 Lee HL, Chen KW, Chi CH, Huang JJ, Tsai LM (2000) Clinical presentations and prognostic
factors of a glyphosate — Surfactant herbicide intoxication a review of 131 cases. Academic
Emergency Medicine 7: 906-910

Lynch JD, Arroyo SB (2009) Risks to Colombian amphibian fauna from cultivation of coca
(Erythroxylum coca): A geographical analysis. Journal of Toxicology and Environmental
Health Part A 72: 974-985

Maibach HI (1986) Irritation, sensitization, photoirritation and photosensitization assays with
a glyphosate herbicide. Contact Dermatitis 15: 152-156

Maltby L, Blake N, Brock TCM, Van den Brink PJ (2005) Insecticide species sensitivity
distributions: Importance of test species selection and relevance to aquatic ecosystems.
Environmental Toxicology and Chemistry 24: 379-388

Menzie CA, Booth PN, Kane Driscoll SB (2009) Evaluation of chemicals used in Colombia's

aerial spraying program, and risks presented to people, plants, animals, and their environment
in Ecuador. Report prepared for Government of Ecuador. Exponent, Inc., Alexandria,
Virginia, USA

Mohamed AI, Nair GA, Abbas HL, Kassam HH (1992) Effects of pesticides on the survival,
growth and oxygen consumption of Hemilepistus reaumuri (Audouin & Savigny 1826)
(Isopoda Oniscidea). Tropical Zoology 5: 145-153

Mullin LS, Ader AW, Daughtrey WC, Frost DZ, Greenwood MR (1990) Toxicology update
isoparaffinic hydrocarbons: A summary of physical properties, toxicity studies and human
exposure data. Journal of Applied Toxicology 10: 135-142

Murray JA, Vaughan LM (1970) Measuring pesticide drift at distances to four miles. Journal

of Applied Meteorology 9: 79-85

Nowak JT, Harrington TB, Berisford CW (2003) Nantucket pine tip moth development and
population dynamics: Influence of nitrogen fertilization and vegetation control. Forest
Science 49: 731-737

Oliveira AG, Telles LF, Hess RA, Mahecha GA, Oliveira CA (2007) Effects of the herbicide

Roundup on the epididymal region of drakes Anas platyrhynchos. Reproductive Toxicology
23: 182-191

Relyea R (2005a) The lethal impacts of Roundup and predatory stress on six species of North
American tadpoles. Archives of Environmental Contamination and Toxicology 48: 351-357

Relyea R (2005b) The lethal impact of Roundup on aquatic and terrestrial amphibians.
Ecological Applications 15: 1118-1124

Robinson E, Fox LL (1978) 2,4-D herbicides in central Washington. Journal of the Air
Pollution Control Association 28: 1015-1020

Santillo DJ, Leslie DM, Brown PW (1989) Responses of small mammals and habitat to

glyphosate application on clearcuts. Journal of Wildlife Management 53: 164-172

34

550Smith GR (2001) Effects of acute exposure to a commercial formulation of glyphosate on the
tadpoles of two species of anurans. Bulletin of Environmental Contamination and Toxicology

67: 483-488

Solomon KR, Anadon A, Cerdeira AL, Marshall J, Sanin LH (2005) Environmental and
human health assessment of the aerial spray program for coca and poppy control in
Colombia. Report prepared for the Inter-American Drug Abuse Control Commission
(CICAD) section of the Organization of American States (OAS), Washington, DC, USA

Talbot AR, Shiaw MH, Huang JS, Yang SF, Goo TS, Wang SH, Chen CL, Sanford TR
(1991) Acute poisoning with a glyphosate-surfactant herbicide ('Roundup'): A review of 93
cases. Human and Experimental Toxicology 10: 1-8

Temple WA, Smith NA (1992) Glyphosate herbicide poisoning experience in New Zealand.
New Zealand Medical Journal 105: 173-174

Tenuta M, Beauchamp EG (1996) Denitrification following herbicide application to a grass
sward. Canadian Journal of Soil Science 76: 15-22

US Department of State (2002) Chemicals used for the aerial eradication of illicit coca in
Colobia and conditions of application. US Department of State, Washington DC, USA
(http://www.state.gov/p/inl/rls/rpt/aeicc/13234.htm)

WHO (1994) Glyphosate. Environmental Health Criteria 159. World Health Organization,
Geneva

WHO (2003) Ethylene oxide. Concise International Chemical Assessment Document 54.
World Health Organization, Geneva

Williams GM, Kroes R, Munro IC (2000) Safety evaluation and risk assessment of the
herbicide Roundup and its active ingredient, glyphosate, for humans. Regulatory Toxicology
and Pharmacology 31: 117-165

35

552 LIST OF ANNEXES

VOLUME II

DIPLOMATIC DOCUMENTS AND CORRESPONDENCE

Annex 1 Note Nº E-067 from the Colombian Ambassador
in Quito to the Presidential Adviser for
Coexistence, National Security, and Fight against
Crime, 18 January 2001.

Annex 2 Note Nº E-934 from the Colombian Ambassador
in Quito to the Ecuadorian Defence Minister,
16 August 2001.

Annex 3 Note Nº E-931 from the Colombian Ambassador
in Quito to Valerio Greffa Uquiña, Ecuadorian
Congressman, 20 August 2001.

Annex 4 Note Nº E-962 from the Colombian Ambassador
in Quito to the President of the National Congress
of Ecuador, 23 August 2001.

Annex 5 Diplomatic Note Nº 72523/2001-SG/SSN from
the Ecuadorian Foreign Minister to the
Colombian Foreign Minister, 12 September 2001.

Annex 6 Note Nº E-1313 from the Colombian Ambassador
in Quito to the Ecuadorian Minister of
Agriculture, 15 November 2001.

Annex 7 Note Nº E. 455/90 from the Colombian
Ambassador in Quito to the Colombian Foreign
Minister, 26 March 2003, enclosing the non-
paper given to the Ambassador by Ecuadorian
authorities at the meeting held on the same date.

553Annex 8 Note Verbale VRE Nº 32759 from the Colombian
Foreign Ministry to th e Ecuadorian Embassy in
Bogotá, 18 September 2003.

Annex 9 Note Verbale DBR/CAL unnumbered from the
Colombian Foreign Ministry to the Ecuadorian
Embassy in Bogotá, 13 November 2003.

Annex 10 Diplomatic Note Nº 4-2-336/03 from the
Ecuadorian Foreign Ministry to the Colombian
Foreign Ministry, 10 December 2003.

Annex 11 Note Nº CMFS-46-2004 from the President of the
Ecuadorian Scientific and Technical Commission
to the Acting Director of the Anti-Narcotics
Direction of the Colombian National Police,
18 March 2004.

Annex 12 Diplomatic Note DPM/CDR Nº 20125 from the
Colombian Acting Foreign Minister to the
Ecuadorian Foreign Minister, 7 April 2004.

Annex 13 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, 14 April 2004.

Annex 14 Diplomatic Note VRE/DAA/CAL Nº 18067 from
the Colombian Vice-Minister of Foreign Affairs
to the Ecuadorian Vice-Minister of Foreign
Affairs, 14 April 2004.

Annex 15 Diplomatic Note VR/ST/CVICE Nº 27776 from
the Colombian Vice-Minister of Foreign Affairs
to the Ecuadorian Ambassador in Bogotá, 9 June
2004.

Annex 16 Diplomatic Note VRE/DAA/CAL Nº 28511 from
the Colombian Vice-Minister of Foreign Affairs
to the Ecuadorian Vice-Minister of Foreign
Affairs, 17 June 2004.

Annex 17 Joint Declaration of the Presidents of Colombia
and Ecuador, Esmeraldas, Ecuador, 15 October
2004.

554Annex 18 Note Nº DSF 40.1/3.1.3-4-00421 from the
Executive Secretary of the Inter-American
Commission on Human Rights to the Colombian
Foreign Minister, 12 September 2005.

Annex 19 Diplomatic Note Nº DDH. 58003 from the
Colombian Foreign Ministry to the Executive
Secretary of the Inter-American Commission on
Human Rights, 18 September 2005.

Annex 20 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 November 2005.
Annex 21 United Nations, Terms of Reference of the
“Scoping Mission for a Detailed Assessment
Study on the Poten tial Impact to Human Health,

the Environment and Agriculture that Occurs in
Ecuador’s Territory as a Result of Glyphosate
Fumigations Done by the Government of
Colombia in the Border Area with Ecuador”,
January 2006. (Enclosure to Note Nº E-104/032
from the Colombian Ambassador in Quito to the
Colombian Foreign Minister, 27 January 2006).

Annex 22 Minutes of Meeting of th e Bi-national Technical
Commission on Health, Atacames, Esmeraldas,
2-3 March 2006.

Annex 23 Health and Environment Integrating Borders,
Proposal for Technical Cooperation between
Countries Colombia-Ecuador: Epidemiological
Surveillance, Pesticides, and Water Quality,

2007.
Annex 24 Diplomatic Note DM/VRE/DPM/CDR/DDA/CAL
Nº 44664 from the Colombia n Foreign Minister
to the Ecuadorian Foreign Minister, 5 September

2006.
Annex 25 Diplomatic Note Nº39064/06-VM/SSNDF/DGRFC
from the Ecuadorian Foreign Minister to the

CICAD Executive Secretary, 20 September 2006.
555Annex 26 Diplomatic Note from the Colombian Foreign
Ministry to the Ecuadorian Embassy in Bogotá,
20 December 2006.

Annex 27 Note from the President of Colombia to the
President of Ecuador, 21 December 2006.

Annex 28 Diplomatic Note DM/VRE Nº 6454 from the
Colombian Foreign Minister to the Ecuadorian
Foreign Minister, 9 February 2007.

Annex 29 Minutes of the Bilateral Evaluation Meeting,
“Health and the Environment Integrating
Borders: Strengthening of Surveillance in Public
Health, Pesticides, and Water Quality”, Ipiales,
Nariño, 12 December 2007.

COLOMBIAN OFFICIAL DOCUMENTS

Annex 30 Colombian Decree 1843 of 1991, Articles 87,
102.

Annex 31 Communiqué of the National Narcotics Council
of Colombia to public opinion on the eradication
of illicit poppy crops, 31 January 1992.

Annex 32 Colombian Law 99 of 1993, Article 57.

Annex 33 Cosmo-Flux 411F, Sales Registration Nº 2186 of
19 April 1993.
Annex 34 Provisional Toxicological Opinion LP-0593-93
(Cosmo-Flux 411F), Colombian Health Ministry,

30 July 1993.
Annex 35 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 October 1993.

556Annex 36 Note from the Colombian Health Minister to the
Director of the National Narcotics Directorate, 11
October 1993.

Annex 37 Resolution Nº 001 of 11 February 1994 of the
National Narcotics Council of Colombia.

Annex 38 Colombian Decree 1753 of 1994, Articles 1, 38.
Annex 39 Roundup SL: Sales Registration Nº 0756 of
11 August 1997.

Annex 40 “Environmental Impact Caused by Chemical
Substances, Illicit Crops and Related Activities”,
National Narcotics Directorate of Colombia,

Strategic and Research Division, 2000.
Annex 41 Resolution Nº 005 of 11 August 2000 of the
National Narcotics Council of Colombia.

Annex 42 Communiqué of the Ministry for the
Environment of Colombia to the Public, October
2000.

Annex 43 Resolution Nº 017 of 4 October 2001 of the
National Narcotics Council of Colombia.

Annex 44 Toxicological Opinion Nº 0685, regarding the
toxicological classification of the mix Glyphosate
+ POEA + Cosmo-Flux(1%), Colombian Health
Ministry, 8 October 2001.

Annex 45 Resolution Nº 1065 of 26 November 2001 of the
Ministry for the Environment of Colombia,
Reasoning, Section 3, Technical Visit.

Annex 46 GLY-41: Sales Registration Nº 4294 of 2 July
2002.
Annex 47 Note 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 January 2003.

557Annex 48 Resolution Nº 099 of 31 January 2003 of the
Ministry for the Environment of Colombia
(excerpts).

Annex 49 Resolution Nº 013 of 27 June 2003 of the
National Narcotics Council of Colombia.

Annex 50 Resolution Nº 1054 of 30 September 2003 of the
Ministry for the Environment of Colombia.
Annex 51 Note Rec’d. Nº 29157 from the Director of the

National Narcotics Directorate (Head of the
Colombian Delegation to the Joint Commission)
to the Colombian Foreign Minister, 6 April 2004.

Annex 52 Report from the Colombian Delegation, Joint
Scientific and Technical Commission on Aerial
Sprayings, Orito, Putumayo, 25-28 May 2004.
Annex 53 Note Nº SARE-321 from the Director of the

National Narcotics Directorate to the Colombian
Foreign Minister, 11 August 2004.
Annex 54 Council of State of Colombia, Judgment on
appeal from the Admi nistrative Tribunal of

Cundinamarca, 19 October 2004.
Annex 55 Note Nº 001727 from the General Director of the
National Police of Colombia to the Colombian

Foreign Minister, 2 November 2004.
Annex 56 Aide-Mémoire “Aerial Spraying Issue with
Ecuador”, Ministry of Foreign Affairs of
Colombia, Division of Multilateral Political

Affairs, Sub-division for Drug Affairs, September
2005.
Annex 57 “Plan Colombia Progress Report 1999-2005”,

National Planning Department of Colombia,
Office of Justice and Security, September 2006.
Annex 58 Communiqué of the Minist ers of Foreign Affairs
and National Defence of Colombia, 12 December

2006.

558Annex 59 Addendum Nº 1 to Tender Nº 02 of 2007, with
the purpose of “Contracting the Audit to the
Program for the Eradication of Illicit Crops by
Aerial Spraying with Glyphosate Herbicide”,

National Narcotics Directorate of Colombia.
Annex 60 Press Release of the Colombian Foreign Ministry,
8 February 2007.

Annex 61 Resolution Nº 008 of 2 March 2007 of the
National Narcotics Council of Colombia.

Annex 62 Communiqué of the Colombian Foreign Ministry,
12 July 2007.

Annex 63 “Expenditures of the Colombian State in the
Fight against Drugs, 2005-2006”, National
Narcotics Directorate and National Planning
Department, Bogotá, 2008.

Annex 64 “Investigation Regarding Possible Secondary
Effects on Human Health, Allegedly Derived
from Glyphosate Spraying in the Rural Area of
Villanueva of the Orito Municipality, Putumayo
on 6 August 2007”, Final Report National Health
Institute of Colombia, June 2008.

Annex 65 Report by the Colombian Agriculture and
Livestock Institute (ICA), 29 January 2010.

Annex 66 Report by the National Narcotics Directorate
(DNE), 4 February 2010.

• Appendix: External Environmental Audit.
Annex 67 Report by the Anti-Narcotics Direction of the
Colombian National Police (DIRAN), 8 February
2010.

Annex 68 Report by the National Health Institute (INS),
“Summary of Activitie s Carried Out by the
Ministry of Social Protection and the National

Health Institute with Regard to the Program for
the Eradication of Illi cit Crops with Glyphosate
Herbicide – PECIG”, February 2010.

559Annex 69 Report on Complaints submitted to the
Colombian National Health Institute with regard
to the Program for the Eradication of Illicit Crops
with Glyphosate Herbicide – PECIG, February

2010.
Annex 70 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 th e spraying operations
conducted between September 2008 and
February 2009)

• Appendix 2: Sample of results of soil
analyses taken prior to / immediately
following the spraying (N° 1972/08).

• Appendix 3: Sample of results of soil
analyses taken 60 days after spraying
(N° 2125/08).

• Appendix 4: Sample of results of water
analyses taken prior to / immediately
following the spraying (N° 1951/08).

• Appendix 5: Sample of results of superficial
water analyses taken 60 days after spraying
(N° 2194/09).

Annex 71 Spray Mix Quality Control, National Laboratory
for Agricultural Use Products (LANIA),
Colombian Agriculture a nd Livestock Institute
(ICA), 18 February 2009.

Annex 72 Note N° 0958 ARECI/JEFAT from the Anti-
Narcotics Direction of the Colombian National
Police (DIRAN) to the Colombian Foreign
Ministry, 18 February 2010.

560Annex 73 Note N° 20103291383181 from the Presidential
Agency for Social Action and International
Cooperation ( Acción Social ) to the Colombian
Foreign Ministry, 23 February 2010.

ECUADORIAN OFFICIAL DOCUMENTS

Annex 74 Main Fertilizers Import and Manufacturing

Companies, Ministry of Agriculture, Livestock,
Aquaculture and Fishing of Ecuador.
Annex 75 Siona Indigenous Nation, Web page of the

Council for the Development of the Nationalities
and Peoples of Ecuador.
Annex 76 Ai’cofan Indigenous Nati on, National Institute
for Statistics and Censuses of Ecuador.

Annex 77 Sucumbíos Provincial Government Homepage,
History, Political Organization.

Annex 78 “Social Development and Poverty in Ecuador,
1990-2001”, Chapter 8, Poverty Profile according
to Unmet Basic Needs, Social Front’s Technical
Secretariat Information and Analysis Unit –
SIISE, Government of Ecuador, Social Report

2003, Quito, December 2003.
Annex 79 “Model Project for the Planting, Production and
Sustainable Agro-industry of the Guanábana
(Soursop)”, Federation fo r Integral Development

of Peasants and Afro-descendants of Esmeraldas
(FEDICAE), Esmeraldas, January 2004.
Annex 80 Press Bulletin Nº 388 of the Ecuadorian Foreign

Ministry, “No glyphosate re sidues exis t in the
waters of the rivers of the Sucumbíos province”,
25 June 2004.
Annex 81 Press Bulletin Nº 480 of the Ecuadorian Foreign

Ministry, “No glyphosate re sidues were found in

561 Esmeraldas, border with Colombia”, 26 August
2004.

Annex 82 Press Bulletin Nº 721 of the Ecuadorian Foreign
Ministry, 24 December 2004.

Annex 83 Memorandum of the Ecuadorian Foreign
Ministry, 24 December 2004.
Annex 84 Memorandum of the Ecuadorian Foreign
Ministry, 29 December 2004.

Annex 85 Press Bulletin N° 732 of the Ecuadorian Foreign
Ministry, “Commission of the Foreign Ministry
finds no evidence of sprayings following

verifications in the entire area where sprayings
were claimed to have been conducted”,
30 December 2004.
Annex 86 Press Bulletin Nº 027 of the Ecuadorian Foreign

Ministry, 24 January 2005.
Annex 87 Press Bulletin Nº 419 of the Ecuadorian Foreign
Ministry, 2 September 2005.

Annex 88 Press Bulletin Nº 593 of the Ecuadorian Foreign
Ministry, 18 November 2005.

Annex 89 “Manual for the Identification and Management
of Weeds on the Galapagos Islands”, Charles
Darwin Foundation and Galapagos National Park
Services, 2006.

Annex 90 “Health baseline of th e Ecuadorian Northern
Border Provinces”, Public Health Ministry of
Ecuador, Pan-American Health Organization
(PAHO), Quito, Ecuador, 2006.

Annex 91 “Esmeraldas Health and Environment Program
(SYMAE)”, Municipality of Esmeraldas,
1 August 2006.

Annex 92 “Ten Main Morbidity Causes per Province”,
Public Health Ministry of Ecuador, Epidemiology
Sub-process, Quito, 2007.

562Annex 93 Cofan Indigenous Nation, Webpage of the
Council for the Development of the Nationalities
and Peoples of Ecuador, 6 November 2008.

Annex 94 “The Living Conditions of Ecuadorians. Results
of the Survey of Living Conditions – Fifth
Round.Poverty and Inequality”, National
Institute for Statistics and Censuses of Ecuador,
Quito, 2009.

UNITED NATIONS DOCUMENTS

Annex 95 Description of the Illicit Crops Integrated
Monitoring System (SIMCI).
Annex 96 International Programme on Chemical Safety

(IPCS), International Programme on Chemical
Safety, Environmental Health Criteria (EHC)
159, Glyphosate”, Geneva, 1994.
Annex 97 United Nations International Drug Control

Programme (UNDCP), “Economic and Social
Consequences of Drug Abuse and Illicit
Trafficking”, Technical Series 0101, 1998.

Annex 98 World Health Organization (WHO), “Guidelines
for Drinking-Water Quality”, Second Edition,
Addendum to Volume 1, Recommendations,
Geneva, 1998.

Annex 99 General Assembly Resolution S-20/2, “Political
Declaration”, 10 June 1998. UN doc. A/S-20/4
Annex 100 United Nations Children's Fund (UNICEF),

“Baseline of the Province of Esmeraldas, Final
Consulting Report”, April 1998.
Annex 101 Food and Agriculture Organization of the United
Nations (FAO), “Specifications and Evaluations

for Plant Protection Products, Glyphosate N-
(phosphonomethyl)glycine”, 2000/2001.

563Annex 102 United Nations Office for Drug Control and
Crime Prevention (ODCCP), “Global Illicit Drug
Trends 2001”, New York, 2001.

Annex 103 United Nations Office on Drugs and Crime
(UNODC), “Colombia Coca Cultivation Survey
2003”, June 2004.

Annex 104 UNODC, “Colombia Coca Cultivation Survey
2004”, June 2005.

Annex 105 World Health Organization (WHO), “The WHO
Recommended Classification of Pesticides by
Hazard”, 2005.

Annex 106 “Report of the United Nations Technical
Preliminary Mission to propose studies on the
impact of the aerial sprayings and complementary
actions in the northern border of Ecuador”, Quito,
Ecuador, April 2006.

Annex 107 UNODC, “Colombia Coca Cultivation Survey
2006”, June 2007.
Annex 108 UNODC, “Colombia Coca Cultivation Survey

2008”, June 2009.
Annex 109 UNODC, “World Drug Report 2009”, Vienna,
2009.

Annex 110 Diplomatic Note N° 57491 from the Colombian
Vice-Minister for Foreign Multilateral Affairs to
the United Nations Office on Drugs and Crime
(UNODC) Representative to Colombia,

19 October 2009.
Annex 111 Diplomatic Note from the United Nations Office
on Drugs and Crime (UNODC) Representative to

Colombia to the Colombian Vice-Minister for
Foreign Multilateral Affairs, 20 January 2010.

564 VOLUME III

O.A.S. DOCUMENTS

Annex 112 Annual Report of the Inter-American Drug Abuse
Control Commission (CICAD) to the General
Assembly of the Organization of American States
nd
at its 32 Regular Session, 12-15 November
2001, Caracas, Venezuela, 3 April 2002.
OEA/SER.L/XIV.2.30 CICAD/doc.1123/01 rev.1

Annex 113 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
the environment, 4 February 2004.

Annex 114 Annual Report of the Inter-American Drug Abuse
Control Commission (CICAD) to the General
Assembly of the Organization of American States
at its 34thRegular Session, 17-20 November
2003, Montreal, Canada, 3 May 2004.
OEA/Ser.L/XIV.2.34 CICAD/doc.1264/03 rev. 2

Annex 115 35thRegular Session of CICAD, Washington,
D.C., 27-30 April 2004, Session Documents.

Annex 116 CICAD, K.R. Solomon et al. “Environmental and
Human Health Assessment of the Aerial Spray
Program for Coca and Poppy Control in

Colombia”, OAS, Wash ington, D.C., 31 March
2005, (CICAD I).
Annex 117 37thRegular Session of CICAD, Santo Domingo,

Dominican Republic, 26-29 April 2005,
Document 1421.

565Annex 118 CICAD, R.A. Brain et al., “The Toxicology of
Substances Used in th e Production and Refining
of Cocaine and Heroin: A Tier-Two Hazard

Assessment”, OAS, Washington, D.C., 31 July
2005, (CICAD I).
Annex 119 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, 23 May 2006.
Annex 120 Annual Report of the Inter-American Drug Abuse
Control Commission (CICAD) to the General

Assembly of the Organization of American States
at its 36th Regular Session, Santo Domingo,
Dominican Republic, 4-6 June 2006.
Annex 121 Annual Report of the Pe rmanent Council to the

General Assembly of the Organization of
American States 2006-2007. OAS,
AG/doc.4698/07.

Annex 122 Annual Report of the Inter-American Drug Abuse
Control Commission (CICAD) to the General
Assembly of the Organization of American States
at its 39thRegular Session, San Pedro Sula,
Honduras, 2-3 June 2009. OEA/Ser.G,
CP/doc.4395/09 corr. 1, 26 May 2009.

SCIENTIFIC AND TECHNICAL REPORTS

Annex 123 SGS (Societé Génerale de Surveillance, S.A)
Colombia S.A., “Report of Contamination
Control for glyphosate application at the Sierra of
Santa Marta”, 1987.

566Annex 124 J.P. Giesy, S. Dobson S & K.R. Solomon,
“Ecotoxicological risk assessment for Roundup
herbicide”, Reviews of Environmental
Contamination and Toxicology 167: 35-120,

2000.
Annex 125 G. M. Williams et al., “Safety evaluation and risk
assessment of the herbicide Roundup® and its
active ingredient, glyphosate, for humans” in

Regulatory Toxicology aPndarmacology
31:117-165, 2000.
Annex 126 Embassy of the United States of America -

Clínica de Toxicología ‘Uribe Cualla’, Alleged
effects of Glyphosate on Human Health , Bogotá,
December 2001.
Annex 127 R.E. Ramos C., J.P. Ramos B., Environmental

Assessment of the Impact of coca crops and the
processing of coca leaf, Universidad de los
Andes, Bogotá, Colombia, 2002.
Annex 128 O. Saavedra, Laboratorio Inmunopharmos Ltda.,

Toxicity Study on Laboratory Animals for two
concentrations of Glyphosate 44% + Cosmoflux
1% + Water 55%, Bogotá, 15 February 2002.

Annex 129 M. San Sebastián & A-K. Hurtig, “Oil
exploitation in the Amazon Basin of Ecuador: a
public health emergency”, Pan Am J Public
Health, 15(3):205-211, 2004.

Annex 130 K.R. Solomon et al., A Response to Tomás León
Sicard et al., CICAD, 2005.
Annex 131 Journal of Toxicology and Environmental Health,

Part A: Current Issues, Volume 72, Numbers 15-
16, August 1 and August 15 2009, Special Issue:
Production of Illicit Drugs , the Environment and
Human Health, Taylor & Francis, London, 2009.
As follows:

• Annex 131-A: K.R. Solomon et al., “Human
Health and Environmental Risks from the Use

567 of Glyphosate Formulations to Control the
Production of Coca in Colombia: Overview
and Conclusions”, in Journal of Toxicology
and Environmental Health , Part A, 72:914-

920, 2009.
• Annex 131-B: A.J. Hewitt et al., “Spray
Droplet Size, Drift Potential, and Risks to

Nontarget Organisms from Aerially Applied
Glyphosate for Coca Control in Colombia”, in
Journal of Toxicology and Environmental
Health, Part A, 72:921-929, 2009.

• Annex 131-C: E.J.P. Marshall et al, “Coca
(Erythroxylum coca ) Control is Affected by
Glyphosate Formulations and Adjuvants”, in
Journal of Toxicology and Environmental
Health, Part A, 72:930-936, 2009.

• Annex 131-D: 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 Colombia”, in Journal of

Toxicology and Environmental Health, Part A,
72:937-948, 2009.
• Annex 131-E: L.H. Sanin et al., “Regional

Differences in Time to Pregnancy Among
fertile Woman from Five Colombian Regions
with Different use of Glyphosate”, in Journal
of Toxicology and Environmental Health , Part
A 72:949-960, 2009.

• Annex 131-F: M.H. Bernal et al., “Toxicity of
Formulated Glyphosate (Glyphos) and Cosmo-
Flux to Larval Colombian Frogs 1. Laboratory
Acute Toxicity”, in Journal of Toxicology and
Environmental Health , Part A 72:961-965,

2009.
• Annex 131-G: M.H. Bernal et al., “Toxicity
of Formulated Glyphosate (Glyphos) and

568 Cosmo-Flux to Larval and Juvenile
Colombian Frogs 2. Fi eld and Laboratory
Microcosm Acute Toxicity”, in Journal of
Toxicology and Environmental Health , Part A

72:966-973, 2009.
• Annex 131-H: J. D. Lynch et al., “Risks to
Colombian amphibian fauna from cultivation

of coca (Erythroxylum coca ): A geographical
analysis” in Journal of Toxicology and
Environmental Health , Part A, 72:974-985,
2009.

• Annex 131-I: C. Bolognesi et al.,
“Biomonitoring of Genotoxic Risk in
Agricultural Workers from Five Colombian
Regions: Association to Occupational
Exposure to Glyphosate”, in Journal of
Toxicology and Environmental Health , Part A

72:986-997, 2009.
Annex 132 Centre for Toxicology and Environmental Health,
L.L.C., University of Arkansas for Medical
Sciences, GLYPHOSATE Frequently Asked

Questions, 2009.

OTHER DOCUMENTS

Annex 133 Safety Data Sheet for Roundup SL.
Annex 134 Label and Safety Data Sheet for GLY-41.

Annex 135 Andean Regulation for the Registration and
Control of Chemical Pesticides for Agricultural
Use, Andean Community, Decision 436 of 2000,
Arts. 57-59; Annex 1 - Glossary, “Mean Lethal
Dose, LD 50”.

Annex 136 Report of the Fifth Meeting of Vice-Ministers of
Foreign Affairs of the Andean Community,
Caracas, 16-17 October 2000.

569Annex 137 Andean Cooperation Plan for the Control of
Illegal Drugs and Related Offenses, Andean
Community, Decision 505 of 2001.

Annex 138 COSMO-FLUX® 411-F Tec hnical Data Sheet,
Cosmoagro, June 2002.

Annex 139 Press item: “Ecuador accepts sprayings to
continue”, El Comercio (Ecuadorian newspaper),
Quito, 4 August 2004.

Annex 140 US Department of State, Bureau for International
Narcotics and Law Enforcement Affairs,
Memorandum of Justific ation Concerning the
Secretary of State’s 2002 Certification of
Conditions Related to the Aerial Eradication of
Illicit Coca in Colombia, Washington D.C., 2002.

Annex 141 United States Environmental Protection Agency
(EPA), Addendum to memorandum. Subject:
Description of Glyphosat e Use in the U.S. as a
Basis for Comparison to Glyphosate Use in
Colombia for Coca Eradication , From: Virginia

Werling and Timothy Kiely (EPA, BEAD
Biological and Economic Analysis Division)
to Jay Ellenberger (EPA, BEAD Biological and
Economic Analysis Division), 21 August 2002.

Annex 142 United States Environmental Protection Agency
(EPA), Details of the Consultation for
Department of State: Use of Pesticides for Coca
and Poppy Eradication Program in Colombia ,
August 2002.

Annex 143 US Department of State, Bureau for International
Narcotics and Law Enforcement Affairs,
Memorandum of Justific ation Concerning the
Secretary of State’s 2003 Certification of
Conditions Related to the Aerial Eradication of

Illicit Coca in Colombia, Washington D.C., 2003.
Annex 144 US Department of State, Bureau for International
Narcotics and Law Enforcement Affairs,

Memorandum of Justific ation Concerning the
570 Secretary of State’s 2004 Certification of
Conditions Related to the Aerial Eradication of
Illicit Coca in Colombia, Washington D.C., 2004.

Annex 145 Andean Regulation for the Control of chemical
substances used in the illegal manufacture of
narcotic drugs and ps ychotropic substances,
Andean Community, Decision 602 of 2004.

Annex 146 US Department of State, Bureau for International
Narcotics and Law Enforcement Affairs,
Memorandum of Justific ation Concerning the
Secretary of State’s 2005 Certification of

Conditions Related to the Aerial Eradication of
Illicit Coca in Colombia, Washington D.C., 2005.
Annex 147 US Department of State, Bureau for International
Narcotics and Law Enforcement Affairs,

Memorandum of Justific ation Concerning the
Secretary of State’s 2006 Certification of
Conditions Related to the Aerial Eradication of
Illicit Coca in Colombia, Washington D.C., 2006.

Annex 148 Press item: “Defence Minister welcomes the new
Super Tucano airplanes of the Air Force”,
14 December 2006.

Annex 149 US Department of State, Bureau for International
Narcotics and Law Enforcement Affairs,
Memorandum of Justific ation Concerning the
Secretary of State’s 2007 Certification of
Conditions Related to the Aerial Eradication of
Illicit Coca in Colombia, Washington D.C., 2007.

Annex 150 Press item: “Ecuador will sue Colombia over
sprayings in the border”, Revista Caminos, 2 July
2007.

Annex 151 Colombian Association of Autonomous Regional
Corporations (ASOCARS), “Shared
Responsibility: The World Drug Problem from a
Green Perspective”, Periódico Virtual, Issue
Nº 14, 2008.

571Annex 152 Embassy of the United States in Bogotá, Fact
Sheet 2008, “Cocaine production and cultivation:
Colombia”, 6 November 2009, enclosure to Press
Item: “Official U.S. Colombia Survey Shows

Sharp Drop in Coca Cultivation and Cocaine
Production”.
Annex 153 Arias v. Dyncorp , Plaintiffs’ Motion to dismiss
Three Individual Plaintiffs, 23 December 2009.

Annex 154 Arias v. Dyncorp , Defendants’ Response, 6
January 2010.

Annex 155 Arias v. Dyncorp , --- F. Supp. 2d ---, 2010 WL
94563 (DDC, 2010, Roberts J).

Annex 156 Arias v. Dyncorp , Defendants’ Motion for
Sanctions Against the Aria s/Quinteros Plaintiffs
for Violations of Discovery Orders, 26 January
2010.

572 LIST OF DOCUMENTS DEPOSITED WITH THE
REGISTRY

- H. I. Maibach, “Irritation, se nsitization, photoirritation and
photosensitization assays with a glyphosate herbicide”,
Contact Dermatitis, vol. 15, 1986, pp. 152-156.

- D. A. Goldstein, J. F. Acquavella, R.M. Manion & D.R.
Farmer, “An Analysis of Glyphosate Data from the
California Environmental Protection Agency Pesticide Illness
Surveillance Program”, Journal of Toxico logy Clinical

Toxicology, vol. 40, No. 7, 2002, pp. 885-892.

573

Document Long Title

Counter-Memorial of Colombia

Links