Reply of Romania

Document Number
14701
Document Type
Date of the Document
Document File
Document

INTERNATIONAL COURT OF JUSTICE

CASE CONCERNING MARITIME

DELIMITATION IN THE BLACK SEA

(ROMANIA v. UKRAINE)

REPLY

SUBMITTED BY

ROMANIA

22 DECEMBER 2006

I TABLE OF CONTENTS

CHAPTER 1 INTRODUCTION 1
(A) Purpose of the Reply 1
(B) The Court’s jurisdiction over the present dispute
1
(C) Structure of the Reply 7

CHAPTER 2 THE APPLICABLE LAW 10
(A) Introduction 10

(B) The 1949, 1963 and 1974 ProcŁs-Verbaux and the 1954 Act as delimitation 11
agreements
(C) The 1997 Additional Agreement underwrites the equitable principles-relevant 12
circumstances approach
(D) Irrelevance of Romania’s “denunciation” of the 1948 Protocol and of the 196113
Border Regime Treaty
(E) Conclusion
14
15
CHAPTER 3 THE GEOGRAPHICAL FRAMEWORK:

THE RELEVANT COASTS AND THE
RELEVANT AREA
(A) Introduction 15

(B) The general geographic setting and its relevance 16
(1) Physical Characteristics of the Parties’ Coasts
16
(a) The Romanian coast 16

(b) The Ukrainian coast 17
(2) The Appropriateness of a Two-Sector Approach 21

(a) The 1997 Agreement foresaw a two sector approach 21
(b) The general geographical situation dictates a two sector approach 22

(c) A two-sector approach involves no double-counting 26
(C) Identifying the relevant coasts and the relevant area
27
(1) The Relevant Coasts 29

(a) The relevant Ukrainian coasts 29
(i) The irrelevance of Serpents’ Island 31

(ii) The Ukrainian coast north of Point S and Cape Tarkhankut bears no 32
relationship to the Romanian coast and is not the source of overlapping
entitlements
(iii) Ukraine’s construction of the supposed coastal projections is 32
methodologically flawed
(iv) Ukraine’s arguments from the jurisprudence of the Court
37
(b) The relevant Romanian coast 47

(i) Relevance of the southern sector to the delimitation 48
(ii) Permissibility of the Sulina Dyke and the Sacalin Peninsula as bas50oints

II (c) Conclusions on relevant coasts 54

(2) The Relevant Area 54
(a) The differences between the Parties as to the relevant area
54
(i) The area north of the line joining Point S and Cape Tarkhankut 55

(ii) The sliver of maritime area off the Romanian and Bulgarian coast 55
(iii) The triangle of maritime area south of Cape Sarych 58

(b) Romania’s position 59
(D) Conclusion
59
CHAPTER 4 THE FIRST STAGE OF THE DELIMITATION 61

AND ITS LEGAL BASIS
(A) The positions of the Parties 61
(B) The maritime boundary around Serpents’ Island as established by agreement 63

(1) Ukrainian Claims as to the Extent and Significance of the Boundary agreed in 1949 63

(a) The character of the boundary agreed in 1949 65

(b) The status and reliability of the sketches included in the individual ProcŁs- 75
Verbaux
(c) The argument that Map 134 depicts the full extent of the boundary 77

(i) The language of the 1949 General ProcŁs-Verbal and its travaux 78
prØparatoires
(ii) Ukraine’s “blank space” argument 80

(iii) Ukraine’s “coincidence of points” argument 83
(iv) Other maps of the boundary annexed to the 1949 General ProcŁs-Verbal
87
(d) The argument that Romania’s 12 nm claim in 1951 changed the function of 92
Map 134
(e) The argument that the term “Soviet marine boundary zone” was adopted 93
because of Romanian “inhibitions” as to areas beyond 6 nm

(f) The argument that the 1949 line was not an all purpose boundary but was 94
limited to areas under the sovereignty of the two States
(g) Ukraine’s reliance on Romanian legislation subsequent to 1949 96
(h) The assertion that Romania’s claim was not previously raised
97
(2) Ukrainian Claims as to Maps and Charts relied on by Romania 98
(a) The assertion that the charts and maps have no evidentiary value 100

(b) The assertion that the charts are irrelevant because they are not part of any103
official text
(c) The assertion that none of the 23 charts and maps forms part of a written 106
agreement establishing a maritime boundary
(d) The assertion that none of the 23 maps depicts the different legal elements 107
which comprise the maritime boundary

(e) The assertion that none of the 23 charts and maps indicates either EEZ or 108
continental shelf belonging to Romania
(f) The assertion that the 23 charts and maps do not make consistent use of 111
maritime chart symbols
(g) The assertion that none of the 23 charts and maps fixes the final point on the
semicircle around Serpents’ Island 116

(h) The assertion that some of the 23 maps are self-evidently not intended to ser117
as maritime boundary maps
(i) The assertion that many of the 23 maps and charts are copies of or are based 118

III upon earlier maps and have no independent significance
(3) Romania’s Map Annexed to its 1997 Notification
119
(C)Conclusions 125

CHAPTER 5 SERPENTS’ ISLAND IN LAW AND FACT 127

(A) Introduction 127

(B) Ukraine’s interpretation of Article 121(3) 129

(1) The meaning of “rock” in the context of Article 121 130

(2) The requirement of “human habitation” 131

(3) The requirement of an “economic life of its own” 132

(4) Romania’s declaration relating to Article 121 133

(C) Serpents’ Island meets all the criteria of Article 121(3) 135

(1) Serpents’ Island is a very small rocky island 135

(2) Serpents’ Island is devoid of water resources and practically devoid of soil, 143
vegetation and fauna
(3) Serpents’ Island cannot sustain human habitation
152
(4) Serpents’ Island cannot sustain any economic life of its own
156

(5) Ukraine’s artificial island 160

(D) Conclusions 165

Addendum to Chapter 5 166
Elements from the Annexes of the Counter-Memorial and from the Ukrainian mass-
media showing the character of Serpents’ Island
CHAPTER 6 THE RELEVANT CIRCUMSTANCES 188

(A) The role of relevant circumstances in maritime delimitation 188

(B) Irrelevance of “geographical factors” and “State activities” as special 191
circumstances
(1) Geographical Factors 191
(2) State Activities in the Relevant Area
192
(C) Serpents’ Island and its potential effect as a relevant circumstance 193
(1) Serpents’ Island as a relevant circumstance 194

(a) Serpents’ Island: overview of the arguments 195
(i) The legal entitlements of Serpents’ Island pursuant to the 1949 ProcŁs-195
Verbaux
(ii ) A rock unable to sustain human habitation or an economic life of its own
195
(iii) The position of Serpents’ Island vis-à-vis the Parties’ coasts 196
(b) Serpents’ Island as a special circumstance in any event
197
(D) The Black Sea as an Enclosed Sea and the Existing Delimitation Agreements in 200
the Black Sea
(a) The existing delimitation agreements 201

(b) The relevance of the enclosed character of the Black Sea in conjunction wit202
the geographical configuration of the delimitation area and the existing
delimitation agreements

IV(E) Conclusion 207
Addendum to Chapter 6
208
State practice cited in Romania’s Memorial and criticized in Ukraine’s Counter-Memorial
CHAPTER 7 OIL AND FISHERIES ACTIVITY 246

(A) Introduction 246
(B) Oil and Gas Concessions in the Delimitation Area 246

(1) General Considerations 246
(2) Paragraph 4(f) of the Additional Agreement 248

(3) Ukrains Concession Practice is Without Influence on the Delimitation 249
(a) The Olympiiska and the Gubkina blocks 250

(b) The Delphin block 251

(4) Romania consistently objected to Ukrainian hydrocarbon activity 252
(5) Romania’s Practice Regarding Oil and Gas Concessions 255

(C) Fisheries practice in the Delimitation Area 260
(1) Ukrainian Claims have never been Recognised
260
(2) Ukraine’s limited practice does not sustain any de facto line 263

(3) Judicial Consideration of Fisheries Issues 264
(4) Limited Dependence of the Parties on Black Sea Fisheries 267

(D) Conclusions 268

CHAPTER 8 THE DELIMITATION LINE 270

(A) The positions of the Parties 270
(B) Ukraine’s delimitation line 271
(C) Arriving at a delimitation line: a principled approach
273
(1) Serpents’ Island as a basepoint 273

(2) Ignoring small islands in defining a provisional equidistance line 275
(3) Adjusting the provisional equidistance line 281

(D) The delimitation line 282
(1) The delimitation between adjacent coasts (Sector 1)
282
(2) The delimitation between opposite coasts (Sector 2) 287
(E) Conclusion 289

CHAPTER 9 THE EQUITABLE CHARACTER OF THE 291
DELIMITATION
(A) Introduction
291
(B) The proportionality test in international case-law 291
(C) The equitableness of the line 296

(1) The Calculation of Proportionality 296
(2) The Principle of Non-encroachment
299
(3) The Impact on Security Interests of the Parties 303

(D) Conclusions 304

VCHAPTER 10 SUMMARY 305
SUBMISSION 312

APPENDIX THE DIPLOMATIC HISTORY RELATIVE TO 313

THE DISPUTE
LIST OF ANNEXES 326

LIST OF FIGURES 330

VI CHAPTER 1

INTRODUCTION

A. PURPOSE OF THE REPLY

1.1. This Reply is filed by Romania in accordance with the Court’s Order of 30 June

2006.

1.2. Taking into account the provisions of Article 49 of the Rules of Procedure of the

Court, this Reply will focus on those issues that divide the Parties in light of the Counter-

Memorial submitted by Ukraine.

1.3. Romania will not reiterate in this Reply the various errors in the Ukrainian

Counter-Memorial, which have already been detailed in the letter and attachment dated
1
25 September 2006, addressed to the Registrar.

B. THE COURT’S JURISDICTION OVER THE PRESENT DISPUTE

1.4. An initial area of disagreement concerns the competence of the Court to decide on

a maritime boundary delimiting the territorial sea of a Party, a competence which Ukraine

denies.

1.5. Ukraine stresses that Romania’s claimed maritime boundary in its first segment

(which coincides which the maritime boundary already established by the 1949

Romanian/Soviet agreements) divides the exclusive economic zone and continental shelf
of Romania from the territorial sea of Ukraine around Serpents’ Island. 2 Since, in its

view, the language of paragraph 4(h) of the 1997 Additional Agreement allows the Court

1
See Letter of the Agent of Romania to the Registrar of the International Court of Justice dated 25
September 2006 (Annex RR 1). By the time of the completion of Romania’s Reply, Ukraine forwarded the
answer that “it is in the process of verifying the questions raised by Romania” (by a letter dated 23
November 2006 sent to the Registrar by the Ukrainian Embassy in the Hague)
2 See RM, paras. 4.39, 11.54.

1to rule on a delimitation only between the exclusive economic zone and continental shelf

of one Party and the exclusive economic zone and continental shelf of the other Party,

Ukraine argues that to delimit the maritime boundary in accordance with Romania’s

claim would exceed the Court’s jurisdiction in this case. 3 In effect, it says, even if

Romania’s arguments concerning the Romanian/Soviet agreements are correct as a matter

of substantive law, the Court cannot give effect to those agreements. In other words,

according to Ukraine, the Court has jurisdiction to decide in its favour but not in favour

of Romania.

1.6. Ukraine relies on the 1977 Anglo-French Continental Shelf Arbitration case,

where the Court of Arbitration did not establish the maritime boundary between the

British Channel Islands and the French opposing coast but only the “outer” maritime

boundary between these islands’ maritime areas and the French continental shelf lying

seawards up to the median line from the French and British mainland. 4 Ukraine

misinterprets the Court of Arbitration’s decision. 5 The Special Agreement in that case

instructed the Court of Arbitration to delimit the continental shelf, which it held was by

definition an area lying seaward of the territorial sea. The Court of Arbitration did indeed

refer to its lack of competence to “settle differences between the Parties regarding the

boundary of their respective zones of territorial sea or of their respective fishery zones”, 6

but this referred to the fact that it lacked jurisdiction to delimit one Party’s territorial sea

or fishery zone from the other Party’s territorial sea or fishery zone, not that it lacked

jurisdiction to delimit one Party’s continental shelf from the other Party’s territorial sea

or fishery zone.

1.7. In fact while the Court of Arbitration refrained from establishing the eastern

(“inner”) maritime boundary between the maritime areas of the Channel Islands and the

French maritime areas lying in front of the French coasts of Normandy and Brittany, it

3
See UCM, paras. 2.19-2.20.
4 The Channel Islands were enclaved within the French continental shelf. See Delimitation of the
Continental Shelf (United Kingdom of Great Britain and Northern Ireland and the French Republic),
Decision of 30 June 1977, 54 ILR p. 6, at pp. 102-3 (para. 202).
5 See UCM, para. 2.12.
6
Decision of 30 June 1977, 54 ILR, p. 6, at p. 33 (para. 13).

2did establish the western (“outer”) maritime boundary between the maritime areas of the

Channel Islands and the French continental shelf lying seawards, even though this
boundary delimited the fishery zone of the Channel Islands from the French continental

shelf and even though the Court of Arbitration itself contemplated the UK’s right to

extend its then 3 nm territorial sea around the Channel Islands to 12 nm, thereby

transforming the boundary into one between a British territorial sea and the French

continental shelf. This situation is shown in Figure RR1 at page 4 of this Reply.

1.8. The reason the Court of Arbitration refrained from ruling on the “inner” maritime

boundary between the Channel Islands and France was that such a boundary would have

effectively delimited the territorial sea of one Party from the territorial sea of the other.

In this context, the Court of Arbitration stated that:

“the narrowness and rock-strewn character of these sea passages, which
the Court has already described, brings into question its competence to

make any such delimitation in the area in question. Although the United
Kingdom at present claims only a three-mile territorial sea around the
Channel Islands, the French Republic has established a 12-mile territorial
sea off all its coasts, including those of Normandy and Brittany. Again, the

United Kingdom claims, and has claimed before the Court, the right to
extend its three-mile territorial sea to one of 12 miles; and it already has a
12-mile fishery limit around the Channel Islands, established in
conformity with the European Fisheries Convention of 9 March 1964. To

this has to be added the fact that many of the French and British islets and
rocks scattered along the sea passages provide possible base-points for
advancing seawards the limits established or claimed by each Party. As a

result, the ‘continental shelf’ boundary which the Parties invite the Court
to delimit in the areas between the Channel Islands and the coasts of
Normandy and Brittany must traverse over almost its whole length waters
either claimed by France as part of its territorial sea, or by the United

Kingdom as par7 of its actual or potential territorial sea and of its existing
fishery zone.”

7
Ibid., p. 33 (para. 14).

3 4

i e o
n s o
t a e
r n s
b i h
r n (
f c d
t c l
u e I
o e r e
e a e n
t C th h
b e n C d
d S a h r
h a a f i
l n pe a e
R a n y s r
R e t a a m
r s o n o i
g y C o r r
F a c b t da
n r n l n
o - o t o
b l u a eb
m n r o t
i A d e
a h r t
m t o p
c C o
e h p
F , s
h e i
l s B
n r e
E l t
h c o
T A f1.9. The possibility that the UK could extend the territorial sea of the Channel Islands

from 3 to 12 nm turned into reality in the case of Jersey in 1987, when its territorial sea

was extended by the British authorities to 12 nm. The breadth of Jersey’ territorial sea

having been extended to 12 nm, the Anglo-French Agreement concerning the

Establishment of a Maritime Boundary between France and Jersey, St. Helier, 4 July
8
2000 (in force 1 January 2004) confirmed the maritime boundary between what became

the territorial sea of Jersey and France, as envisaged by the Court of Arbitration in 1977.

1.10. The situation in the present case does not resemble the situation of the “inner”

maritime boundary of the Anglo-French Arbitration. The situation is not geographically

complex and relevant basepoints are clear and not in dispute. There is no possibility of

the further extension of Romania’s territorial sea, which already has a breadth of 12 nm.

Thus the maritime boundary on the 12 nm arc around Serpents’ Island between Point F

and Point X can never mutate from a boundary separating the Romanian continental shelf

and exclusive economic zone from the Ukrainian territorial sea into a boundary

separating the Romanian and Ukrainian territorial seas (as was the case with the “inner”

boundary in the Anglo-French Arbitration).

1.11. The Black Sea delimitation situation resembles more the situation of the “outer”

maritime boundary between the Channel Islands and the French continental shelf, which
9
was duly established by the Court of Arbitration .

1.12. The case-law supports the conclusion that international courts do not consider

themselves inhibited from establishing maritime boundaries separating, on the one hand,

the continental shelf (or the exclusive economic zone) of one party and, on the other

hand, other maritime areas (including the territorial sea) of the other party.

1.13. Examples include the Dubai/Sharjah case where the Tribunal established a

maritime boundary which, in its final segment, delimited the continental shelf of Dubai

8 Agreement between the United Kingdom of Great Britain and Northern Ireland and the French
Republic concerning the establishment of a maritime boundary between France and Jersey, St. Helier, 4
July 2000, 2269 UNTS 288.
9
See para. 1.6 and Figure RR1 above.
5from the territorial sea of the island of Abu Musa, belonging to Sharjah; 10 the second

phase of the Eritrea/Yemen arbitration, where the maritime boundary established by the

Arbitral Tribunal delimited the Eritrean continental shelf and exclusive economic zone

from the Yemeni territorial sea around the islands of the Jabal al-Tayr and the Zubayr

group, 11and the St. Pierre et Miquelon arbitration, where the Tribunal awarded only a 12

nm belt of territorial sea to the south-eastern coast of the French islands of St. Pierre,

neighbouring the Canadian continental shelf. 12

1.14. In any event, the practical consequences of the Parties’ divergent approach on the

Court’s jurisdiction are nil. The maritime boundary in the area around Serpents’ Island

was already established by a succession of Romanian/Soviet agreements beginning in
13
1949, on the 12 nm arc surrounding Serpents’ Island (between Points F and X). Even if

(quod non) the Court had no jurisdiction to delimit the continental shelf and exclusive

economic zone of one Party from the territorial sea of the other (which is not the case), it

would still have to take into account the agreements in force between Romania and

Ukraine and the resulting maritime boundary. Those agreements did not cease to be in

force or to have legal effect by reason of the conclusion of the 1997 Additional

Agreement. Nor did the conclusion of the 1997 Agreement involve a waiver or

renunciation by either Party of legal positions which were by then already articulated and

well-known to the other. Thus, in such a hypothetical situation, the Court would have

jurisdiction to perform the delimitation between the Romanian and Ukrainian respective

continental shelves and exclusive economic zones starting from the final point of the

boundary already established in 1949 on the 12 nm line around Serpents’ Island (i.e.,

Point X).

1.15. The consequences would be the same: instead of starting the delimitation from

Point F, confirming the first sector of the maritime boundary as established in 1949 (on

the 12 nm arc around Serpents’ Island) and then ruling on the delimitation seawards, the

10
11 See Dubai-Sharjah Boundary Arbitration, (1981) 91 ILR, p. 543, at pp. 675-7.
Government of the State of Eritrea and Government of the Republic of Yemen (Phase Two:
Maritime Delimitation), Award of 17 December 1999, 119 ILR, p. 417, at p. 454 (para. 119).
12 See Delimitation of Maritime Areas between Canada and France (St. Pierre et Miquelon), Award
of 10 June 1992, 31 ILM, pp. 1145, 1148, 1169-71.
13 See RM, paras.11.3-11.57 and see further Chapter 4.

6Court would start to rule on the delimitation between the outer maritime areas of the two
States directly from Point X seawards.

C. STRUCTURE OF THE REPLY

1.16. This Reply contains 10 chapters and an Appendix.

1.17. In its Counter-Memorial, Ukraine disputes the necessity to take into account the

five principles recognised by the Parties in the 1997 Additional Agreement. In addition,

while acknowledging the validity of the 1949, 1954, 1963 and 1974 Romanian-Soviet

ProcŁs-Verbaux, it denies that these are agreements relating to the delimitation of
maritime boundaries beyond 12 nm in the sense of 1982 UNCLOS. Chapter 2 responds

briefly to these arguments concerning the applicable law.

1.18. Although it is evident that the delimitation must be conducted in two sectors,

dominated first by the adjacent east-facing coasts of the Parties and subsequently by their

opposite coasts, this conclusion is challenged by Ukraine based on a mistaken

interpretation of the geography of the delimitation area. In Chapter 3 Romania reinstates

the correct approach and justifies its envisaged limits of the area relevant to the
delimitation.

1.19. In its Memorial, Romania demonstrated that the first segment of the delimitation

line (i.e. the maritime boundary on the 12 nm arc of circle around Serpents’ Island up to

the point referred by Romania as Point X) had already been agreed as a result of the

1949, 1954, 1963 and 1974 Romanian-Soviet agreements. This was subsequently

reflected in charts issued by Romania, Ukraine, the Soviet Union and third parties.

Ukraine rejects the legal consequences of the Romanian-Soviet instruments, as well as

the evidentiary value of the annexed maps and charts (though it presents no
countervailing maps and charts supporting its position). Chapter 4 shows that Ukraine’s

attacks both on the Romanian-Soviet ProcŁs-Verbaux and the map evidence are

unfounded.

71.20. As one of the main elements in dispute, Serpents’ Island needs further
consideration. Chapter 5 (which is supplemented by an Addendum) underlines its

status as a rock in the sense of Article 121(3) of the 1982 UNCLOS, not entitled to its

own continental shelf and exclusive economic zone. In spite of the continuous efforts of

the Ukrainian authorities to artificially change its character, Serpents’ Island is a small

rocky feature, unable to sustain human habitation or an economic life of its own: it will

be shown that the evidence annexed by Ukraine to its Counter-Memorial eloquently

supports this conclusion.

1.21. The relevant circumstances which have to be taken into account so that an

equitable solution may be reached are essentially the following: the presence and

character of Serpents’ Island and the aspect of the Black Sea as an enclosed sea in which

what are essentially mainland coast equidistance lines have already formed the basis for

delimitation agreements. Ukraine denies the relevance of each of these: its arguments are

considered in Chapter 6, which is supported by a further analysis of State practice by

way of an Addendum.

1.22. By contrast Ukraine argues that its oil and fisheries activities should be treated as
relevant if not decisive for the delimitation: Chapter 7 rebuts these arguments.

1.23. In theory Ukraine accepts the equitable principles-relevant circumstances method;

in practice it does not apply it. Chapter 8 reemphasizes the need for a two-sector

approach, starting with a provisional equidistance line drawn from the mainland coasts of

the Parties. It also rebuts Ukraine’s theory according to which Serpents’ Island is a

basepoint for the purposes of delimiting the maritime spaces of the two States, effectively

occluding most of the Romanian coasts in both sectors.

1.24. In Chapter 9, Romania will demonstrate the inequitable character of the

Ukrainian approach, as opposed to the equitableness of its proposed delimitation. In

particular it will show that the Ukrainian proposed line encroaches upon the Romanian

maritime areas and has a pronounced cut-off effect on Romania’s entitlements.

81.25. Chapter 10 presents a summary of Romania’s case.

1.26. The Appendix gives a short review of the diplomatic history, correcting

allegations made in that regard in the Counter-Memorial.

1.27. Filed with this Reply are a volume with documentary annexes and maps and a

volume with translations into English of the documents issued originally in Romanian,

Ukrainian or other languages.

9 CHAPTER 2

THE APPLICABLE LAW

A. INTRODUCTION

14
2.1. In its Memorial, Romania set out its approach regarding the law applicable to

the present dispute. In carrying out the task entrusted to it by the Parties, the Court

should apply the relevant articles (Articles 74 and 83) of the 1982 UNCLOS, to which

both Romania and Ukraine are parties. These specifically state that:

“4. Where there is an agreement in force between the States
concerned, questions related to the delimitation of the exclusive economic

zone/continental shelf shall be determined in accordance with the
provisions of that agreement.”

Thus existing agreements in force between the Parties apply. This is the case of the

agreements establishing a maritime boundary between Romania and Ukraine, namely the
15
Romanian-Soviet ProcŁs-Verbaux of 1949, 1963 and 1974. These agreements are

controlling as to the respective entitlements of the Parties in the area neighbouring

Serpents’ Island. Beyond that the principles of delimitation recognised by the Parties in

the 1997 Additional Agreement are also applicable and govern the delimitation where the
boundary has not already been agreed.

2.2. The principles of delimitation referred to in the 1997 Additional Agreement are

consistent with the jurisprudence of the Court concerning maritime delimitation. The

resulting approach may be referred to as the equitable principles-relevant circumstances

approach (under customary law as embodied in the 1982 UNCLOS); it is substantially

equivalent to the equidistance-special circumstances approach under the 1958

Convention (also referred to in Article 15 of the 1982 UNCLOS in respect to the

delimitation of territorial waters).

14
15 See RM, Chapters 7-8.
Annexes RM 13, 14, 15, 19, 20, 21 & 22.
102.3. At a general level, Ukraine seems to agree with the equidistance-relevant

circumstances approach, 16 in marked contrast with the position it took during the bilateral

negotiations. 17 On the other hand Ukraine denies the relevance of the principles

identified in the 1997 Additional Agreement. It also denies that the 1949, 1963 and 1974

ProcŁs-Verbaux and the 1954 Act represent delimitation agreements in the sense of

articles 74(4) and 83(4) of the 1982 UNCLOS, and, thus, denies their relevance to the

present dispute.

B. THE 1949, 1963 AND 1974 PROC¨S-VERBAUX AND THE 1954 ACT AS

DELIMITATION AGREEMENTS

2.4. While accepting the binding character of the 1949, 1963 and 1974 ProcŁs-

Verbaux and the 1954 Act, Ukraine asserts that they are not applicable because they

“were not agreements delimiting the continental shelf and/or EEZ” in the sense of articles
18
74(4) and 83(4) of the 1982 UNCLOS.

2.5. The 1949 ProcŁs-Verbaux established the Romanian/Soviet (now Ukrainian)

maritime boundary from the last point of the land/river border between the two countries,

going through points 1438 and 1439 and then continuing on the 12 nm arc of circle

surrounding Serpents’ Island. This boundary was confirmed subsequently by the 1954

Act and the 1963 and 1974 ProcŁs-Verbaux. Convergent graphical representations of this

boundary, up to a point on the 12 nm arc of circle around Serpents’ Island situated due

east of it, are given by Soviet and Romanian charts (as well as third States’ charts)

produced since 1951. 19 This will be demonstrated in further detail in Chapter 4.

2.6. Therefore, the 1949, 1963 and 1974 ProcŁs-Verbaux and the 1954 Act, which

establish a maritime boundary between Romania and Ukraine in the area neighbouring

Serpents’ Island, are controlling as to the respective entitlements of the Parties.

16
See UCM, Chapter 6, Section 6.
17 See Note Verbale no. 72/16-446-119 dated 29 May 2002 of the Ministry of Foreign Affairs of
Ukraine to the Embassy of Romania in Kiev (Annex RM 26).
18 UCM, para. 6.24.
19 RM, paras. 11.26-11.38.

11C. THE 1997 ADDITIONAL AGREEMENT UNDERWRITES THE

EQUITABLE PRINCIPLES-RELEVANT CIRCUMSTANCES APPROACH

2.7. In Romania’s view, the principles mentioned by the 1997 Additional Agreement

are to be applied by the Court, since it is for the Court to complete what the Parties have

not been able to achieve through direct negotiations.0 As a Chamber of the Court said in

the Gulf of Maine case:

“The application of the rules of international law and the methods of

delimitation considered the most appropriate in this case might present the
Chamber with the temptation to adopt another starting-point of the line to
be drawn, or to draw a line terminating at a point outside the triangle.

However, even disregarding the somewhat improbable nature of this
hypothesis, the decisive reason why such solutions should not be pursued
is the fact that for the delimitation of a maritime boundary – whether it

concern[s] the territorial sea or the continental shelf or the exclusive
economic zone – both conventional and customary international law
accord priority over all others to the criterion that this delimitation must

above all be sought, while always respecting international law, through
agreement between the parties concerned. Recourse to delimitation by
arbitral or judicial means is in the final analysis simply an alternative to
21
direct and friendly settlement between the parties.”

2.8. Ukraine argues that “it would have been impertinent to seek to tell the tribunal
22
what ‘principles’ to apply”. But spelling out the task entrusted to a court or tribunal is

common. For example in the Case concerning the Continental Shelf (Tunisia/Libyan

Arab Jamahiriya), the Court was requested to decide which were the principles and rules

of international law to be applied in the delimitation of their continental shelf and “in

rendering its decision, to take account of equitable principles and the relevant

circumstances which characterise the area, as well as the recent trends admitted at the

Third Conference on the Law of the Sea.” 23 The Court had no difficulty in acting in

20 See RM, paras. 7.7-7.8.
21
Delimitation of the Maritime Boundary in the Area of the Gulf of Maine (Canada/United States of
America), ICJ Reports 1984, p. 266 (para. 22). See also North Sea Continental Shelf Cases, I.C.J. Reports
1969, p. 47 (para. 87), citing Free Zones of Upper Savoy and the District of Gex, PCIJ, Series A, No.22
(1932), p. 13.
22 UCM, para. 6.16.
23 Special Agreement, Tunis, 10 June 1977, 1120 UNTS 104, Article 1, quoted by the Court in its
judgment: Case concerning the Continental Shelf (Tunisia/Libyan Arab Jamahiriya), I.C.J. Reports 1982,
p. 21 (para. 2).

12accordance with this mandate: it expressly accepted that States parties in delimitation

cases may stipulate a binding lex specialis. 24

2.9. The Parties to the 1997 Additional Agreement understood that the principles it

identified as relevant were to be applied also by the Court, if the issue of the maritime

delimitation in the Black Sea were to be submitted to it. This is clear from the Additional

Agreement itself. Article 4, which specifies the applicable principles and procedures,

goes on to provide in paragraph (h) for reference to the Court if the parties fail to agree: it

cannot be supposed that the Court would ignore the very principles which in Article 4 the

Parties committed themselves to take into account. Article 4 deals with an overall

process for “reaching a solution concerning the delimitation of the continental shelf” (see

paragraph (f)); it does not envisage two processes with distinct purposes to which

different considerations would be relevant. It is on this basis that the standstill agreement

in paragraph (f) is applied also while the case is pending before the Court. 25

2.10. In conclusion, it is clear that, in accordance with international law, the principles

of delimitation recognised by the Parties in the 1997 Additional Agreement are to be

applied by the Court as governing the delimitation of the maritime areas in the Black Sea

where the boundary has not already been agreed.

D. IRRELEVANCE OF ROMANIA’S “DENUNCIATION” OF THE 1948

PROTOCOL AND OF THE 1961 BORDER REGIME TREATY

2.11. In its Counter-Memorial Ukraine refers to Romania’s “denunciation” of the 1948

Protocol to Specify the Line of the State boundary between Romania and the USSR and

of the 1961 Treaty on the Romanian/Soviet State Border Regime. 26 But Ukraine does not

seem to draw from this episode any conclusions relevant to the present proceedings.

24
25 Ibid, p. 38 (para. 24).
Para. (f) of Article 4 provides for the commitment of the two Parties to refrain from exploiting the
mineral resources of the continental shelf (unless by a joint exploitation) until the issue of the delimitation
is solved; evidently this provision is applicable not only during the negotiations, but also if the negotiations
are suspended or terminated and the case is submitted to legal settlement by a third party, until a solution is
decided on. To interpret the provisions of the Additional Agreement as meaning that either Party would be
free to start unilateral exploitation of the mineral resources of the undelimited area in case that the bilateral
negotiations, for whatever reasons, come to an end, cannot be correct.

13Romania for its part agrees that these issues have no present relevance. However, in the

interest of an accurate presentation of the facts, in the Appendix Romania responds
27
briefly to the particular points made.

E. CONCLUSION

2.12. To summarise:

(1) The Romanian-Soviet agreements establishing a maritime boundary between

Romania and Ukraine in the area neighbouring Serpents’ Island (i.e., the

ProcŁs-Verbaux concluded in 1949, 1963 and 1974 and the 1954 Act) are

controlling as to the respective entitlements of the Parties;

(2) The principles of delimitation recognised by the Parties in the 1997 Additional

Agreement fall to be applied by the Court as governing the delimitation where
the boundary has not already been agreed;

(3) The so-called “denunciation” by Romania of the 1948 Protocol to Specify the

Line of the Romanian/Soviet Boundary and the 1961 Romanian/Soviet Border

Regime Treaty has no consequence regarding these agreements and the

maritime boundary they established.

26 UCM, paras. 5.39, 5.93.
27
See below, paras. A27-A36
14 CHAPTER 3

THEGEOGRAPHICALFRAMEWORK:

THE RELEVANT COASTS AND THE RELEVANT AREA

A. INTRODUCTION

28
3.1. As Romania stated in its Memorial, the correct approach in delimiting the

maritime areas of Romania and Ukraine in the Black Sea is first to draw a provisional
equidistance/median line between the mainland coasts of the Parties, and subsequently to

adjust it, if necessary, to take into account (a) any applicable agreements and (b) the

relevant circumstances of the delimitation area, in order to reach an equitable result.

3.2. In accordance with international practice in this regard, cited by Romania in its

Memorial, and as specifically accepted by Ukraine in its Counter-Memorial, 29before

drawing the provisional equidistance/median line it is necessary to identify the coasts of

the Parties relevant to the delimitation, as well as the relevant maritime area. This can

only be done taking into account the geographical situation of the general area in which

the delimitation is to occur. Further, the general geographical setting, in particular the

coastal relationships of the Parties, may lead the Court to conclude that the area to be

delimited should be regarded as composed of different sectors.

3.3. The present Chapter is concerned with these issues. Given the importance of the

general geographic situation, Romania will show that the relevant area falls into two
sectors, correcting the erroneous treatment contained in Ukraine’s Counter-Memorial in

that regard. It will briefly discuss the correct approach to the identification of the

relevant coasts and relevant areas, as disclosed by the case-law of this Court and

international tribunals, outlining the ways in which the approach adopted by Ukraine

diverges from this approach. It will show how Ukraine falls into error in identifying

respectively the relevant coasts and relevant areas for the delimitation, while addressing

the criticisms raised by Ukraine of Romania’s approach.

28
29 See RM, Chapter 8 and also paras. 9.7, 9.15.
See UCM, Chapter 7, Section 2.
15B. THE GENERAL GEOGRAPHIC SETTING AND ITS RELEVANCE

(1) Physical Characteristics of the Parties’ Coasts

3.4. In its Memorial, Romania demonstrated that the coastal situation of the Parties

clearly dictates that the construction of the provisional equidistance line be carried out in

two sectors; Ukraine disputes this approach. 30

3.5. Ukraine purports to divide its own coast into three sectors. It then adopts a

“projection method” in order to justify the position that its entire coastline bordering on

the western basin of the Black Sea constitutes a legally relevant coast for the present

delimitation. Furthermore, it includes Serpents’ Island in its discussion of its relevant

coast, even though it is situated a substantial distance out to sea, more than 20 nm (over

37 km) from the Ukrainian baselines as deposited with the UN Secretary-General in
31
accordance with Article 16 of UNCLOS. The incorrectness of Ukraine’s approach in

using this “projection” method is discussed in this Chapter. The weight (if any) to be

given to Serpents’ Island is discussed in Chapters 5 and 8.

(a) The Romanian coast

3.6. The Romanian coast comprises two distinct segments. Starting from the border

with Ukraine, the coast of Romania stretches in a broadly southerly direction for a

distance of about 70 km (70.25 km) until the Sacalin Peninsula. The only major features

in this stretch of coast are the Sulina Dyke, an extended harbour work stretching some 7,5

km out to sea a short distance south of the Romania/Ukraine land-river boundary, and the

mouth of the Sfântu Gheorghe arm of the Danube, located slightly to the north of the

Sacalin Peninsula. The Sacalin Peninsula forms the southern limit of this section, and is

a narrow promontory extending in a roughly south-westerly direction from the point at

which the general direction of the coast changes direction. 32

30 See RM, paras 9.1, 9.4, 9.5; see also UCM, paras 4.4-4.32.
31 See Annex RM 27.
32 For an image of the Sacalin Peninsula, see UCM, Figure 4-1 facing p. 38.

163.7. From the Sacalin Peninsula, the coast, having altered its course to turn westwards,
proceeds in that direction until it reaches the area of the Razim Lake, a brackish lake on

Romanian territory separated from the sea by a relatively narrow strip of land. The coast

then gradually curves to the south, and subsequently proceeds in a broadly southerly

direction until it reaches the Romanian-Bulgarian land border. Seen together, the

Romanian coast from the Sacalin Peninsula in the north to the Romania/Bulgarian land

border in the south forms a single segment.

3.8. This second, southern sector recedes to the west behind the first, northern segment
which lies adjacent to a segment of the Ukrainian coast. The northern sector, which

corresponds to the Romanian coast of the Danube Delta, projects straight on into the

Black Sea. Its extremities (the end of the Sulina Dyke in the north and the end of the

Sacalin Peninsula in the south) constitute the points of the Romanian coastline located

furthest to the east and they are basepoints for drawing the median line between the

opposing coasts of the Parties. Nevertheless, the whole of the Romanian coast lies in a

relationship of oppositeness as regards the sector of the Ukrainian coast located on the

Crimean Peninsula. The Romanian coast and its sectors is presented in Figure RR9 at

page 49 below.

(b) The Ukrainian coast

3.9. The Ukrainian coast in the western basin of the Black Sea is extremely irregular.

It is characterised by a number of deep indentations and reverses its course sharply
several times, with segments of its coast facing one another.

3.10. From the land-river border with Romania, the Ukrainian coast proceeds broadly

northwards for a short distance until it reaches the northernmost tip of the Danube Delta

appertaining to Ukraine (the end of the Prorva canal). From this point, its course veers

westwards for a relatively short distance, then in a north-easterly direction until the

coastline is broken by the large Nistru/Dniester Firth. Taking into account the

geographical characteristics of the coasts, the Nistru/Dniester Firth and, more concretely,
the point where its southern bank meets the coast (referred to as Point S) marks the end

17of the section of Ukraine’s coast which has a relation of adjacency with the Romanian

coast.

3.11. The Nistru/Dniester Firth has an average width of 8,8 km. To the north of the

estuary the Ukrainian coast changes direction, proceeding in a north-north-easterly

direction until it reaches Odessa. At Odessa it adopts a roughly northern course until, at a

point north-east of Odessa, it once again markedly alters its course, turning eastwards

until it reaches the Dnieper Firth. Here the coastline becomes extremely irregular, with

two large land protrusions flanking the Yahorlitska Gulf. The general direction of the
coast is first a southerly one, and then, from the bottom of the Yarholitska Gulf, the

direction is an easterly one, until it comes to the bottom of the Karkinitska Gulf. From

here it turns back on itself sharply, extending south-westwards along the southern coast

of the Karkinitska Gulf, until it reaches Cape Tarkhankut. A final segment comprises the

coast of Crimea between Cape Tarkhankut and Cape Sarych. This last sector is concave,

its general direction being interrupted by a significant protrusion, the western-most point

of which is Cape Khersones. 33

3.12. Contrary to the position taken in Ukraine’s Counter-Memorial, the Ukrainian

coast fronting onto the north-western part of the Black Sea is composed of as many as

eight distinct segments, determined by marked changes in the direction of the coast:

- first, from the end of the Romanian-Ukrainian land boundary to the northern tip of

the Danube Delta;

- second, from this point to the Nistru/Dniester Firth;

- third, from the Nistru/Dniester Firth to a point north of Odessa;

- fourth, from that point north of Odessa to the Dnieper Firth;

- fifth, from the Dnieper Firth to the bottom of the Yahorlitska Gulf;

- sixth, a sector from the Yahorlitska Gulf to the bottom of the Karkinitska Gulf;

- seventh, a sector comprising the southern coast of the Karkinitska Gulf to Cape

Takhankut; and

33 Ukraine’s description of its coast is replete with comments on the size of its cities and their main
industrial activities: see e.g., UCM, paras. 3.17, 3.18; see also UCM Figure 3-5. These descriptions are

irrelevant, as the Court observed in Continental Shelf (Tunisia/Libyan Arab Jamahirya), ICJ Reports 1982
p. 18, at p.77 (para. 107).
18 - eighth, a further segment, composed of the Crimean coast between Cape Tarkhankut
and Cape Sarych.

The Ukrainian coast and its various segments are represented in Figure RR2 at page 20

below.

19 0
2

s
i
a
V
i
d
R a ts
R s e
r o m
g C e
F a S
i
r
k
e
h
T (2) The Appropriateness of a Two-Sector Approach

3.13. It is evident from the geographical situation of the coasts of the two Parties that

the delimitation should be approached on the basis of two sectors. Ukraine argues that
Romania’s approach is illegitimate as it involves “double counting” of the portion of the

Romanian coast between the Romanian-Ukrainian land/river boundary and the basepoint

utilised by Romania on the Sacalin Peninsula. However, the Parties expressly foresaw a

two-sector approach in the 1997 Additional Agreement, unsurprisingly given the

geographical realities of the area. Further Ukraine’s allegation that the Romanian

technique involves “double-counting” is unsustainable.

(a) The 1997 Agreement foresaw a two sector approach

3.14. The Parties already envisaged that the delimitation area should be treated in two

different sectors. Paragraph 4(b) of the 1997 Additional Agreement establishes that the

Parties are to apply:

“the principle of the equidistance line in areas submitted to delimitation
where the coasts are adjacent and the principle of the median line in areas
where the coasts are opposite.”

Clearly the Parties envisaged that they would have first to distinguish those “areas

submitted to delimitation where the coasts are adjacent” from those “areas where the

coasts are opposite” and secondly to apply the corresponding principles of either “the

equidistance line” or “the median line”.

3.15. The approach required by the 1997 Additional Agreement is a direct reflection of

the geographical situation of the area to be delimited. It is also fully in line with the

decisions of international courts and tribunals. But Ukraine, having agreed to paragraph

4(b) of the 1997 Additional Agreement, now argues that it is “artificial” and
34
“disingenuous”.

34 See UCM, paras. 4.2-4.3.

21 (b) The general geographical situation dictates a two sector

approach

3.16. It is evident that the Romanian coast at the same time is both adjacent to a

segment of the Ukrainian coast of the Ukrainian region (oblast) of Odessa and opposite

the west-facing Ukrainian coast of the Autonomous Republic of Crimea. This situation is

presented in Figure RR3 at page 23 of this Reply (reproduced from Romania’s

Memorial, page 137). This fact makes it appropriate that the delimitation be approached

on the basis of two sectors. 35 In this context, it is irrelevant that “Romania’s coast may

itself be divided into two sectors”. 36

3.17. The distinction between coastal relationships of adjacency and oppositeness has

always been recognised in maritime delimitation. The articles of the 1982 UNCLOS

concerning the delimitation of territorial seas (Article 15), exclusive economic zones

(Article 74) and continental shelf (Article 83) expressly mention “States with opposite or

adjacent coasts” (emphasis added). The same was true of the 1958 Geneva
37
Conventions.

35 See RM, Chapter 9, Section 1.
36 Ukraine alleges that this was the criterion invoked by Romania to justify its two-sector approach.
See UCM, para. 4.5; this paragraph stands in contrast with the preceding paragraph (para. 4.4), where by
reference to the Tunisia/Libya and Qatar-Bahrain cases Ukraine apparently takes the position (identical to

37at of Romania) that the relevant criterion is the relationship between the coasts of the two States.
Convention on the Continental Shelf, Geneva, 29 April 1958, 499 UNTS 312, art. 6; Convention
on the Territorial Sea and the Contiguous Zone, Geneva, 29 April 1958, 516 UNTS 205, art. 12.

22 Figure RR3
(Reproduced from RM, Figure 11)
The relevant coasts of the two States
As seen, the Romanian coast is both adjacent and
opposite to the Ukrainian coast

233.18. When approaching questions of maritime delimitation, it has been the constant

practice of the Court and of arbitral tribunals to characterise the relationship of the coasts

of the parties in terms of adjacency or oppositeness. The Court has never hesitated to

conduct an in-depth examination of the relationship of the coasts of the parties, in some

cases concluding that the factual relation of adjacency between part of the coasts of the

parties at a certain point gave way to one of oppositeness or vice versa. In such

circumstances, the Court has consistently concluded that the delimitation was to be

conducted in several sectors in order to respond to the geographical reality of the

situation.38

3.19. Thus the Court has been called upon to decide cases in which different segments

of the relevant coasts of the parties stood in differing relations of adjacency or

oppositeness. For example, in the Qatar-Bahrain case the Court held that in the southern

part of the delimitation area the coasts were opposite to each other while to the north the

coasts were no longer opposite but rather in a situation of adjacency. 39

3.20. Similarly in the Anglo-French Continental Shelf Arbitration, the Tribunal

considered the coasts of the two parties in the English Channel as opposite, while their

coasts as the delimitation line moved out into the Atlantic Ocean were to be regarded as

standing in a relation of adjacency. 40

3.21. In the Tunisia/Libya Continental Shelf case, the Court identified a transformation

of a situation of adjacency although it stopped short of saying that the relationship of the

coasts had actually changed from a relationship of adjacency to one of oppositeness. The

Court determined that the geographical configuration of the Tunisian coast, which

initially was undoubtedly one of adjacency with the coast of Libya at the terminus of the

land boundary changed along its length:

38 See RM, Chapter 9 for a review of the authorities.
39 Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain),

40J Reports 2001, p. 40, at pp. 91-93 (paras. 169-170).
Delimitation of continental shelf (United Kingdom of Great Britain and Northern Ireland and the
French Republic), Decision of 30 June 1977, 54 ILR, p. 6.

24 “While the initial part of the Tunisian coast, westwards from Ras Ajdir,
runs for some distance in approximately the same direction as the Libyan
coast, the most marked characteristic of the coast, discussed at length by

the Parties, is that it subsequently changes direction, so as to run roughly
southwest-northeast. This aspect of the geographical situation as it exists
in the area relevant to the decision is legally significant, in the context of

the present examination of the application of equitable principles, as one
of the relevant circumstances which characterize the area. The change in
direction may be said to modify the situation of lateral adjacency of the

two States, even though it clearly do41 not go so far as to place them in a
position of legally opposite States.”

Despite the fact that the coasts were held not to be in a situation of “legal” oppositeness,

the Court nevertheless concluded that effect had to be given to the changed relationship

of the coasts as a result of the physical geographic situation, and that the delimitation had

to be approached on the basis of division of the delimitation into two sectors:

“in the view of the Court, the proper appreciation and taking into account
of the ‘relevant circumstances which characterise the area’ call for the area
close to the coasts of the parties to be treated differently from the area

further offshore. The Court will therefore deal with the area as divided
into two sectors. It must, however, be emphasised that such difference of
treatment is ultimately dictated by the primordial requirement of achieving
42
an overall equitable result.”

3.22. The geographical situation in the present case is even clearer than that in the

Tunisia/Libya case. The coast of Romania is at one and the same time both adjacent to

one segment of the Ukrainian coast and opposite to another segment. As can be readily

seen from Figure RR3 at page 23 above, the Romanian coastal segment represented by

the Black Sea façade of the Danube Delta, situated between the last point of the land/river

border between Romania and Ukraine and the outer limit of the Sacalin Peninsula, stands

in a relationship of adjacency with the segment of the Ukrainian coast situated to the

north of the boundary with Romania. At the same time, the entire length of the

Romanian coast, from the land boundary with Ukraine in the north to the land boundary

with Bulgaria in the south, faces across the western basin of the Black Sea a separate

41
Continental Shelf Case (Tunisia/Libya), ICJ Reports 1982, p. 18, at p. 63 (para. 78).
42 Ibid, p. 82 (para. 114).

25segment of the Ukrainian coast – the western coast of the Crimea Peninsula – with which
it is in a relationship of oppositeness.

3.23. The situation in the present case is particularly striking in that, due to the

configuration of the coasts of Romania and Ukraine, there is a clearly identifiable point

on a provisional equidistance/median line drawn between the closest relevant basepoints

on the coastlines of Romania and Ukraine at which the line changes from being one

governed by portions of the coasts of the Parties standing in a relationship of adjacency to

being governed by the coasts of the Parties standing in a relationship of oppositeness to
each other. That point corresponds to the point referred to by Romania in the Memorial

as “Point T”. To the west of Point T, the provisional line is an equidistance line based

upon points on the adjacent coasts of Romania and Ukraine; to the south of Point T, the

line becomes a median line between the segments of the coasts of Romania and Ukraine

which are opposite.

3.24. Thus the maritime delimitation in the Black Sea clearly falls to be effected in two

sectors: Sector 1, characterised by the relationship of adjacency between the relevant

segments of the Parties’ coasts, and Sector 2, where it is the relationship of oppositeness
between the relevant segments of Parties’ coasts which prevails.

(c) A two-sector approach involves no double-counting

3.25. Ukraine argues that Romania’s approach involves double counting of the same
coasts. But it is a fact that a certain segment of the Romanian coast finds itself, at the

same time, in a situation of adjacency with one segment of the Ukrainian coast and

(together with the rest of the Romanian coast) in a situation of oppositeness in relation to

another segment of the Ukrainian coast. This can in no way affect the appropriateness of

a two-sector approach.

26 43
3.26. This does not mean that this segment is “counted” twice in the delimitation but

only that, as this same segment is relevant to both sectors of the delimitation area, it is

utilised in relation to both sectors. Thus, while the length of the Ukrainian relevant coast

is calculated by adding the length of the relevant adjacent coast to the length of the

relevant opposite coast, the total length of the Romanian coast is not calculated in the

same way. The Romanian relevant coast includes those segments relevant for

delimitation, each of which is counted only once, irrespective of whether the segment is
44
relevant for more than one sector.

C. IDENTIFYING THE RELEVANT COASTS AND THE RELEVANT AREA

3.27. In the Tunisia-Libya Continental Shelf case, the Court discussed the approach to

be taken in identifying the relevant coasts and the relevant area for the delimitation of

continental shelf between the Parties. The Court emphasised that entitlement to

submarine areas of jurisdiction is premised on the possession of a coast bordering onto
45
any given area. Thus, as the Court affirmed:

“The coast of each of the Parties, therefore, constitutes the starting line

from which one has to set out in order to ascertain how far the submarine
areas appertaining to each of them extend in a seaward direction, as well
as in relation to neighbouring States situated either in an adjacent or

opposite position.”

As the Court observed:

“The only areas which can be relevant for the determination of the claims
of Libya and Tunisia to the continental shelf in front of their respective
coasts are those which can be considered as lying either off the Tunisian or

off the Libyan coast. These areas form together the area which is relevant
to the decision of the dispute. The area in dispute, where one claim
encroaches on the other, is that part of this whole area which can be

considered as lying both off the Libyan coast and off the Tunisian coast.”

43
Romania is somewhat intrigued by the obsessive manner in which Ukraine accuses Romania of
44uble counting the same coasts in its approach to the delimitation of the maritime areas.
45 See RM, paras. 9.24, 9.25. See also below, paras 3.30, 9.26-9.27.
Continental Shelf Case (Tunisia/Libya), ICJ Reports 1982, p. 18, at p. 61 (para. 73).

27However, not all of those “potentially relevant” areas will be relevant to a delimitation of

the boundary between them. Certain lengths of the coast, even though they may generate

entitlement to submarine areas, will not be treated as relevant for the purposes of the

delimitation, as the Court went on to explain:

“Nevertheless, for the purpose of shelf delimitation between the Parties, it

is not the whole of the coast of each Party which can be taken into
account; the submarine extension of any part of the coast of one Party
which, because of its geographic situation, cannot overlap with the
extension of the coast of the other, is to be excluded from further

consideration by the Court. It is clear from the map that there comes a
point on the coast of each of the two Parties beyond which the coast in
question no longer has a relationship with the coast of the other Party

relevant for submarine delimitation. The sea-bed areas off the coast
beyond that point cannot therefore constitute an area of overlap of the
extensions of the territories of the two Parties, and are therefore not
46
relevant to the delimitation.”

On the facts of the case, the Court held that those points were Ras Kaboudia on the

Tunisian coast, and Ras Tajoura on the Libyan coast. 47

3.28. Thus if the submarine extension of the coast of one Party is, due to its geographic

situation, incapable of overlapping with the submarine extension of any part of the coast

of the other party that coast will be irrelevant to the delimitation. The relevant area for

the delimitation is that area consisting of the overlapping submarine extension or

prolongation of the coasts of each Party.

3.29. That this is the correct approach is confirmed by the decision in the Jan Mayen

case. In identifying the relevant area, the Court observed that:

“The ‘area of overlapping claims’… between the two lines representing
the Parties’ claims, is of obvious relevance to any case involving opposed

boundary claims. But maritime claims have the particular feature that
there is an area of overlapping entitlements, in the sense of overlap
between the areas which each State would have been able to claim had it

not been for the presence of the other State; this was the basis of the
principle of non-encroachment enunciated in the North Sea Continental
Shelf cases (I.C.J. Reports 1969, p.36, para. 57; p.53, para. 101(C)(1)). It

46
Ibid., pp. 61-62 (para. 75).
47 Ibid., pp. 61-62 (para. 75).

28 is clear that in this case a true perspective on the relationship of the
opposing claims and the opposing entitlements is to be gained by

considering both the area of overlap48ng claims and the area of
overlapping potential entitlement.”

The Court used the terminology of “overlapping entitlements”, rather than the language

of overlapping “submarine extensions” of the coast as in Tunisia/Libya. But these are

two ways of expressing the same idea.

(1) The Relevant Coasts

3.30. Despite Ukraine’s complain about “double counting”, the entirety of the

Romanian coast should be treated as relevant for the purposes of the present delimitation.

By contrast, given the geographical configuration of the area not all of the various

segments of the Ukrainian coast are relevant for present purposes.

(a) The relevant Ukrainian coasts

3.31. In its Memorial, Romania carefully identified those portions of the Ukrainian

coast which are to be regarded as legally relevant for the purposes of the present

delimitation and why. By contrast, Ukraine gives a detailed description of its entire coast

which borders the western basin of the Black Sea, including areas which face each other
49
and can have no possible effect on delimitation vis-à-vis Romania.

3.32. Evidently a segment of Ukrainian coast adjacent to Romanian coast, north of the

land-river boundary between the two States constitutes a relevant coast for the

delimitation, being relevant for the construction of the equidistance line in Sector 1.

Similarly a segment of the Ukrainian opposite coast located on the Crimean Peninsula is

relevant for the delimitation, being relevant for the construction of the median line in

Sector 2.

48
See UCM, para. 3.25, Maritime Delimitation in the Area between Greenland and Jan Mayen
49enmark v. Norway), ICJ Reports 1993 p. 38, at p. 64 (para. 59).
See UCM, paras. 3.15-3.28.

293.33. The portion of coast whose relevance for the delimitation is disputed is that
portion lying to the north of a line drawn between two points on Ukraine’s coast. These

are, on the one hand, the point where the southern bank of the Nistru/Dniester Firth meets

the coast, on the segment of Ukraine’s coast north of, and adjacent to, Romania’s coast

(Point S) and on the other hand, Cape Tarkhankut, lying at the tip of a promontory,

forming part of the Crimean peninsula. The former is the point at which the Ukrainian

coast changes direction and loses its relationship of adjacency to the Romanian coast.

Similarly, Cape Tarkhankut is the most northerly point on the Ukrainian coast which is in

a relationship of oppositeness with the Romanian coast: it lies slightly north of due east
of the terminus of the Ukrainian/Romanian land-river boundary which is the most

northerly point on the Romanian coast.

3.34. Ukraine argues that the entirety of its coast fronting onto the western basin of the

Black Sea is relevant to the delimitation. It argues that its mainland coast should be

regarded as being divided into three segments, all three of which are relevant for the

delimitation. It attempts to show that due to their “general direction”, all three segments

of its coast “project” into what it alleges is the “relevant area”.

3.35. Ukraine’s approach is seriously flawed. The segment of coast in question cannot

be relevant since it does not generate areas of overlapping entitlement. In addition,

Ukraine adopts a dubious methodology in order to construct the supposed “projections”

of each of the three segments, constructing straight lines which purport to depict the

overall direction of each of the three segments of the coast, and then not even adopting

directions of projection which are perpendicular to those lines. Finally, various

arguments are made by Ukraine on the basis of the configuration of the coasts of the

parties in previous decisions of the Court. These issues will be addressed in turn. First,
however, it is necessary to make certain brief comments on the discussion of Serpents’

Island in Chapter 3 of the Counter-Memorial.

30 (i) The irrelevance of Serpents’ Island

3.36. Most of the discussion of Ukraine’s coasts in Chapter 3 (“The Relevant Coasts of

the Parties”) is devoted to a description of the three sectors put forward by Ukraine, their

lengths (measured in varying ways) and their alleged projections. By contrast Section 2,
50
the sub-section on Serpents’ Island, is of an entirely different character. It is entitled

“The Geological Characteristics of Serpents’ Island” and it purports to describe “the
51
physical and geographical characteristics” of the formation. In fact it addresses quite

different matters.

3.37. The omission of any discussion of the relevance of the coast of Serpents’ Island to

the delimitation is striking in light of the position adopted by Ukraine elsewhere in the

Counter-Memorial that Serpents’ Island generates a full-range of maritime zones. In the

light of that position, one would have expected the frontage of Serpents’ Island to have

been included among the relevant coasts. The omission is all the more striking since in

Ukraine’s view Serpents’ Island constitutes by far the most important Ukrainian

basepoint, governing the provisional equidistance line for most of its length.

3.38. As demonstrated in Chapter 4, the Parties have previously agreed the course of

the delimitation in the region of Serpents’ Island, and in particular have agreed that

Serpents’ Island is to be limited to no more than a 12 nm maritime zone. Furthermore, as

demonstrated again in Chapter 5, Serpents’ Island constitutes a “rock” within the

meaning of Article 121(3) of UNCLOS and generates no entitlement to a continental

shelf or an exclusive economic zone. But in any case Serpents’ Island does not form part

of the coastal configuration of the Parties; it constitutes merely a small maritime feature
52
situated at a considerable distance out to sea from the coasts of the Parties.

50 See UCM, paras. 3.15-3.47.
51 See UCM, para. 3.30.
52
See below, para. 8.18; see also UCM Annex 11, according to which Serpents’ Island is “an island
situated lonely amidst the immense water space”, or UCM Annex 48 which refers to Serpents’ island as
situated “in the middle of the sea”.

31 (ii) The Ukrainian coast north of Point S and Cape Tarkhankut bears
no relationship to the Romanian coast and is not the source of overlapping

entitlements

3.39. While certain segments of the Ukrainian coast find themselves in a situation of

adjacency with the Romanian coast, and another is situated opposite to it, the segments

situated to the north of the line joining Point S and Cape Tarkhankut on the coast of the

Crimean Peninsula have a relationship of neither adjacency nor oppositeness with the

Romanian coast. Any entitlement generated by this stretch of coast is eclipsed by that

generated by the coast south of point S and east of Cape Tarkhankut. In the

circumstances it is irrelevant whether this stretch of coast is 20 or 200 miles long, and

whether it runs west-east or north-south.

3.40. An issue of delimitation can arise only in the case of coasts that are either adjacent

or opposite, since it is only the seaward projection of such coasts that can actually meet
53
and overlap. There is no casus omissus, no geographical situation where the

delimitation is to be effected between coasts that are neither adjacent nor opposite. As

the Court said in the Tunisia/Libya Continental Shelf case:

“it is not the whole coast of each Party which can be taken into account;

the submarine extension of any part of the coast of one Party which,
because of its geographic situation, cannot overlap with the extension of
the coast of the other, is to be excluded from further consideration by the
54
Court.”

(iii) Ukraine’s construction of the supposed coastal projections is
methodologically flawed

3.41. Ukraine contends that “the various Ukrainian coastal segments create three

distinct but converging coastal projections.” 55 Ukraine constructs its projections of the

three segments of coast with little respect for geography. A number of observations may

be made in this regard.

53
Delimitation of the Continental Shelf (United Kingdom of Great Britain and Northern Ireland and
54e French Republic), Decision of 30 June 1977, 54 ILR, p. 6, at pp. 118-9 (para. 237).
55 See RM, para. 9.18.
See UCM, para. 8.21, and see UCM, Figures 3.4 & 3.10.

323.42. First, Ukraine does not create projections of all the different sectors of its coast

(there are at least eight such segments, established in accordance with the general coastal
orientation). Instead, it attempts to use a broad-brush technique to determine the

“general” direction of its coast, which grossly oversimplifies the actual situation.

3.43. Second, the method used to select the division between the different segments of

coast and to construct the baselines is not explained and appears arbitrary. For instance,

the first, western, segment line (referred to by Ukraine as “Sector 1”) does not terminate

at the starting point of the second, central segment (“Sector 2”). Rather, the western

segment appears to terminate on the south facing coast to the north of Odessa, some

distance to the east of the starting point of the central segment. The Counter-Memorial

attempts to disguise this fact by referring to the terminus of the western segment as being
56
“a point located just north of the city of Odessa”; however, Figure 3-1 in the Counter-
Memorial shows otherwise.

3.44. It is also striking that Ukraine’s third (eastern) segment, which runs from the end

of the Karnkinitska Gulf to Cape Sarych, in fact has a sharp elbow in it, amounting to a

change of direction of almost 90°, with Cape Tarkhankut located at the elbow, the two

arms being formed by the southern coast of the Karkinitska Gulf and a line drawn

between Cape Tarkhankut and Cape Sarych across the Kalamitska Gulf, respectively.

Ukraine suggests that the projection of this segment of coast is based solely on the latter

of these arms, in effect ignoring the north-west-facing arms. Nevertheless it purports to

count the entire length of both arms of the segment as part of its relevant coast, despite
the fact that the northwest facing arm of the segment faces nothing other than the south

facing coast of the Karkinitska Gulf, and plays no role either in the projection of that

segment or in the delimitation itself.

3.45. Third, the graphical construction of these projections in the figures in the Counter-

Memorial is distorted. For instance, the projection of the western segment of coast

identified by Ukraine as Sector 1 is constructed at an angle of approximately 102°, while

56 See UCM, para. 3.17.

33the projection of the eastern segment of the coast of the Crimean peninsula between Cape
Sarych and Cape Tarkhankut (Sector 3) the angle used is of 93°. See Figure RR4 at

page 35 below.

3.46. Despite this, Ukraine does not succeed in “squeezing” large segments of its coast

so as to fit into these projections. Almost the entire coastline of the Karkinitska Gulf, as

well the segment between Odessa and Yahorlytska Gulf are omitted – showing that the

seaward extension of these segments of the shoreline cannot be seen as overlapping with

the projection of the Romanian coast. Nonetheless Ukraine counts the length of these
segments of coast as relevant for the purpose of its purported proportionality test. An

accurate representation of Ukraine’s coastal projections is presented in Figure RR5 at

page 36 below. Out of the eight Ukrainian coastal segments, only four project on the

relevant area of delimitation (segments 1, 2, 4 and 8); segment 6 projects only partially on

to the delimitation area (and partially on Ukrainian coast of Crimea), while segments 3, 5

and 7 project onto Ukrainian territories only. Even if they project upon the delimitation

area, segments 4 and (partially) 6 are in no relation (either of adjacency or of

oppositeness) with the Romanian coasts, and in fact are situated between segments that

do not project on the delimitation area (3 and 5, and 5 and 7 respectively). Consequently,
out of the eight Ukrainian coastal segments, only three (1, 2 and 8) are relevant for the

present delimitation.

34 Figure RR4
The projections of the Ukrainian coast,

as presented by UCM, Figure 3-4
The angles used are not 90°, thus making
the projections inaccurate

35 6
3

h
f
s
o s
5 c o
R j n
r r i
u e i
F a r
u U
c
e
h
T (iv) Ukraine’s arguments from the jurisprudence of the Court

3.47. The first dictum quoted by Ukraine in support of its position regarding the

determination of the relevant coasts of the Parties is a fragment of a passage from the

judgment of the Court in the Jan Mayen case. 57 However, the fragment cited refers not to

relevant coasts, but to “overlapping claims” and “overlapping entitlements”.

Nevertheless, elsewhere in its judgment the Court did discuss the approach to be taken to

the identification of the relevant coasts of Greenland and Jan Mayen. Despite the

different geographical situation in the Jan Mayen case these passages are helpful with

respect to the general approach to be adopted when identifying relevant coasts.

3.48. Jan Mayen involved delimitation between opposite coasts, with a difference in

lengths which the Court eventually found that amounted to a striking disproportion,

sufficient to justify an adjustment of the provisional equidistance line. The Court

identified the relevant coasts as follows:

“It is appropriate to treat as relevant the coasts between points E and F and

between points G and H on sketch-map No. 1, in view of their role in
generating the complete course of the median line provisionally drawn
which is under examination.” 58

It was, thus, not all of the east-facing coast of Greenland that was considered to form part

of the relevant coast by the Court. Points H and G constituted points at which the coast

of Greenland changed direction. As a consequence segments situated, respectively, to the

north of Point H and to the south of point G bore no relation to the coast of Jan Mayen.

Consequently they were not considered as being relevant. In addition, the maritime areas
situated in front of them were treated as not forming part of the relevant area of

delimitation.

3.49. The segments of Greenland’s coast considered by the Court not to be relevant

(north of Point H and south of Point G) may be compared with the segments of the

Crimean coast which are not relevant for the present delimitation (north of Cape

57
See UCM, para. 3.25, quoting Maritime Delimitation in the Area between Greenland and Jan
Mayen (Denmark v. Norway), ICJ Reports 1993 p. 38, at p. 64 (para. 59).
58 Ibid., p. 68 (para. 67).

37Tarkhankut and east of Cape Sarych). The situation in the Jan Mayen case is presented in

Figure RR6 below.

3.50. Ukraine also refers to the Continental Shelf (Tunisia/Libya) case. In an attempt to

support its position that the whole of its coast in the north of the western basin of the

Black Sea is relevant, including those segments that do not bear any relationship of

adjacency or oppositeness in relation with the Romanian coast, it twice underlines that

the entire Tunisian coast facing into the Gulf of Gabes was treated as forming part of the

relevant coast.59

3.51. It is true that the Court considered that segment of the Tunisian coast to be

relevant. However, it reached that conclusion for a reason entirely different to that

suggested by Ukraine. The Court held that the entire coast of the Gulf of Gabes

maintains its relation of adjacency with the Libyan coast. As the Court observed:

“The change in direction may be said to modify the situation of lateral
adjacency of the two States, even though it clearly does not go so far as to
place them in a position of legally opposite States.” 60

59 See UCM, paras. 3.26, 4.21.
60 Continental Shelf Case (Tunisia/Libya), ICJ Reports 1982, p. 63 (para. 78).

38 Figure RR6
Relevant v. irrelevant coasts as
defined by the Court in the Jan
Mayen case

393.52. At the same time, the Court did not treat as relevant the whole Tunisian and

Libyan coasts. Thus the Court stated that

“[n]evertheless, for the purpose of shelf delimitation between the Parties,
it is not the whole of the coast of each Party which can be taken into

account; the submarine extension of any part of the coast of one Party
which, because of its geographic situation, cannot overlap with the
extension of the coast of the other, is to be excluded from further
consideration by the Court. It is clear from the map that there comes a

point of the coast of each of the two Parties beyond which the coast in
question no longer has a relationship with the coast of the other party
relevant for submarine delimitation. The sea-bed areas off the coast
beyond that point cannot therefore constitute an area of overlap of the

extensions of the territories of the two Parties, and are therefore not
relevant to delimitation. In the view of the Court, in the present context
that point on the Tunisian coast is Ras Kaboudia; on the Libyan coast is
61
Ras Tajoura.”

This coastal configuration is represented on Map No.1 of the Tunisia/Libya judgment.

For convenience, it is reproduced below, with the relevant and non-relevant coasts

indicated, as Figure RR7.

3.53. There are similarities between the geographical situation in the Tunisia/Libya case
and in the present case. First, there is substantial similarity between, on the one hand, the

two points identified by the Court (Ras Kaboudia and Ras Tajoura), and, on the other,

Point S on the Ukrainian coast, all of which are points marking changes in the direction

of the respective coasts “beyond which the coast in question no longer has a relationship

with the coast of the other party relevant for submarine delimitation”. Further there is

substantial similarity between the coastal segments of Tunisia and Libya situated beyond,

respectively, Ras Kaboudia and Ras Tajoura, which were considered by the Court not to

constitute relevant coasts for the purposes of the delimitation, and the Ukrainian coastal

segments situated to the north of Point S and Cape Tarkankhut, which in Romania’s view

are not relevant for the present delimitation.

61
Ibid., pp. 53-55 (para.75).

40 Figure RR7
Relevant v. irrelevant coasts as

defined by the Court in the
Tunisia/Libya case

413.54. If the criterion of coastal projection were to be used, the projection of a segment

on the Tunisian coast beyond Ras Kaboudia (the segment between the Gulf of

Hammamet and Cape Bon) would have overlapped with the projection of the coast of

Libya in the delimitation area. But as this segment had no actual relationship with the

Libyan coast, the Court excluded it from the area considered to be relevant for the

delimitation.

3.55. The third case invoked by Ukraine as supporting its inclusion of its whole coast is

the Gulf of Maine case. Ukraine emphasises that the Chamber rejected the distinction put

forward by the United States between “primary” and “secondary” coasts and found the

Canadian coast bordering the Bay of Fundy to form part of the relevant coast of Canada;

equally the waters within the bay were held to form part of the relevant area for the

purposes of the delimitation. 62

3.56. It must be noted that these two issues (the division between primary and

secondary coasts and the possible exclusion of the coasts of the Bay of Fundy) were not

linked in any coherent way. Indeed, the New Brunswick coast of the Bay of Fundy was

argued by the United States of America to constitute a primary coast. This criterion

proposed by the United States to justify its distinction between “primary” and

“secondary” coasts was not one linked to the relevance of the coast for the delimitation

but according to the measure in which the coasts bordering the delimitation area followed

the general direction of the mainland coast as a whole. 63 In other words, the United States

attempted to introduce a categorisation of relevant coasts, a categorisation rejected by the

Chamber. 64 But Romania does not seek to establish a hierarchy between the relevant

coasts of the Parties.

3.57. A second argument relied upon by Ukraine relates to the inclusion of the

Canadian coast facing onto the Bay of Fundy as part of the Canadian relevant coast,

62 See UCM, para 3.26.
63
Delimitation of the Maritime Boundary in the Area of the Gulf of Maine (Canada/United States of
64erica), ICJ Reports 1984, p. 246, at pp. 298, 318 (paras. 108, 170).
Ibid., p. 298 (para. 109).

42“although this bay was bounded exclusively by Canadian coasts”. 65 The general map of

the region of the Gulf of Maine was graphically represented on Map No. 1 of the

judgment of the Chamber, and is included below as Figure RR8 (see page 44 below).

The Chamber analysed the relevant coasts of the parties, comparing their lengths in order

66
to identify any possible disproportion caused by its provisional delimitation. The

relevant United States coast was found to start at the “elbow” of Cape Cod and to run to

the terminus of the American/Canadian land border, thereby excluding the coast situated
67 68
westwards of Cape Cod. Contrary to Ukraine’s assertion, the Canadian relevant coast

was found to include only a part of the coasts bordering the Bay of Fundy, as well as the

Nova Scotia coast up to Cape Sable, thereby excluding the coast to the east of Cape

Sable:

“The overall length of the Canadian coastline, as similarly calculated
along the coastal fronts from the terminal point of the international
boundary to the point on the New Brunswick coast off which there cease

to be any waters in the bay more distant than 12 miles from a low-water
line (45°16'31''N and 65°41' 01''W), then from that point across to the
corresponding point on the Nova Scotian coast (44°53' 49''N and

65°22'47''W), thence to Brier Island, and from there to Cape Sable, is
approximately 206 nautical miles.” 69

The points on the coast of the Bay of Fundy beyond which the coast was considered not

to be relevant for the delimitation were not graphically illustrated in the Chamber’s

judgment; however, they are illustrated by Romania on Figure RR8 below.

65 See UCM, para. 3.69, fn. 42.
66 Delimitation of the Maritime Boundary in the Area of the Gulf of Maine (Canada/United States of

67erica), ICJ Reports 1984, p. 246, at pp. 334-5 (para. 218).
68 Ibid., pp. 335-6 (para. 221).
See UCM, para. 3.26.
69 Delimitation of the Maritime Boundary in the Area of the Gulf of Maine (Canada/United States),
ICJ Reports 1984, p. 246, at pp. 335-6 (para. 221).

43 Figure RR8
Relevant v. irrelevant coasts as
defined by the Court in the Gulf
of Maine case

443.58. Apart from underlining that the Bay of Fundy was a part of the Gulf of Maine and

that the fact that its coasts were both Canadian was not a reason to disregard them, the

Chamber did not explain why segments of the coasts of the Bay of Fundy were included

in the relevant coasts. Nor did it elaborate on why those segments situated beyond the
two specified points were excluded, apart from stating that beyond those points there

were no waters more than 12 nm distant from the low-water line. Nevertheless, the

geographical situation of the area is striking. The two coasts of the Bay of Fundy clearly

lie in relations of adjacency and oppositeness with the American coast: the New

Brunswick segment lies in a relationship of adjacency to the American coast immediately

to the west of the terrestrial boundary (i.e. the coast of the State of Maine), while the

Nova Scotia segment lies opposite the same segment of American coast. Failure to take

account of the two segments would have amounted to an impermissible refashioning of

the geography of the area.

3.59. The situation in the present case is quite different. The exclusion of the coastal

segments situated north of Point S and Cape Tarkhankut as relevant coasts for the

purposes of the delimitation does not affect the coastal relationship between the two

Parties for the simple reason that those portions of the Ukrainian coast do not bear any

relation, whether of adjacency or oppositeness, to the Romanian coast.

3.60. In addition, it is helpful to note some other cases where the issue of “relevant

coasts” has been given consideration by the Court. In the Libya/Malta case, in the

context of assessing the disparity between the lengths of the coasts of the two parties, the

Court again determined first which were the relevant portions of the coasts of the parties

to be compared; in the Courts words, “[s]uch a test [of proportionality] would be

meaningless in the absence of a precise definition of the ‘relevant coasts’ and the
70
‘relevant area’, of the kind which the Court carried out in the Tunisia/Libya case.” The

Court went on to determine that:

70 Continental Shelf (Libyan Arab Jamahiriya/Malta), ICJ Reports 1985, p. 13, at pp. 49-50
(para.67).

45 “[o]n the Libyan side, Ras Ajdir, the terminus of the frontier with Tunisia,

must clearly be the starting point; the meridian 15°10' E which had been
found by the Court to define the limits of the area in which the Judgment
can operate crosses the coast of Libya not far from Ras Zarruq…” 71

The Court thus found that the Libyan relevant coast consisted of the segment between

Ras Ajdir and Ras Zarruq, excluding any segment situated beyond the latter point. In the

case of Malta, the Court treated as relevant the coast “from Ras il-Wardija to Delimara

Point, following straight baselines but excluding the islet of Filfla”. 72

3.61. Similarly, the Court considered the question of the “relevant coasts” in the

Cameroon-Nigeria case. The two parties held sharply conflicting views on the matter.

Thus, Nigeria took the position that:

“the coasts to be taken into account in the construction of a maritime
delimitation line must be ‘adjacent’ or ‘opposite’. Moreover they must be

coasts of the parties, and not those of a third State. In this regard Nigeria
considers that the relevant coast of Nigeria is that running west from its

boundary with Cameroon as far as the Akasso (where it changes direction
north-westwards, turning its back to the Gulf of Guinea), and that of
Cameroon is the coast running east from the boundary between the two

States and then south, as far as the Debundsha Point, which marks the
beginning of the blocking effect of Bioko Island.” 73

Cameroon took a different view, wishing to include within the relevant coasts further
74
coastal segments bordering the Gulf of Guinea.

3.62. The Court agreed with the Nigerian position, holding that the part of the

Cameroon coastline beyond Debundsha Point was not relevant: “it cannot therefore be

treated as facing Nigeria so as to be relevant to the maritime delimitation between
75
Cameroon and Nigeria”. That was sufficient for the purposes of that case:

“in the present case, whichever coastline of Nigeria is regarded as
relevant, the relevant coastline of Cameroon… is not longer than that of

71
Ibid., p. 50 (para.68).
72 Ibid., para. 68.
73 Land and Maritime boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial
Guinea intervening), ICJ Reports 2002, p. 303, at p. 435 (para. 278).
74 Ibid., p. 433 (para. 272).
75
Ibid., p. 443 (para. 291).

46 Nigeria. There is therefore no reason to shift the equidistance line in
76
favour of Cameroon on this ground.”

Evidently the Court considered the relevant coasts of the two parties to be roughly

comparable in length, or at least not wholly out of proportion. In the same way that only

a portion of Cameroon’s coast was considered to be relevant by the Court, so, it may be

inferred, only a portion of Nigeria’s entire coast was considered as relevant.

3.63. In Eritrea/Yemen (Maritime Delimitation), the Arbitral Tribunal also considered

the relevant coasts of the parties, who “disagreed strongly… over the respective lengths
77
of their coasts for the purposes of the calculation.” Even though both parties had
78
supported a different conclusion, the Tribunal gave preference to the criterion of the

actual relation (of oppositeness) between the two coasts; thus, in relation to the northern

extent of the Eritrean coast, it expressed doubt as to:

“the appropriateness of employing a horizontal line of latitude to divide,
for the purposes of the proportionality test, waters of the Red Sea which
lie at an angle of roughly 45°. The Tribunal has therefore considered the

relevant proportion of the Eritrean coast, which can be said to be
‘opposite’ that of Yemen as ceasing where the general direction of that
coast meets a line drawn from what seems to be the northern terminus of

the Yemen land frontier at right angles with the general direction of the
Yemen coast. In the same way the Tribunal determined the southern end

point to79e considered for the computation of the length of the Yemen
coast.”

The coastal segments situated beyond the limits thus defined were excluded from the

80
relevant coasts of the parties for the purposes of the proportionality test.

(b) The relevant Romanian coast

3.64. In spite of certain conflicting comments to be found in the Ukrainian Counter-

Memorial, Ukraine does not question that the whole length of the Romanian coast is

76
Ibid., pp. 446-7 (para. 301).
77 Government of the State of Eritrea and Government of the Republic of Yemen (Phase Two:
Maritime Delimitation), Award of 17 December 1999, 119 ILR, p. 417, at p. 466 (para.166).
78 Ibid., p. 466 (para. 167).
79 Ibid., p. 466 (para. 167).
80
For the Arbitral Tribunal’s application of the proportionality test, see ibid., p. 466 (para. 168).

47relevant for the purpose of the present delimitation. 81Nonetheless certain paragraphs

from Ukraine’s Counter-Memorial seem inconsistent with this position, or at least cast

doubt on the selection of the Romanian basepoints on this coast. 82

(i) Relevance of the southern sector to the delimitation

3.65. Ukraine seems to question the relevance of the Romanian coast situated to the

south of the Sacalin Peninsula, arguing that the segment does not play any role in the

construction of the Romanian delimitation line given that all basepoints used by Romania

for that purpose are situated at the northern end to this segment. 83 But the criterion for

determining the relevance of a specific coastal segment for delimitation is not whether it

contributes with basepoints from which a provisional equidistance line is measured.

Rather it is whether the prolongation of the coast generates a zone of entitlement to

84
extended maritime zones in the area relevant to the delimitation. The whole Romanian

coast faces the opposite Ukrainian coast of Crimea, generating an entitlement to maritime

areas that overlaps with the entitlement generated by that opposite coast. The fact that in

relation to Sector 2, where the coasts of the Parties lie in a relationship of oppositeness,

the points on the Romanian coast from which the provisional median line is constructed

are located on the Sulina Dyke and on the Sacalin Peninsula does not mean that the coast

south of the Peninsula is irrelevant. Otherwise the existence of minor coastal projections

and promontories could determine, in an arbitrary way, whole coastal lengths. The

Romanian coast and its two sectors are presented in Figure RR9 below.

81
See UCM, paras. 10.14, 10.21, Figures 3-7, 3-8, 10-2
82 See UCM, paras. 3.53, 3.57, 4.13, 4.26, 7.13, 10.12
83 See UCM, paras. 3.57, 4.5, 10.12.
84 See most recently Barbados v Republic of Trinidad and Tobago, Arbitral Award of 11 April 2006,
45 ILM, p. 800, at p. 837 (para. 224).

48 9
4

a
c
R n
R i
r a
g o
F R
h
T (ii) Permissibility of the Sulina Dyke and the Sacalin Peninsula as

basepoints

3.66. In its Counter-Memorial, Ukraine refers to the Romanian basepoints represented

by the outer end of the Sulina Dyke and by the south-eastern end of the Sacalin

Peninsula. It accepts their relevance, 85 but refers to them in a way that seems to question

the appropriateness of their use. 86

3.67. A graphical description of the two points is included in the Counter-Memorial

(Figure 4-1). This is an extract from a Romanian chart and presents accurately both the

Sulina Dyke and the Sacalin Peninsula.

3.68. The Sacalin Peninsula first appeared as a sand islet at the end of the 19 century:

it was referred to by Romanian geographers as “Sacalin Island” until relatively recently.

Due to natural modifications of the Romanian coast of the Danube Delta (mainly as a

result of repeated deposits of alluvia from the Sfântu Gheorghe arm of the Danube), the

island became connected to the mainland coast, and nowadays represents a long sandy

promontory extending from the delta, having broadly a north-east/south-west orientation.

As UCM Figure 4.1 clearly shows, the Sacalin Peninsula is fully integrated with the

Romanian coastline.

3.69. The Sulina Dyke is a permanent harbour work extending into the sea from the
87
Romanian port of Sulina. Its construction started shortly after the establishment of the

European Danube Commission in 1856, and it underwent major extension work from the

1950s until the 1980s. At present it extends into the sea approximately 7.5 km. By

contrast, in 1949 the dyke extended only some 4.5 km. A recent satellite image of the

dike is presented by Romania in Figure RR10 at page 52 of this Reply.

85 See UCM para.7.90 which defines the Ukrainian view on the relevant basepoints of the Parties.
86 See UCM, paras. 3.53, 4.13, 4.14.
87 The fact that the dyke is a permanent work of the Sulina harbour is also mentioned in the bilateral
Romanian/Soviet documents – e.g. the ProcŁs-Verbal no.10 of the Romanian-Soviet Commission for

demarcation of the State Border between the People’s Republic of Romania and the Union of the Soviet
Socialist Republics, Ismail, 5 December 1948 (Annex RR 2) refers to “the dyke of the port of SULINA”

503.70. In 1997, Romania deposited with the United Nations Secretary-General, in

accordance with Article 16 of UNCLOS, the list of baselines used for measuring the
88
breadth of its territorial sea. Included in that list were the coordinates of both the outer

limit of the Sulina Dyke and the south-eastern tip of the Sacalin Peninsula – then known

as the “Sacalin Island”. Ukraine did not raise any question vis-à-vis the Romanian Note

and did not protest. Nor did any other State.

3.71. The calculation of straight baselines and the choice of basepoints for those

baselines are regulated by Article 7 of the 1982 UNCLOS: Romanian choice is in full
compliance with this provision. Article 11 expressly allows for the use of permanent

harbour works as base points for the establishment of the breadth of territorial sea:

“For the purpose of delimiting the territorial sea, the outermost permanent
harbour works which form an integral part of the harbour system are

regarded as forming part of the coast. Off-shore installations and artificial
islands shall not be considered as permanent harbour works.”

As shown above, the Sulina Dyke is a permanent harbour work forming an integral part

of the harbour system and is neither an “off-shore installation” nor an “artificial island”.

88 Letter of the Romanian Permanent Mission to the United Nations to the Secretary General, dated
18 June 1997, annexing the geographical coordinates of the points between which Romania has drawn its
straight baselines (Annex RR 3).

51 2
5

6
0
k 1
y .
a 2
l e
0 u e
R e c
R t ,
r o t
g g a
F a e
i g
i o
e G
a c
S u
S3.72. This Court and international arbitral tribunals have seldom had the occasion to

rule on the use of permanent harbour works as basepoints. However, in the Dubai-

Sharjah delimitation, the tribunal referred to certain cases where permanent harbour

works had been used as basepoints for the delimitation of maritime boundaries, and stated

that:

“there is a body of practice, and of conventional law, in which full effect

has been given to harbour works in the construction of frontal maritime
boundaries as between opposing States. The same principles apply to the
construction of lateral maritime boundaries as between adjacent States as
to the construction of frontal maritime boundaries between opposing

States. The Court is satisfied that, in the light of the provisions of Article 8
of the 1958 Geneva Convention on the Territorial Sea and the Contiguous
Zone and of Article 11 of the 1980 Draft Convention on the Law of the

Sea the permanent harbour works of both Dubai and Sharjah must be
treated as a part of the coast for the purpose of drawing the base lines from
which the lateral sea boundary between them is constructed.” 89

3.73. On earlier occasions Ukraine has accepted the use of the point located on the

outer end of the Sulina Dyke as a basepoint for calculating the breadth of the Romanian
90
territorial sea. The 2003 Border Regime Treaty established the position of the final

point of the maritime boundary separating the territorial seas of Romania and Ukraine. In

Article 1, the 2003 Border Regime Treaty describes the course of the Romanian-

Ukrainian land, river and maritime frontier, specifying that, at sea, it continues…

“from the border sign 1439 (buoy) on the outer limit of Ukraine’s
territorial waters around the Serpents’ Island, up to the point of 45°021''

north latitude and 30°02' 27''east longitude, which is the meeting point
with the Romanian State border passing on the outer limit of its territorial
sea.”

This point (referred to by Romania as Point F) which marks the outer limit of the

Romanian and Ukrainian territorial seas lies 12 nm from the basepoint located on Sulina

Dyke, and 12 nm from Serpents’ Island. Its position was calculated by experts of the two

Parties on the basis of a basepoint located on the end of the Sulina Dyke. That point has

thus been accepted by Ukraine in the recent past as a relevant basepoint for calculating

the breadth of the Romanian territorial sea.

89 Boundary Dispute between Dubai and Sharjah (Emirate of Dubai v. Emirate of Sharjah), Award

90 19 October 1981, 91 ILR, p. 543, at pp. 662-3.
See RM, paras. 4.36-4.40.
53 (c) Conclusions on relevant coasts

3.74. In conclusion, the criterion for determining the relevance of any given coast is the

actual relation of adjacency or oppositeness between the coasts of the parties, as well as

the ability of those coasts to generate overlapping entitlements. Applying this conclusion

to the present case, the relevant coasts of the Parties are as follows:

- As regards Romania, the whole Romanian coast, of which the coastal

segment situated between the last point of the Romanian-Ukrainian land/river

border and the outer extremity of the Sacalin Peninsula is relevant for both sectors

of the delimitation area (characterised respectively by situations of coastal

adjacency and coastal oppositeness), while the segment situated south of the

Sacalin peninsula to the last point of the Romanian/Bulgarian land border is

relevant only for Sector 2 of the delimitation area (characterised by a coastal
situation of oppositeness);

- As regards Ukraine, the Ukrainian coast between the last point of the

Romanian-Ukrainian land/river border and Point S, as relevant for Sector 1 of the

delimitation, and the western-facing coast of the Crimean Peninsula between Cape

Tarkhankut and Cape Sarych as relevant for Sector 2.

(2) The Relevant Area

(a) The differences between the Parties as to the relevant area

3.75. A comparison of the relevant area defined by Romania in its Memorial 91 with the

relevant area put forward by Ukraine in its Counter-Memorial reveals that the two

approaches differ in three respects (not two as suggested by Ukraine): 92

- in the north, Ukraine’s position as to the relevant area includes the

maritime spaces situated to the north of the line uniting the extreme northern

points of the relevant coasts (i.e. the line between Point S and Cape Tarkhankut);

- in the south-west, Ukraine includes in the relevant area a slim triangle in

the region off the Romanian and Bulgarian coasts;

91
92 See RM, paras. 9.26-9.29.
See UCM, paras. 3.66-3.71.
54 - in the south-east, Ukraine excludes from the relevant area a triangular area

situated off the coast of Crimea, south of Cape Sarych.

These differences are illustrated on Figure RR11 at page 57 below, which follows Figure

3-12B of Ukraine’s Counter-Memorial.

(i) The area north of the line joining Point S and Cape

Tarkhankut

3.76. As far as the maritime spaces situated north of the line uniting the points

constituting the northern extremities of the relevant coasts are concerned (i.e. the line
93
between Point S and Cape Tarkhankut), as previously explained that area does not

represent an area subject to overlapping Romanian maritime entitlements or claims.

3.77. As a result of the configuration of the coasts of Romania and Ukraine, Romania

has no entitlement to maritime zones in that zone, nor has it ever made such a claim. In

conclusion that area should not be included in the relevant area of overlapping

entitlements.

(ii) The sliver of maritime area off the Romanian and
Bulgarian coast

3.78. The second difference between the approaches of the Parties relates to the

inclusion by Ukraine in its relevant area of a triangular sliver of maritime area situated off

the Romanian and Bulgarian coasts. Even though the south-western limit of Ukraine’s

relevant area is defined by Ukraine as “a line drawn roughly perpendicular from the

mainland coast from the point where the Bulgarian/Romanian land boundary reaches the

Black Sea until a point between the Romanian and Ukrainian coasts where the interests of

third States potentially come into play”, 94 Ukraine does not explain on what basis it

maintains this limit. Nor does it refer to the resulting difference between its approach and

that put forward by Romania, despite the fact that the difference is shaded green on

Figure 3-12B of the Counter-Memorial.

93
94 See paras. 3.39-3.40 above.
See UCM, para. 3.62.
553.79. The south-western limit of the relevant area as proposed by Romania is

represented by the equidistance line between the adjacent Romanian and Bulgarian

coasts. Had a maritime boundary to delimit the exclusive economic zones and the

continental shelf of Romania and Bulgaria been agreed between Romania and Bulgaria,

that boundary would have been taken as representing the south-western limit of the

relevant area. 95 However, such a boundary has not yet been established. In the absence

of any agreement, in fixing the south-eastern limit of the relevant area for the present

delimitation, Romania has followed the jurisprudence of this Court and other arbitral

tribunals in utilising a line which does not prejudice the interests of the third State in

question. Whenever the potential interests of third States have been involved, this Court

and international arbitral tribunals have been cautious to avoid adversely affecting their

interests .

3.80. The south-eastern limit of the relevant area proposed by Ukraine is situated south

of the equidistance line between Romania and Bulgaria and prejudges the interests of

Bulgaria as to its potential entitlements and claims to maritime zones. For this reason it

cannot be accepted, and, consequently, this sliver cannot be considered as part of the

relevant area.

95 This is also why the agreed boundary resulting from the USSR-Turkish agreement should form the
southern boundary of the relevant delimitation area in the south-east.
96 See, for example, the cases concerning the Continental Shelf (Libyan Arab Jamahiriya/Malta), ICJ
Reports 1985, pp.24-26, paras. 20-23, and the Land and Maritime Boundary between Cameroon and

Nigeria (Cameroon v. Nigeria; Equatorial Guinea intervening), ICJ Reports 2002, p. 117, para.238

56 7
5

e
t
g
d
a
e o
s t
t i
1 a i
R P e
R h fd
r n a
g e r
F tw t
b a
e e
n e
r r
f
d
h
T (iii) The triangle of maritime area south of Cape Sarych

3.81. The third difference between the two approaches relates to the south-eastern

part of the relevant area. Romania considers that the limit of the relevant area in this

sector is represented by the median line between the opposite Romanian and Turkish

coasts up to the point which is equidistant from Turkey, Romania and Ukraine, and

thereafter follows the maritime boundary established by the USSR-Turkish Agreement
of 1978 until the junction of that boundary with a meridian passing through Cape

Sarych. Accordingly, in Romania’s view, in its south-eastern part the relevant

delimitation area is bounded by two almost perpendicular lines.

3.82. Ukraine for its part considers that the south-eastern part of the delimitation area

should be limited by a straight line connecting the “point between the Romanian and

Ukrainian coasts where the interests of third States potentially come into play” to Cape
Sarych. The area whose relevance is disputed by Ukraine has the shape of a near right-

angled triangle, with the Ukrainian-proposed limit of the delimitation area as its

hypotenuse and the limits advocated by Romania as its other two sides. See Figure

RR11 at page 57 above.

3.83. To justify excluding this triangular area, Ukraine explains that it has “nothing to

do” with Romania, concerning a maritime area lying between Ukraine and either
97
Turkey or Bulgaria. But the maritime areas within this triangle represent overlapping
entitlements of Romania and Ukraine (and of Turkey and in part of Bulgaria). 98

Accordingly it is a relevant area in the present case.

3.84. If the area in this triangle has “nothing to do” with Romania, constituting an

area of overlapping entitlement of Ukraine and Turkey and/or Bulgaria, it is difficult to

see how Ukraine can consistently argue that the south-western corner area forms part of

the relevant area. Following the logic of Ukraine’s argument, the south-west corner
would be an area of overlapping Romanian, Bulgarian and Turkish entitlement which

has nothing to do with Ukraine.

97 See UCM, para. 3.71.
98 Using the approach of this Court in the Jan Mayen case, ICJ Reports 1993 p. 38, most of the
southern part of the western basin of the Black Sea is in fact an area of overlapping entitlements of at
least two of the four littoral States (Turkey, Bulgaria, Romania and Ukraine).

58 (b) Romania’s position

3.85. In accordance with the approach of the Court, given that the relevant coasts

have already been identified, the identification of the relevant area for the delimitation

is relatively simple.The relevant area consists of those areas which involve
overlapping areas of entitlement of each of the Parties.

3.86. The relevant area is as follows:

- in the north, the area is bordered by a line uniting Point S (46°020'N

30°28'30'E) and Cape Tarkhankut (45°50'N, 32°2943'E);

- in the south, the area is bordered by the line equidistant between the adjacent
Romanian and Bulgarian coasts, the median line between the opposite Romanian

and Turkish coast and the delimitation line agreed upon by the USSR and Turkey,

to which Ukraine has succeeded;

- in the south-east the limit of the area is constituted by the meridian uniting the

southern extremity of the Crimean Peninsula (Cape Sarych) with the delimitation

boundary between Ukraine and Turkey;

- in the west and in the east the limits of the area are formed by the respective

coasts of Romania and Ukraine, which are the relevant coasts, as described above.

D. Conclusion

3.87. For these reasons, the relevant coasts and areas are as depicted in Figure RR12

at page 60 below.

59 6

a
r
a
i
2 a
R i
R l
r d
g n
F v
l
r
h
T CHAPTER 4

THE FIRST STAGE OF THE DELIMITATION
AND ITS LEGAL BASIS

A. THE POSITIONS OF THE PARTIES

4.1. In its Memorial, Romania discussed the delimitation already agreed upon with the
99
Soviet Union in the initial portion of the first sector. This was effected by a series of
agreements, commencing with the 1949 ProcŁs-Verbaux (the General 1949 ProcŁs-Verbal

and the Individual ProcŁs-Verbaux of Description of Border Signs 1438 and 1439). The

boundary thereby established was substantially confirmed by the 1954 Act and the 1963 and

1974 General and Individual ProcŁs-Verbaux. These agreements established the course of
the initial portion of the maritime boundary around Serpents’ Island, leaving a 12 nm band of

maritime jurisdiction to it. They are binding on Ukraine as a successor State of the Soviet

Union.

4.2. The course of the agreed maritime boundary extends from the last point of the

land/river border (Point 1437) passing along a straight line to Point 1438, and from there

passing along straight line on azimuth 102°30" until it reaches the point on that line located

12 nm from Serpents’ Island (Point 1439). Thence the boundary continues around the arc of
circle based on Serpents’ Island having a radius of 12 nm, to a point situated to the east of

Serpents’ Island. The terminal point of the boundary was referred to by Romania in its

Memorial as “Point X”.

4.3. Confirmation as to the understanding of the Parties as to the course and extent of the

boundary is to be found in concordant Soviet, Romanian and subsequently Ukrainian

navigational charts, issued by the relevant authorities over a considerable period of time, as
well as in navigational charts produced by third States.

4.4. UNCLOS Articles 74(4) and 83(4) provide that “where there is an agreement in force

between the States concerned, questions relating to the delimitation of [the exclusive
economic zone/the continental shelf] shall be determined in accordance with the provisions

99
See RM, Chapter 11.

61of that agreement”. Accordingly Romania asks the Court to confirm the initial segment of
the boundary between the respective maritime areas of the Parties, as clearly established by

the succession of agreements between Romania and the Soviet Union which are binding on

both Parties.100 This boundary, lying between Point F and Point X, falls entirely within

Sector 1 of the delimitation, i.e. that portion of the delimitation characterised by the situation

of adjacency of the mainland coasts of the two Parties.

4.5. In its Counter-Memorial, Ukraine accepts the validity of the 1949 ProcŁs-Verbaux,

the 1954 Act and the 1963 and 1974 ProcŁs-Verbaux. It does not deny that each of these

instruments is binding upon it as successor to the Soviet Union. But it argues that the
boundary agreed in 1949 only extended as far as the final point shown on Map 134, which

was annexed to the 1949 General ProcŁs-Verbal. In this regard, Ukraine makes much of the

fact that this point approximately corresponds with Point F, the current point of intersection

of the outer limits of the 12 nm territorial seas of the Parties as agreed in the 2003 Border

Regime Treaty.

4.6. As will be shown in this chapter, Ukraine’s reasoning is based on a fundamental flaw.

The approximate correspondence between the final point shown on Map 134 and Point F is

coincidental and has no significance for the present dispute. Moreover, Romania has

managed to locate two further schematic maps, which were part of the bundle of maps
agreed in 1949 (of which Map 134 itself was part too) and annexed to the 1949 General

ProcŁs-Verbal. Both show the boundary around Serpents’ Island continuing further than

Point F, and clearly refute Ukraine’s argument based on Map 134.

4.7. There is no dispute between the Parties as to the course of the boundary line dividing

their respective territorial seas up to Point F, as agreed in the 2003 Border Regime Treaty,

and indeed that issue has not been referred to the Court. Similarly, the Parties agree that

Point F marks the intersection of the outer limits of their respective territorial seas. But

although the Court is not asked to determine these issues, it can and should take them into

account in its reasoning and draw the appropriate conclusions.

100 See above paras. 1.14-1.15, 2.7-2.10 for the Court’s jurisdiction in this regard.

624.8. In this Chapter, Romania will address the comments and assertions contained in

Ukraine’s Counter-Memorial in relation to the effect of the 1949 and subsequent agreements.

In the process, Romania will further explain the course of the initial portion of the

delimitation up to Point X, which the Court is asked to confirm. From Point X, there is no
agreed boundary; that point constitutes the starting point for the remainder of the

delimitation.

B. THE MARITIME BOUNDARY AROUND SERPENTS’ ISLAND AS

ESTABLISHED BY AGREEMENT

(1) Ukrainian Claims as to the Extent and Significance of the Boundary

agreed in 1949

4.9. Ukraine argues that the agreements concluded between 1949 and 1974 only resulted

in an agreed boundary up to a particular point, located on the 12 nm arc around Serpents’

Island, depicted on one of the maps attached to the 1949 General ProcŁs-Verbal (Map 134).

It also argues that they do not constitute delimitation agreements in the sense of Articles
74(4) and 83(4).101

4.10. The following assertions are made by Ukraine in support of its position:

a. The boundary established by the 1949 ProcŁs-Verbaux had the character of an

international boundary only up to the point situated at a 6 nm distance from the coast,

referred to by Romania as Point A; the second segment of the line, situated beyond

Point A, represented only the outer limit of the Soviet territorial sea, separating
102
Soviet territorial waters from the “high seas”. It would have been impossible for

Romania and the USSR in 1949 to have delimited anything other than their territorial

seas, since the Romanian legislation in force at that time did not purport to regulate
103
any other maritime zone.

b. The sketches of the maritime boundary included in the various editions of the

Individual ProcŁs-Verbal of Border Sign 1439 were not expressly incorporated into
104
these agreements, are not drafted to scale and should be disregarded.

101 For Romania’s comments on this assertion, see above, paras. 2.4-2.6.
102 See UCM, paras. 5.33(i)-(ii), 5.59-5.67.
103 See UCM, para. 5.83.
104 See UCM, paras. 5.54, 5.55, 5.97.

63 c. The only reliable representation of the extent of the boundary established in

1949 is Map 134 annexed to the 1949 General ProcŁs-Verbal; it depicts the boundary
105
as extending only up to Point F.

d. After Romania extended its territorial sea to 12 nm in 1951, the whole of the

boundary which Ukraine alleges was agreed in 1949 (i.e., the boundary up to the last

point shown on Map 134) became relevant as the boundary delimiting the territorial

seas of Romania and the USSR, and this boundary was subsequently confirmed by
106
the 1954 Act, as well as by the 1963 and 1974 ProcŁs-Verbaux.

e. The terminology used by the 1949 ProcŁs-Verbaux to indicate the Soviet

waters around Serpents’ Island as a “Soviet marine boundary zone” does not indicate

that the maritime boundary was an all-purpose boundary; rather it was used by the

drafters of the ProcŁs-Verbaux as a consequence of Romania’s “inhibition” to accept

any reference to the “Soviet territorial sea”, given that at the time Romania only

claimed a territorial sea with a breadth of 6 nm. 107

f. Nothing in the 1949 ProcŁs-Verbaux, or in the sketches included therein, or

the maps attached thereto, indicates that the agreed boundary continues on the 12 nm

arc surrounding Serpents’ Island beyond the final point shown in Map 134, dividing

areas not subject to the sovereignty of the Parties. 108

g. The conclusion that in 1949 Romania and the Soviet Union did not delimit

anything other than the boundary of the territorial sea between them (to a point 6 nm

from the Romanian coast), and thereafter only agreed the boundary between the
Soviet territorial sea and the high seas, is confirmed by the fact that the 1956

Romanian legislation on Romania’s territorial waters refers to a “delimitation by

agreement with the USSR” only in relation to territorial waters, and that subsequent

legislation on Romania’s exclusive economic zone enacted in 1986 did not refer to

any delimitation line agreed with the USSR. It is also argued that the chart annexed

by Romania to its 1997 notification to the Secretary-General of the United Nations of

the straight baselines used to measure the breadth of its territorial sea did not indicate

105
106 See UCM, paras. 5.55, 5.132-5133.
See UCM, paras. 5.81, 5.82, 5.92, 5.94, 5.95.
107 See UCM, paras. 5.46, 5.92, 5.102.
108 See UCM, paras. 5.69, 5.75, 5.102.

64 any boundary delimiting the continental shelf or the exclusive economic zone of

Romania with areas appertaining to Ukraine. 109

h. In the bilateral negotiations Romania did not argue that the first segment of

the boundary of the continental shelf and the exclusive economic zones between

Romania and Ukraine had already been delimited by the 1949 and subsequent
110
ProcŁs-Verbaux.

Romania will deal with each of these assertions in turn.

(a) The character of the boundary agreed in 1949

4.11. Ukraine accepts that the boundary agreed between Points 1437 and 1438 (located,

respectively in the mouth of the Musura/Musuna channel and to the south-east) “is a true

State boundary between the territorial sea and/or internal waters of Romania and the Soviet
111
Union”. Similarly, Ukraine does not deny that the portion of the boundary running

beyond Point 1438 out to the point located 6 nm from the relevant basepoint on the

Romanian coast, was also a “true State boundary” having a similar character, given that it
112
divided the respective territorial seas claimed by the Parties at that time.

4.12. But despite the clear depiction of the boundary, undifferentiated along its entire

length, on Map 134 annexed to the 1949 General ProcŁs-Verbal, and the sketches contained

in the Individual ProcŁs-Verbaux for Border Points 1438 and 1439, Ukraine denies that the

remainder of the boundary (i.e. between the point located 6 nm from the Romanian territorial

sea baseline to Point 1439 and from Point 1439 along the 12 nm arc around Serpents’

Island), has such a character. In its view it constitutes the boundary

“not between Romania’s and the Soviet Union’s sovereign territorial seas,
but rather the boundary of the Soviet Union’s sovereign territorial sea and
between it and adjacent high seas (the term ‘boundary’ being consistent
113
with both senses).”

109 See UCM, paras. 5.83, 5.123-5.126.
110
111 See UCM, para. 5.117
112 See UCM, para. 5.64.
113 See UCM, para. 5.65.
See UCM, para. 5.66 (emphasis in original).

654.13. Ukraine’s suggestion that there was a change in the character of the boundary
between Points 1438 and 1439 is inconsistent with the clear concordant terms of the various

ProcŁs-Verbaux concluded by Romania and the Soviet Union in 1949, as well as the

sketches included therein and Map 134 or other relevant maps 114annexed to the 1949

General ProcŁs-Verbal. In all depictions the boundary was unchanged in character along its

entire length. It reflected the agreement on a maritime boundary between Romania and the
Soviet Union extending on a 12 nm arc “surrounding Serpents’ Island”.

4.14. That this is the case results clearly from the language of the ProcŁs-Verbaux

concluded in 1949. All contain language in similar terms which makes clear that that the
entire length of the boundary between Points 1438 and 1439, as well as subsequently on the

12 nm arc around Serpents’ Island, has the same character of a boundary separating areas

appertaining to Romania on the one hand and the USSR on the other.

4.15. The description of the maritime boundary in the relevant area is couched in similar
terms in both the September 1949 General ProcŁs-Verbal and in the 1949 Individual ProcŁs-

Verbaux of Border Signs 1438 and 1439, all of which consistently refer to “the State

boundary line” in relation to the entirety of the boundary so described. The complete title of

the 1949 General ProcŁs-Verbal is “ProcŁs-Verbal of the Description of the State Border

Line between the People’s Republic of Romania and the Union of the Soviet Socialist
Republics, demarcated in 1948-1949”. The complete title of Map 134 annexed to the 1949

General ProcŁs-Verbal is “Map of the State Border between the Union of the Soviet Socialist

Republics and the People’s Republic of Romania from border signs no. 1438 to no. 1439”.

No reference is made anywhere to a boundary between the territorial sea of the USSR and

the high seas.

4.16. Equally the preamble to the 1949 Individual ProcŁs-Verbal of Border Sign 1439

reads, in part, as follows:

“the Soviet-Romanian Commission for the demarcation of the State Border
between the Union of the Soviet Socialist Republics and the People’s
th
Republic of Romania, hereby certifies that the 8 Soviet-Romanian Joint
Sub-Commission, demarcating the State border between the Union of the

114 See below, paras. 4.65-4.68.

66 Soviet Socialist Republics and the People’s Republic of Romania, settled…
the border sign no. 1439 (beacon).”

4.17. All various ProcŁs-Verbaux refer to a boundary between Romania and the USSR.
This language is inconsistent with the suggestion that what the Parties had in fact agreed was

a two-part boundary, first between the territorial seas of the Parties, then between the

territorial sea of the USSR and the high seas.

4.18. It bears noting that Border Sign 1439, situated approximately 9.6 nm from the

Romanian coast and therefore outside the 6 nm territorial sea claimed by Romania in 1949, is

expressly stated to be one of the markers demarcating the “State Border” between Romania

and the USSR. The language of the 1949 Individual ProcŁs-Verbal for Border Sign 1439 is
clear. It cannot be read as meaning that Border Sign 1439 was a marker demarcating a

turning point of the boundary between the USSR and the high seas, as Ukraine suggests.

4.19. Neither in the 1949 General ProcŁs-Verbal nor the 1949 Individual ProcŁs-Verbal of

Border Sign 1439 is there any suggestion that a change of regime was envisaged at any point

along the boundary so described, as suggested by Ukraine. On the contrary, the texts of all

the ProcŁs-Verbaux, as well as the title of Map 134, are consistent and clear: they describe

the land, river and maritime boundary between the two States. By the terms of these
agreements Border Sign 1439, although situated outside the territorial sea claimed by

Romania at that time, constitutes one of the signs demarcating this Romanian/Soviet

maritime boundary.

4.20. The sketches forming part of the Individual ProcŁs-Verbaux, as well as Map 134 and

the other maps annexed to the 1949 General ProcŁs-Verbal, all use a single form of marker

along the entire length of the boundary depicted. There is no change in the symbol used to
depict the boundary at the point representing the outer limit of Romania’s 6 nm territorial sea

as claimed in 1949.

4.21. That the maritime areas situated south of the boundary described by the ProcŁs-
Verbaux appertained to Romania and were not regarded as high seas is clearly visible from

the sketch included in the 1949 Individual ProcŁs-Verbal of Point 1439 and from Map 134.

Both feature, south of the boundary, the abbreviations “PHP”/“RPR” – as opposed to the

67texts located to the north of the boundary: “CCCP”/“URSS”. These abbreviations are used

on all sketches and maps included in or annexed to the 1949 ProcŁs-Verbaux to indicate the

appurtenance of the areas delimited by the boundary. The abbreviations “CCCP” and

“URSS” are the Russian and Romanian variants for “USSR”. The other two abbreviations
stand for the official acronym of Romania at that time, “People’s Republic of Romania”: in

Russian “PHP” (▯▯▯▯▯▯▯▯▯ ▯▯
▯▯▯▯ ▯▯ ▯
▯▯▯▯▯▯), in Romanian “RPR” (Republica

Popular▯ Român▯). The Romanian abbreviation for Romania (“RPR”) is located south of

the maritime boundary opposite the Romanian abbreviation for “USSR” located to north of
the boundary in the outer segment formed by the 12 nm arc around Serpents’ Island. This is

obviously inconsistent with the Ukraine thesis that the “boundary” was merely between the

Soviet territorial sea and the high seas: see Figures RR13, RR14 at pages 69, 70

respectively. If this Ukrainian thesis were correct, then the “RPR” abbreviation would be

placed on the high seas!

4.22. The conclusion to be drawn from these agreed sketches and maps is clear. The areas

south of the boundary were agreed, as between Romania and the USSR, to appertain to
Romania, even though they lay beyond the outer limit of the 6 nm territorial sea claimed by

Romania at that time.

4.23. In this context it is irrelevant that the continental shelf was not then firmly

established. No issue arises as to the interests of any third State or of the international

community in areas beyond national jurisdiction. Whatever the USSR and Romania agreed

in 1949 obtains as between Ukraine and Romania today. As to what they agreed the sketch

maps and agreements are concordant and unequivocal: a 12 nm maritime zone appertained to
Serpents’ Island; areas to the south appertained to Romania.

68 Figure RR13
Extract from the sketch included in the individual
ProcŁs-Verbal of Point 1439
The maritime areas situated south of the boundary are
clearly indicated as appertaining to Romania

(“PHP”, respectively “RPR”)

69 Figure RR14
Reproduction of Map 134
The maritime areas situated south of the boundary are
clearly indicated as appertaining to Romania
(“PHP”, respectively “RPR”)

704.24. Ukraine’s explanation that Romania and the USSR intended merely to indicate the outer

limit of the Soviet territorial sea is not credible. It would have been very unusual for a bilateral
agreement to have purported to determine the outer limit of the territorial sea of one of the

Parties from the high seas, and to have defined in detail the course of that outer limit, including

by specifying the exact geographical coordinates of one turning point used to define that outer

limit (the point represented by Border Sign 1439). It would have been even more unusual, if

this was in fact what Romania and the USSR intended to do, that this was not made explicit in

the various instruments.

4.25. There does not appear to be a single example of such an agreement in State practice;

none has been cited by Ukraine. On the contrary, State practice relating to maritime

delimitation offers a number of examples where States have established maritime boundaries

purporting to delimit maritime spaces appertaining to them even when the internal legislation

of one or both parties did not purport to regulate or lay claim to such areas. One example is the

1970 USA/Mexico Boundary Treaty, Article V(A) and (B) of which established the
international maritime boundary between the two States in the Gulf of Mexico and the Pacific

Ocean extending “to a distance of 12 nautical miles from the baselines used for its delineation”,

in spite of the fact that at that time the United States of America only laid claim to a 3 mile
115
territorial sea while Mexico had already established a breadth of 12 nm.

4.26. Similarly, the 1978 USA-Mexico Treaty on Maritime Boundaries, which fixed the

boundary between 12 nm and 200 nm in both the Caribbean Sea and Pacific Ocean, simply

referred to the establishment of the “maritime boundaries” between the parties, with no

reference to the regime or regimes to which the area so divided was subject. 116

4.27. It is not surprising that States conclude such agreements even in the absence of any
legislation or proclamation claiming a particular maritime area. As to the territorial sea, a

treaty such as the 1970 USA/Mexico Treaty delimits maritime zones as they will be claimed in

the future, and not only as they exist at the time of its conclusion. As to the continental shelf,

Article 77(3) of the 1982 UNCLOS specifies that the rights of a coastal State over its

continental shelf are inherent and do not depend upon “occupation, effective or notional, or on

115 Treaty to Resolve Pending Boundary Differences and Maintain the Rio Grande and Colorado River as the

International Boundary Between the United States of America and the United Mexican States, Mexico City, 23
116ember 1970, 830 UNTS 55.
Treaty on Maritime Boundaries between the United Mexican States and the United States of America
(Caribbean Sea and Pacific Ocean), Mexico City, 4 May 1978, 2143 UNTS 405.
71 117
any express proclamation”. The Ukrainian argument that the USSR and Romania could not

have agreed a delimitation of anything other than the limit of the Soviet territorial sea given

that Romania did not claim anything other than a 6 nm territorial sea at the time is baseless.

4.28. In short the 1949 ProcŁs-Verbaux were intended to establish an all-purpose

international boundary between maritime areas appertaining to Romania and the USSR

respectively. In its first section, that boundary divided the respective territorial seas of the
Parties, up to the point on the line between Points 1438 and 1439 representing the intersection

of the outer limit of the Parties’ territorial seas at that time, located at a distance of 6 nm from

the Romanian coast. From that point, in its second section the boundary continued along the

line between Points 1438 and Point 1439, and from Point 1439 it followed the 12 nm arc of

circle around Serpents’ Island. In this second sector, the boundary was intended to divide areas
on the one hand appertaining to the USSR, and on the other appertaining to Romania, in

relation to which it was agreed that no Party would make any claim as against the other. The

practical effect was that Serpents’ Island was restricted to no more than a 12 nm band.

4.29. The course of the boundary as fixed by the 1949 ProcŁs-Verbaux was confirmed by
every later bilateral Romanian/Soviet agreement on the subject (the 1954 Act, the 1961 Treaty

on the Border Regime, and the 1963 and the 1974 ProcŁs-Verbaux).

4.30. The change of legal regime which resulted from Romania’s claim to a 12 nm territorial
sea in 1951 had no influence on the course of the boundary or its character as an all-purpose

maritime boundary between the two States. The fact that the legal regimes to which the areas

delimited by this boundary were subject were not initially the same on the two sides of the

boundary, and that the character of some of those areas changed over the years, in no way calls

into question the validity of the all-purpose boundary established in 1949. None of the later
agreements included any language to reflect the Romanian extension of its territorial sea.

Rather the boundary was consistently described as continuing on the 12 nm arc of circle around

Serpents’ Island leaving the island on the Soviet side, and the parties to those agreements

consistently confirmed that Serpents’ Island was to be restricted to a 12 nm band, while areas

to the south of that boundary were agreed to appertain to Romania.

117
Cf North Sea Continental Shelf Cases (Federal Republic of Germany/Denmark; Federal Republic of
Germany/Netherlands), ICJ Reports 1969, p. 3, at p. 22 (para. 19). Elsewhere Ukraine recognises the point:
UCM, para. 5.74.

724.31. The practice of establishing an all-purpose maritime boundary is well-known in

international practice. One of the arguments for establishing such boundaries is that “the all-

purpose formula allowed States to avoid any risk that boundaries might be reopened if new

forms of jurisdiction emerged” in a period when the law of the sea was in a state of rapid

evolution. 118 Agreements intended to establish all-purpose boundaries may do so in several

ways, one technique being to identify the boundaries using terms which are not associated with

any particular jurisdictional regime. Colson provides the following examples of such an

approach:

“the 1957 Norway-USSR Agreement in the Varangerfjord uses the expression

‘sea frontier’; the 1986 Colombia-Honduras agreement describes a ‘maritime
frontier’; and agreements between Abu Dhabi and Dubai and Qatar and Abu
Dhabi speak of the ‘offshore’ boundary. The encompassing nature of these
119
terms suggests the boundary’s use for all jurisdictional purposes.”

4.32. Most such agreements date from the 1950s or 1960s, when the law of the sea regarding

the various areas of maritime jurisdiction was not fully established; yet no-one questions their

validity or the suitability of their language to create a stable delimitation notwithstanding

changes in the extent or character of the maritime zones covered.

120
4.33. Of relevance in this regard is the Norway-USSR Agreement concluded in 1957, only

8 years after the conclusion of the 1949 General and Individual ProcŁs-Verbaux between
Romania and the USSR. The Agreement established the maritime boundary between Norway

and the USSR in the mouth of the Varangerfjord. At the time, Norway claimed a 4 nm

territorial sea and purported to draw a closing line across the Varangerfjord which acted as a

straight baseline, while the Soviet Union claimed a 12 nm territorial sea. The agreed boundary

has a length of 24.35 nm, composed of a segment dividing the territorial seas claimed by the

two parties at the time, measuring 12.6 nm long, followed by a further segment of boundary

11.75 nm long. 121 Nevertheless, much like the 1949 Romanian-Soviet ProcŁs-Verbal, the

Agreement itself does not refer in terms to the legal regimes of the areas delimited, merely

118
D. Colson, “The Legal Regime of Maritime Boundary Agreements”, in J.I. Charney and L.M. Alexander
(eds.), International Maritime Boundaries (Martinus Nijhoff Publishers, 1996) vol. I, p. 49, and see further ibid.,
pp. 44-50.
119 Ibid., p. 49 and references cited.
120 Agreement between the Royal Norwegian Government and the Government of the Union of the Soviet
Socialist Republics Concerning the Sea Frontier between Norway and the USSR in the Varangerfjord, Oslo, 15
February 1957, 312 UNTS 322; see also the Descriptive Protocol relating to the Sea Frontier between Norway and
the USSR in the Varangerfjord, demarcated in 1957, Moscow, 29 November 1957, 312 UNTS 326.
121 See J.I. Charney and L.M. Alexander (eds.), International Maritime Boundaries, vol. II, (Martinus
Nijhoff Publishers, 1998), pp. 1781-1782.

73referring to the agreed boundary as being the “sea frontier”. Article 1 of the Norway-USSR

Agreement provides that:

“The sea frontier between Norway and the Union of Soviet Socialist
Republics in the Varangerfjord shall follow a straight line from frontier mark

No. 415 (spar buoy), which is the terminal point of the frontier drawn in 1947,
to the intersection of the outer limits of Norwegian and Soviet territorial
waters.

Neither of the Contracting Parties shall extend its territorial waters beyond the
straight line extending from the intersection referred to in the first paragraph
of this article to the median point of the line between Cape Nemetsky and
Cape Kibergnes.”

4.34. The fact that at the time the notion of the continental shelf was not definitively settled
does not affect the validity of the Agreement or the boundary it establishes. The agreement on

a line joining the intersection of the outer limits of the Norwegian and Soviet territorial waters

and the median point joining Cape Nemetsky and Cape Kibergnes de facto delimited the
122
continental shelves of the two States. All authorities appear to be in agreement on this point.

Indeed the way the boundary is described is similar to the language used in the 1949 ProcŁs-

Verbaux. Certain terms in the original Russian texts are the same, even if the English
translations of the 1949 ProcŁs-Verbaux as set out above, and of the 1957 Norway-USSR

agreement in the United Nations Treaty Series, use different terms. Both use the same Russian

term “▯▯▯▯▯▯▯” – in the case of the 1949 ProcŁs-Verbal “▯▯ ▯
▯▯▯▯▯
▯▯▯▯ ▯▯▯▯▯▯▯”,

translated by Romania as “State boundary”, and in the case of the Norway-USSR agreement

“▯▯ ▯▯▯▯▯ ▯▯▯▯▯▯▯”, translated as “sea frontier” – to define the all-purpose boundary
established by the respective agreements.

4.35. In conclusion, the Ukrainian assertions regarding the nature of the maritime boundary

established by the 1949 ProcŁs-Verbaux are groundless. The provisions of the various 1949
ProcŁs-Verbaux are unequivocal: they establish a single all-purpose Romanian/Soviet maritime

boundary. South of that boundary, the maritime areas appertain to Romania, while to the north,

the maritime areas appertain to Ukraine, as they previously appertained to the Soviet Union.

The boundary runs from the last point of the land/river border (Point 1437) in a straight line to

border sign 1438 and along a further straight line between border signs 1438 and 1439, and

thereafter follows the arc of a circle having a radius of 12 nm drawn around Serpents’ Island.

122
See, e.g., Colson, p. 44, and the entry on the Continental Shelf Boundary: Norway-Soviet Union in Limits
in the Seas, Series A, no. 17 (1970), which assumes that the Agreement was a continental shelf boundary
delimitation and states that “[t]o complete the division of the continental shelf of the Varangerfjord, the two
Governments agreed to extend the CSB from the terminal point of the sea frontier between the two countries, to
the median point of the geographical closing line of the Varangerfjord, the Cape Kibergnes-Cape Nemetsky Line.”

74At different points along its length and at various points in time, it has divided areas

appertaining to the Parties which were subject to different legal regimes as the applicable law

and the areas claimed by the Parties have developed.

(b) The status and reliability of the sketches included in the individual ProcŁs-

Verbaux

4.36. Ukraine attempts to diminish the consistency and relevance of these sketches,
123
suggesting that they are “unreliable”. Three arguments are put forward. First, it notes that

“the ProcŁs-Verbaux do not appear to indicate the relationship between the ‘sketches’ and the
124
verbal descriptions, or which prevails in case of discrepancy”. Second, while forced to

admit at least that the sketches “go together” with the verbal descriptions of the boundary in the

ProcŁs-Verbaux, Ukraine suggests that the sketches “do not appear to have been formally made
125 126
integral parts of them”. Finally, the various sketches are said to be unreliable: this is

particularly the case with the sketch forming part of the Individual ProcŁs-Verbal for Point

1439, in relation to which Ukraine observes that it “is not to any consistent scale” and appears

to be “no more than a rough illustrative sketch drawn without reference to illustrations of
127
scale”.

4.37. Romania will first address the Ukrainian observation that the sketches do not appear to

have been made a formal part of each procŁs-verbal. As already indicated, both the 1949

Individual ProcŁs-Verbaux of Border Signs 1438 and 1439, as well as the 1974 Individual

ProcŁs-Verbal of Border Sign 1439 128 included, within the body of the instrument, sketches

illustrating the position of the border signs and the course of the boundary in the area to which

they related. Contrary to the Ukrainian suggestion that these sketches were merely
129
“accompanying” the ProcŁs-Verbaux, they were included in the body of each of the ProcŁs-

Verbaux and clearly formed an integral part of them. True, the reproductions of the sketches

from the ProcŁs-Verbaux included in Ukraine’s Counter-Memorial as Figures 5-7 and 5-8 have

been carefully cropped so that they do not reproduce any of the surrounding text. But an

123 UCM, para. 5.97.
124 UCM, para. 5.54.
125 UCM, para. 5.54.
126 Ukraine also makes the same objection in relation to the sketches in the 1974 Individual ProcŁs-Verbal
for Point 1439, see UCM, paras. 5.97.
127
128 UCM, para. 5.55.
As explained by Romania (see RM, para. 4.18), in 1963 no new individual ProcŁs-Verbal for this border
sign was concluded since no factual modification had occurred since the ProcŁs-Verbal of 1949. See likewise
UCM, para. 5.91.
129 See in relation to the 1949 Individual ProcŁs-Verbaux, UCM, para. 5.54; in relation to the 1974 ProcŁs-
Verbal, UCM, para. 5.97.

75examination of the ProcŁs-Verbaux themselves shows that the sketches are located

immediately after the detailed description of the position of the border sign, are introduced by

the words “The place of the border sign is shown on the sketch below” and are followed by the

usual final clauses establishing the number of the copies, the languages in which the ProcŁs-
Verbal was concluded etc. 130 These portions of the text of the ProcŁs-Verbaux are clearly

visible on the reproductions of the sketches included as Figures 5, 6 and 7 in the Memorial.

4.38. Given the inclusion of the sketches within the body of the ProcŁs-Verbaux, any

specification to the effect that they were “an integral part of them” would have been redundant.
The sketches constitute an integral part of the instrument by reason of their very location: they

are not annexures or addenda. To contest the validity of the sketches would amount to

contesting the validity of the instruments in which they are contained. But Ukraine does not
131
contest the validity of the 1949 and 1974 ProcŁs-Verbaux.

4.39. Turning to the objection that the ProcŁs-Verbaux do not specify the relationship of the

sketches to the text, it follows from the location of the sketches within the body of the ProcŁs-

Verbaux and their status as an integral part of these agreements that there was no need to

specify the relationship of the sketches to the description of the boundary in the text. When

interpreting the ProcŁs-Verbaux “in their context and in the light of their object and purpose”,

full effect must be given to the sketches as with the other parts of the instrument.

4.40. The sketches must be approached in the light of the purpose for which they were

intended and for which they were included in the ProcŁs-Verbaux. They were not intended to

be minutely accurate depictions of the geography of the area they depict but to illustrate the
position of the border sign and the course of boundary when read in conjunction with the

accompanying text.

4.41. Thus any interpretation of the verbal description of the course of the boundary that was

entirely inconsistent with the sketch would have to be rejected. An example of such an
inconsistent interpretation is that put forward by Ukraine, namely that the maritime boundary

agreed constituted an international maritime boundary separating the territorial seas of the

Parties along only part of its length, while along the rest of its length it merely delimited the

outer limit of the Soviet territorial sea vis-à-vis the rest of the world. That interpretation has no

130
131 See Annexes RM 4, 15, 22.
See UCM, para. 5.103.
76foundation in the text itself. It is also inconsistent with the depiction of the boundary in the

sketches, which show the boundary as having the same character along its entire length.

4.42. These considerations provide the answer to the third argument put forward by Ukraine,

based on the inaccuracy of the sketches insofar as their scale is concerned. Once it is

understood that their purpose was to provide an illustration of the border points and the

boundary described in words in the individual ProcŁs-Verbaux, the sketches are clear and
telling: they are entirely reliable for the purpose for which they were included.

4.43. The description of the Romanian/Soviet maritime boundary resulting from the text of

the ProcŁs-Verbaux corroborated by the sketches is unequivocal. From border sign 1439, the
boundary continues on the 12 nm arc of circle around Serpents’ Island, as a boundary between

the two States, and not between one State’s territorial sea and the high seas. The boundary is

intended to have an unchanged character along its entire length, as is shown from the lack of

any change in the way in which the boundary is described and depicted. It is true that the final

point of the boundary following the arc of circle around Serpents’ Island is not specified in any

of the ProcŁs-Verbaux and is not shown on any of the sketch maps. But the sketch forming
part of the 1949 Individual ProcŁs-Verbal of Border Sign 1439 (and that forming part of the

corresponding instrument concluded in 1974) is entirely consistent with the ProcŁs-Verbaux,

namely that the boundary continues around the arc of circle.

4.44. For these reasons, Ukraine’s suggestion that the sketches “must be disregarded as being

wholly unreliable” 132cannot be accepted. On the contrary, the sketches, as integral parts of the

ProcŁs-Verbaux, must be given the weight appropriate to the purpose for which they were

included in the individual ProcŁs-Verbaux.

(c) The argument that Map 134 depicts the full extent of the boundary

4.45. In light of the depiction of the boundary in the sketches contained in the 1949

individual ProcŁs-Verbaux, and in Map 134 annexed to the 1949 General ProcŁs-Verbal,

Ukraine cannot but accept that the agreed boundary extended beyond the junction of the outer

limit of the territorial seas of Romania and the Soviet Union, continuing along the line between
Points 1438 and 1439 and then going at least a certain distance around the arc of circle with a

radius of 12 nm from Serpents’ Island. But Ukraine takes the position that the only reliable

132
UCM, para. 5.97, referring to the sketch forming part of the 1974 Individual ProcŁs-Verbal for Border
Sign 1439. In relation to the 1949 sketches, see, e.g. UCM, para. 5.54.

77representation of the extent of the boundary beyond Point 1439 is that depicted on Map 134. 133

Ukraine places particular reliance on Map 134 on the basis that, unlike the sketches forming

part of the Individual ProcŁs-Verbaux, it is drawn to scale and accurately represents the

geographical features of the zone.

4.46. In considering this issue, three points are relevant. The first is Map 134 itself, and the

clear language of the General ProcŁs-Verbal, which is confirmed by its travaux. Against this

Ukraine makes two main arguments: it relies on the sliver of blank space to the east of the line

shown on Map 134, and on the apparent identity between the end of the line as shown on that

Map and points subsequently identified by the Parties. These points will be dealt with in turn;

finally, reference will be made to two other maps attached to the 1949 General ProcŁs-Verbal.

(i) The language of the 1949 General ProcŁs-Verbal and its travaux
prØparatoires

4.47. Map 134 shows the position of the border sign marking Point 1439 and the maritime
boundary agreed in 1949 in a way very similar to, and entirely consistent with, the

representation on the sketch included in the individual ProcŁs-Verbal, despite the fact that that

sketch was not drawn to scale. Thus (1) the boundary shown is marked using a single symbol

along its entire length; (2) the maritime areas situated south of the boundary depicted are

clearly indicated as appertaining to Romania (again using the abbreviations “RPR”/”PHP”) and

(3) those indications are situated in areas clearly lying further than 6 nm from the coast. Map

134 is thus fully consistent with and supports Romania’s position that the agreed boundary was
an all-purpose international maritime boundary, intended to delimit along the whole of its

length areas appertaining to the Soviet Union from those appertaining to Romania.

4.48. The text of the 1949 General ProcŁs-Verbal, to which Map 134 is annexed, provides

that from point 1439 the boundary continues on the 12 nm arc of circle around Serpents’

Island. The relevant passage provides:

“The border sign no. 1439 (beacon), is fixed in water, in the point where the
State border line going through the Black Sea changes its direction, at the
intersection of the straight line going from the border sign no. 1438 (buoy) on
the azimuth of 102°30,' 0, with the exterior margin of the Soviet marine
boundary zone, of 12 miles, surrounding Serpents’ Island.” 134

133
134 See UCM, paras. 5.55, 5.132-5.133.
RM, para. 4.5; the original is located at Annex RM 13 (emphasis added). Cf. the slightly different
translation put forward by Ukraine, which nevertheless conveys the same suggestion that the boundary continues
around the island: “The state boundary from state boundary mark No. 1439 (pole) passes along external line of a
78The 1949 Individual ProcŁs-Verbal of Border Sign 1439 is in substantially similar terms.

4.49. The Russian expression used in the 1949 General ProcŁs-Verbal and the 1949

Individual ProcŁs-Verbal relating to Border Sign No. 1439 to describe the line following the

arc of circle on which point 1439 lies and on which the boundary passes after point 1439 is

“▯▯▯▯ ▯
▯▯
▯▯ ▯▯▯▯▯▯▯”; the Romanian version reads “în jurul Insulei ▯erpilor”. Both

these expressions may be translated into English as “around Serpents’ Island” or “surrounding

Serpents’ Island”. To consider the point depicted on Map 134 as constituting the final point of

the boundary so established would contradict this text: the short segment of boundary from
Point 1439 up to the point where the drawing terminates cannot be considered to constitute a

boundary going “around” or “surrounding” Serpents’ Island.

4.50. The terms of the 1949 ProcŁs-Verbaux were apparently decided upon at one of the
preparatory meetings of the Romanian-Soviet Commission for Demarcation of the State Border

between the People’s Republic of Romania and the Union of the Soviet Socialist Republics,

which took place in Ismail on 5 December 1948. The agreed minutes provide:

“- In respect to point five on the agenda, at the proposal of the Soviet
delegation, the following decision was reached:

- The State boundary in the Black Sea from the flow of the Musuna (Musura)
channel is traced in a straight line on the prolongation of the boundary line
going through the Musuna (Musura) channel up to the point that divides in
two halves the water expanse from the terminus point of the land on the
Soviet coast (the islet of Limba) to the dyke of the port of SULINA.

- From this point (that divides in two halves the water expanse) the boundary
line passes in a straight line on the azimuth of 102°,5 until it meets the
exterior margin of the marine boundary strip surrounding Serpents’ Island,
with a 12 mile radius, and further on the exterior margin of the marine
136
boundary strip of 12 miles.”

4.51. This clear language shows the consistent approach of the two Parties in establishing

their maritime boundary on a course running from Point 1439 and then along the 12 nm arc of

circle measured from Serpents’ Island and surrounding it, throughout constituting a boundary

between the Soviet Union and Romania.

12-mile boundary strip, leaving Zmiinyi Island on the side of the USSR”. The relevant part of the translation is
135roduced at UCM, para. 5.43. For the full translation, see UCM Annex 28.
136 RM, para. 4.7; Annex RM 14.
See ProcŁs-Verbal no.10 of the Romanian-Soviet Commission for demarcation of the State Border
between the People’s Republic of Romania and the Union of the Soviet Socialist Republics, Ismail, 5 December
1948 (Annex RR 2) (emphasis added).
794.52. The conclusion advocated by Ukraine from Map 134 is particularly incongruous in light

of the evident care and attention with which the two Parties defined the geographical
coordinates of the other 1439 points on their boundary (with a precision of degrees, minutes,

seconds, and in some cases, tenths of seconds, of latitude and longitude). If significance had

been attached to the last point shown on Map 134, there would have been no hesitation in

defining and describing with precision one more point.

(ii) Ukraine’s “blank space” argument

4.53. Ukraine stresses that while the maritime boundary on the 12 nm arc around Serpents’

Island as shown on the sketches forming part of the various individual ProcŁs-Verbaux

terminates at the edge of those sketches, the maritime boundary illustrated on Map 134

terminates immediately before the edge of the map, there being between the end of the

boundary shown and the edge of the map a thin blank space. It argues that the boundary was
intended to terminate at the specific point shown, which it alleges is situated at 22ºof arc from

Point 1439 on the 12 nm circle around Serpents’ Island. 137

4.54. Ukraine relies on the argument that this point practically coincides with Point 2'
indicated on the chart deposited by Romania with the UN Secretary-General in 1997 depicting

the basepoints and baselines from which Romania’s territorial sea and contiguous zone is

measured, and that it also practically coincides with the point representing the outer limit of the

territorial seas of Romania and Ukraine as agreed in the 2003 Border Regime Treaty (Point F).

From the similarity of these points, Ukraine suggests that this point was agreed by the Parties
in anticipation of the extension of Romania’s claimed territorial sea from 6 nm to 12 nm.

4.55. Neither the Ukrainian assertion as to the relevance of the blank space on Map 134, nor
the conclusion that it seeks to draw from it are supportable. Map 134 bears no indication that it

was intended to depict the final point of the boundary, and in any case, such a conclusion

would be inconsistent with the clear terms of the instrument to which it was annexed.

4.56. As to the alleged relevance of the blank space, nothing on the map suggests that the

point where the drawn boundary stops should be regarded as its final point. There is no change

in the graphical representation to support this conclusion (such as a change in colour). If the

boundary was indeed intended to terminate at that point, one would expect it to be defined in

137
See UCM, para. 5.133.

80the text of the 1949 General ProcŁs-Verbal to which the map is annexed or in the relevant

individual ProcŁs-Verbaux. There is no indication to that effect in any of these instruments.

4.57. In contrast to the sketches contained in the individual ProcŁs-Verbaux which, as

explained above, were included as visual illustrations of the positions of the border signs

described in the text and accordingly depict only a limited area, the maps annexed to the 1949

General ProcŁs-Verbal use a smaller scale and depict areas which are much larger than the

immediate area in question. Nevertheless, not all features in that larger area are depicted. For
138
instance, Map 134 depicts only a small part of the Soviet/Ukrainian mainland and its coast,
even though there is ample room to show more. The fact that only a minor part of the Soviet

mainland is shown clearly does not mean that in the blank space situated close to the left edge

and in the large blank space that covers the whole upper half of that map, there does not exist

any territory appertaining to the USSR.

4.58. The fact that the map features only parts of the territories of the Parties, as well as parts

of the boundary is explained by the purpose of the map as reflected by its title: “Map of the

State Border between the Union of the Soviet Socialist Republics and the People’s Republic of

Romania from border signs no. 1438 to no. 1439”. The purpose of Map 134 is to depict the

boundary between points 1438 and 1439. That sector is drawn in full detail, as are all

geographical features relevant to it. At the same time, the boundary sectors situated both
before and beyond point 1438 and 1439 are only partially depicted.

4.59. Study of the other maps annexed to the 1949 General ProcŁs Verbal reveals that this

was the constant practice. Only the physical geographical features directly relevant for the
boundary sector described by the specific map are depicted, irrespective of whether or not

space is available for the depiction of other features. For instance, Romania has reproduced, in

Figure RR15 below, Map 133, entitled “Map of the State Border between the Union of the

Soviet Socialist Republics and the People’s Republic of Romania from border signs no. 1436 to

no. 1438”. It is clearly visible that, in spite of the available place, only parts of the territories of

the two States are depicted.

138
Contrary to the impression which UCM, para. 5.50 attempts to give, by asserting that “the mainland coast
and its accompanying territorial sea area marked up to the bottom edge of the map”. In fact, as seen from Map
134, only a limited sector of the Parties’ mainland areas and territorial seas are marked to the edge of the chart, i.e.
that sector directly relevant for the two border signs representing the object of the map (Points 1438 and 1439).
The rest of the mainland is not marked at all.
81 Figure RR15

Reproduction of Map 133 annexed to the September
1949 ProcŁs-Verbal
In spite of plenty of available space, only the elements

related to the purpose of the map are featured

82 (iii) Ukraine’s “coincidence of points” argument

4.60. Ukraine asserts that the end of the depicted boundary on Map 134 constitutes a point

situated approximately 12 nm from the Romanian coast, and that this fact is evidence of the

will of the Parties to effect only a delimitation between their respective territorial seas. It notes

that Romania, though it had not officially made a claim to a 12 nm territorial sea, was already
139
“contemplating” making such a claim. But, even if this assessment on Romania’s

“contemplation” were true – which is not supported with arguments in Ukraine’s Counter-

Memorial -, this would make it not less likely that “the exterior margin of the marine boundary

strip of 12 miles” was intended to survive changes in the status of the surrounding waters and

to constitute a permanent maritime boundary between the two States.

4.61. In any event, the Ukrainian argument does not fit the facts, since it is based upon the

modern coastal situation and not that of 1949. If the present location of the outer end of the

Sulina Dyke is taken into account, that point is indeed situated at approximately 12 nm from

the Romanian coast, and almost coincides with Point F from the 2003 Border Regime Treaty,

and with Point 2' as shown on the 1997 chart deposited by Romania with the Secretary-General

of the United Nations. But the fundamental point overlooked by Ukraine is that in 1949 the
140
Sulina Dyke was about 1.5 nm shorter. The allegedly “final” point shown on Map 134 was
141
at that time actually situated at about 13.4 nm from the Romanian coast, rather than 12 nm.
142
This is illustrated on Figure RR16 below.

139 See UCM, para. 5.33(iii).
140
141 See RM, para. 11.17.
142 The point was situated at 10.9 nm from the Soviet mainland as well.
A Romanian error may be responsible for this unfounded Ukrainian hypothesis. Unfortunately the
reproduction of Map 134 included in the Romanian Map Atlas was not a copy of the original Map 134 annexed to
the 1949 General ProcŁs-Verbal, but a reproduction of a copy which had been used as a working document during
the negotiations with Ukraine. The copy provided with the Memorial contains not only the original elements of
Map 134, but also subsequent comments, notations and graphical additions written or drawn in by Romania’s
experts in the course of the negotiations. These recent additions include manuscript indications regarding the
distance between Points 1438 and 1439, and between Point 1439 and Serpents’ Island, as well as a depiction of the

prolongation of the Sulina Dyke to reflect developments subsequent to 1949, resulting in the dyke being depicted
as being longer than it actually was in 1949. As all reproductions of Map 134 found in the Ukrainian Counter-
Memorial show these subsequent additions, it appears that they are copies of the version of Map 134 included in
the Romanian Map Atlas. Full-scale copies of the unadorned original Map 134 have been deposited by Romania
with the Registry of the Court together with this Reply; all figures in this Reply related to Map 134 are based on
this accurate reproduction.

83 Figure RR16

The position of various points depicted by Map 134, in
accordance with the factual situation of 1949 (the year
when Map 134 was issued)

844.62. Ukraine stresses that the allegedly final point shown on Map 134 more or less coincides

with Point 2' shown on the chart deposited by Romania with the UN Secretary-General in 1997

(which constitutes one of the turning points on the outer limit of Romania’s territorial sea), and

also with Point F as agreed by Romania and Ukraine in the 2003 Treaty on the Border Regime,

which was agreed by the Parties to represent the junction of the outer limits of their respective

territorial seas.

4.63. The supposed coincidence between the final point of the boundary shown on Map 134

on the one hand, and Point 2' and Point F on the other, is at best approximate. In this regard,

Ukraine contends that the final point of the boundary depicted on Map 134 lies at “about 22° of
arc beyond Point 1439”, and that as depicted on the chart deposited by Romania in 1997 with

the UN Secretary-General the arc between Point 1'and Point 2' , representing a portion of the

outer limit of Romania’s territorial, is equal to “approximately 22°”.143 The use of such vague

terms belies the inaccuracy of Ukraine’s measurement; as shown on Figure RR17 below, the

arc between Point 1439 and the final point shown on Map 134 in fact measures 23°. Similarly,
as between Point 1'and Point 2' , the arc of circle in fact amounts to 24.5° of the 12 nm arc

around Serpents’ Island.

4.64. In short, Ukraine’s argument is anachronistic. The approximate correspondence
between the point shown on Map 134 and Points 2'and F is a coincidence resulting from the

subsequent extension of the Sulina Dyke. No conclusion is to be drawn from that approximate

coincidence as concerns what was agreed in 1949.

143
UCM, para. 5.75.

85 6
8

e t
t i
o P
n l
o v
l c
n p
f e
h ,
f 2
s i
1 n o
R i d
R so n
r p 4
i h 1
F n p
e a
t n
b i
s t
c i
r e
f y
d a
e d
T u
b (iv) Other maps of the boundary annexed to the 1949 General
ProcŁs-Verbal

4.65. In any event Ukraine’s explanation is inconsistent with other depictions of the boundary

around Serpents’ Island which were part of the assemblage of maps annexed to the 1949

General ProcŁs-Verbal. Since completing its Memorial, Romania has discovered a copy of

volume 1 of the 1949 Catalogue of Maps (containing 5 schematic maps and 134 maps

depicting various segments of the Romanian/Soviet boundary, as well as representations of the

border marks used to demarcate the boundary). Some further individual copies of particular
folios have also been located. Two of the schematic sketches – Plates I and V – show

Serpents’ Island: both depict the boundary around it. They are attached as Figures RR18 and

RR19 at pages 88 and 89 of this Reply. Full size replicas have been deposited with the

Registry.

4.66. Plate I is entitled “Schema of the State Boundary between the Union of Soviet Socialist

Republics and the People’s Republic of Romania demarcated in the years 1948-1949”. It

shows the eight sectors of the demarcation, coloured red (to indicate that the demarcation

works on those sectors were the responsibility of the Soviet Union) or green (to indicate that it

was a Romanian responsibility). The area of the Danube Delta and the maritime boundary is

Sector VIII. It is coloured red, indicating that the demarcation works, as well as the
preparation of maps covering that sector, was a Soviet responsibility.144

4.67. Plate V is entitled “Schema of the Assembly of the Folios of the Map of the State

Boundary between the Union of Soviet Socialist Republics and the People’s Republic of

Romania”. It provides a key to the location of specific maps from Nos. 95-134 (although the
border sector covered by Map 134 is not specifically indicated), i.e. those covering Sectors VII

and VIII.

144See also the ProcŁs-Verbal no. 10 of the Romanian-Soviet Commission for the Demarcation of the State
Border between the People’s Republic of Romania and the Union of the Soviet Socialist Republics, signed
in Ismail, on 5 December 1948 (Annex RR 2), where it is specified that the maritime boundary was
established “at the proposal of the Soviet delegation”.

87 Figure RR18

Plate I annexed to the 1949 ProcŁs-Verbal

88 Figure RR19

Plate V annexed to the 1949 ProcŁs-Verbal

894.68. As seen from the detailed extracts from the two Plates (presented below, superposed on

Map 134, as Figure RR20) both show the Romanian/Soviet boundary passing on the arc

around Serpents’ Island. In both cases the arc is longer than that shown in Map 134. Thus,

Plate V depicts the boundary on the 12 nm arc around Serpents’ Island up to a distance of 16.4
nm from the Romanian coastline (according to the factual situation of 1949), while Plate I

depicts it up to a distance of 21.7 nm. In both cases the line is drawn using the same symbol

throughout. In both cases it clearly depicts a boundary between the two States throughout its

length. The two Plates were intended as guides to finding the specific maps; they are part of a
package with the demarcation maps annexed to the 1949 General ProcŁs-Verbal, of which Map

134 is also part, and consequently have the same status. They clearly depict the boundary on

the 12 nm arc around Serpents’ Island, with areas appertaining to Romania on the other side of

the line. It is also clear that these areas could not have been thought of either as high seas or as

Romanian territorial sea. They are not high seas because the line indicates a boundary between
the two States. They are not territorial sea because the line stops, in the case of Sketch Map V,

at the eastern edge of the map about 16.4 nm from the nearest point of the Romanian coast, and

in the case of Sketch Map I further east still, about 21.7 nm offshore. Consequently, they

delimit the Soviet territorial sea from maritime areas appertaining to Romania that can now be
characterised as its exclusive economic zone and continental shelf. The two Sketch Maps

dispose of Ukraine’s arguments based on Map 134 and its final point.

90 1
9

e
f V
o t
t l
o P
l n
a I
f t
h l
f ,
2 o 3
R on 1
R i a
r o M
g e o
F t d
g c
d p
a d
e s
r y
c a
e n
e u
i b
D(d) The argument that Romania’s 12 nm claim in 1951 changed the function of Map

134

4.69. In tandem with the previous allegation, Ukraine asserts that the final point of the
boundary established by the 1949 ProcŁs-Verbaux (which it alleges is the final point of the

boundary depicted on Map 134) was intended to be located “at approximately the point at

which that prospective 12 n.m. limit extending seawards from the Romanian coast would

intersect with the Soviet Union’s 12 nm arc around Serpents’ Island”. 145 Slightly later, having

again drawn attention to the alleged similarity between the final point depicted on Map 134 and

Point 2’ and Point F, Ukraine asserts:

“No other conclusion is possible than that the general 1949 ProcŁs Verbal and
the individual 1949 ProcŁs-verbaux established a boundary which was at the
time intended to be, and was later expressly confirmed by Romania as being,
a territorial sea boundary extending out to sea as far only as the point at which

the outer l146ts of Ukraine’s and Romania’s 12 n.m. territorial seas
diverged.”

4.70. Ukraine bases itself on the assertion that the agreed boundary only extended as far as

the final point shown on Map 134, and that the Parties agreed on the extent of that boundary

due to the fact that Romania was, in 1949, already contemplating the extension of its territorial

sea to 12 nm.

4.71. The first of these assertions has been shown to be without foundation. 147 The final

point on Map 134 was located about 13.4 nm from the relevant basepoint (the end of the Sulina

dyke as it then was). Thus Ukraine’s assertion that the agreed boundary depicted on Map 134

was fixed in light of Romania’s eventual extension of its territorial sea cannot be correct.

Similarly, the assertion that following the Romanian extension of its territorial sea in 1951, the

entire course of the boundary allegedly agreed in 1949 and shown on Map 134 became the

boundary delimiting the Parties’ respective territorial seas is equally incorrect.

4.72. The claim that this boundary was “later expressly confirmed by Romania” to be a

territorial sea boundary is also without foundation. Although the 1963 ProcŁs-Verbaux

referred to the boundary for the first time as running around the “12-mile territorial sea of the

145 UCM, para. 5.47.
146 UCM, para. 5.76.
147 See above, paras. 4.45-4.68.
92USSR, leaving Serpents’ Island on the USSR side”, 148 the Individual ProcŁs-Verbal of Border

Sign 1438 retains the language of “Soviet Maritime Boundary Zone”. Similar language was

used in the 1974 ProcŁs-Verbaux. The change of terminology no doubt amounted to an

acceptation by Romania of the Soviet claim to a 12 nm territorial sea. But it is not evidence

that Romania recognised that the boundary allegedly so agreed was intended only to constitute

the demarcation between the 12 nm territorial seas of the Parties. The Soviet Union gained no

maritime territory in 1963.

4.73. Ukraine’s argument on this point, as on the preceding one, is also inconsistent with the

intention manifested on Plates I and V to depict a boundary line between Romanian and Soviet

maritime areas going around Serpent’s Island much further to the east, and well beyond any

conceivable area of Romanian territorial seas: see Figure RR20 above. These Plates confirm

that the 1949 Individual ProcŁs-Verbal meant what it said when it referred to a “Soviet marine

boundary zone, of 12 miles, surrounding Serpents’ Island”.

(e) The argument that the term “Soviet marine boundary zone” was adopted because

of Romanian “inhibitions” as to areas beyond 6 nm

4.74. In order to counter Romania’s argument based on the actual language of the 1949

procŁs verbal, Ukraine suggests that the use of the term “Soviet maritime boundary” was

included only because Romania, claiming only a 6 nm territorial sea, was “inhibited from
149
referring to this purely Soviet maritime limit as a ‘territorial sea’”.

4.75. This suggestion does not sit comfortably with Ukraine’s argument that Romania was

already “contemplating” a 12 nm territorial sea. 150 That would have been a factor weighing

against any such “inhibition”.

4.76. But in any event the argument is not convincing. If Romania had had any “inhibition”

of this kind which had played any role in the drafting of the 1949 ProcŁs-Verbaux, one would
have expected this to have been reflected in the text. If Romania was inhibited from accepting

the Soviet claim to a 12 nm territorial sea, one would have expected that it would have been

“inhibited” from concluding any maritime delimitation at all beyond 6 nm.

148 Annex RM 19, p. 10 the relevant part of the text is reproduced at RM, para. 4.19.
149 UCM, para. 5.46(i).
150 UCM, para. 5.47.
934.77. Ukraine argues that this Romanian “inhibition” is confirmed by the fact that, after

Romania claimed a 12 nm territorial sea in 1951, the language of the subsequent bilateral

instruments changed. This is simply incorrect. The 1954 Act retained the language of “marine

boundary zone”, as did the 1974 individual ProcŁs-Verbal of Border Sign 1439. It was only the

1963 and 1974 general ProcŁs-Verbaux which used the expression “Soviet territorial sea” in
relation to the area surrounding Serpents’ Island. If Ukraine’s argument were correct, the 1954

Act, concluded three years after the Romanian extension of its territorial sea, as well as the

individual ProcŁs-Verbaux concluded thereafter, would not have preserved the language agreed

upon in 1949.

(f) The argument that the 1949 line was not an all purpose boundary but was limited

to areas under the sovereignty of the two States

4.78. Ukraine, having highlighted that the boundary depicted on Map 134, extends for “about

22° of arc beyond Point 1439”, baldly observes that:

“nothing in the 1949 ProcŁs-Verbaux, either their texts or in the accompanying

maps and sketches, suggests that the relevant part of the 12 n.m. arc around 151
Serpents’ Island extended to the southeast or east around Serpents’ Island.”

Similarly, it asserts that:

“nothing in the ProcŁs-Verbaux (or elsewhere) indicates that the Parties’ were
considering anything other than the limits of the areas over which they possessed
sovereignty. In particular there is nothing to suggest that in 1949 the Parties had

in contemplation anything in the nature of continental shelf or EEZ rights, neither
of which in 1949 an established part of international law. Their agreement as to
the boundaries of their respective sovereign territories and territorial seas cannot
be interpreted as signifying anything in relation to yet-to-be established maritime
152
rights.”

4.79. As shown already, these assertions disregard the clear import of the text of the ProcŁs-

Verbaux, and of the sketches and maps, including Map 134 itself.

4.80. As to the assertion that the Parties were not considering anything other than the “limits

of the areas over which they possessed sovereignty”, the language of the ProcŁs-Verbaux
shows that the Parties intended to establish a single all-purpose maritime boundary, separating

on the one hand areas appertaining to Romania and on the other hand areas appertaining to the

151 UCM, para. 5.75.
152 UCM, para. 5.69.
94USSR. The Parties were not contemplating which particular regime those areas would be

subject to, or the existence or acceptance as part of international law of any particular legal

regime in 1949; rather the agreement was simply that certain areas were to be seen as

appertaining to one or the other and that as a result, Serpents’ Island was to be treated as having
been limited to only a 12 nm zone.

4.81. It appears that the main concern of the Parties under the political circumstances of 1949

was to ensure a 12 nm maritime belt around Serpents’ Island following its annexation by the
Soviet Union. The Parties confined themselves to fixing the course of the maritime boundary

in front of their mainland coasts and in the Serpents’ Island region, thereby assuring a maritime

belt of 12 nm for Serpents’ Island while the breadth of the maritime area of the opposite

Romanian coast was confined to only 9 nm. This was not a short-term arrangement, and it was

not made by reference to the then-current terminology of maritime zones. It defined a 12 mile
belt around Serpent’s Island, leaving to Romania the areas beyond.

4.82. Finally, Ukraine asserts that “nothing in the 1949 ProcŁs-Verbaux, either their texts or

in the accompanying maps and sketches, suggests that the relevant part of the 12 n.m. arc
around Serpents’ Island extended to the southeast or east around Serpents’ Island.” 153 As to the

texts of the ProcŁs-Verbaux this is incorrect: the 1949 texts refer to a boundary, an “exterior

margin” of 12 nm, “surrounding” Serpents’ Island and leaving it on “the USSR side”. A line

confined to a small south-westerly segment of the marine boundary strip would not surround

Serpents’ Island. The travaux are equally clear and confirm this intention: “the boundary line

passes… further on the exterior margin of the marine boundary strip of 12 miles” surrounding
Serpents’ Island. 154 As to the maps, it is true that these do not show the full extent of the

exterior margin. Still, there are three depictions of that margin in the 1949 assemblage of

maps: Map 134, Plate V and Plate I. They all show the boundary line proceeding on the 12 nm

arc around Serpents’ Island, and none of them defines a terminal point. They have different

undefined terminal points, respectively about 13.4 nm, 16.4 nm and 21.7 nm from the nearest
point on the Romanian coast as it then was. Point 1439 was the last demarcated point on the

boundary and there was no need for the maps and sketches to take matters further. Once they

had shown the marine boundary strip of 12 nm surrounding Serpents’ Island, they had

sufficiently illustrated the intention of the Parties as to the further marine boundary. But what

is significant is that when in 1951 the Soviet Union – which had had responsibility for mapping

153
UCM, para. 5.75.

95 155
Sector VIII, the coastal sector, in 1949 – had to produce a detailed chart of the area, it had no

hesitation in depicting a semi-circular line surrounding Serpents’ Island up to Point X, situated
156
due east of it. No explanation is given as to the three different terminal points of the three

depicted lines surrounding Serpents’ Island in the 1949 assemblage, and no significance is to

be attached to the terminal points on those maps.

(g) Ukraine’s reliance on Romanian legislation subsequent to 1949

4.83. Ukraine argues that Romanian maritime zone legislation does not refer to the 1949

ProcŁs-Verbaux or the Romanian/Soviet boundary established by it in a way which is

consistent with the argument now presented by Romania. That being so, the boundary must be

presumed to have delimited only the territorial seas of the two Parties. 157

4.84. The first Romanian law considered is the 1956 Decree No. 39 on “Regulating the

Regime of the Territorial Waters of the People’s Republic of Romania”. According to Ukraine,

article 1 of this Decree provided only that:

“The territorial waters of the People’s Republic of Romania are… delimited by the

territorial waters of the neighbouring countries… and in the north by a line
determined by agreement between the People’s Republic of Romania and the
Union of the Socialist Soviet Republics.” 158

This does not make any provision in relation to delimitation of the Romanian maritime areas

beyond the territorial sea.

4.85. The complete text of the Decree is at RM Annex 81. Its scope is clear both from its title

and its content. It deals with the legal regime of Romanian territorial waters (i.e. Romania’s

territorial sea) and does not purport to make provision for any other maritime area beyond the

territorial sea. In these circumstances, there was no reason to refer to the 1949 ProcŁs-Verbal

as dealing with the delimitation of maritime areas beyond the territorial sea.

154 See above, para. 4.50.
155 See above, para. 4.66 and Figure RR18.
156
See RM, Map RM A 16 in the Map Atlas, which is a subsequent edition of the Soviet map of the Western
157t of the Black Sea, first edition – 1951.
158 UCM, para. 5.83.
See Decree No. 39 of the Presidium of the Great National Assembly of the People’s Republic of Romania
on Regulating the Regime of the Territorial Waters of the People’s Republic of Romania of 21 January 1956
(Annex RM 81).
964.86. The second piece of legislation referred to by Ukraine is the 1986 Decree Establishing

the Exclusive Economic Zone of Romania in the Black Sea. 159 Again Ukraine seeks to draw

inferences from the fact that this Decree contains no reference to a delimitation line having

been agreed with the USSR. The relevant provision is article 2, which reads as follows:

“The exclusive economic zone extends in its exterior side up to a distance of 200
maritime miles from the baselines from where the breadth of the territorial sea is
measured; under the circumstances of the reduced dimensions of the Black Sea, the
actual breadth of the exclusive economic zone of the Socialist Republic of

Romania shall be determined by its delimitation through negotiations with the
neighbouring States whose coasts are adjacent with or opposite to the Romanian
coast of the Black Sea. The delimitation shall be done, taking into account the
legislation of the Socialist Republic of Romania, by agreements with these States,
applying, depending on the specific circumstances of each delimitation sector, the

delimitation principles and criteria generally recognised by International Law and
State practice, in order to reach equitable solutions.”

Apart from this provision, the Decree does not make any other reference to the delimitation of

Romania’s exclusive economic zone vis-à-vis other States. The provision does not purport to
deal with any pre-existing delimitations, and the fact that it does not refer to the various

Romanian/Soviet agreements is hardly surprising. The emphasis of the Decree was not so much

upon the lateral delimitation of Romania’s exclusive economic zone as on the breadth of the

exclusive economic zone, in conditions in which the small size of the Black Sea and the

entitlements of other States meant that Romania’s exclusive economic zone could not extend

out to 200 nm. No inference is to be drawn as to existing bilateral agreements.

(h) The assertion that Romania’s claim was not previously raised

4.87. Even if it were true, Ukraine’s assertion that the agreed boundary was not raised during
160
the delimitation negotiations would have no relevance to the Court’s task. Nothing obliges a

party to disclose all the legal arguments in support of its position. Equally, a party may use
certain arguments in bilateral negotiations in an attempt to achieve agreement but remains

entitled to supplement those arguments before the Court.

159 Decree no. 142 of 25 April 1986 of the Council of State concerning the establishment of the Exclusive
Economic Zone of the Socialist Republic of Romania in the Black Sea See (Annex RR 4).
160 See UCM, para. 5.117
974.88. But Ukraine’s assertion is untrue. Romania did invoke this argument, both in the

negotiations with the Soviet Union and subsequently in the discussions with Ukraine.

4.89. As far as the Romanian/Soviet negotiations are concerned, up to 1974 the boundary

established by the 1949 ProcŁs-Verbal was constantly confirmed in each ProcŁs-Verbal or

other border document concluded between Romania and the USSR. But this was not a factor

conducive to the successful finalisation of the bilateral talks on the delimitation of the Parties’

continental shelf (which had begun in 1967). In its Memorial, Romania quoted from the
161
minutes of the Romanian-Soviet negotiations. From these extracts it is clear that the

Romanian negotiators constantly promoted a delimitation based on the limited effect agreed to
be given to Serpents’ Island.

4.90. For example, the approach of the Soviet Union led the Romanian chief negotiator at the

tenth round (1-2 October 1987) to emphasise again that:

“in 1949, our governments established a sui-generis delimitation line, which
confirmed the pass-over of Serpents’ Island to the USSR and allocated to it, in part
explicitly and in part implicitly, a semicircular maritime space, with a radius of 12

miles, whose exterior limit on the segment separating Romanian wa162s from
Soviet waters received the characteristics of a State boundary.”

This is a succinct presentation of Romania’s position now before the Court.

(2) Ukrainian Claims as to Maps and Charts relied on by Romania

4.91. In its Memorial Romania invoked various charts issued by the Romanian, Soviet and

Ukrainian authorities, as well as by third States, as evidence supporting its position as to the

extent of the Romanian/Soviet maritime boundary established in 1949. These charts

consistently depict the maritime boundary from Point 1439 as continuing around the 12 nm arc
around Serpents’ Island to a point situated on that arc approximately due east of Serpents’

Island (Point X).

4.92. In its Counter-Memorial Ukraine suggests that the cartographic evidence relied upon by

Romania has no evidential value. It argues that “none of the 23 maps adduced by Romania
falls into the category of ‘maps [which] are annexed to an official text of which they form an

161 See Annexes RM 28, RM 29, RM 30 and RM 31, as quoted in RM, paras. 5.12-5.15.
162 Extract of the minutes of the 1987 Romanian-Soviet negotiations (Annex RM 31); also reproduced in
UCM, para. 5.109.
98integral part’”. 163 In accordance with its position that the 1949 ProcŁs-Verbaux and subsequent

agreements did not establish a boundary beyond that shown on Map 134, Ukraine argues that

“no map can be regarded as visually depicting a continental shelf or EEZ boundary already

164
formally agreed verbally by the States concerned” and that “there is nothing which
165
depictions of such a boundary on maps can be said to be corroborating or confirming”.

These last two observations can be briefly disposed of: to the extent that they are premised on

the correctness of the Ukrainian position as to the boundary agreed in 1949, they must stand or

fall with Ukraine’s position as to the scope and character of the agreed boundary.

4.93. In addition Ukraine identifies various features of the charts introduced by Romania

which, in its view, demonstrate their irrelevance and inconsistency. In an Appendix to Chapter

5 of its Counter-Memorial, Ukraine purports to analyse in detail each of the charts, in order to

reach this conclusion. The arguments identified by Ukraine are as follows:

166
a. The charts and maps have no evidentiary value;
167
b. The charts are irrelevant because they are not part of any official text;

c. None of the 23 charts and maps forms part of a written agreement establishing a
168
maritime boundary;

d. None of the 23 charts and maps depicts the different legal elements which

comprise the maritime boundary; 169

e. None of the 23 charts and maps indicates either EEZ or continental shelf

belonging to Romania; 170

f. The 23 charts and maps do not make consistent use of maritime chart

171
symbols;

g. None of the 23 charts and maps fixes the final point on the semicircle around
172
Serpents’ Island;

163 See UCM, para. 5.131(i).
164 See UCM, para. 5.131(ii).
165
166 See UCM, para. 5.131(iii).
167 See UCM, paras. 5.163, 5.167.
See UCM, para. 5.138.
168 See UCM, para. 5.138.
169 See UCM, para. 5.139.
170 See UCM, para. 5.140.
171
172 See UCM, paras. 5.141-5.149.
See UCM, paras. 5.150-5.153.
99 h. Some of the 23 maps are self-evidently not intended to serve as maritime

boundary maps. 173

i. Many of the 23 maps and charts are merely copies of or are based upon earlier
174
maps and have no independent significance.

Again, Romania will deal with each of these assertions in turn.

(a) The assertion that the charts and maps have no evidentiary value

4.94. In this regard Ukraine quotes the well-known passage from the judgment of the
Chamber of the Court in the Frontier Dispute (Burkina Faso/Mali) case where the Chamber set

out its approach in relation to maps. 175 Romania agrees with the approach of the Chamber,

which is consistent with that of the Court as a whole as well as of arbitral tribunals.

Nevertheless, Romania does not agree with the interpretation which Ukraine places on the

dictum of the Chamber or the conclusions it seeks to draw from it.

4.95. In the words of the Chamber the charts, the first of which was issued in 1951, constitute
“corroborative evidence” as to the course of the boundary fixed by the Parties in 1949. All the

charts placed before the Court depict the maritime boundary exactly as described by the 1949

ProcŁs-Verbaux and subsequently confirmed by the other relevant bilateral documents. As the

1949 and later ProcŁs-Verbaux did not establish the boundary beyond Serpents’ Island’s, the

visual depiction of the boundary on all charts stops at a point situated on the 12 nm arc of circle

around Serpents’ Island, and situated almost exactly due east of it. In this context, it is

significant that the same graphical representation of the extent of the boundary is featured on

official charts produced by both Parties.

4.96. It is true that unlike the ProcŁs-Verbaux themselves some of the charts depict the

entirety of the boundary on the 12 nm arc of circle around Serpents’ Island. But this does not

mean that the charts depart from the ProcŁs-Verbaux. The charts depict areas far larger than

the area in the immediate vicinity of the coasts of the Parties or of Serpents’ Island. They show

geographical features more accurately than the sketches, Plates I or V or Map 134 whose

purpose was to illustrate the course of the boundary in the location of particular border signs or

along a particular sector. Given these elements, it is not surprising that the charts depict the

173 See UCM, para. 5.154.
174 See UCM, para. 5.155.
175 UCM, para. 5.129.
100agreed boundary up to the final point envisaged at that time – Point X, situated due east of

Serpents’ Island.

4.97. As to the position of Point X, admittedly it is not situated on the line equidistant from

the Romanian and Ukrainian adjacent mainland coasts given the present coastal situation.
176
Nevertheless, its position can readily be explained and is not at all “fictitious”. The point is

consistently shown in the same position on all the post-1949 charts that depict the maritime

boundary around Serpents’ Island. Given the coastal configuration existing in 1949 (i.e., the

extension at that time of the Sulina Dyke) Point X represents the intersection of the 12 nm arc
around Serpents’ Island with a line drawn from the last point of the Romanian/Soviet land-river

boundary on a bearing perpendicular to the segment closing the Musura Bay and uniting its two

salient entrance points (the south-easternmost Soviet islet, and the outer end of the Sulina

Dyke).

4.98. Such a method of delimitation, using perpendicular lines (whether constructed on the

general direction or the coast or on the closing lines of coastal indentations) has been used in
177
delimitation of maritime areas, both in State practice and in international jurisprudence. The

method used for the construction of Point X is illustrated on Figure RR21 below.

176
177 See UCM, para. 4.19.
For instance, in the Gulf of Maine Case, the Chamber of the Court decided that the third segment of the
maritime boundary, dividing the Atlantic area seaward of the Gulf of Maine, was to be represented by the line
perpendicular to the closing line of the Gulf. Delimitation of the Maritime Boundary in the Area of the Gulf of
Maine (Canada/United States of America), ICJ Reports 1984, p. 246, at pp. 337-8 (para. 224).

101 Figure RR21

The Construction of Point X

102(b) The assertion that the charts are irrelevant because they are not part of any

official text

4.99. Ukraine argues that the cartographic evidence has no evidentiary value because none of

the 23 maps adduced by Romania falls into the category of maps which are “annexed to an
178
official text of which they form an integral part”. Here Ukraine misinterprets the Chamber’s

dictum in the Frontier Dispute case. Referring to the value of maps, the Chamber declared
that:

“in some cases maps acquire such legal force [of a territorial title], but where this
is so the legal force does not arise solely from their intrinsic merits: it is because
such maps fall into the category of physical expressions of the will of the State or

States concerned. This is the case, for example, when179ps are annexed to an
official text of which they form an integral part.”

The Chamber’s approach (confirmed by other judgments of the Court) is clear: maps acquire

legal force when they “fall into the category of physical expressions of the will of the State or

States concerned”. One example of such a situation is when maps form an integral part of a

treaty. But the Chamber did not say that this is the only situation where maps may constitute a

“physical expression” of the will of a State. The Chamber also envisaged that a map could

constitute the physical expression of the will of a single State alone, thereby excluding the
possibility that the category is limited solely to maps annexed to treaties.

4.100. Ukraine refers to the Chamber’s “example” as if it were the only category of maps

which acquire legal force as a physical expression of the will of the State or States concerned.

But there are other cases where maps issued by the authorities of a State have been considered

to reflect the will of that State. In the Minquiers and Ecrehos case, the Court took into account

two maps annexed to a French document, on which the Minquiers group and certain islands
180
from the Ecrehos group were indicated as belonging to the United Kingdom. The Court
considered that the maps and the French document which they accompanied “must therefore be

considered as evidence of the French official view at that time”. 181

4.101. In the Temple of Preah Vihear case, most of the evidence considered by the Court

consisted of maps submitted by both sides which showed the temple lying on one or other side

of the border. Cambodia grounded its case, inter alia, on maps drafted by French officials

178
See UCM, para. 5.131(i).
179 Frontier Dispute (Burkina Faso/Republic of Mali), ICJ Reports 1986, p. 554, at p. 582 (para. 54)
(emphasis added). This paragraph is quoted in UCM, para. 5.129.
180 Minquiers and Ecrehos case (France/United Kingdom), ICJ Reports 1953, p. 47, at pp. 66-7, 71.
181 Ibid., p. 71.
103which depicted the temple as being on Cambodian territory. Thailand contended that these

maps were to be disregarded on various grounds but the Court disagreed. In particular it held

that it was…

“difficult to overlook such a fact as, for instance, that in 1937, even after
Thailand’s own survey in 1934-1935, and in the same year as the conclusion of a

Treaty with France in which, as will be seen, the established common frontiers
were reaffirmed, the Siamese Royal Survey Department produced a map showing
Preah Vihear as lying in Cambodia. … That this map may have been intended for

internal military use does no182eem to the Court to make it any less evidence of
Thailand’s state of mind.”

183
4.102. Another relevant case was the Beagle Channel arbitration. After a thorough analysis

of the cartographic material, the tribunal reached the conclusion that the cartography submitted

by Argentina did not support its contentions, while the cartographic materials submitted by
184
Chile stood “in marked contrast”, and fully supported its submissions.

“In the present case it is not a matter of setting up one or more maps in opposition
to certain Treaty attributions or boundary definitions, but of the elucidation of the
latter – in which task map evidence may be of assistance. The problem involved in

the present dispute arises from the difficulties created by the structure and the
language of the 1881 Treaty already discussed, not from its incompatibility with
some map, or vice versa – and the solution has to be found through the ordinary
185
process of interpretation, to which cartography may contribute.”

The Tribunal affirmed that:

“maps published after the conclusion of the Treaty can throw light on what the

intentions of the Parties in respect of it were and, in general, on how it should be
interpreted. But the particular value of such maps lies rather in the evidence they
may afford as to the view which the one or the other Party took at the time, or

subsequently, concerning the settlement resulting from the Treaty, and the degree
to [which] the view now being asserted by that Party as the correct one is
consistent with that which it appears formerly to have entertained.” 186

4.103. Analysing the Argentine contention that cartographic materials could have probative

force only if they forming part of the treaties in force between the parties or if they themselves
187
had been agreed by the parties, the tribunal concluded that “these views are too restrictive”:

182 Ibid., pp. 27-28.
183 Beagle Channel Arbitration between the Republic of Argentina and the Republic of Chile, Report and

184ision of the Court of Arbitration, 18 February 1977, 21 RIAA, p. 57.
185 Ibid., p. 182 (para. 162).
186 Ibid., pp. 163-4 (para. 137).
187 Ibid., p. 164 (para. 137).
Ibid., p. 166 (para.141).
104 “It cannot be the case that non-agreed maps, produced, acted upon or adopted

unilaterally by a Party… must, merely on account of their unilateral provenance, be 188
regarded as devoid of all value. They can have such value, in varying degree…”

4.104. In accordance with this approach, cartographic evidence may be relied upon in

situations other than where the maps or charts form part of an agreement between the parties.

In addition to the situation where maps or charts are included in or annexed to treaties, maps

may fall into the category of physical expressions of the will of the State concerned provided

they are issued by the official authorities of the State. But even if the map or chart does not

possess such a character, it may be relevant as reflecting general opinion and repute as to the

effect of a treaty, or the positions of the parties to a treaty as to its effect.

4.105. Thus maps issued by third parties have been considered to be potentially relevant as

demonstrating informed opinion as to the existence and location of a boundary. In the Beagle

Channel case the Tribunal stated that maps produced by third States…

“are significant relative to a given territorial settlement where they reveal the
existence of a general understanding in a certain sense, as to what that settlement
is, or, where they conflict, the lack of any such general understanding.” 189

4.106. In this context, the following observations can be made:

- The cartographic evidence submitted by Romania supports the legal situation

established by treaties concluded between Romania and the Soviet Union, by which

Ukraine is bound by way of succession, and is consistent in its depiction of the situation

which Romania says derives from those instruments.

- A significant part of the cartographic evidence was issued by official authorities of the
190
USSR and subsequently by the Ukraine, which during a period from 1951 onwards

issued charts featuring the maritime boundary up to Point X. In the words of the Court

in the Temple of Preah Vihear case, these charts represent evidence of the Soviet and

subsequently Ukrainian “official view” and “state of mind” as to the course and extent
191
of this boundary.

188 Ibid., p. 167 (para. 141).
189
190 Ibid., p. 167 (para. 142(2)).
The cartographic evidence put forward includes a brochure issued in 2004 by an entity which is described
as “the only specialized and authorized enterprise in State [Ukraine] the main goal of which is fulfilment of State
tasks on production of nautical charts”: see RM, para. 11.38 & Annex RM 86. In its Counter-Memorial, Ukraine
does not contest Romania’s conclusions in relation to this brochure.
191 Temple of Preah Vihear (Cambodia v. Thailand), Merits, ICJ Reports 1962, p. 6, at p. 27.
105 - Further, charts showing the same course and extent of the boundary were issued by the

competent Romanian authorities.

- This practice of both parties to the 1949 ProcŁs-Verbaux and subsequent instruments
evidences the understanding of the Parties as to the meaning of the various agreements.

To adapt the observations of the Arbitral Tribunal in the Beagle Channel case, “clearly,

a map emanating from [Ukraine] showing certain [maritime areas] as belonging to

[Romania] is of far greater evidential value in support of [Romania’s] claims to that

[maritime area] than a map emanating from [Romania] itself, showing the same thing.

Yet that is not the whole story – for […] a consistent or very general emission from

[Romania] of maps favouring its claim will at least show a settled belief in the validity
192
of that claim.”

4.107. Romania has also placed before the Court charts issued by third parties, depicting the

same course and extent of the boundary. These charts evidence informed opinion as to the

course of the boundary, namely as to the existence and extent of the agreed international

maritime boundary between Romania and the USSR (and subsequently Ukraine).

4.108. For these reasons the charts submitted by Romania should be given the proper

evidential value consistent with the purpose for which they have been introduced. They

corroborate and support the course and extent of the boundary agreed by Romania and the

Soviet Union in 1949.

(c) The assertion that none of the 23 charts and maps forms part of a written

agreement establishing a maritime boundary

4.109. Ukraine argues that none of the 23 maps forms part of a written agreement establishing

a maritime boundary. 193 This is true, but it misses the point of the reliance placed on the maps.

4.110. In this context, it is convenient to mention Ukraine’s statement that Map 134

“surprisingly was not included by Romania in its catalogue of 23 relevant maps”, even though

“it is the only cartographically reliable map which is annexed to an agreement establishing a
194
Soviet-Romanian or Ukraine-Romanian border.” In fact Map 134 was annexed to the 1949
General ProcŁs-Verbal, and was discussed at length in the Memorial. On the other hand, the

192 Beagle Channel Arbitration between the Republic of Argentina and the Republic of Chile, Report and
Decision of the Court of Arbitration, 18 February 1977, 21 RIAA, p. 57, at p. 167 (para. 142(1)).
193 See UCM, para. 5.138.
194 See UCM, para. 5.132 (emphasis in original).
106section of the Memorial in which Romania deals with the other 23 relevant maps was devoted

to maps and charts issued after the conclusion of the 1949 ProcŁs-Verbaux, which are evidently

invoked in corroboration of the maritime boundary thereby established. Ukraine’s observation

as to the omission of Map 134 is unfounded.

4.111. Although not attached to, or included in, an agreement between Romania and the Soviet

Union or Ukraine, the chart included as RM A15 was attached to and formed part of the 1978

Turkish/Soviet Continental Shelf Agreement. Ukraine contends that the chart constitutes res

inter alios acta, so that it cannot be invoked to the detriment of Ukraine. However, the chart

was annexed to the Turkish/Soviet agreement and it clearly falls into the category of “physical

expressions of the will of the State or States concerned”, even if the relevant boundary is not
one the Parties agreed between themselves. It also constitutes evidence of Soviet (and Turkish)

views as to the legal situation. The chart is an official document produced jointly by the Soviet

Union and Turkey, attached to a treaty registered with the UN Secretary-General.

(d) The assertion that none of the 23 maps depicts the different legal elements which

comprise the maritime boundary

4.112. Ukraine asserts that none of the maps put forward by Romania correctly depicts the

different legal elements comprising the maritime boundary to the south of Serpents’ Island. 195

4.113. It is true that the maritime boundary depicted on the 23 charts is made up of two

sectors: in relation to the first of these, the Parties agree that it represents the maritime

boundary between the Romanian and Ukrainian territorial seas. The parties differ as to the
characteristics of the boundary in the second sector. Despite this disagreement, there is no

indication on any of the charts as to the position of the point where the first sector ends end the

second sector starts.

4.114. But whether separating the territorial seas of the two States or areas subject to different
legal regimes, the boundary agreed in 1949 remains an international maritime boundary along

its entire length. It is consistently depicted as such on the charts, and on all but one 196of the

charts using the appropriate symbols.

195 UCM, para. 5.139.
196 For this exception, see below, para. 4.128.
1074.115. According to the International Hydrographic Organisation, there is only one symbol

used to graphically represent international maritime boundaries, whether they separate

territorial seas, exclusive economic zones or continental shelves appertaining to different
197
States.

4.116. On the other hand, a completely different symbol is used specifically for the purpose of

indicating the seaward limit of the territorial sea of one State when that limit divides the State’s

territorial sea from its outer maritime areas (such as its exclusive economic zone or continental

shelf). This symbol is not used on any chart (including Soviet and Ukrainian charts) in order to
198
depict the boundary between Point F and Point X. It appears on only one Ukrainian chart

where it is used to indicate, not the boundary running along the 12 nm arc to the south of

Serpents’ Island but the seaward limit of the Ukrainian territorial sea around Serpents’ Island to

the north of Point X, in the portion where Serpents’ Island faces the Crimean Peninsula to the

east and the Ukrainian mainland to the north. The consistent use of symbols on the various
maps and charts invites the conclusion that their makers understood the boundary from the

mainland coasts of the two States up to Point X as continuing along its entire length and

separating maritime areas of the two States.

(e) The assertion that none of the 23 charts and maps indicates either EEZ or

continental shelf belonging to Romania

4.117. Ukraine asserts that none of the 23 maps contains any boundary or other marking
indicating that to the south of the arc around Serpents’ Island the maritime areas are to be

regarded as being areas of either exclusive economic zone or continental shelf appertaining to

Romania. But the charts are clear about the State to which the areas lying outside the 12 nm

Romanian territorial sea to the south of Serpents’ Island appertain. It is sufficient to look at the

German chart, which contains clear indications by words as to which State the maritime areas
199
south of Serpents’ Island belong.

4.118. Ukraine asserts that “there are accepted chart symbols for continental shelf and EEZ

boundaries”, suggesting that the failure to use these symbols equals their failure to indicate the

appurtenance of the maritime areas on either side of the boundary. This assertion in itself is

based on a false premise, since there are no different symbols for international maritime

boundaries separating territorial seas, exclusive economic zones or continental shelves

197 See Figure RR 23 below, RM, Figure 25 at p.209, as well UCM, Annex 42.
198 See RM, Figure 25, also in the Map Atlas, maps RM A 23 and RM A 24.
199 See RM, Figure 26, also in the Map Atlas, maps RM A 41 and RM A 42.
108appertaining to different States. There is only one accepted symbol which indicates all types of

international maritime boundaries. 200

4.119. Ukraine also asserts that, since certain maps use “markings” in relation to other
boundaries they depict (for instance, the USSR/Turkey maritime boundary), the fact that the

common markers for continental shelf and exclusive economic zone boundaries are not used on

the charts submitted by Romania to show that the maritime areas situated immediately south

and south-east of the 12 nm arc around Serpents’ Island appertain to Romania “is highly

significant”.201 The implication would seem to be that this supports the Ukrainian position that

the line shown on the charts (albeit using the symbol for international maritime boundaries)

does not represent the course and extent of the boundary agreed in 1949.

4.120. Once again, this is not the case. Reference need only be made to the German chart

already mentioned (RM A 41). All the charts placed before the Court consistently use symbols

for depicting the Romanian/Soviet (and subsequently Romanian/Ukrainian) boundary. No

matter whether the same symbol is used for the entire length of the land/river boundary and the

maritime boundary, or different symbols are used for, respectively, the land/river portion of the

boundary and the maritime boundary (both approaches being in conformity with the

international uses of symbols as recommended by the International Hydrographic

Organisation), the boundary depicted is a single, continuous one, having the same character
along its entire length, dividing areas appertaining to the Soviet Union or Ukraine on the one

hand and Romania on the other.

4.121. Twelve charts clearly indicate, by words, the States situated on the two sides of the

boundary, to which the respective land or maritime areas appertain:

- Certain charts contain the names of the States, written several times north and

south of both the land/river and the maritime boundary: this is the case of the charts

included in the Romanian Map Atlas as Maps RM A 21 (an enlargement of the area

around Serpents’ Island is reproduced as RM A 22), RM A 35, RM A 36, RM A 41

(with detail of Serpents’ Island area reproduced as RM A 42).

200 Although, as discussed below, there do exist internationally accepted symbols indicating the outer limits
of the continental shelf and exclusive economic zone: see para. 4.126.
201 See UCM, para. 5.140. Ukraine also refers to a “fishery zone off Romania St. Gheorghe” depicted on the
charts included in Romania’s Map Atlas as map RM A28. This zone, depicted using the international symbol for
“limits of fishery zones”, is situated within the Romanian territorial sea.
109- Other charts contain the names of the States written only north and south of the

land/river boundary (on the corresponding land territories of the two States); where that
boundary is continued at sea by the maritime boundary, no marking or symbol is used

which might denote that the appurtenance of the areas situated to the north and south of

the boundary is any different from the appurtenance of the land territories to the north

and south of the boundary on land. This is the case the charts included in the Romanian
Map Atlas as Maps RM A 15, RM A 25, RM A 31, RM A 32, RM A 33, RM A 34, RM

A 37, RM A 39 (the detail of the area around Serpents’ Island is reproduced as RM A

40).

- Eleven charts do not bear any indication regarding the appurtenance of either the

land or maritime areas situated to the north or south of the boundary they depict. See

RM A 16 (with detail of the area of Serpents’ Island reproduced as RM A 17), RM A

18, RM A 19, RM A 20, RM A 23 (enlargement of the area around Serpents’ Island
reproduced as RM A 24) RM A 26, RM A 27, RM A 28, RM A 29, RM A 30, RM A

38. This does not imply, however, that the land or maritime areas do not belong to any

State or that the maps are without value as evidence of the state of mind of their makers

as to the situation, in particular the course and extent of the boundary. On the contrary,
one can readily discern the course and extent of the boundary, which are the States

depicted and which are the land or maritime areas appertaining to each of them. Once

again, the continuity of the land/river and maritime boundary is not broken in any way.

110(f) The assertion that the 23 charts and maps do not make consistent use of maritime

chart symbols

4.122. Ukraine asserts that the 23 charts and maps do not make consistent use of maritime
202
chart symbols. In fact, the use of symbols on the charts and maps submitted by Romania is

consistent and in accordance with the guidelines of the International Hydrographic

Organisation. Although different symbols are used to depict different elements on different
charts, the meaning of each of the charts is clear and the use of symbols consistent.

4.123. In this context, Romania reproduces, as Figure RR22 at page 112 of this Reply, the

relevant page of the volume edited by the International Hydrographic Organisation “Symbols,
203
Abbreviations, Terms Used on Charts”. The same page was also in Ukraine’s Annex 42.

The Ukrainian reproduction is in black-and-white but the original is coloured, so that the

Ukrainian reproduction fails to present accurately the various symbols.

4.124. According to this authoritative IHO publication, there are two symbols used to indicate

“international maritime boundaries”: the symbol now in use consists of a succession of

alternate red crosses and dashes (+ – + – + – + – + – +); the older symbol consists of a

succession of black dashes (▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯). While the new symbol for “international

maritime boundaries” differs from the new symbol for “international boundaries on land”

(which consists of a succession of black crosses: + + + + + + + + +), the old symbol for

international maritime boundaries was also occasionally used to indicate “international

boundaries on land”. In other words, while the modern symbols used to indicate, respectively,

land and maritime boundaries differ, in the past the same symbol was sometimes used to
indicate all types of international boundaries – a succession of black dashes. A further

variation of this older symbol, which is also fairly common on older maps and charts, consists

of a succession of alternate black dashes and dots (▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯ ▯). This symbol is

still frequently used, for instance, in the illustrative maps included in the judgments of the

Court. 204

202 UCM, para. 5.141.
203 See International Charts Series INT1 Symbols, Abbreviations, Terms Used on Charts (Bundesamt für
Seeschiffahrt und Hydrographie, Hamburg/Rostock, 1996), p. 51.
204 See, e.g., the representations of the land boundaries between Tunisia and Libya and Libya and Egypt in
the Case Concerning the Continental Shelf (Libya/Malta), of the land boundaries between Tunisia and Algeria and
Tunisia and Libya in the Case Concerning the Continental Shelf (Tunisia/Libya), of the agreed land boundaries in
the Case Concerning the Land, Island and Maritime Frontier Dispute (El Salvador/Honduras; Nicaragua
intervening).

111 Figure RR22

Extract from International Charts Series INT1 Symbols,
Abbreviations, Terrms Used on Charts, published by

Bundesamt fur Seeschiffahrt und Hydrographie,
Hamburg/Rostock, 1996
1124.125. But there are no different symbols to represent maritime boundaries between different

types of maritime areas. For instance, there is not one symbol for the boundary separating

territorial seas, another for the boundary separating exclusive economic zones and yet a third

for boundaries separating continental shelves. There exists only one symbol depicting any kind

of “international maritime boundary”, either the (new) red crosses-dashes symbol, or the (old)
205
black dashes symbol and its variant, the black dashes-dots symbol .

4.126. There also exist different symbols used to indicate either the “seaward limit of

territorial sea” (a red line consisting of an alternation of long dashes and two crosses + +

—— + + —— ), or the “seaward limit of contiguous zone” (a red line consisting of an
alternation of long dashes with one cross —— + —— + ——) where beyond them there are

outer maritime areas (exclusive economic zone or continental shelf) belonging to the same

State. There also exist symbols to indicate the outer limits of a State’s exclusive economic

zone or continental shelf, when they face the high seas and not maritime areas belonging to

other States (these limits are designated by a continuous red line with “EEZ” written above it).

4.127. According to the symbols used to depict the land and maritime boundary, all but one of

the charts submitted by Romania may be grouped as falling into one of two categories:

(1) The first category uses to depict the maritime boundary the variation of the

“old” symbol consisting of alternate black dashes and dots; these charts are, with only a
few exceptions, Romanian and Soviet charts edited before 1990 (the charts included in

Romania’s Map Atlas as RM A 15, 16-17, 18, 19, 20, 26, 27, 28, 29, 30, 37 and 38).

With the exception of charts RM A 20, 26 and 28, which do not feature the mainland

area of Romania or the USSR/Ukraine, all other charts use the same symbol to indicate

the international boundary on land.

(2) The second category, comprising charts issued after the 1990, depicts the

international maritime boundary using the “new” red crosses-and-dashes symbol (charts

RM A 21-22, 23-24, 25, 31, 32, 33, 34, 35, 36, 39-40); all of these charts
correspondingly depict the international boundary on land using the modern black

crosses symbol.

205
See, for example, the chart on the Use of symbols and abbreviation as recommended by the
International Hydrographic Organisation, attached to International Charts Series INT1 Symbols,
Abbreviations, Terms Used on Charts (Bundesamt für Seeschiffahrt und Hydrographie, Hamburg/Rostock,
1996).

1134.128. Only one chart, the German chart included in the Romanian Map Atlas as map RM A

41-42, departs from the internationally endorsed symbols: this chart uses, in order to depict the

entirety of the Romanian/Ukrainian boundary (both the land/river boundary and the maritime

boundary), a symbol consisting in a succession of red crosses ( + + + + + + + + ); this does not,
206
as Ukraine asserts, constitute a symbol appropriate for depiction of a land boundary. Despite

the unusual use of symbols on the German chart, there can be no doubt in relation to the

international character of the maritime boundary depicted, as well as to which State the land

and maritime areas on either side of the boundary belong, since these aspects are clearly

indicated by the names of the States. The names (“UdSSR” and, respectively, “Rumänien”) are
placed on the land territories situated, respectively, north and, south of the river boundary;

secondly, the same names are twice placed on either side of the international maritime

boundary – first, on its straight sector just off the adjacent coasts of the two States, and second,

on its terminal sector situated on the 12 nm arc of circle around Serpents’ Island to the

southeast (“UdSSR” being written inside this circle, while “Rumänien” being written

immediately outside). 207 This chart, issued by a third State, represents, together with the

Russian, French and Bulgarian charts submitted by Romania, evidence of informed opinion as

to the course of the Romanian/Ukrainian maritime boundary.

4.129. Thus all but one of the charts are consistent in using internationally recognised symbols

to depict the international maritime boundary from the coast up to a point on the 12 nm arc of

circle around Serpents’ Island, and the meaning of the one chart which does not use the

internationally recognised symbols cannot be doubted. The course, extent and nature of the

boundary depicted on these charts confirm the meaning of the provisions of the 1949 ProcŁs-

Verbaux and subsequent agreements.

4.130. Ukraine relies on the fact that the seaward limits of the Romanian and (with one
208
exception) Ukrainian territorial sea are not depicted on the chart. This argument must be
seen in the light of Ukraine’s assertion that the second portion of the maritime boundary

depicted on all these charts represents not an international maritime boundary but only the

seaward limit of Ukraine’s territorial sea around Serpents’ Island.

206 See UCM, para. 5.194.
207 Details of this chart were presented in RM, para. 11.35. Ukraine tries to explain why this chart marks
twice (including in the controversial region) the two sides of the boundary with “USSR” and “Romania” by
considering the second marking as a mere repetition of the first: UCM, para. 5.195. It admits that this is nothing
more than “speculation”: UCM, para. 5.196.
208 See UCM, paras. 5.148, 5.149.

1144.131. In order to explain why all of the charts depict the boundary on the 12 nm arc around

Serpents’ Island between Point F and Point X using the same symbol along its entire length

(that used for international maritime boundaries), Ukraine builds a complicated explanation. It

argues from the hypothesis that most of the maps derived directly or indirectly from maps
produced by the “Defence authorities of the Soviet Union”, and postulates that the depiction of

the boundary up to Point X was necessary for the Soviet defence authorities to know where the

area under Soviet sovereignty (i.e. the territorial sea) stopped. 209 By contrast Ukraine had

elsewhere postulated that the depicted boundary on the charts terminated at Point X “for

convenience”, since north of the final point, indication of the outer limit of the territorial sea

was unnecessary as “on that side the Island and its waters faced ‘inwards’ onto Ukraine’s
mainland territory”. 210

4.132. The argument is unconvincing. First, it is not clear why the Soviet defence authorities

should have needed to know the boundary between the outer limit of the Soviet territorial sea

and what, on Ukraine’s case, were unapportioned areas lying to the south of the 12 nm arc
around Serpents’ Island, while at the same time those authorities had no similar need to know

the outer limit of the Soviet territorial sea and the Soviet outer maritime areas to the north of

Serpents’ Island, or indeed elsewhere. A coherent approach would have implied either the

depiction of the entirety of the length of the outer limit of the Soviet territorial sea (even using

a wrong symbol instead of the correct one) or no depiction of that outer limit at all.

4.133. Second, the argument that the depiction of the supposed seaward limit of the Soviet

territorial sea around Serpents’ Island was necessary only up to Point X as from there the

waters of Serpents’ Island faced “inwards” onto the Soviet (Ukrainian) mainland, has no

support in the geography of the area. The waters bounded by a fairly long segment of the 12

nm arc east and north of Point X are no more orientated “inwards” towards the Crimean
Peninsula or the Ukraine mainland territory than the waters situated east of the boundary

immediately south of Point X; they would equally have constituted areas facing onto the “high

seas”. Portions of the waters to the north of Point X are located a considerable distance away

from the Crimean Peninsula and the mainland coast, and if there were a need to depict the outer

limit of the territorial sea in relation to waters south of Point X, it is hard to understand why

there did not exist a corresponding need to do so along at least a portion of the 12 nm arc
around Serpents’ Island to the north of Point X.

209
210 See UCM, para. 5.153.
See UCM, para. 5.149.
1154.134. In short Ukraine’s explanation of the depiction of the boundary around Serpents’ Island

is an ex post attempt to rationalise from its own angle the situation depicted on the various

charts. It involves piling assertion upon unproved assertion. First, the Soviet or Ukrainian

authorities felt it necessary, for vaguely specified reasons, to depict the outer limit of the
territorial sea. Second, despite the overall consistency in the use of symbols on the charts,

nevertheless every one of the authorities issuing the maps and charts used the wrong symbol in

delimiting the boundary beyond Point X. Third, despite the strange course taken in depicting

the outer limit of the territorial sea at all, an eccentricity compounded by the use of the wrong

symbol, the cartographers did not carry that exercise to its logical conclusion by depicting the

outer limit of the territorial sea around the entire island and in other areas, but rather simply
stopped at Point X, which on Ukraine’s case has no relevance. Faced with this combination of

hypotheses, the attractive explanation is simply that the boundary on the various maps and

charts depicts the course of an agreed maritime boundary from the coasts of the Parties up to

Point X on the 12 nm arc around Serpents’ Island as established by the 1949 ProcŁs-Verbaux.

4.135. In analysing the use of symbols in the charts submitted by Romania, the 2001 Ukrainian

map included as RM A 23 (with enlarged detail of Serpents’ Island region as RM A 24)

deserves particular attention. It was published by the Ukrainian State Hydrographic Institution

Branch “Ukrmorcartographia”, “the only official Ukrainian institution competent to issue
211
nautical charts”. It is the only one of the 23 charts submitted by Romania which, in addition

to featuring the Romanian/Ukrainian maritime boundary up to Point X (using the red crosses-
and–dashes symbol), also depicts the seaward limit of Ukraine’s territorial sea, using the

correct symbol. According to the symbols used, from the last point of the Romanian/Ukrainian

land/river boundary up to Point X there is an international maritime boundary, while beyond

Point X to the north, continuing on the 12 nm arc around Serpents’ Island, there is the seaward

limit of (Ukraine’s) territorial sea. The clear import is that up to Point X there exists an

international maritime boundary, while north of Point X what is depicted is the outer limit of
the Ukrainian territorial sea.

(g) The assertion that none of the 23 charts and maps fixes the final point on the

semicircle around Serpents’ Island

212
4.136. This assertion of Ukraine is unsustainable. All charts introduced by Romania draw

the boundary up to a point situated due east of Serpents’ Island, situated on the 12 nm arc of

211
212 See Annex RM 86.
See UCM, paras. 5.150-5.153.
116circle surrounding it, i.e. Point X. All charts are concordant in depicting Point X in the same

position.

4.137. Ukraine argues that if the maritime delimitation agreed upon in 1949 were as Romania
contends, this would mean that:

“the Soviet Union agreed in 1949 to give up to Romania a large area of
continental shelf to which it had an automatic entitlement and a large area of
EEZ to which it would later acquire sovereign rights.” 213

Ukraine seeks to infer that the maritime areas allocated to Romania by the 1949 ProcŁs-

Verbaux (i.e. situated south and south-east to the 12 nm arc around Serpents’ Island) would

have otherwise belonged to the USSR, if the USSR had not ceded them to Romania, and that

the USSR thereby suffered a significant detriment. But of course the 1949 ProcŁs-Verbaux

reduced the breadth of the Romanian maritime areas facing Serpents’ Island to less than 10 nm,

while the maritime zone around Serpents’ Island was 12 nm in breadth. In addition, the USSR

was allocated an area of territorial sea situated south of the equidistance line between the
214
adjacent Romanian and Soviet coasts.

(h) The assertion that some of the 23 maps are self-evidently not intended to serve as
maritime boundary maps

4.138. Ukraine asserts that several of the maps submitted by Romania are “self-evidently not

intended to serve as maritime boundary maps”, given that they were published after other

international boundaries between the riparian States in the Black Sea were established, but yet
215
do not depict these boundaries. But the fact remains that, whether or not intended to serve as

maritime boundary maps, the charts submitted by Romania, whatever their origin, clearly
depict the maritime boundary established in 1949 and confirm the course and extent of the

boundary.

4.139. The very fact that the boundary is depicted upon these maps, whether or not other

existing maritime boundaries are also shown, constitutes evidence of the “point of view” of the

States producing them as to the course and extent of the boundary. Three of the maps were

produced by the Romanian authorities (Maps RM A 35, 36 and 31). This position is reflected
in various other charts dating from earlier periods. Of particular importance is the fact that the

213 UCM, para. 5.151.
214 See RM, Chapter 5, paras. 5.2-5.4.
215
See UCM, para. 5.154.
117fourth map of sufficient scale to depict other maritime boundaries (Map RM A 21, and the

enlargement of the detail around Serpents’ Island Map RM A 22) was produced by the relevant

Ukrainian authorities in 2000.

(i) The assertion that many of the 23 maps and charts are copies of or are based upon

earlier maps and have no independent significance

4.140. Ukraine asserts that many of the 23 maps are copies of or based on earlier maps and

have no independent significance of their own. 216 Indeed, it seems probable that most Soviet,

Ukrainian and even Romanian charts of the western basin of the Black Sea were based on a

Soviet chart issued in 1951, two years after the 1949 ProcŁs-Verbaux. But this does not alter

the fact that all of the charts consistently depict the maritime boundary as extending around
Serpents’ Island in the same way. Even when maps are reissued in later editions the elements

featured by the maps are normally checked for accuracy in the circumstances existing at the

time of their publication. The fact that the maritime boundary around Serpents’ Island was

depicted in the same way on all of the charts – as extending on the 12 nm arc as far as Point X–

does not mean that they are to be treated as mere repetitions of the position shown in the

original chart. On the contrary, the depiction of the boundary on a chart, even if based on an

earlier chart, is to be taken as strong evidence that its maker was of the view that it

corresponded to the legal situation, as verified by the body issuing it obtaining at the moment
of issue or reissue.

4.141. Moreover in order to criticise the inferences to be drawn from a certain body of map

evidence, it would be normal to produce contrary map evidence. Given the varied criticisms

made by Ukraine, one might have thought it would produce at least one map emanating from
Ukraine, or from its predecessor, the Soviet Union, which accorded with the position it now

takes as to the extent of the 1949 boundary.

4.142. In fact Ukraine has submitted five maps. None emanate from either Ukraine or the

Soviet Union. Three come from the same source (Turkey). None of them features a boundary

supporting Ukraine’s position. Indeed, none of them depicts any international maritime
boundary at all.217

216
217 See UCM, para. 5.155.
Some of the charts depict only the Romanian/Ukrainian river boundary, while others depict also segments
(varying in length) of the boundary separating the interior waters of the two States immediately in front of their
river boundary.
1184.143. Ukraine argues that the handful of maps it has been able to find “carry no depiction of
218
an arc to the south of Serpents’ Island” and omit “any line around Serpents’ Island”. This is

to make something out of nothing. To conclude from the fact that no boundary is depicted that

no boundary exists is absurd. The legal science does not accept negative evidence. Not only is
no boundary on the 12 nm arc around Serpents’ Island depicted but neither is the entirety of the

boundary separating the territorial seas of the two States. Nor is any other boundary in the

Black Sea shown.

4.144. In short, Ukraine has not placed before the Court a single map or chart depicting
maritime boundaries in the relevant area in a way different than the charts submitted by

Romania, still less confirming the Ukrainian position as to the course and extent of the

boundary agreed in 1949.

(3) Romania’s Map Annexed to its 1997 Notification

4.145. In its Counter-Memorial Ukraine relies on the map annexed to the Note deposited by

the Permanent Mission of Romania to the United Nations with the UN Secretary-General on 18

June 1997. In accordance with the provisions of the 1982 UNCLOS, Romania thereby notified
the coordinates of the straight baselines used by it for measuring the breadth of its territorial

sea.

4.146. Ukraine asserts:

(a) that point 2'depicted on this chart coincides with the position of Point F from

the 2003 Border Regime Treaty and the point on map 134 annexed to the 1949 ProcŁs-Verbal,

showing that no maritime boundary was established beyond this point;

(b) that beyond the outer limit of the Romanian territorial sea there is open sea; and
(c) that the chart makes no reference to any Romanian area of exclusive economic

zone or continental shelf and does not depict any boundary between the Ukrainian territorial

waters surrounding Serpents’ Island and the Romanian maritime areas beyond the territorial

sea.

219
4.147. The first assertion was already dealt with. The similar positions of point 2' and of the
point on map 134 annexed to the 1949 ProcŁs-Verbal where the drawing of the boundary stops

are purely coincidental, while the similar positions of points 2' and F are normal having in view

218
219 See UCM, para. 5.156.
See above, paras. 4.62-4.64.
119that the two points bear the same meaning and are calculated on the basis of the same

geographical situation.

4.148. As far as the third assertion is concerned, the map is attached to a Romanian piece of

legislation whose scope was to regulate the regime of the Romanian territorial sea and

contiguous zone. At the time the legal regime of Romanian maritime areas was regulated by

several laws, among which the 1990 Act notified by Romania to the UN Secretariat in 1997
referred only to the two mentioned areas. The map that accompanied this Act has consequently

to be read in this context. The purpose of the map was to illustrate the baselines established to

calculate the breadth of the territorial sea, as well as the outer limits of the Romanian territorial

sea and contiguous zone. The map did not go beyond the substantial scope of the Act to which

it was annexed: it depicts “the national maritime areas of Romania and their geographic

coordinates” which fall within the scope of the Act to which it is annexed, i.e., the Romanian

territorial sea and its contiguous zone.

4.149. At the same time, while providing a complete graphical representation of the whole
outer limit of Romania’s territorial sea and an almost complete 220representation of the outer

limit of its contiguous zone, the map does not depict the lateral maritime boundary between

Romania and its northern or southern neighbours. 221 It is not only the outer segment of the

northern lateral boundary between Romania and Ukraine established by the 1949 ProcŁs-

Verbal which is not drawn on the chart. The maritime boundary separating the territorial sea of

the two countries from the final point of the land/river border to points 1438 and 1439 is not

depicted either.

4.150. As to the second Ukrainian assertion, Ukraine affirms that the sector of arc between
points 1' (corresponding to point 1439) and 2' (corresponding to point F from the 2003 Treaty

on the Border Regime) “is depicted by a line using symbols which are the same as those used

for the line running out from the shore to point 1' , but different from the symbols used from the

outer limit running generally southwards from point 2' ”.222 The difference between the

symbols used for the segment 1' -2'and those used for the stretch south to point 2'is easily

explainable: the first segment delimits the Romanian territorial sea from the Ukrainian

220 The chart used to depict the two Romanian maritime areas represents only a limited area of the western
part of the Black Sea. The northern sector of the outer limit of the Romanian contiguous zone passes over the edge
of the chart.
221 These boundaries are not depicted by lines but their position is readily visible as a consequence of the
different colours used to mark maritime areas with a different regime: see below, paras. 4.152-4.153, 4.155
222
UCM, para. 5.125.
120territorial sea around Serpents’ Island, the second is nothing more than the outer limit of the

Romanian territorial sea, beyond which there are areas representing the contiguous zone of

Romania, as well as its exclusive economic zone and continental shelf. But the suggestion that

the symbols used for the segment 1' -2'and for the “segment uniting the shore to point 1' ” are

the same is surprising. The chart does not depict in symbols any lateral boundary (of which the

segment uniting the point 1'to the shore would have been a part). But even if the chart had

depicted the lateral Romanian-Ukrainian maritime boundary, the symbols used would have
been the same, taking into account the same character of the two “segments”.

4.151. Ukraine notes that the line southwards to point 2'is “a boundary of Romania’s

sovereign territorial sea, not a boundary between Romanian and Soviet/Ukrainian territorial
223
sea”. This is correct: eastwards of the line on the Romanian map uniting points 2' to 9'there

is no Ukrainian territorial sea but Romanian maritime areas (contiguous zone, exclusive

economic zone and continental shelf). The chart depicts not only the outer limit of the

Romanian territorial sea but also the outer limit of Romania’s contiguous zone. The chart

actually transmitted by Romania to the UN Secretariat features different colours to indicate the
different Romanian maritime areas it depicts (internal waters, territorial sea and contiguous

zone). Unfortunately, the reproduction included in Annex 41 of Ukraine’s Counter-Memorial

is in black and white, while the reproductions included as figures in the Counter-Memorial

feature only yellow nuances for all the maritime areas beyond the baselines. 224

4.152. An accurate reproduction of the chart is presented in this Reply as Figure RR23 below.

Romania has deposited with the Registrar copies of this chart in its original size. As can be

seen, internal waters are depicted in blue, the territorial sea in light pink, the contiguous zone in

light green. Yellow is used to indicate maritime areas situated beyond the Romanian territorial
sea or contiguous zone, whether they are Romanian maritime areas situated beyond the outer

limit of the contiguous zone or maritime areas belonging to other States. 225

223 UCM, para. 5.125.
224 See UCM, Figures 5-4, 5-5, 5-13, 5-14.
225 As mentioned above in para. 4.148, the representation of exclusive economic zones and continental shelf
were outside the scope of the Romanian Act notified to the UN Secretariat, to which this chart is annexed.

121 Figure RR23

Accurate representation of the map annexed to the 1997

Romanian notification to the UN
The different Romanian maritime areas, as well as the boundary
between the Romanian areas and the Ukrainian ones, are clearly
indicated by the various colours used

1224.153. As the contiguous zone is within the scope of the Act to which the chart is attached, it is

accurately represented in a separate colour. Because of the limited area of the Black Sea which

the chart covers, the northernmost sector of the outer limit of the Romanian contiguous zone is

not drawn; still, the extent of this Romanian area (represented on the map in light green) is

clear and leaves no doubt regarding its geographic position: in front of the Romanian territorial
sea, including immediately south to the 12 nm arc of circle surrounding Serpents’ Island, in

conformity with the maritime boundary established by the 1949 ProcŁs-Verbal. Although the

chart does not depict lateral boundaries, the difference of colour is apparent: the yellow

indicating third State maritime areas is situated north of the 12 nm arc of circle surrounding

Serpents’ Island, while the light green indicating the Romanian contiguous zone is immediately

south of this arc.

4.154. The 1997 Map together with the Romanian law, were deposited with the UN Secretary-

General, and were registered by the Division for Ocean Affairs and the Law of the Sea of the

Office of Legal Affairs under reference M.Z.N. No. 15.1997.LOS of 7 August 1997. Both Law

no.17/1990 and the accompanying 1997 Map were given due publicity; the Law was published
in the Law of the Sea Bulletin shortly after its adoption (albeit with some errors, in particular in

relation to the coordinates of the basepoints). 226 The illustrative map was reproduced in the

Law of the Sea Information Circular. 227 The map featured in the Law of the Sea Information

Circular is reproduced as Figure RR24 at page 124 of this Reply. As seen from this figure, the

outer limits of Romania’s maritime zones covered by the 1990 Law are clearly indicated.

4.155. The 1997 Map clearly depicts the boundary between the Romanian territorial sea

(coloured in light pink on the map) and the Ukrainian maritime areas (coloured in yellow) in

full conformity with the provisions of the 1949 ProcŁs-Verbaux. In addition, the 1997 Map

shows the boundary between the Romanian contiguous zone (coloured light green) and the

Ukrainian maritime areas to the north of it (coloured yellow). The boundary is clearly visible,

constituting the continuation of the 12 nm arc of circle around Serpents’ Island to the east of

Point 2'.

226 DOALS/OLS, Law of the Sea Bulletin, No. 19 (October 1991), at pp. 9-20.
227 DOALS/OLS, Law of the Sea Information Circular, No. 9 (February 1999) at pp. 46-47.

123 4
1

e
t
i
d 7
c .
d ,
r 9
e 9
s y
, a
s r
i e
l (
e 9
4 i o
2 ir ,
R a a
e M u
u a i
i n C
F a o
o a
e m
t f
f I
p a
a S
e e
i f
a o
s a
l L
e
h
T4.156. In the text of its Counter-Memorial Ukraine analyses the 1997 Map in monochromatic

detail, including no less than four reproductions of it (Figures 5-4, 5-5, 5-13 & 5-14).

Although analysing almost all elements of the 1997 Map, Ukraine does not refer at all to the

map’s representation of the Romanian maritime areas. The inference might be that Ukraine

does not contest the manner in which these areas are depicted. And this is consistent with its
earlier conduct: at the time of its publication, Ukraine did not react in any way to the 1997 Map

and its representation of Romanian maritime areas, including the outer limits of those areas in

the region of Serpents’ Island and the lateral boundaries indicated by the differences in colours

used to depict these various maritime areas.

C. CONCLUSIONS

4.157. To summarise, the following conclusions may be drawn from this discussion of the

agreements and maps and of Ukraine’s treatment of them:

(1) Ukraine’s thesis of an agreed maritime boundary only up to Point F fails
entirely. So do the eight arguments marshalled by Ukraine against Romania’s

demonstration of an agreed boundary around Serpents’ Island.

(2) Despite Ukraine’s criticisms, the cartographic evidence introduced by Romania

is relevant and reliable and has probative value as to the views of the makers of those

charts (including the Parties) as to the course and extent of the boundary agreed in

1949. The charts are consistent in depicting, through appropriate internationally-

accepted symbols, the maritime boundary from the last point of the Romanian-

Ukrainian land/river boundary, passing through border signs 1438 and 1439, and then

passing on the 12 nm arc of circle around Serpents’ Island, up to a point situated almost
due east of it (Point X). Not only do all of the charts confirm the course and extent of

the boundary but they are also clear as regards the appurtenance of the maritime areas

to the north and south of that boundary. Moreover many of these charts were produced

by the USSR and Ukraine themselves.

125(3) By contrast there is no cartographic evidence – in particular none of Russian or
Ukrainian provenance – supporting Ukraine’s thesis of an agreed maritime boundary

stopping at Point F.

(4) The map notified by Romania to the UN Secretary-General in 1997 (especially

when viewed in its original colours) is entirely consistent with Romania’s position as to

the maritime boundary around Serpents’ Island. Ukraine’s criticisms of that map are

based on misleading copies and likewise fail entirely.

126 CHAPTER 5

SERPENTS’ ISLAND IN LAW AND FACT

A. INTRODUCTION

5.1. In its Memorial, Romania gave an extended presentation of Serpents’ Island,

demonstrating its characteristics as a rock unable to sustain human habitation or

economic life of its own. Romania based its presentation on sources dating from the

period before World War II, as well as on Soviet, Ukrainian and third country authors. In

its Counter-Memorial, Ukraine does not question these sources and is silent as to most of

the conclusions drawn from them.

5.2. Indeed Ukraine begins its Counter-Memorial with the assertion that the impact of

Serpents’ Island “is not the central issue in this case”. 228 But this protestation is

contradicted by the extensive attention subsequently paid to Serpents’ Island and its

status.229 Ukraine argues at length that Serpents’ Island does not fall within Article

121(3) of the 1982 UNCLOS on the basis that it is able to support human habitation and

an economic life of its own. In short it seeks to demonstrate the character of this maritime

feature as a fully-fledged centre of human and economic activity. 230

5.3. The information on which Ukraine bases its argument does not support its

conclusions. Indeed a careful reading of the Counter-Memorial and its Annexes reveals

much information supporting the contrary view, as well as a good deal which is irrelevant

to the issue. The Addendum to this Chapter presents substantial evidence from the

Annexes of the Counter-Memorial and from the Ukrainian mass-media proving that, in

reality, Serpents’ Island falls under the parameters provided for under Article 121(3) of

the 1982 UNCLOS.

228 UCM, para. 1.6.
229
Under the title “Serpents’ Island is Necessarily One of the Basepoints”, Chapter 7 devotes 53
paragraphs (paras. 7.35-7.88) out of 60 to the argument that “Serpents’ Island is an ‘Island’, not a ‘Rock’”
and that “Serpents’ Island is a Significant Island”. Half the UCM annexes (56 out of 112) are concerned
with Serpents’ Island.
230 See, for instance, UCM, paras. 7.86, 7.88.
1275.4. Amongst the latter category are arguments seeking to show the strategic

significance of Serpents’ Island. 231 But there is no connection between the status of an

island entitled to continental shelf and exclusive economic zone and its strategic

significance. Depending on their location rocks may be strategically significant while

large islands may have little or no strategic significance. It is true that Serpents’ Island

had and continues to have strategic importance, which explains its tumultuous history,

including its seizure from Romania in the circumstances mentioned in the Memorial. 232

This is also why it is mentioned on most maps of the Black Sea, another point

emphasised by Ukraine. 233

5.5. The existence of Serpents’ Island, its presence on maps, its past (or present)

strategic significance are not arguments in favour of an entitlement to continental shelf

and exclusive economic zone; they are not indicators of human habitation or economic

life. The military status it had for more than half a century, involving a small military

contingent on a rotational basis and entirely dependent on external supply, was the most

this rock could offer in terms of “human habitation” or “economic life”.

234
5.6. Ukraine devotes much space to the historical importance of Serpents’ Island.

Again there is no connection between the history of a maritime feature and its status as an

island entitled to continental shelf and exclusive economic zone. In fact virtually all

historical sources note its lack of any capacity to sustain human habitation and economic

life of its own. They show the hostile and arid conditions that always impeded normal

human habitation and any kind of economic activity. The ancient sources, as well as the

historical and archaeological studies (of Russian and, especially, Ukrainian origins)

quoted in the Counter-Memorial do little but stress the religious and sacred significance

of Serpents’ Island, linked to the legend of Achilles, or its occasional use by sailors in

difficulty due to the stormy weather.

231 See UCM, paras. 3.32-3.34.
232 See RM, paras. 3.1-3.32. See also Appendix 1 of this Reply.
233
UCM, para. 3.32 asserts that “[c]learly, Serpents’ Island is an important feature” because “every
single one of the maps contained in Romania’s Map Atlas identifies and depicts this island”. There is, of
course, another series of maps of the Black Sea without a representation of the Serpents’ Island – see, for
instance, the maps included by Romania as Annex RR 5. Following the line of interpretation of the
Counter-Memorial, the absence of a representation of this maritime feature on those maps would imply that
it never existed in the Black Sea.

1285.7. There has never been any domestic settlement on the island. The only relics

found are of a religious kind, such as tribute or sacrifice pottery or objects such as coins
235
offered to Achilles. No trace of domestic settlements has been discovered. By contrast

a large number of anchors have been retrieved around Serpents’ Island. 236 Far from

indicating economic or commercial activity, they are a testimony to the large number of

shipwrecks. 237 In short it was a home only for gods and serpents. 238

5.8. So much for the irrelevant arguments deployed in the Counter-Memorial. But

Ukraine also addresses the issues made relevant by the principle embodied in Article

121(3), i.e. whether Serpents’ Island can sustain human habitation and an economic life

of its own, both as a matter of the interpretation of this provision and – at greater length –

as a matter of fact. The interpretative issues are dealt with in Section B of this Chapter

(paragraphs 5.9-5.26), the factual ones in Section C (paragraphs 5.27-5.106). These

arguments will be discussed as far as possible using the material provided in Ukraine’s

own annexes.

B. UKRAINE’S INTERPRETATION OF ARTICLE 121(3)

5.9. The Counter-Memorial offers a minimalist interpretation of Article 121(3), which

239
is said to be “vaguely worded” and “not without difficulties”. Likewise Ukraine

asserts that the terms used – “human habitation” and “economic life of its own” – have
240
“considerable ambiguity”. But paragraph 3 uses ordinary words in their ordinary

sense, and does not raise special difficulties of interpretation. Moreover it is an inherent

part of a provision in a major multilateral convention which must be given its due effect.

234 See UCM, paras. 7.50–7.71, plus ca 23 Annexes.
235
See UCM, Annex 11, Appendix to letter No. 11/295 of the Ukraine National Academy of Science
dated 30 March 2005. According to this Annex, the artefacts discovered on Serpents’ Island were not the
result or an indication of human habitation or economic life on that territory, but they were only brought
there occasionally for religious purposes.
236 See UCM, para. 7.53 and Annex 56, S.B. Ochotnikov & A.S. Ostroverkhov, “Anchorage of

Achilles’ Sanctuary on Leuke (Zmiinyi) Island”, in Archeology Science Journal, Ukrainian Academy of
237ence – Institute of Archeology, Kyiv, 2002, No. 2.
See RM, para. 10.88 & Annex RM 5.
238 Even the serpents which gave the Island its name are now extinct, according to UCM, Annex 85,
List of Island’s Flora, Fauna and Fish Species.
239 See UCM, para. 7.38.
240
UCM, para. 7.41.
1295.10. In short, paragraph 3 defines an exception in respect to the legal situation set forth
241
in paragraph 2. Its purpose is to institute an exceptional regime for very small
242
islands. Their legal situation is defined by three elements: the feature must be

considered as falling into the class of “rocks”, and it must be incapable of sustaining

human habitation or an economic life of its own.

(1) The meaning of “rock” in the context of Article 121

5.11. Ukraine starts with an evidently incorrect assertion:

“To some extent Serpents’ Island’ s status as an island is the counterpoint
to it not being a rock as the term is understood in Article 121(3) of the
243
UNCLOS.”

Yet it subsequently comes to the opposite conclusion: “a rock is therefore implicitly a

kind of island”, 244and correctly so. All natural features permanently above the sea at

high tide are islands, whether large or small, wet or dry, arid or fruitful. But some islands
245
are rocks covered by Article 121(3), with the consequences there stated. Indeed
246
Article 121 says so explicitly.

5.12. Ukraine accepts that the word “rock” must bear “its ordinary meaning in its

context and in the light of the object and purpose of UNCLOS”. It attributes to Romania

the argument that “since Serpents’ Island is composed of rocky material it must be

regarded as a ‘rock’, but this cannot be so since most islands are to a greater or lesser
247
extent composed of rocky material…”

5.13. For its part Romania accepts that the size of a maritime feature is relevant to its

classification as a “rock” for the purposes of Article 121(3) and it does not confuse the

241
See RM, para. 10.3.
242 See, for a detailed presentation, RM, paras. 8.6-8.19.
243 See UCM, 7.38.
244
245 UCM, para. 7.39.
246 RM, para. 10.3.
But see UCM, para. 7.39: “… a ‘rock’ must be understood… in contradistinction to an ‘island’
which is an altogether larger and more substantial feature.” This is not what Article 121 says. The title of
Article 121 is “The regime of islands”. Had the intention been to consider “rocks” as something different
than “islands”, the title of the article would necessary have been “The regime of islands and rocks”.
247
UCM, para. 7.39.
130rocky composition of a feature with its legal status. 248 The purpose of the demonstration

in the Memorial was to show that Serpents’ Island is not entitled to exclusive economic

zone or continental shelf because it is a very small rocky island which cannot sustain

human habitation or an economic life of its own. But there is no basis in the text or the

travaux of paragraph 3 to hold that it only applies to rocks which are of minimal area. If

this had been intended, the further qualifications in paragraph 3 – concerning human

habitation and economic life – would have been unnecessary. No tiny isolated rock could

do either of these things.

(2) The requirement of “human habitation”

5.14. Ukraine asserts that “‘human habitation’ is not the same as a permanent resident
249
population”. But this ignores the point that paragraph (3) is concerned with the

capacity of an island to sustain a population. Whether the population is in fact itinerant is

not to the point. For example a fishing community may inhabit an island during the

fishing season over generations, gaining their living from it: this would qualify as human

habitation.

5.15. The normal sense of the term “human habitation” is that of a stable community, a

local population, perhaps small but nonetheless sustainable across time. People have

camped on Rockall and stayed for days on the moon, but no one would say that either

place is capable of sustaining human habitation. 250 The notion of human habitation does

not extend to persons sent to an island temporarily for military service in the Army or the

border guard, or to maintain a lighthouse. (Many lighthouses are located on what are

indisputably rocks.) Such persons are not inhabitants of the island; they are ordered to go

there and have neither the right nor the desire to stay after their term is up.

248 For the purposes of the present case it is not necessary for the Court to decide if the term covers
small sandy features. The language of Art. 121(3) is clear in covering, at least, small rocky islands.
249
250 UCM, para. 7.41.
The United Kingdom conceded as much in relation to Rockall at the time it ratified the 1982
Convention: see 299 House of Commons Debates, Written answers, col. 482, 31 July 1997, reprinted in
(1997) 68 BYIL pp. 599-600.
131 (3) The requirement of an “economic life of its own”

5.16. Ukraine also states that “‘economic life’ is not the same as viability as an

independent, self-contained and self-sufficient economy involving the development of
251
natural resources”. In doing so it ignores the “ordinary meaning to be given to the

terms of the treaty” and it is not a “good faith” interpretation; indeed it deliberately

ignores the essential qualification of the economic life: this must be economic life “of its
252
own”. That means that the natural conditions on a feature such as Serpents’ Island

must support the development of economic activities. Such conditions cannot be

artificially created or injected from the mainland: although links with the mainland are

permissible, nonetheless the economic life must be real and not contrived, local and not

merely imported. 253

5.17. Ukraine perverts the sense of this term: it relies on activities directed exclusively

from the mainland, aimed at artificially creating an appearance of economic life, in order

to prove the existence of that very same economic life. This is, of course, unacceptable

and cannot be supported by a good faith interpretation, taking into account the ordinary

meaning to be given to the terms of Article 121(3).

5.18. Furthermore, this interpretation is contrary to the meaning of the Russian official

text of the formula “of its own” in Article 121(3), which can be literally translated as
254
“self-sustaining economic activity”.

5.19. Ukraine argues that the two conditions – human habitation and an economic life
255
of its own – “are to be read conjunctively”, even if the text of paragraph 3 uses “or”.

In other words it is sufficient that a rock either sustain human habitation or an economic

life of its own; both are not required. It may be accepted that the two conditions are

interconnected: it is difficult to conceive an economic life without social life, especially

since the economic life has to be an economic life of the rock itself. In fact, economic

251 UCM, para. 7.41.
252 As Kolb, “L’interprØtation de l’article 121, paragraphe 3, de la Convention de Montego Bay sur le
droit de la mer: les « rochers qui ne se prŒtent pas à l’habitation humaine ou à une vie Øconomique
propre… » “, AFDI, XL, 1994, p.907 states: “Ceci est un critŁre essentiel.”
253
254 Kolb, op.cit., pp. 907-908: “L’autonomie ne doit pas Œtre totale, mais la dØpendance non plus.”
255 This argument, put forward in RM, para. 8.18, is not addressed in UCM.
UCM, para. 7.40.
132life is the expression of human habitation; also, without human habitation, there cannot

be an economic life. 256

(4) Romania’s declaration relating to Article 121

5.20. In interpreting Article 121, of relevance is also the Declaration made by Romania

upon its signature and ratification of the 1982 UNCLOS. Paragraph 3 of the Romanian

Declaration with respect to Article 121 concerning the regime of islands provides that:

“the uninhabited islands without economic life can in no way affect the delimitation of

the maritime spaces belonging to the mainland coasts of the coastal States”. 257 Neither

the former USSR nor Ukraine made any reaction to that Declaration either at the time it

was made or subsequently.

5.21. Ukraine seeks to minimize the significance of the Declaration and of its failure to

object, on the basis that “there is no need to object to a declaration” and that “no legal

effect for a State can be derived from its abstention to do so”. 258

5.22. It is not suggested that unilateral declarations, not amounting to reservations, are

submitted to the general rØgime applicable to reservations under Article 20(5) of the 1969

Vienna Convention, nor is it suggested that the Romanian Declaration constituted a

reservation. Reservations to the 1982 UNCLOS are not permitted. Nonetheless,

declarations made on becoming a party to a multilateral treaty constitute a statement of

position of the declaring State to which objections may be expected to be made by any

other State specifically affected. An appropriate inference that may be drawn from the

failure to object is that the State in question accepts the point being made, or at least that

it does not object to it. Indeed (though Ukraine carefully overlooks this point), in

particular situations a State can by its inaction create a situation of acquiescence with

respect to the new situation created or interpretation proposed unilaterally by another

256 See J.M. Van Dyke, R. Morgan, J. Gurish, “The Exclusive Economic Zones of the Northwestern

Hawaiian Islands. When Do Uninhabited Islands Generate an EEZ?”, (1988) 25 San Diego Law Review p.
437; B. Kwiatkowska & A. Soons, “Entitlement to Maritime Areas of Rocks which Cannot Sustain Human
257itation or Economic Life of Their Own”, (1990) 21 NYIL p. 365.
RM, paras. 8.20-8.30.
258 UCM, para. 7.31.
133State. In the Temple case, the Court clearly underlined that States have a duty to react to

acts encroaching upon what they consider to constitute their rights:

“the circumstances were such as called for some reaction, within a
reasonable period, on the part of the Siamese authorities, if they wished to

disagree with the map or had any serious questions to raise in regard to it.
They did not do so, either then or for many years, and thereby must be
held to have acquiesced. Qui tacet consentire videtur si loqui debuisset ac
259
potuisset.”

5.23. In the Anglo-Norwegian Fisheries case, the Court regarded the absence of protest

by the British authorities with regard to the unilateral enforcement of Norway' s base line

system to be decisive. It considered that:

“her prolonged abstention would in any case warrant Norway’s
260
enforcement of her system against the United Kingdom.”

5.24. The Arbitral Tribunal in the Dubai-Sharjah Border Arbitration, after having
261
quoted these and other relevant decisions, came to the same conclusion:

“…a State must react, although using peaceful means, when it considers

that one of its rights is threatened by the action of another State. Such a
rule is perfectly logical as lack of action in a situation like this can only
mean two things: either the State does not believe that it really possesses

the disputed right, or for its own private reasons, it decides not to maintain
it.”262

5.25. The USSR and in its turn Ukraine had several opportunities to react to the

Declaration made by Romania on 10 December 1982 and confirmed upon ratification on

17 December 1996. On each of these occasions, they failed to do so. In particular there

was no reaction when in 1996 Romania confirmed the Declaration in its instrument of

ratification, although Ukraine was at that time a signatory to the 1982 UNCLOS. Further,

Ukraine made no comment relating to the Declaration when concluding the 1997

259 Temple of Preah Vihear (Cambodia v. Thailand), ICJ Reports 1962, p 6, at p. 23.
260
261 Fisheries (United Kingdom v. Norway), ICJ Reports 1951, p. 116, at p. 139.
Grisbadarna case, Arbitral Award of 23 October 1909, RIAA, vol. XI, p. 147, at p. 161; Island of
262mas, Arbitral Award of 4 April 1928, RIAA, vol. II, pp. 829, 851, 868.
Dubai-Sharjah Border Arbitration, Award of 19 October 1981, 91 ILR, p.543, at p. 623.
134Additional Agreement which expressly referred to Article 121. Nor did it comment on

the Declaration when ratifying the 1982 UNCLOS on 26 July 1999. 263

5.26. These repeated silences are the more eloquent in that Ukraine was well aware of

the situation of Serpents'Island and the potential impact of the Declaration. In 1997 it

acknowledged expressly the application of Article 121 in full knowledge of the

Romanian interpretation of this provision. This is a clear situation where Ukraine should

have objected to the Declaration. Its failure to do so should be taken into account, in case

of doubt, in applying Article 121(3) to the present situation.

C. SERPENTS’ ISLAND MEETS ALL THE CRITERIA OF ARTICLE

121(3)

5.27. In its Memorial Romania cited an extensive range of materials describing the

reality on Serpents’ Island during the period before 1948, as well as statements from the

Ukrainian media expressing the position of the Ukrainian authorities in connection with
264
the conditions on Serpents’ Island. The Ukrainian Counter-Memorial does not contest

the Romanian sources or the authenticity of the Ukrainian public statements. Instead it

seeks to stress the historic and present-day significance of Serpents’ Island. But as this

Section as well as the Addendum to this Chapter will show, Ukraine’s own evidence

reinforces the conclusion that Serpents’ Island is a rock which cannot sustain human

habitation or an economic life of its own.

(1) Serpents’ Island is a very small rocky island 265

5.28. It is clear that Serpents’ Island qualifies as a rock from the geological point of
266
view. Because it is a natural rocky feature of small size, Serpents’ Island is to be

considered a “rock” within the meaning of the term in Article 121(3).

263
This continued silence is in contrast with the fact that Ukraine normally reacts promptly,
especially in border-related issues. Cf. UCM, Annex 46, Statute of Ukraine Concerning the State Frontier
dated 4 November 1991, Bulletin of the Verkhovna Rada, 1992, No.2, p.5 et seq.), “Any violations of the
State frontier will be firmly suppressed.”
264 See RM, Chapter 10.
265
See also RM, paras. 10.12-10.27. For more evidence in this respect, see the Addendum to this
266pter, paras. 5.109-5.129.
See inter alia RM, paras. 10.12-10.27.
1355.29. In its quest to demonstrate that “Serpents’ Island” is an “island” as opposed to a

“rock”, Ukraine refers to its name: Serpents’ Island was known under that name since

ancient times (and was never named “Serpents’ Rock” or “Serpents’ Islet”). 267 But the

status of a rock under Article 121(3) depends on its natural characteristics, not on its

name. For the purposes of the law of the sea the terms “island” and “rock” are not

disjunctive – “rocks” are a species of island. The position may be different historically:

as, for example, in the “Rock of Gibraltar” or the “Rocher de Monaco”, where the term

“rock” denominates an area of mainland.

5.30. Issues of nomenclature aside, Ukraine does not deny the rocky character of

Serpents’ Island. For example Annex 9 to the Ukrainian Counter-Memorial mentions the

sterile reality on the Serpents’ Island:

“The island is a piece of land 268h steep shores. ... It features rocky ground
and poor grassy vegetation.”

5.31. According to the opinion of Ukrainian National Academy of Science on the

matter:

“…in fact Zmiinyi Island [is] … a small (with an area of around 18

hectares) rocky plot of land... situated lonely amidst the immense water
space.” 269

5.32. Ukraine annexes extracts from the Black Sea and Sea of Azov Sailing Directions

(St. Petersburg, 1903) and Black Sea Sailing Directions (Leningrad, 1931) which are in

the same sense:

“Fidonisi or Zmeinyi Island, Ilan-Adasy in Turkish, lying 24 miles to OtN
(sic) of Sulina, is rocky, broken-off and raised in the center. The shape of
the island is irregular ...”70

267
See UCM, para. 7.35.
268 UCM, Annex 9, Information Report on Research of Drinking Underground Waters on Zmiinyi
Island in 2002-2004, Dnipropetrovsk, 2004. This also contains a quite detailed description of the rocky
geology of Serpents’ Island.
269 UCM, Annex 11, Appendix to letter No.11/295 of the Ukraine National Academy of Science, 30

270ch 2005.
UCM, Annex 12, Black Sea and Sea of Azov Sailing Directions, Fourth Edition, St. Petersburg,
1903, pp.64-65.
136 “Fidonisi or Zmeinyi Island, Ilan-Adasy in Turkish, lying 24 miles to OtN

(sic) of Sulina; it is small ...; the shores thereof are rocky and broken-
off.”271

5.33. The Ukrainian scientist S.B.Okhotnikov is of the opinion that

“It is a small island measuring only 18 ha in area. Its surface is rocky and
its steep shores drop from 4 to 25 meters to the sea..... Rufius Festus
Avienus, an author of the 4 th c BC, provides a remarkably accurate

description of the island. He wrote, ‘Leuke is gray (in color) from its
mountains…where a hollowed cliff forms a series of curved roofs along its

broad face, where cliffs with curved roofs ha272over the sea.’... A small
piece of land, in the middle of the sea ...”

5.34. According to an authority published by the Russian Imperial Academy of

Sciences in St Petersburg:

“L’île de Leuce n’est qu’un seul bloc de roche calcaire; ses bords sont

ØlevØs, escarpØs et perpendiculaires en plusieurs endroits .... Depuis le
sommet de ses bords escarpØs le reste de Leuce n’est qu’un rocher
273
convexe ....”

5.35. A Ukrainian scientific expedition which visited Serpents’ Island in August 2003

was struck by the same arid appearance:

“Serpents’ Island is a rocky formation ... [T]he soil on the island is very-
very poor and represents the product of the erosion of different types of

rocks, or as the soil specialists name it, alluvial stratum. The main
vegetation of the island – wild herbage.” 274

5.36. Ukraine tries to assert that the inter-war pictures in the Romanian Memorial (for
275
instance, Figures 15, 16, 17) “convey a misleading image as to its physical conditions”.

To the contrary, the evidence of the Counter-Memorial is to the same effect: see Photos

271
272 UCM, Annex 13, Black Sea Sailing Directions, Seventh Edition, Leningrad, 1931, pp.47-48.
UCM, Annex 48, Ancient Greek Sites on the Northwestern Coast of the Black Sea, Hellenic
Foundation for Culture, Odesa Branch.
273 UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes a Achille dans le Pont-
Euxin’ and attached map ‘carte de l’île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie
ImpØriale des Sciences de St. Petersbourg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
274
275 See the Ukrainian site http://www.astronomy.odessa.ua/zmej.htm, last visited on 17July 2006.
UCM, para. 3.45.
137 276
A, B, C between pages 22 and 23 and those included in Annex 72. They are no

different to those taken by Romanian visitors on Serpents’ Island in the inter-war period.

Picture taken by the Ukrainian
scientific expedition that visited

Serpents’ Island in August 19-23,
2003; the rocky image is identical to
those in the pictures included in the
1931 volume of R.I. C▯linescu, Insula

▯erpilor. Schi▯▯ monografic▯”(see
RM, Figure 13).

Picture taken by the Ukrainian
scientific expedition that visited

Serpents’ Island in August 19-23,
2003; the rocky image is identical to
those in the pictures included in the
1931 volume of R.I. C▯linescu,

“Insula ▯erpilor. Schi▯▯
monografic▯”(see RM, Figure 15).

5.37. In fact, the same suggestive images can be found in the pictures included in a

promotional material on Serpents’ Island, publicly available at the exhibition organised
rd 277
on the occasion of the 3 Investment and Innovations Forum (Odessa, 1–3 June 2006).

They coincide with the images of the pictures inserted in the Romanian Memorial and

considered as “misleading” by the Ukrainian Counter-Memorial and also with the first

two photos found on the Ukrainian site mentioned above. The fact that the Ukrainian

authorities included them in an official exhibition presided over by the Interim Governor

276
See also the Ukrainian site http://www.astronomy.odessa.ua/zmej.htm, last visited on 17 July
2006. Two of the pictures posted there are reproduced below on page 139 of this Reply.
277 See Promotional materials on Serpents’ Island, publicly available at the exhibition organised on
the occasion of the 3 Investment and Innovations Forum (Odessa, 1 – 3 June 2006) (Annex RR 6).

138of the Odessa Region indicates that they may be considered to be representative of the

reality of Serpents’ Island.

Pictures included in a promotional material on Serpenrd’ Island, publicly available
during the exhibition organised on the occasion of the 3nvestment and Innovations
Forum (Odessa, 1-3 June 2006). They feature the same rocky appearance.

5.38. Serpents’ Island is not only rocky and sterile but also very small. According to

the Romanian scientists that studied it before World War II, it has an area of 17 hectares

139 278
(and a natural tendency to erosion). Although Ukrainian figures vary, most of its
279
annexes mention 17 or 18 hectares.

5.39. Precise calculations are not to the point: as a 19 century MØmoire sur les îles et

la course consacrØes à Achille dans le Pont-Euxin notes, authors who wrote about

Serpents’ Island in antiquity exaggerated its dimensions; in truth Serpents’ Island is “un
280 th
rocher solitaire et isolØ”. Another traveller of the 19 Century, E.D. Clarke, wrote:

“It is so small, that as we passed, we could view its whole extent .... It is
quite bare covered with a little grass, and very low herbage... A secluded
281
spot...”

Yet another book of the 19 century speaks about “la petite île des Serpents”. 282 And

according to the Hydrographic Agency of the Military Navy of the USSR “[t]he island is
283
small and bordered with a ridge of submerged stones”.

5.40. It is not surprising then that the Odessa Regional State Administration in 2004

stated that:

“In view of the small size of the island and extreme vulnerability of its
284
ecosystem there is no possibility to develop mass uncontrolled tourism.”

On 3 May 2006, the head of the Odessa Regional Department for ecology and natural

resources declared that:

“... in the narrowest parts of Serpents’ Island the width is not bigger than

60 metres, and unless urgent measures are taken in order for the shores to
be consolidated, this spot of land might seize to exist as a single unit”. 285

278 See RM, paras. 10.5-10.6, 10.21-10.27.
279
UCM, para. 3.39 fudges the issue (“about 20 hectares”). See also UCM, Annexes 10, 60 (17 ha);
11, 48 (18 ha); 7 (20.5, but also 19.6 ha); 67 (19.6 ha).
280 UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes a Achille dans le Pont-
Euxin’ and attached map ‘carte de l’ île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie

281Øriale des Sciences de St. Petersbourg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
UCM, Annex 50, E. D. Clarke, Travels in Various Countries of Europe, Asia and Africa, Third
Edition, Printed for T. Cadell and W. Davies Strand, London, 1813, pp. 648-650. This was also quoted by
Romania in Annex RM 47. This description of 1813 corresponds closely to that portrayed in the 2003
pictures taken by Ukrainian scientists.
282
UCM, Annex 54, E. Taitbout de Marigny, Portulan de la Mer Noire et de la Mer d’Azov ou
Description des côtes de ces deux mers à l’usage des navigateurs, Odesa, 1830, pp. 112-114.
283 UCM, Annex 14, Black Sea Sailing Directions, 1954, p.76.
284 UCM, Annex 81, Direction No. 535/a-2004 of the Odessa Regional State Administration, 12 July

2004.
1405.41. The same perception as to the smallness of the Serpents’ Island can be read in the

Ukrainian press. An article titled “You win Serpents’ Island or you will be defeated in the

fight for it”, published in February 2006 in the weekly newspaper “Argument. Vlada”,

refers to Serpents’ Island as “small piece of land in the sea” and mentions that “[i]n fact,

the importance of this island-rock in the history of the Ukrainian-Romanian diplomacy
286
proved to be bigger than its surface.”

5.42. A satellite picture of the Danube Delta, enclosed in this Reply, shows very clearly

that the smallness of Serpents’ Island is not only a subjective perception, but a reality.

The rocky protuberance can hardly be seen in the upper side of the picture, its

insignificance in the context of the geographic configuration of the area being more than

evident.

285 Article titled “Recycling the garbage – one of the biggest problems on Serpents’ Island”,
published by the Ukrainian Press Agency “Reporter” on 3 May 2006 (Annex RR 7).
286
Article titled “You win Serpents’ Island or you will be defeated in the fight for it”, published by
the Ukrainian weekly newspaper “Argument. Vlada”, the issue of the week 10 – 16February 2006; author
Stella Maks (Annex RR 8).
141Satellite picture of the Danube Delta, publicly available on the Internet site
http://earth.jsc.nasa.gov/sseop/EFS/lores.pl?PHOTO=NASA7-720-6. The smallness of Serpents’ Island is
obvious – it is insignificant in the context of the geography of the area.

5.43. Ukraine compares Serpents’ Island and various other islands. 287 In fact the

examples invoked are islands of considerable areas if compared to Serpents’ Island: they

2
range from Pitcairn Island which has an area of 4.53 km (approximately 26 times more
2
than Serpents’ Island) to South Sandwich Islands (311 km , 1,829 times larger than
2 2
Serpents’ Island), Crozet Island (520 km ), South Georgia Islands (3,590 km ) and

287 UCM, para. 7.37.

142Kerguelen Islands (7,000 km ). 2 The latter is more than 41,000 times larger than

Serpents’ Island.

(2) Serpents’ Island is devoid of water resources and practically devoid of

soil, vegetation and fauna 288

5.44. There are other parameters which underline the fact that Serpents’ Island cannot

sustain either human habitation or an economic life of its own. Key amongst them is

fresh water.

5.45. The Counter-Memorial asserts the existence of fresh water on Serpents’ Island. 289

It relies on two arguments in this respect: two Romanian books of 1925 and 1931 and the

fact that in 2003 the Ukrainian specialists discovered “fresh water” after drilling in the

rocky soil of Serpents’ Island.

5.46. According to Ukraine:

“The existence of wells and cisterns was recorded by Romanian writers in
1925 and 1931. Writing in 1925, the Romanian author Popa-Lisseanu

mentioned two fresh water wells located on Serpents’ Island and recalled
that the Roman historian Ammianus Marcellinus had stated about the
island: ‘Ibi et aquae sunt’ (i.e., where there is also water), thus indicating

the presence of fresh drinking water, rather than salt water. The same
author mentions the existence of two water reservoirs, one of which was
situated in the vicinity of the temple of Achilles.”290

5.47. The translation provided by Ukraine in Annex 8 is inaccurate; so is its

interpretation of this passage. The original Romanian text mentions the word “pu▯”,

which can be defined as a relatively deep hole in the ground – in our case, according to

the text, “in the rock”. The Romanian word “pu▯” does not designate a “spring”, a

natural source of fresh water. At the same time, the original Romanian text uses the

wording “ap▯ dulce” –”sweet water” (in opposition to “salt water”). “Ap▯ dulce” (“sweet

water”) is not equivalent to “fresh water”, as the Ukrainian translation suggests. Because

288
See also RM, paras. 10.28-10.49. For more evidence in this respect, see the Addendum to this
289pter, paras. 5.130-5.145.
UCM, paras. 3.40-3.42.
290 See UCM, para. 3.40, citing UCM, Annex 8.
143“pu▯” was translated in English as “well”, which also has the secondary meaning of

“spring” or “fountain”, the association by Ukraine of the translation of the word “pu▯” as

“well” with the incorrect translation of the word “dulce” with “fresh” (instead of “sweet”)

leads to the misleading impression that the said text refers to natural springs with fresh

water. In reality, the original text in Romanian mentions two holes with sweet water
291
(resulting from occasional rainfall).

5.48. Ukraine invokes another Romanian author (R.I. C▯linescu) who referred to the

“considerable rainfall” on Serpents’ Island, thus suggesting a favourable hydrological

292
climate there. In fact, the study of this author (which can be found in Annex RM 6)

never uses the words “considerable rainfall”. He states that Serpents’ Island “is situated

in the driest area of the East-European steppes” and that “the clime of the island is very

dry during summer”. He also referred to the extraordinary variations in climate on

Serpents’ Island: rainfall in spring, 293strong storms and hurricanes in winter, 294 terrible

295 296 297
frosts, fog in winter and spring, great aridity in summer. As C▯linescu concludes,

the

“... total amount of annual precipitations is 366 mm – which is very low,
and it is unequally distributed during the year, as it rains mostly during

291
The original meaning of the text is reinforced by another sentence, from the same text, referring to
the third hole mentioned there: “It probably served as reservoir for water.” The author was not sure of the
destination of the holes; that is why he used the word “probably”. The following sentence is in the same
personal probabilistic line: “It seems that in front of the temple it was another hole (“pu▯”) with the same

destination.” Thus the interpretation of UCM, para. 3.40 does not correspond to the true sense of the
Romanian text quoted in Annex 8. Also the interpretation of the words of the ancient Roman historian
Ammianus Marcellinus is, clearly, an exaggeration of the Romanian author; in fact, the text of Lisseanu
quoted by the Ukrainian Counter-Memorial is quasi-identical to a paragraph of Koehler, “’MØmoire sur les
îles et la course consacrØes a Achille dans le Pont-Euxin’ and attached map ‘carte de l’ île de Leuce,

aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie ImpØriale des Sciences de St. Petersbourg, Tome X,
p.606 (UCM, Annex 49). As he knew Koehler’s text, he simply copied it. Anyway, “Ibi et aquae sunt”
cannot be seen as a confirmation of the presence of “fresh drinking water”. Water was (and still is) always
brought from the mainland and stored on Serpents’ Island. Occasionally, the holes cut in the rock may have

been used as cisterns. The Romanian author does not mention that the water in those holes is drinkable: if it
were, he would have mentioned it.
292 UCM, para. 3.41.
293 “Particularly spring seems to be the most rainy season of Serpents’ Island.”
294 “Hurricanes are not rare at all on Serpents’ Island especially during winter.”... “These hurricanes,

295ompanied by strong sea storms, will last up to May.”
“During some hard winters, Serpents’ Island will experience the most terrible frosts.”
296 “Fog is a frequent phenomenon in winter and spring, presenting great hazards for navigation.”
297 “That is why the island is mainly dominated exactly by the most unpretentious species, the most

adjusted to aridity.”
144 spring and 298e autumn, thus determining an extremely dry and droughty
summer.”

5.49. Ukraine mentions that in 2003 a geological expedition drilled 3 boreholes on

Serpents’ Island “to a depth of 40-60 metres” and that each borehole

“produced fresh water. The approximate hourly flow is about 2 cubic
metres from each borehole. The water is usable after purification, and
299
equipment has been installed for that purpose.”

5.50. These assertions are not sustained by the document annexed, entitled “Information

about research on drinking underground waters on Zmiinyi Island in 2002-2004, State

Committee of the Natural Resources of Ukraine, Dnipropetrovsk, 2004”. It states that:

“There are no permanent water courses. There are no sources of drinking
water.” 300

As far as the results of drilling for water, the “Information…” is contradictory. It

mentions 6 boreholes, not 3. 301 Of these 6 boreholes, only in 3 cases was anything found.

It seems that “well” 1 was soon taken out of operation due to cross flow of muddy fluid,

the hole being cement plugged. Also, according to the tables enclosed as annexes to the

study from Annex 9 of the Counter-Memorial, it seems that only the fluid extracted from

“well” 2 was “treated”. The information regarding the flow rate is also contradictory, and

does not correspond either to the conclusions of the study or to the text of the Counter-

Memorial. It appears to give not an hourly flow but an annual flow: thus well no. 1

produces “a flow rate of 1.4 m /year”, 302 well no. 2 “a flow rate of 2.07 m /year”, 303and

298 Annex RM 6, p. 13 (translated by Romania) of the Monographic Study of C▯linescu. Ukraine also
invokes, in UCM para. 7.67, an article published by the Romanian Military History Magazine, which can
be found in Annex RM 10, about a group of Romanian soldiers sent on Serpents’ Island in June 1944 to

rebuild the lighthouse damaged by German bombing. Ukraine infers from the fact that “the water supplies
were getting shorter” that “water was available on the island”. This Ukrainian reasoning is misleading,
since it is clear from the text of Annex RM 10 that the respective supplies were brought there from the
mainland. The fact that the small number of Romanian soldiers were suffering because of lack of water
(and food) cannot be used as an argument as to the existence of fresh water on Serpents’ Island, but to the

299trary (see also RM, para. 10.86).
300 UCM, para. 3.42.
UCM, Annex 9, Information about research on drinking underground waters on Zmiinyi Island in
2002-2004, State Committee of the Natural Resources of Ukraine, Dnipropetrovsk, 2004, para. 5.
301 As in UCM, para. 3.40 of the Counter-Memorial. In fact, the text speaks about “wells” 1, 2, 4, 4-A
and 5. Information about “well” 3 was not included in the study; no explanation was provided.
302
See the study included in UCM, Annex 9, p. 5.
145well no. 4 produces “a flow rate of 0.8-1.0 m /year”. 3 304 But section 4.2 of the

“Information…” concludes:

“Daily water intake varies from 0.5-1.5 m to 10-12 m . From 16 July till
3
30 November 2004 there were taken 915 m of the water (on average it
3
makes 7.7 m /day), from 1 De3ember 2004 till 2 January 2005 this figure
made on average 2.2 m /day. The water level lowers by 1.5-5.0 m. During
3
the whole period of the well operation more than 2,000 m of water were
taken. This water was used for household purposes (construction, bath-
3
house, watering etc). 90 m of the purified water were for drinking and
cooking purposes.” 305

5.51. But if the flow rate of the so called “water” produced by the 3 boreholes is around

3
4.47 (1.4 +2.07+1.0) m per year it must be asked how the affirmation of section 4.2 of
3 3
the “Information…” that “[d]aily water intake varies from 0.5-1.5 m to 10-12 m ” can be

true. Moreover if the “[d]aily water intake varies from 0.5-1.5 m to 10-12 m ”, it must 3

be asked how this data can correspond to the assertion that the “approximate hourly flow
306 3 307
is about 2 cubic metres from each borehole” (that is, 144 m per day).

5.52. The same study underlines the difficulty of drilling: many times during the

308
process a “complete loss of the flushing fluid” at all three boreholes was experienced.

It also recognises that the quantity of the reserves of “water” is uncertain, despite the

drilling.309

303 Ibid., p. 7.
304 Ibid., p. 8.
305
Ibid., p. 9.
306 UCM, para. 3.40.
307 Another contradiction derives from the statement that in the drilling period of 6 months a total
3
quantity of around 2000 m was obtained, which is q3ite impossible if taking into account that this would
correspond to a constant quantity of around 11 m of fluid per day; or, the data provided for by Ukraine
mentions that “[d]aily water intake varies from 0.5-1.5 m to 10-12 m ”. It is quite clear that both the
statements of Ukraine in the Counter-Memorial and the “evidence” from the Annexes on which it is

308sed” are unreliable.
Ibid., pp. 6-8.
309 Ibid., p. 12.

1465.53. The chemical composition of the “water” found, as described by the

“Information…”, is also of interest: it cannot be considered fresh. The following table

shows a comparative presentation of the chemical composition of the fluid extracted by

Ukraine, before and after treatment, 310 using the data provided for in Annex 9, and the

composition of Black Sea seawater, as indicated by the same Ukrainian study: 311

Chemical Ukrainian “water” Ukrainian Black Comments resulting
substances reportedly “water” Sea from the

found on the reportedly water comparative
Serpents’ Island found on the (mg/l) presentation

(mg/l) Serpents’
Island, after

treatment
(mg/l)
NH + well 1 –124.8 well 2 – 1.80; 0.10 The sea water
well 2 – 0.32; 10.00; 11.00; 18.00; concentration of this

50.00; 25.00; 40.00 10.00; 8.00 substance is about 10
well 4 – 116.06; 50.00; times less than in the
well 5 – 24.50; 30.00 “water” reportedly found,
even after treatment.

NO 2 well 1 – 24 well 2 – 0.30; 1.20 The sea water
well 2 – 3; 5.4; 4; 0.90; 0.00; 0.05 concentration of this
0.60; 1.50 substance is, with some
well 4 – 0.24; 4 accidental exceptions, less
well 5 – 0.60; 0.50 than in the “water”

reportedly found.
SiO 2 well 2 – 18; 13; 40; 20; 2;well 2 – 0; 4 0 The sea water has no sand
well 4 – 14; 13 in suspension in it, while
well 5 –0; 1 the “water” reportedly
found has it, even after

treatment.
Fe tot well 1 – 0,33; >0.05 well 2 – 0.13; 0.14;0.10 The sea water
well 2 – >0.05; 0.08; 0.04;0.20; 0.05; 0 concentration of this
0.13; 0.20; 0 substance is, with some
well 5 – 0.30; 0.15 accidental exceptions, less

than in the treated “water”.
Ca 2+ well 2 – 24.40; 83.00; well 2 – 6.00; 2.00;180.0
9.00; 31.00; 27.00; 6.00; 1.00
12.00

well 4 – 22.05;3.00
well 5 – 53.00; 199.00
Mg 2+ well 2 – 43.00; 8.00; well 2 – 1; 3; 2; 524.0
53.00; 19.00; 17.00;62.00;
57.00

well 4 – 35.23; 47.00
well 5 – 47.00; 7.00

310
Only the “fluid” extracted from “well” 2 was treated. The concentration of substances in this
treated fluid varies considerably.
311 See last row of Supplement 2, “Summary Table: Results of the Water Chemical Analysis”, UCM,

Annex 9.
147 +
K well 2 – 230; 200; 170; well 2 – 7.00; 250 The sea water
140; 180 13.00; 5.00; concentration of this
well 4 – 200 substance is almost the
well 5 – 325; 280 same, and sometimes less

than in the “water”
reportedly found.
Na + well 1 – 504.4; 564.4 well 2 – 45.00; 433 The sea water
well 2 – 526.5; 493; 552; 51.00; 64.00; 20.00 concentration of this

593; 404; 587; 577 substance is, with some
well 4 – 208; 575 accidental exceptions, less
well 5 – 565; 1216 than in the “water”

reportedly found.
Cl well 1 – 552.1; 774 well 2 – 43; 6.40; 6806
well 2 – 671.4; 661; 780; 14
716; 688; 723

well 4 – 574; 766
well 5 – 539; 170
NO 3 well 1 – 19.27; 13.45 well 2 – 66; 64; 78; 8.00 The sea water
well 2 – 82; 144; 174; 22; 27 concentration of this
156; 153 substance is roughly

well 4 – 43.6; 207 between 2.5 and 35 times
well 5 – 183; 282 less than in the “water”
reportedly found, and
between 3 and 10 times

less than in the treated
“water”. The values of
concentration of this
substance in the treated
“water” are higher than in

the “GOST” Standard used
by Ukraine .
well 2 – 93.4; 85.92; well 2 – 0; 0.19; The sea water
Oxidation 8.58
substances 27.00; 34.88; 28.84 0.59 concentration of this
well 4 – 57.6; substance is even more
well 5 – 9.62; 93.44 than 10 times less than in
the “water” reportedly
found.

5.54. Certain conclusions can be drawn from this table. With three exceptions (Mg , 2+

2+
Cl, Ca ), all the substances have concentrations higher than the concentrations of the

same substances in seawater. Sometimes, the concentrations of these substances even in

the treated water are higher than in seawater. It seems that the water extracted on

Serpents’ Island is seawater with higher concentrations of certain chemicals, in which

there are further harmful substances not allowed in potable water (e.g., humic acid is a

perfect medium for bacteria; it cannot be eliminated efficiently by using the method

employed on Serpents’ Island). Moreover, the “Information…” mentions “a great content
312
of oil products” and a “cross-flow of muddy water”. It also confirms that the water is

312
See the study included in UCM, Annex 9, p. 6.
148 313 314
salty, and before purification “straw-color and brown”. The study is eloquent: “the

natural water from the well cannot be used for drinking purposes because of its chemical

characteristics”; 315 “it features high content of organic and nitrous compounds” 316 and

“[t]aking into account the high content of organic matters and potassium in the

underground waters, it is necessary to study the possibility of their use for medical
317
purposes.” Even after purification “according to sanitary and bacteriological

parameters the purified water, in individual cases, does not meet standard requirements”

and “after purification in individual 318 probes” there are still “nitrous compounds”. 319 It

would have been easier for Ukraine to try purifying seawater, which is plentiful in the

locality.

5.55. That the water found is basically infiltrated seawater with additives is not

surprising. The drilling took place, according to the same “Information…” at great depth:

almost 40 metres for well no. 1, 56 metres for well no. 2, 42 metres for well no. 4-A, 97

metres for well no. 5. This is below sea level: since the maximum absolute height of

Serpents’ Island is 41.0 m. 320

313
Ibid., pp. 6, 9.
314 Ibid., p. 10.
315 Ibid., p. 11.
316
317 Ibid., p. 12, para. 5.
Ibid., p. 13. It is interesting to note yet another inconsistency: the results presented in the table
enclosed in Supplement 3 of Annex 9 to the Counter-Memorial were obtained after purifying the samples
of “water” drilled in September 2003. They show that the treated “water” has concentrations of nitrous
components within the “GOST” standard used by Ukraine. This is surprising if taking into account that,

“[i]n 2005, it [wa]s planned to install an additional unit [on Serpents’ Island] for reducing the content of
nitrous components to permissible concentrations.” (as there were still “nitrous compounds” “after
purification in individual probes”, by using the devices for purification installed on Serpents’ Island). The
conclusion is that the so-called treated “water” was not “purified”, as Ukraine asserts, on Serpents’ Island –

where there were no technical conditions to reduce the content of nitrous components – but on the
mainland.
318 This statement about “individual probes” is not supported in any way, as the tables included as
annexes of this study from Annex 9 of the Counter-Memorial contain no relevant information about the

results of the microbiological tests. One could presume that the reality is worse. Not only that
microbiological testing results are missing, but the chemical analysis is far from being complete: certain
pollutants are not measured, like pesticides, trihalomethanes, heavy metals; radioactivity is not measured
either.
319
320 Ibid., p. 10.
The connection with the sea water is recognized in section 4.2 of the study included in Annex 9:
“Under conditions when circular bonds of surface sea waters directly connected with the exploited water
bearing horizon get closer, a steady lowering of the underground water level to values below the sea level
will result in irreversible approaching of the sea waters that is unacceptable.”

1495.56. In any event the quantity of “purified water” is insufficient. Between July 2004

and January 2005, 90 m‡ of purified “water” was obtained, whereas the necessary daily

quantity of water, as assessed by Ukrainian authorities, was 27 m in 1997. 321Thus,

taking into account this assessment, the amount obtained in six months would have

sufficed only for three days.

5.57. Interesting evidence as to the lack of water is to be found in Annex 68 to the

Ukrainian Counter-Memorial. This Annex is composed by official documents for a

meeting of the Cabinet of Ministers of Ukraine regarding problems of development of

Serpents’ Island. On the first (cover) page the date (“March 16, 2006”) and hour

(“16.00”) of the meeting are mentioned. After the first table included in this Annex, a

page entitled “Proposals to the procedure of holding a conference presided over by the

First Vice Prime Minister of Ukraine Dubnaya O. V. on problem issues of Zmiinyi

Island” is inserted. It contains several notes in Italics, probably a hand-written instruction

by the First Vice Prime Minister Dubnaya O. V. The last words of this page are

extremely eloquent as to the officially assumed situation on Serpents’ Island:

322
“d) Water. To search water.”

5.58. The lack of water on Serpents’ Island is also confirmed by recent statements of

Ukrainian officials as reflected by articles in Ukrainian media. For example, on 3 May

2006, the head of the Odessa Regional Department for ecology and natural resources,

Vitali Primak, declared that “…the island continues to suffer because of the lack of

potable water.” 323

5.59. The arid climate of Serpents’ Island and the lack of water produce an extremely

sparse flora and no fauna. Further to the arguments of the Romanian Memorial, the

Annexes of the Ukrainian Counter-Memorial provide clear evidence of this.

321
322 See UCM, Annex 91
Italics in the English text of the translation. UCM, Annex 68, Documents for a Meeting of Cabinet
of Ministers of Ukraine Regarding Problems of Development of Zmiinyi Island, and Enclosure (1998
Information Concerning the Implementation of the Cabinet of Ministers of Ukraine Regulation No. 1114 of
8 October 1997).

1505.60. According to Koehler’s Memoire:

“La surface de Leuce n’est recouverte que d’une couche de terre d’à peine

un ou deux pieds de profondeur, de sorte que presque par tout le rocher se
montre à dØcouvert... Mais il n’y avoit dans toute l' île ni arbres ni

arbustes, pas mŒme ces espŁces de buissons qui croissent ordinairement
dans les terrains pierreux.” 324

th
5.61. According to E. D. Clarke, who visited it also in the 19 century, Serpents’ Island

“[…] is qui325bare, being covered only with a little grass, and very low
herbage.”

5.62. The reality nowadays is not different from what the earlier sources describe.

Annex 9 of the Ukrainian Counter-Memorial mentions that Serpents’ Island “features

rocky ground and poor grassy vegetation. During a hot summer season all vegetation gets

withered.” 326

5.63. The nowadays photographs are eloquent: no trees, little vegetation and that

withered and burnt. This situation is also confirmed by the “List of Island’s Flora, Fauna

and Fish Species” annexed by Ukraine. According to this list, the only mammals on the

Island are a “Norway rat”, a “House mouse” – both species introduced by humans – and

two species of bat. 327

323 Article titled “Recycling the garbage – one of the biggest problems on Serpents’ Island”,

324lished by the Ukrainian Press Agency “Reporter” on 3 May 2006 (Annex RR 7).
UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes à Achille dans le Pont-
Euxin’ and attached map ‘carte de l’île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie
ImpØriale des Sciences de St. Petersburg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
325 UCM, Annex 50, E. D. Clarke, Travels in Various Countries of Europe, Asia and Africa, Third

326tion, Printed for T. Cadell and W. Davies Strand, London, 1813, pp. 648-650.
UCM, Annex 9, Information about research on drinking underground waters on Zmiinyi Island in
2002-2004, State Committee of the Natural Resources of Ukraine, Dnipropetrovsk, 2004.
327 This scientific document may be contrasted with the image of the conditions on Serpents’ Island –
see for instance the picture of a pig in UCM, Annex 75.

151 (3) Serpents’ Island cannot sustain human habitation 328

5.64. Serpents’ Island has never had a permanent resident population due to the very

difficult conditions that cannot sustain human habitation.

5.65. A small group of persons temporarily residing, on a rotational basis, on an island

cannot qualify as a situation of human habitation under Article 121(3); they cannot be

considered even as a “temporary resident population”. A population, even temporary, has

different characteristics. There are many examples of islands temporarily inhabited in

certain more favourable seasons of the year, for instance for fishing purposes, by a

community over time. This is clearly not the case of Serpents’ Island. Individuals are

sent to perform certain tasks ordered by the authorities. Even if Article 121(3) included a

temporary population in the concept of “human habitation” (which it does not), the

situation on Serpents’ Island would not correspond to this requirement: the heterogeneous

group based there does not qualify as a “population” in the proper sense of this word.

5. 66. The difficult conditions on Serpents’ Island are evidenced by the story, told by

Ovid, about the fate of the ruler of Amastria who was exiled on Serpents’ Island by his
329
father-in-law Mitridates and who “died from hunger and cold”.

5.67. A Russian study of 2002, included in Annex 57 of the Counter-Memorial, recalls

the “testimonies of antique authors who wrote about forbiddance (sic) to settle on Leuke”

and the opinions of certain historians who “believe that the island could not be inhabited

by ‘somewhat stable population in the true sense of the word’”. 330 Unfortunately, the

Court will not learn who these (well informed) scientists are because the respective

footnotes are missing both from the English, and the Russian texts.

328 See also RM, paras. 10.50-10.79. For more evidence in this respect, see Addendum to this
Chapter, paras. 5.146-5.165.
329
UCM, Annex 52, S.B.Ochotnikov, Odesa National University, Notes of the Faculty of History,
330sa, 1996, pp. 46-59.
UCM, Annex 57, I. V. Tunkina, Russian Science on Classical Antiquities of the Russian South
(XVIII-Mid XIX Centuries), St. Petersburg, 2002, p.401-450.
1525.68. Annex 50, at its turn, is also clear of the situation in the 19 century: “No human

being dwells there…” 331

5.69. The Black Sea Sailing Directions, published in Leningrad in 1931, are

representative. They state:

“Except civilian lighthouse watches and the sentry composed of four

Romanian frontier guards nobody dwells on the island. Foodstuff for the
guards is delivered from Sulina.” 332

5.70. Documents from the official archives of the European Danube Commission,

annexed to this Reply, show how the lighthouse personnel used to find pretexts to avoid

service on Serpents’ Island, and how difficult it was to find replacements:

“Je crois devoir relever à cette occasion qu’il est fort difficile à trouver des
candidats pour un post de gardien du dit phare avec la rØtribution de 875

lei par mois. D’ailleurs, les gens ne veulent pas s’engager par le motif que
devant Œtre isolØs du commerce des hommes, ils n’ont pas le confort
333
nØcessaire au logement à l’île, lequel laisse beaucoup à dØsirer.”

5.71. The captain of the Sulina Port reported to the European Danube Commission that

it was impossible to find appropriate candidates and that he was obliged to appoint the

first person willing to do such a job in such difficult conditions:

“J’ai du donc choisir entre les deux maux le moindre soit celui de recruter
n’importe quel ØlØment pour faire face aux exigences momentanØes les
plus urgentes du service du phare en question. ”334

5.72. The difficulties continued: on 9 March 1923 the Captain wrote again:

Łme
“Le Sieur Dumitru IFTODI, nommØ 3 gardien du phare de l’île des
Serpents par Votre dØcision No.4190/2233 du 16 DØcembre dernier, ayant
abandonnØ son service sous prØtexte qu’il ne peut pas vivre en isolement

331 UCM, Annex 50, E. D. Clarke, Travels in Various Countries of Europe, Asia and Africa, Third

332tion, Printed for T. Cadell and W. Davies Strand, London, 1813, pp. 648-650.
UCM, Annex 13, Black Sea Sailing Directions, Seventh Edition, Leningrad, 1931, pp.47-48. The
same information can be read in UCM, Annex 12, comprising a Russian document of 1903 - Black Sea and
Sea of Azov Sailing Directions, Fourth Edition, St. Petersburg, 1903, pp. 64-65.
333 Document No. 452, sent by the captain of the Sulina Port to the European Danube Commission, 7
May 1921 (Annex RR 9).
334
Document No. 1131, sent by the captain of the Sulina Port to the European Danube Commission,
6 December 1922 (Annex RR 10).
153 continuel, j’ai l’honneur de Vous informer que j’ai dß le remplacer
provisoirement et d’urgence…” 335

5.73. The situation nowadays is no different on Serpents’ Island. According to the

Ukrainian press, “(a)t present, it (the Serpents’ Island) is inhabited by border guards,
336
meteorologists and ecologists.” Another piece of information published on the 10

February 2006 by the Ukrainian Press Agency “ProUA” mentions that “[o]n the island

there are border guards’ detachments, representatives of the Ministry of Interior and of

the anti-aircraft intervention troops”. 337 An article in the newspaper Golos Ukraini

enumerates the persons temporarily working there as “12 border guards, one lighthouse
338
keeper and one postal worker, two ornithologists and few scientific researchers”. The

border guards are hired on a contract basis and are rotated monthly; the researchers stay

only 10 month per year, due to the difficult life conditions. 339

5.74. Ukraine refers to an inter-war plan of “the Romanian Government” to build a

340
“monastery” or later to “build a prison for political prisoners with 1500 cells”. It also

quotes the Romanian author C▯linescu who wrote about the idea to establish a balneal
341
station on Serpents’ Island.

5.75. In fact, the “monastery” was the idea of a Romanian bishop who visited Serpents’

342
Island in 1936. His idea was to build a small chapel (in Romanian, “paraclis”) or

335 Document No. 339, sent by the captain of the Sulina Port to the European Danube Commission, 9

336ch 1923 (Annex RR 11).
Article titled “By the year 2008, a town will be settled on Serpents’ Island”, published by the
Ukrainian Press Agency ProUA on 1 June 2006 (Annex RR 12); see also the article titled “A town will be
settled on Serpents’ Island”, published by the Ukrainian Press Agency PODROBNOSTI.UA on 1 June 2006

337nex RR 13).
Article titled “Ehanurov initiates a contest regarding the development of Serpents’ and Tuzla
Islands”, published by the Ukrainian Press Agency ProUA on 10 February 2006 (Annex RR 14).
338 Article titled “There, where the wind always blows from the sea”, published by the Ukrainian
newspaper Golos Ukraini (official newspaper of the Supreme Rada), issue of 29 September 2006; author:

339stasia Sirotkina (Annex RR 15).
Ibid.
340 UCM, para. 7.64.
341 UCM, para. 7.66.
342
UCM, Annex 61.
154 343
hermitage (in Romanian, “schit”). These plans came to nothing. The same is true of

the plan to “build a prison for political prisoners with 1500 cells”. Ukraine provides only
344
partial or no translations of the relevant annexes, but it will be seen from reading the

full texts that this proposal was not feasible and was never implemented. 345

5.76. As to the balneal station, C▯linescu explains why this idea too was abandoned:

“But the very tough and irregular winds, which blow terribly from autumn

to spring, the difficulty to bring supplies and especially the costs of
building were as many reasons to give up the making of this project.” 346

5.77. A recent article in the Ukrainian press, published in September 2006, confirms the

toughness of the environment on Serpents’ Island:

“It is said that the work here is not different than that at any other border
post. In my view, it is not true. According to the statements of the head of

the border post, captain Vitalii Balchiuk, the main problems faced by the
border guards here are of medical type – the colds. The climate dictates its

own conditions. So, it is explainable why there are permanent deposits of
supplies for 3-4 months. In his turn, the physician of the island, major

Vladimir Sanoed, has to fight not only the traditional gastritis, but also347e
psychological problems, especially in the autumn-winter period.”

343 See the Romanian original of UCM, Annex 61, P. Dogaru, Serpents’ Island in the Way of Sharks,
Bucharest, 1996, p.69.
344
See UCM, Annexes 60, 62, 63. These are selectively translated and in some cases untranslated
(see Annex 63). The complete text of the accurate translation of the newspaper article included in Annex 60
can be found in Annex RM 42; as for the other two documents, Romania encloses to this Reply the missing
English translations; see Annexes RR 16 (article titled “Serpents’ Island, asylum for deportees” published

by the Romanian newspaper Curentul, issue of 21 March 1938) and RR 17 (article titled “A prison with
1500 cells for political convicts is going to be built on Serpents’ Island”, published by the Romanian
newspaper Informa▯ia, issue of 22 March 1938).
345 The failed inter-war “plan” to build a small chapel is no argument in favour of the habitability of
Serpents’ Island, as neither the very recent initiative of a similar kind of a Ukrainian deputy in the Odessa

Regional Council is: “Vasili Chernolutzky, the head of the fraction of the Regions Party in the Regional
Council, forwarded this idea (the proposal for a church to be build on Serpents’ Island)” to the permanent
regional commission for culture, historic patrimony, spirituality and national minorities. See the article
titled “Vasili Chernolutzky proposed the building of a church on Serpents’ Island”, published by the
Ukrainian on-line publication Reporter on 19 October 2006 (Annex RR 18).
346
See p. 51, last paragraph, not translated into English, continued on p. 52 of the Romanian original
text included in UCM, Annex 10, R.I. C▯linescu, ‘Insula ▯erpilor. Schi▯▯ monografic▯’, in Analele
Dobrogei, Anul XII, Cern▯u▯i, 1931, pp. 47, 53, 57.
347 Article titled “There, where the wind always blows from the sea”, published by the Ukrainian
newspaper Golos Ukraini (official newspaper of the Supreme Rada), issue of 29 September 2006; author:

Anastasia Sirotkina (Annex RR 15).
1555.78. Finally Ukraine refers to a “Natural Reserve” of 232 hectares created on a

unspecified part of Serpents’ Island and in the surrounding waters. 348 What natural

values are being protected, and from whom, is not specified.

5.79. To summarise, not merely is there no civil regime on Serpents’ Island – which is a
349
locked territory, there is naturally no possibility of sustaining a human population

there. It should be noted that in no picture included in the UCM Annexes are any
350
“inhabitants” present. All pictures indicate a deserted territory, lacking the conditions

for human habitation.

(4) Serpents’ Island cannot sustain any economic life of its own 351

5.80. Ukraine claims that Serpents’ Island can sustain economic life, but again much of

the evidence in its Annexes shows the contrary.

5.81. The lack of easy access to Serpents’ Island is an initial impediment for any

economic life there. An official Ukrainian document of 2005 mentions the newly built

berth on the north-eastern part of Serpents’ Island, which proves to be unsatisfactory in

providing for safe access:

“the berth is unreliable as the ground holds the anchor badly and it is
insufficiently protected from disturbance.” 352

5.82. The situation was exactly the same more than 100 years before. An official

Russian document of 1903 presents the same problem:

“Anchorage at the island is inconvenient due to the bad ground; however
vessels sailing to Sulina with the NO 353 wind sometimes land at the

348 UCM, Annex 78, Order No. 54 of the Ministry of Natural Environment protection of Ukraine, 25
January 2005, and Regulation No. 54 on the general Zoological Reserve of National Importance ‘Island
Zmiinyi’, 27 January 2005.
349 “Entry to and temporary stay” there can only be “exercised on the ground of a permit granted by

the Southern Regional Department of the State Frontier Service of Ukraine”; see UCM, Annex 89, Decree
No. 700/A-2003 of the Odessa Regional Administration dated 29 August 2003.
350 There is only one picture (photo F, p.196) showing some persons, probably just arrived with the
ship docked at the berth.
351 See also RM, paras. 10.80-10.100. For more evidence in this respect, see Addendum to this
Chapter, paras. 5.166-5.188.
352
UCM, Annex 5, Sailing Directions of the Black Sea and the Sea of Azov for Waters of Ukraine,
Kiev, 2005.
156 southern side at 12 sajenes 354and 13 sajenes although the island due to its

tiny stretch provides little protection during the roughness. ... The
disembarking on the island is not allowed. ... the NW side is not deep
355
enough for boats and abundant in stones.”

5.83. Another official Soviet document of 1954, when Serpents’ Island was already

under Soviet authority, depicts the same reality:

“The island is ... bordered with a ridge of submerged stones… This
anchoring site is not reliable since the ground within it holds poorly the
356
anchor and the site is inadequately protected from the roughness.”

Access to Serpents’ Island was always very difficult and continues to be so, despite

constant efforts to provide access.

5.84. Not only is the geography of Serpents’ Island an obstacle to economic life, the

adverse climate is another difficulty. For example:

“Maximum air temperature is observed from June to August and reaches
40°C. Average number of frosty days during a year is approximately 100

days. Maximum depth of frost penetration in ground is 51 cm. The sea
becomes frozen in December-January. In second half of February the ice
begins to thaw and the sea becomes ice-free in March.

... A period from May to October is considered the most favourable one
for navigation of ships to Zmiinyi Island. During the rest of the period the
357
island can be reached only with helicopter.”

5.85. The extremity of the climate means that only during half of the year navigation is

possible. Because of the strong winds, a helicopter cannot land all the time. As to the
358
frequency of maritime transport, from June-December 2004 there were 16 voyages, an

353
As such in the English translation provided by Ukraine in Annex 12; it probably indicates the
direction of the wind.
354 Russian measurement unit of distances (1 sajen = 2.133 metres), roughly equivalent to the fathom.
355 UCM, Annex 12, Black Sea and Sea of Azov Sailing Directions, Fourth Edition, St. Petersburg,

3563, pp.64-65.
UCM, Annex 14, Black Sea Sailing Directions, 1954, p.76.
357 UCM, Annex 9, Information about research on drinking underground waters on Zmiinyi Island in
2002-2004, State Committee of the Natural Resources of Ukraine, Dnipropetrovsk, 2004.
358 UCM, Annex 96, Letter from Ostrivne Odesa Oblast Municipal Enterprise to Oleksandr

Mychaylovych Hordiyenko, Chief of the Main Department of Economy, Oblast State Administration, 5
March 2005, Enclosing Transport Communications with Serpents’ Island for 2004.
157average of 2.3 per month. Even this “regular” transportation was established only in

2004. 359

5.86. Serpents’ Island is entirely dependent from the supplies brought from the

mainland.

“The said voyages were made with the purpose of ensuring vital activities

on the island, personnel rotation, and delivery of building materials and
food.” 360

5.87. The situation is no different from that on Serpents’ Island more than a century

before. An official Russian document of 1903 pictures the same dependence:

“foodstuff… is delivered from Sulina.” 361 The same thing was true in 1931. 362

5.88. Without these supplies the few officials on Serpents’ Island could not survive.

This is the case because Serpents’ Island cannot produce anything, either in order to

sustain basic human habitation or to stimulate economic life. There are no natural

resources except for rocks. There are no useful minerals and the rocks are not even good

for construction purposes, as the Russian captain P.A. Drozzhin noted in 1836:

“... all rocks on the island are made up, like on purpose, out of regular

block stones abundant in quantity, the stone in fact is siliceous, reddish
and bluish hard and absolutely unsuitable for the lighthouse
construction.” 363

359 UCM, Annex 73, Information concerning the Fulfillment of the Complex Program of the Further
Development of the Infrastructure and the Realization of the Economic Activities on Zmiynyi Island and
the Continental Shelf for the Period of 2002-2004, The State Regional Administration Odesa, the Central
Administrative Board of Economy, November 2004.
360
UCM, Annex 96, Letter from Ostrivne Odesa Oblast Municipal Enterprise to Oleksandr
Mychaylovych Hordiyenko, Chief of the Main Department of Economy, Oblast State Administration, 5
March 2005, Enclosing Transport Communications with Serpents’ Island for 2004.
361 UCM, Annex 12, Black Sea and Sea of Azov Sailing Directions, Fourth Edition, St. Petersburg,
1903, pp.64-65.
362 UCM, Annex 13, Black Sea Sailing Directions, Seventh Edition, Leningrad, 1931, pp.47-48.
363
UCM, Annex 57, I.V. Tunkina, Russian Science on Classical Antiquities of the Russian South
(XVIII-Mid XIX Centuries), St. Petersburg, 2002, pp.401-450.
1585.89. All other efforts to create economic activity have failed. One example is the

attempt made to cultivate mussels near Serpents’ Island. 364

5.90. It is true that efforts have been made to try to stimulate economic life. A Regional

Utility Company called “Ostrivne” (from “ostriv” – “island” in Ukrainian) has been

established

“with the purpose of maintenance of life sustenance system and stirring up
the process of integration of the island with the economy of the region
(province) and settling matters concerning its adjustment” 365

But these artificial efforts have so far simply involved the expenditure of money,

resources and bureaucratic time – without anything remotely approaching an economic

life proper to Serpents’ Island. 366 The “integration of the island with the regional

economy” is a modern myth, to go with the ancient ones with which this small rocky

island is associated.

5.91. Ukraine places considerable stress on the opening of the branch of the “Aval

Bank” which is “testimony to the growing needs of the people living on the Serpents’
367
Island”. Ukraine admits that the branch was inaugurated immediately after the deposit
368
of the Romanian application in the present case, but argues that it was in the planning

stage since 2004. In reality, the likelihood of a case before the Court was known well

before then: indeed it is explicit in the Additional Agreement.

5.92. Photographs of the branch included in the Counter-Memorial (photo I) and in

Annex 94 show the same image of an unused and deserted barrack with closed doors. No

customers are shown. It is doubtful whether there is anything inside. There is no

364 See for example UCM, Annex 73, Information concerning the Fulfillment of the Complex
Program of the Further Development of the Infrastructure and the Realization of the Economic Activities
on Zmiynyi Island and the Continental Shelf for the Period of 2002-2004, The State Regional

365inistration Odesa, the Central Administrative Board of Economy, November 2004.
UCM, Annex 87, Resolution No. 111-XXIV of the Odessa Regional Council, 10 January 2003.
366 See also UCM, Annex 88, Articles of Association of Ostrivne, Odesa Regional State
Administration, 21 March 2003.
367 UCM, para. 7.84.
368 “Indeed, the opening of a new branch of a bank is ... something which follows a longer period of

consideration… the board of Aval Bank decided to open a regional branch on Serpents’ Island on 11 May
2004, i.e. long before Romania filed its application in this case.” UCM, para. 7.84.
159mention of any bank clerk on Serpents’ Island. 369 Attempts to contact the branch on

Serpents’ Island at the phone numbers provided by the website of the Odessa

headquarters of Aval Bank were fruitless. Nobody ever answered at the first number.

The second number proved to be wrong: it corresponded to a private person in Kylia

town of the Odessa Region; later on this number remained un-answered too. The records

of the phone conversations were deposited by Romania with the Court’s Registry; their
370
transcripts are annexed.

5.93. Alternatively Ukraine argues that Serpents’ Island might acquire an economic life

if Ukraine were to win the case! It says:

“the potential economic importance that an island might have if it
generated an EEZ or its own continental shelf; and when applied to

Serpents’ Island, the e371tence of substantial natural resources ‘of its own’
cannot be excluded.”
372
This is a neat reversal of Article 121(3), a splendid example of circular reasoning.

Evidently, however, a State cannot claim that, if a rock may (with the help of an

exclusive economic zone) acquire an economic life of its own, therefore it already has an

exclusive economic zone. Article 121(3) lays down a pre-requisite, not a post-requisite.

(5) Ukraine’s artificial island

5.94. No doubt conscious of the weakness of the argument that Serpents’ Island is now

capable of sustaining human habitation or economic life, Ukraine has been attempting to

make it so for the future. On 1 June 2006 the Interim Governor of Odessa Region
rd
publicly declared on the occasion of the 3 Investment and Innovations Forum (Odessa)

the intention of the Ukrainian authorities to create, by 2008, a “town” on Serpents’ Island

with the name “Bilyi” (“White”) with a population of 100 persons. Funds have been

369 Neither the most recent articles from the Ukrainian media presenting detailed information on the

370pulation” of Serpents’ Island include any reference to the presence of bank employees (see para. 5.73).
Annex RR 19.
371 UCM, para. 7.41.
372 See Kolb, “L’interprØtation de l’article 121, paragraphe 3, de la Convention de Montego Bay sur
le droit de la mer: les « rochers qui ne se prŒtent pas à l’habitation humaine ou à une vie Øconomique
propre… » “, AFDI, XL, 1994, p.907: “Car autrement on tomberait dans la circularitØ qui consiste à

subordonner d’un cotØ l’existence de ces zones à une vie Øconomique, mais de l’autre cotØ à dØduire une
vie Øconomique (la condition) de l’existence de ces zones (le conditionnØ).”
160 373
allocated and the first tenders called. Reportedly, the officially-declared purpose is “to

annul the controversies as to the status of Serpents’ Island as an island”. As revealed, at

least proleptically, in the Ukrainian Press:

“So, on the island a real settlement appeared, which, according to the
decision of the Odessa Regional State Administration, recently got the

name ‘Bilyi’. ... So, the i’s were dotted and the t’s were crossed in the
dispute between the two parties.” 374

This is perhaps the first time in history that a town has been created because of a treaty on
375
the law of the sea.

5.95. In its Memorial Romania took note of these activities and argued that legally they

could not change the situation. 376 No response is made in the Counter-Memorial to these

arguments. 377

5.96. When Romania filed its Memorial it had no access to the various official decrees

issued by Ukraine for the purpose of artificially transforming Serpents’ Island. Indeed, as

shown by the Annexes to the Ukrainian Counter-Memorial, these texts were

confidential. 378 If the activities undertaken by Ukraine on Serpents’ Island are, as

Ukraine contends, “part of legitimate sovereign activity of Ukraine, and arise from a
379
natural development of the island’s long established uses”, it may be asked why the

decisions regarding these measures were classified.

373
See Article titled “By the year 2008, a town will be settled on Serpents’ Island”, published by the
Ukrainian Press Agency ProUA on 1 June 2006 (Annex RR 12); see also the article titled “A town will be
settled on Serpents’ Island”, published by the Ukrainian Press Agency PODROBNOSTI.UA on 1 June 2006
(Annex RR 13).
374
Article titled “The Settlement ‘Bilyi’ will appear on Danube”, published by the Ukrainian
newspaper Slovo, on 29 September 2006; author: Tamila Cernetzkaia (Annex RR 20).
375 The artificial character of such demarches was noticed by the Ukrainian press: “It is true, recently,
a new administrative entity appeared on the island – the settlement Bilyi, and, in consequence, (there

should be) a president of it. But, for the time being, there is no one to collect even the garbage…”; see the
article titled “There, where the wind always blows from the sea”, published by the Ukrainian newspaper
Golos Ukraini (official newspaper of the Supreme Rada), issue of 29 September 2006; author: Anastasia
Sirotkina (Annex RR 15).
376 See RM, paras. 10.101-10.131.
377
378 See UCM, para. 7.87.
See, for instance, the mention “Not intended for Publishing.” on p. 1 of the 1995 Resolution of the
Cabinet of Ministers of Ukraine included in UCM, Annex 67, Resolution of the Cabinet of Ministers of
Ukraine No. 1009, 18 December 1995.
379
See UCM, para. 7.86.
1615.97. It emerges that in 2002 the Cabinet of Ministers of Ukraine approved a

“Comprehensive Programme of the Further Development of the Infrastructure and

Conduct of the Economical Activity on Serpents’ Island and on the Continental Shelf”.

The Comprehensive Programme is clear about its objective:

“Objective and Task

The Programme objective is to create the appropriate conditions for
human habitation and conduct of the economical and other activities on

the island.

The main tasks of the Programme are:
To complete the work connected with documentary legitimation of the
380
Island status as an administrative and territorial formation…”

5.98. Similarly in 1995 the Cabinet of Ministers of Ukraine directed that:

“5. The Ministry of Foreign Affairs shall proceed with a package of
measures aimed at legitimation of the state boundary with the Republic of

Romania proceeding from that that in conformity with the United Nations
Convention on the Law of the Sea the island has the exclusive economic
zone and continental shelf of its own.” 381

5.99. In its turn, the Analytical Brief on the implementation of the above

Comprehensive Programme stated that the said program was approved

“with the purpose of implementation of a package of measures aimed
at ensuring proper conditions for human habitation and for conduct
of economic and other activities on the island.” 382

5.100. In January 2006, Ukrainian Prime Minister Ehanurov visited the Odessa Region.

During the visit it became public that the Ukrainian State budget had allotted 11 million

hryvnyas (about 2.2 million dollars) to be spent in 2006 for the development of Serpents’

Island. The press report mentions that:

380 UCM, Annex 74, Resolution No. 713 of the Cabinet of Ministers of Ukraine, 31 May 2002
regarding the approval of the “Comprehensive Programme of the Further Development of the Infrastructure
and the Conduct of the Economic Activities on Zmiynyi Island and the Continental Shelf”.
381
382 UCM, Annex 67, Resolution of the Cabinet of Ministers of Ukraine No. 1009, 18 December 1995.
UCM, Annex 84, Analytical Brief on the Implementation of the Comprehensive Program for
Further Development of the Infrastructure and Conduct of Economic Activities on Zmiinyi Island and on
the Continental Shelf (bold in the original text).
162 “the capital problem of the delimitation of the continental shelf and the

exclusive economic zones of Ukraine and Romania in the Black Sea is
represented by the juridical status of Serpents’ Island, which Ukraine
considers to be an island entitled to its own territorial waters, while
383
Romania - a rock without this right.”

After this visit, on the occasion of a meeting with students and professors of National

Polytechnic University of Lvov, the Ukrainian Prime Minister Ehanurov underlined the

fact that important quantities of hydrocarbons were identified in the Black Sea

continental shelf. In this context, he evidenced the

“importance of establishing the status of island for the Serpents’ Island,

this being the 384son for which its infrastructure will have to be developed
in the future.”

5.101. On 18 April 2006, the Ukrainian Ministry of Transport and Communications

announced “the process of creating a special free economic zone on Serpents’ Island”.

The purpose of this measure is to extend the system of simplified customs formalities for

economic agents willing to invest in this zone. Another proposal was to apply, on

Serpents’ Island, a simplified procedure to register vessels under the Ukrainian flag. If

this last measure is approved by the Ukrainian Parliament, an office of the Ukrainian

Register for international vessels will be opened on Serpents’ Island. The real aim of

these proposals is revealed by the Ukrainian Minister of Transport:

“... these measures ensure the protection of the status of Serpents’ Island,

allow for it385uture development, guarantee the inviolability of Ukraine’s
borders ...”

5.102. On 3 May 2006, the head of the Odessa Regional Department for ecology and

natural resources declared that

“…in order for this statue [of island] to be recognized b386he international
community, ‘the infrastructure must be developed’.”

383 Article titled “11 million hryvnyas were allotted for the development of Serpents’ Island”,
published by the Ukrainian Press Agency ProUA, on 24January 2006 (Annex RR 21).
384
Article titled “Ehanurov initiates a contest regarding the development of Serpents’ and Tuzla
385ands”, published by the Ukrainian Press Agency ProUA on 10 February 2006 (Annex RR 14).
Article titled “Serpents’ Island might become an off-shore zone”, published by the Press Agency
ITAR-TASS on 18 April 2006 (Annex RR 22).
1635.103. Annex 68 to the Ukrainian Counter-Memorial is also illustrative for the political

target of the measures taken by Ukraine. This Annex is composed by official documents

prepared for a meeting of the Cabinet of Ministers of Ukraine regarding problems of

development of Serpents’ Island. After the first table included in this Annex, a page

entitled “Proposals to the procedure of holding a conference presided over by the First

Vice Prime Minister of Ukraine Dubnaya O. V. on problem issues of Zmiinyi Island” is

inserted. It contains several notes in Italics, probably a hand-written instruction by the

First Vice Prime Minister Dubnaya O. V. The first words in italics on this page are

extremely eloquent:

“O.V. Dubnaya.

1. A political issue.
Request
2. Development of Zmiinyi Island” 387

5.104. Romania has continued to point out that these artificial activities do not create a

new situation opposable to Romania. A number of Notes Verbales, annexed to this
388
Reply, conveyed this message to Ukraine.

5.105. For example by a Note Verbale (no. H2/232) of 25 January 2006, the Romanian

Ministry of Foreign Affairs

“...reiterated its constant position according to which any actions
undertaken by the Ukrainian side having as exclusive aim to prove the

status of island capable to sustain habitation and economic life of its own
(in the sense of article 121 of the Convention on the Law of the Sea of
Montego Bay) for Serpents’ Island cannot produce any effects opposable

to Romania and cannot affect the proceedings taking place in front of the

386 Article titled “Recycling the garbage – one of the biggest problems on Serpents’ Island”,
published by the Ukrainian Press Agency “Reporter” on 3 May 2006 (Annex RR 7).
387
UCM, Annex 68, Documents for a Meeting of Cabinet of Ministers of Ukraine Regarding
Problems of Development of Zmiinyi Island, and Enclosure (1998 Information Concerning the
Implementation of the Cabinet of Ministers of Ukraine Regulation No. 1114 of 8 October 1997).
388 See Note Verbale no. H2/232 of 25 January 2006 of the Romanian Ministry of Foreign Affairs
addressed to the Ukrainian Embassy in Bucharest (Annex RR 23), Note Verbale no. H2/1191 of 19 April
2006 of the Romanian Ministry of Foreign Affairs addressed to the Ukrainian Embassy in Bucharest
(Annex RR 24), Note Verbale no. H2/1416 of 10May 2006 of the Romanian Ministry of Foreign Affairs

addressed to the Ukrainian Embassy in Bucharest (Annex RR 25), Note Verbale no. H2/1724 of 2 June
2006 of the Romanian Ministry of Foreign Affairs addressed to the Ukrainian Embassy in Bucharest
(Annex RR 26).
164 International Court of Justice in the case regarding the delimitation of the
389
maritime spaces of Romania and Ukraine.”

5.106. To summarize, recent Ukrainian attempts to create the artificial appearance of

conditions supporting human habitation are devoid of any legal effect, in so far as they

are undertaken exclusively with this aim. Indeed they constitute an abus de droit, 390 to

which Romania has objected. More importantly, they amount to an admission by the

Ukrainian authorities that at the time of the Additional Agreement – which was, at the

latest the critical date in relation to this case – Serpents’ Island did not have the status

which is now belatedly and artificially sought to be conferred on it.

D. CONCLUSIONS

5.107. According to its natural characteristics, Serpents’ Island is a rock unable to sustain

human habitation or economic life of its own. Consequently it has no entitlement to

continental shelf and exclusive economic zone. This conclusion is strongly supported by

the documentary evidence annexed by Ukraine to its Counter-Memorial. It is in effect

admitted by the recent Ukrainian demarches aimed at artificially changing the character

of Serpents’ Island – measures which in the light of Article 121(3) can have no such

effect.

389 Note Verbale no. H2/232 of 25 January 2006 of the Romanian MFA addressed to the Ukrainian

390assy in Bucharest (Annex RR 23).
See e.g., R. Kolb, “L’interprØtation de l’article 121, paragraphe 3, de la Convention de Montego
Bay sur le droit de la mer: les «rochers qui ne se prŒtent pas à l’habitation humaine ou à une vie
Øconomique propre…»”, (1994) 40 AFDI pp. 876, 903-904: “... il faut retenir la notion d’abus de droit
qui… a ØtØ insØrØe dans la Convention (de Montego Bay) comme principe directeur de son interprØtation et
de son application. Il permettra par exemple d’Øliminer les ilots oø des travaux sont entrepris mala fide

165 Addendum to Chapter 5

Elements from the Annexes of the Counter-Memorial and from the Ukrainian mass-

media showing the character of Serpents’ Island

5.108. As presented above, the Annexes attached by Ukraine to its Counter-Memorial

include a great deal of information showing that, as a matter of fact, Serpents’ Island fully

meets the criteria of Article 121(3). Recent articles and pieces of news published in the
391
Ukrainian media are in the same line.

A. Serpents’ Island is a very small rocky island

5.109. Annex 5 to the Counter-Memorial provides that:

“The Zmiiny Island ... is situated opposite the middle part of the Danube
River Delta, for a distance of 19 miles from the coast. Its coasts are rocky,
392
especially on the West and on the South.”

5.110. Annex 9 to the Counter-Memorial mentions the sterile reality on Serpents’ Island:

“The island is a piece of land with steep shores... It features rocky
ground...” 393

5.111. Annex 11 includes the position of the Ukraine National Academy of Science. The

document mentions Serpents’ Island as “a small (with an area of around 18 hectares)

rocky plot of land ... situated lonely amidst the immense water space”. 394

5.112. Annex 14 of the Counter-Memorial reflects the official view of the Soviet

authorities as to the size of Serpents’ Island. The document, issued by the Hydrographic

pour les rendre habitables, oø des colons sont installØs à la seule fin de dØtournement, ou à propos desquels
toute autre mesure artificielle est prise en vue de contourner l’effet du paragraphe 3.”
391 The examples and quotations included in this Addendum are not exhaustive.
392
UCM, Annex 5, Sailing Directions of the Black Sea and the Sea of Azov for Waters of Ukraine,
Kiev, 2005.
393 UCM, Annex 9, Information Report on Research of Drinking Underground Waters on Zmiinyi
Island in 2002-2004, Dnipropetrovsk, 2004. It also contains a quite detailed description of the rocky
geology of Serpents’ Island.
394
UCM, Annex 11, Appendix to letter No.11/295 of the Ukraine National Academy of Science
dated 30 March 2005.
166Agency of the Military Navy of the USSR considers that “[t]he island is small and
395
bordered with a ridge of submerged stones”.

5.113. According to Annex 48 to the Counter-Memorial, which expresses the opinion of

the Ukrainian scientist S.B.Okhotnikov, Serpents’ Island

“... is a small island measuring only 18 ha in area. Its surface is rocky and

its steep shores drop from 4 to 25 thters to the sea..... Rufius Festus
Avienus, an author of the 4 c BC, provides a remarkably accurate
description of the island. He wrote, ‘Leuke is gray (in color) from its

mountains…where a hollowed cliff forms a series of curved roofs along its
broad face, where cliffs with curved roofs hang over the sea.’... A small
396
piece of land, in the middle of the sea ...”

5.114. The same description is to be found in Annex 49 of the Counter-Memorial:

“L’ile de Leuce n’est qu’un seul bloc de roche calcaire; ses bords sont

ØlevØs, escarpØs et perpendiculaires en plusieurs endroits .... DØpuis le
sommet de ses bords escarpØs le reste de Leuce n’est qu’un rocher
397
convexe...”

5.115. The author quoted here by Ukraine, H. Koehler, mentions that all authors who had

written about Serpents’ Island in Antiquity exaggerated its dimensions. He considers
398
Serpents’ Island “un rocher solitaire et isolØe”.

5.116. According to Annex 50 of the Counter-Memorial, another traveller of the 19 th

century, E. D. Clarke, quoted in, considered Serpents’ Island “... so small, that as we
399
passed, we could view its whole extent .... ... A secluded spot ...”

395 UCM, Annex 14, Black Sea Sailing Directions, 1954, p.76.
396
UCM, Annex 48, Ancient Greek Sites on the Northwestern Coast of the Black Sea, Hellenic
397ndation for Culture, Odesa Branch.
UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes a Achille dans le Pont-
Euxin’ and attached map ‘carte de l’ île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie
ImpØriale des Sciences de St. Petersburg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
398
UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes a Achille dans le Pont-
Euxin’ and attached map ‘carte de l’ île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie
ImpØriale des Sciences de St. Petersburg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
399 UCM, Annex 50, E. D. Clarke, Travels in Various Countries of Europe, Asia and Africa, Third
Edition, Printed for T. Cadell and W. Davies Strand, London, 1813, pp. 648-650. It was also quoted by

Romania in Annex RM 47.
1675.117. In its turn, Annex 54 of the Counter-Memorial speaks about “la petite île des

Serpents”. 400 It describes “Ses bords ... accores et rocailleux ...“ 401

5.118. Annex 57 of the Counter-Memorial is in the same line:

“Outwardly ‘rocky, chopped-off and elevated in the middle’ island had

inconvenient anchorage; ...

P A Drozzhin noted that “... all rocks on the island are made up, like on purpose, out of

regular block stones abundant in quantity, the stone in fact is siliceous, reddish and bluish

402
...’”

5.119. The same Annex 57 mentions the “negligible dimensions” of the small Serpents’

Island that”could hardly protect them (the ships) from ‘roughness’”. It also quotes a

memorandum of a certain Lazarev (Chief Commander of Russian Black Sea Fleet and

Ports around 1835):

“Lazarev writes that, ‘23 miles to the east from Sulina mouth lies the small
Fidonisi island, which in spite of attempts of the seafarers ... in order to
avoid danger, however the stream and the low-lying banks of the Danube

and the negligible size of the island make their efforts during the night
time sometimes futile’.” 403

5.120. Annex 60 includes an article published by the Romanian journalist D.L. St▯hiescu

in the newspaper”Ac▯iunea” (“The Action”), on the 25 March 1938. The passages which

were not included in the translation provided by Ukraine refer to Serpents’ Island as

“... a huge white fang, thrust into the sea water ...a white rock, not too high
above sea waters, seems from far, the top foamy wave turned into stone ...

A remainder of a huge mountain, that fell down under the waves of the sea
a long time ago, it is obviously a rock, covered only with a thin layer of

400
UCM, Annex 54, E. Taitbout de Marigny, Portulan de la Mer Noire et de la Mer d’Azov ou
Description des cotes de ces deux mers a l’usage des navigateurs, Odesa, 1830, pp. 112-114.
401 UCM, Annex 54, E. Taitbout de Marigny, Portulan de la Mer Noire et de la Mer d’Azov ou
Description des cotes de ces deux mers a l’usage des navigateurs, Odesa, 1830, pp. 112-114.
402 UCM, Annex 57, I. V. Tunkina, Russian Science on Classical Antiquities of the Russian South
(XVIII-Mid XIX Centuries), St. Petersburg, 2002, pp.401-450.
403
UCM, Annex 57, I. V. Tunkina, Russian Science on Classical Antiquities of the Russian South
(XVIII-Mid XIX Centuries), St. Petersburg, 2002, pp.401-450.
168 soil strangely kept and remained there. ... this secluded small plot of
land.” 404

405
5.121. In Annex 62 of the Counter-Memorial, the first of the omitted (in the

translation provided by Ukraine) original paragraphs in Romanian reads as follows:

“Situated at 40 km from the Sulina mouth, the island has a height, which
goes from the coast to the centre, from 15 to 40 metres. Its surface is about

14 hectares. The ground is rocky, and the awful climate is under a
permanent salted breeze.” 406

5.122. The article included in Annex 63 of the Counter-Memorial mentions “the rocky

ground, the un-favourable climate” on Serpents’ Island, “this isolated corner of land”,
407
which is “a territory unfitted for a longer stay”.

5.123. Annex 81 to the Counter-Memorial, including an official normative act of the

Odessa Regional State Administration, adopted in 2004, mentions “... the small size of

the island and extreme vulnerability of its ecosystem ...” 408

5.124. The Ukrainian media have the same perception of the smallness and rocky

character of Serpents’ Island. A piece of news of 1 June 2006, released by the press

agency “Podrobnosti”, reads that “Serpents’ Island is, as such, a rocky protuberance ...

situated in the North-Western part of the Black Sea ...” 409

5.125. An article titled “You win Serpents’ Island or you will be defeated in the fight for

it”, published in February 2006 the weekly newspaper “Vlada”, refers to Serpents’ Island

404
405 See, for the full text of the accurate translation of this article, Annex RM 42.
See above, para. 5.75.
406 UCM, Annex 62, Informa▯ia, 22 March 1938. In this Annex, after the fourth paragraph of the
English translation, Ukraine omitted to mark the fact that no less than 9 (nine) other paragraphs were

missing from the original Romanian text. The full translation in English is enclosed by Romania as Annex
RR 17.
407 UCM, Annex 63, Curentul, 21 March 1938. As the Ukrainian Annex omits to include any
translation of the text of this article, Romania translated and included it in Annex RR 16.
408 UCM, Annex 81, Direction No. 535/a-2004 of the Odessa Regional State Administration dated 12

409y 2004.
See the article titled “By the year 2008, a town will be settled on Serpents’ Island”, published by
the Ukrainian Press Agency ProUA on 1 June 2006 (Annex RR 12); see also the article titled “A town will
be settled on Serpents’ Island”, published by the Ukrainian Press Agency PODROBNOSTI.UA on 1 June

2006 (Annex RR 13).
169as a “small piece of land in the sea” and mentions that “[i]n fact, the importance of this

island-rock in the history of the Ukrainian-Romanian diplomacy proved to be bigger than
410
its surface.”

5.126. The Ukrainian scientific expedition that visited Serpents’ Island between 19 and

23 August 2003 was struck by the same arid appearance of this maritime formation:

“Serpents’ Island is a rocky formation .... As demonstrated, the soil on the
island is very-very poor and represents the product of the erosion of
411
different types of rocks, ....”

5.127. Ukrainian officials have mentioned several times the urgent need to consolidate

the shores. For instance, the Interim Governor of the Odessa Region, Boris Zviaghintev,

rd
declared on 1 June 2006 on the occasion of the 3 Investment and Innovations Forum

(Odessa) that “for 2006, 11.6 million hryvnyas (around 2.8 million Euros) were allotted
412
for the consolidation of the shores” of Serpents’ Island. On 3 May 2006, the head of

the Odessa Regional Department for ecology and natural resources declared that

“... in the narrowest parts of Serpents’ Island the width is not bigger than
60 meters, and unless urgent measures are taken in order for the shores to
be consolidated, this spot of land might seize to exist as a single unit”. 413

5.128. This piece of information is confirmed by an article published in October 2006,

which mentions that consolidation of shores

“... is the main problem of the island. And without proper financing this
problem will only aggravate. ‘In the narrowest part a rift appeared’, said

410 Article titled “You win Serpents’ Island or you will be defeated in the fight for it”, published by
the Ukrainian weekly newspaper “Argument. Vlada”, the issue of the week 10 – 16February 2006; author
Stella Maks (Annex RR 8).
411
412 See the Ukrainian site http://www.astronomy.odessa.ua/zmej.htm, last visited on 17July 2006.
See the article titled “By the year 2008, a town will be settled on Serpents’ Island”, published by
the Ukrainian Press Agency ProUA on 1 June 2006 (Annex RR 12); see also the article titled “A town will
be settled on Serpents’ Island”, published by the Ukrainian Press Agency PODROBNOSTI.UA on 1 June
2006 (Annex RR 13).
413
Article titled “Recycling the garbage – one of the biggest problems on Serpents’ Island”,
published by the Ukrainian Press Agency “Reporter” on 3 May 2006 (Annex RR 7).
170 Tamara Penaki, the chief-architect of the project regarding the ‘General
414
Plan for the Development of Serpents’ Island’.”

B. Serpents’ Island has no water resources and is practically devoid of soil,
vegetation and fauna

5.129. Annex 9 of the Counter-Memorial, comprising a study titled “Information about

research on drinking underground waters on Zmiinyi Island in 2002-2004”, undertaken

by the State Committee of the Natural Resources of Ukraine of Dnipropetrovsk in 2004,

was analysed in paragraphs 5.49-5.56 of this Reply.

5.130. The lack of water on Serpents’ Island is also confirmed by Annex 57 of the

Counter-Memorial which includes information regarding a certain archaeological

expedition that took place in the month of June 1841. After only a couple of days (3-5

June),

“The lack of fresh water and the hardly endurable heat made the travellers
hasten their departure.” 415

5.131. Annex 62 includes a reference which underlines the total lack of drinkable water

on Serpents’ Island in the inter-war period:

“In order for the prison to be built a lot of plans and preliminary works are
to be done, among them the most important being providing the island
416
with potable water.”

5.132. This piece of information, regarding the absence of potable water (in the context

of the difficult climate on Serpents’ Island) is confirmed by Annex 63 of the Counter-

Memorial:

414
Article titled ”Snakelike Investments”, published by the Ukrainian newspaper Ekonomicheskye
Izvestia, issue of 24October 2006 (Annex RR 27).
415 UCM, Annex 57, I. V. Tunkina, Russian Science on Classical Antiquities of the Russian South
(XVIII-Mid XIX Centuries), St. Petersburg, 2002, pp.401-450.
416 See UCM, Annex 62, Informa▯ia, 22 March 1938. In this Annex, after the fourth paragraph of the

English translation, Ukraine omitted to mark the fact that no less than 9 (nine) other paragraphs were
missing from the original Romanian text. The full translation in English is enclosed by Romania in Annex
RR 17.
171 “Immediately after the temperature will allow, in April for instance, a
commission of hydrology engineers will decide the way to provide the
deportees with abundant potable water supplies.” 417

5.133. Annex 68 to the Counter-Memorial is composed by official documents prepared

for a meeting of the Cabinet of Ministers of Ukraine regarding problems of development

of Serpents’ Island. On the first (cover) page the date (“March 16, 2006”) and hour

(“16.00”) of the meeting are mentioned. After the first table included in this Annex, a

page entitled “Proposals to the procedure of holding a conference presided over by the

First Vice Prime Minister of Ukraine Dubnaya O.V. on problem issues of Zmiinyi Island”

is inserted. It contains several notes in italics, probably a hand-written instruction by the

First Vice Prime Minister Dubnaya. The last words of this page are eloquent as to the

officially assumed situation on Serpents’ Island: “d) Water. To search water.” 418

5.134. The absence of water is confirmed by Annex 73 of the Counter-Memorial, which

admits that:

“[d]ue to uncertain volumes of water stocks on the island, ther419s always
the reserve stock of potable water keeping in reservoirs.”

5.135. Annex 84 states that because the water reserves are not clear, “a reserve quantity
420
of potable water is constantly delivered to the island and kept in tanks”. This is

another admission of the lack of fresh water on Serpents’ Island, indicating that the

authorities are forced to bring water to Serpents’ Island from the mainland:

417 See UCM, Annex 63, Curentul, 21 March 1938. As the Ukrainian Annex omits to include any

418nslation of the text of this article, Romania translated and included it in Annex RR 16.
Italics in the English text of the translation. UCM, Annex 68, Documents for a Meeting of Cabinet
of Ministers of Ukraine Regarding Problems of Development of Zmiinyi Island, and Enclosure (1998
Information Concerning the Implementation of the Cabinet of Ministers of Ukraine Regulation No. 1114 of
8 October 1997).
419
UCM, Annex 73, Information concerning the Fulfillment of the Complex Program of the Further
Development of the Infrastructure and the Realization of the Economic Activities on Zmiynyi Island and
the Continental Shelf for the Period of 2002-2004, The State Regional Administration Odesa, the Central
Administrative Board of Economy, November 2004.
420 UCM, Annex 84, Analytical Brief on the Implementation of the Comprehensive Program for
Further Development of the Infrastructure and Conduct of Economic Activities on Zmiinyi Island and on

the Continental Shelf.
172 “The Regional Administration completed a package of works connected
421
with the water supply to the island (delivered potable water)...”

5.136. At its turn, Annex 91 of the Counter-Memorial contains inter alia an “Act of

transfer of barrack housing and property of military station No. 1 on Zmiinyi Island” of

20 June 2002. It shows how vital is to maintain in good conditions the reservoirs in

which the water brought exclusively from the mainland is kept (this being the only

source):

“Reservoir for water (R-10, 4 pieces) and water-supply system need
substitution, in winter it is impossible to organize washing for
personnel.” 422

The same Annex 91 contains another document with the same title dated 26 June 2002
423
which mentions that “Buildings ... are not equipped with water-supply system.”

5.137. Another document of the same Annex, a table entitled “General Plan of measures

for providing and development of Zmiinyi Island for 1997”, includes the following

references under the heading “Name of measures and works”:

“3. Providing of drinking water according to norms ...

10. Provide for quick transportation of ... water ...
424
14. Provide supply of water using for building by water-tanks boats.”

5.138. The lack of water on Serpents’ Island is also confirmed by recent statements of

Ukrainian officials as reflected by articles in Ukrainian media. For example, on 3 May

2006, the head of the Odessa Regional Department for ecology and natural resources

declared that:

“... the island continues to suffer because of the lack of potable water. ...

421
UCM, Annex 84, Analytical Brief on the Implementation of the Comprehensive Program for
Further Development of the Infrastructure and Conduct of Economic Activities on Zmiinyi Island and on
the Continental Shelf.
422 UCM, Annex 91, Direction No. 277-r of the Cabinet of Ministers of Ukraine dated 30 May 2002
and Subsequent Resolutions.
423
424 Ibid.
Ibid.
173 As far as the supply with potable water ... for technical necessities water is
425
brought from the mainland.”

5.139. Regarding the quasi-absence of vegetation and the tough climate on Serpents’

Island, Annex 9 of the Counter-Memorial reads that it:

“... features poor grassy vegetation. During a hot summer season all

vegetation gets withered.”...

The Black Sea western part, where Zmiinyi island is situated, features
temperate climate with a prevailing cyclonic type, continental and sea air

mass circulation. It results in frequent changeability of all meteorological
characteristics and the weather on the whole... Considerable fluctuations

of the air temperature and n426uniform distribution of precipitation by
season prove the above.”

5.140. Annex 49 of the Counter-Memorial contains a description of both soil and flora

on the Serpents’ Island:

“La surface de Leuce n’est recouverte que d’une couche de terre d’a peine

un ou deux pieds de profondeur, de sorte que presque par tout le rocher se
montre a dØcouvert... Mais il n’y avoit dans tout ile ni arbres ni arbustes,

pas mŒme ces espŁces427 buissons qui croissent ordinairement dans les
terrains pierreux.”

5.141. Annex 50 of the Counter-Memorial mentions that Serpents’ Island “... is quite
428
bare, being covered only with a little grass, and very low herbage.”

5.142. The Ukrainian scientific expedition that visited Serpents’ Island between 19 and

23August 2003 produced the same description:

425 Article titled “Recycling the garbage – one of the biggest problems on Serpents’ Island”,

426lished by the Ukrainian Press Agency “Reporter” on 3 May 2006 (Annex RR 7).
UCM, Annex 9, Information about research on drinking underground waters on Zmiinyi Island in
2002-2004, State Committee of the Natural Resources of Ukraine, Dnipropetrovsk, 2004.
427 UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes a Achille dans le Pont-
Euxin’ and attached map ‘carte de l’ île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie
ImpØriale des Sciences de St. Petersburg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
428
UCM, Annex 50, E. D. Clarke, Travels in Various Countries of Europe, Asia and Africa, Third
Edition, Printed for T. Cadell and W. Davies Strand, London, 1813, pp. 648-650.
174 “As demonstrated, the soil on the island is very-very poor and represents
the product of the erosion of different types of rocks, or as the soil
specialists name it, alluvial stratum. The main vegetation of the island –
429
wild herbage.”

5.143. Annex 85 of the Counter-Memorial includes the “List of Island’s Flora, Fauna and

Fish Species”, according to which there are no trees, little vegetation (withered and burnt

in the most part of the year), and no species of mammals, except for two species of rats

and two of bats.

C. Serpents’ Island cannot sustain human habitation

5.144. Annex 10 of the Counter-Memorial includes a description of the rough climate on

430
Serpents’ Island (which explains why the idea of a balneal station on Serpents’ Island

was abandoned). The text reads as follows:

“But the very tough and irregular winds, which blow terribly from autumn
to spring, the difficulty to bring supplies and especially the costs of
building were as many reasons to give up the making of this project.” 431

5.145. Annex 12 of the Counter-Memorial, including a Russian document of 1903, reads

that “... nobody dwells on the island.” 432 There is identical language in Annex 13 of

Ukraine’s Counter-Memorial. 433

5.146. Annex 49 of the Counter-Memorial mentions, in the same line, that

429
See the Ukrainian site http://www.astronomy.odessa.ua/zmej.htm, last visited on 17July 2006.
430 See UCM, para. 7.66.
431 See page 51, last paragraph, not translated into English, continued on page 52 of the Romanian

original text included in UCM, Annex 10, R.I. C▯linescu, ‘Insula ▯erpilor. Schi▯▯ monografic▯’, in Analele
432rogei, Anul XII, Cern▯u▯i, 1931, pp. 47, 53 and 57.
UCM, Annex 12, Black Sea and Sea of Azov Sailing Directions, Fourth Edition, St. Petersburg,
1903, pp. 64-65.
433 UCM, Annex 13, Black Sea Sailing Directions, Seventh Edition, Leningrad, 1931, pp.47-48. The
same message is conveyed by UCM, Annex 10, R.I. Calinescu, ‘Insula Serpilor. Schita monografica’, in

Analele Dobrogei, Anul XII, Cernauti, 1931, pp. 47, 53 and 57.
175 “Aucun home n’habitoit Leuce... les anciennes auteurs nous disent que
l’île d’Achille Øtoit dØserte et sans habitants ....” 434

The same Annex 49 invokes the difficult environment on Serpents’ Island:

“...le sol est peuplØ d’un grand nombre de serpents assez longues et de
435
couleur noire ... Cette île est en effet infestØe par ces reptiles... ”

th
5.147. Annex 50 of the Counter-Memorial mentions the situation in the 19 century: “No
436
human being dwells there...“

5.148. In a footnote of the text enclosed in the same Annex 50 another traveller, count

Jean Potocky, who had passed near the Serpents’ Island in 1784, is quoted:

“L’on me repondait alors qu’il Øtoit difficile d’y aborder; tant parceque la

cote Øtoit dangereuse que parceque la terre Øtoit couverte de serpents
vØnimeux.“ 437

The story of Potocky is also mentioned in Annex 57 of the Counter-Memorial, in a
438
Russian study of 2002.

5.149. Annex 52 of the Counter-Memorial presents the difficult conditions on Serpents’

Island which are not fit for any kind of human habitation; they are also evidenced by the

story, told by Ovid in one of his poems and included in this Annex, about the fate of the

ruler of Amastria who was exiled on Serpents’ Island by his father-in-law Mitridathes:

“... he died from hunger and cold.” 439

434
UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes a Achille dans le Pont-
Euxin’ and attached map ‘carte de l’ île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie
ImpØriale des Sciences de St. Petersburg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
435 UCM, Annex 49, M. Koehler, ‘MØmoire sur les îles et la course consacrØes a Achille dans le Pont-

Euxin’ and attached map ‘carte de l’ île de Leuce, aujourd’hui Ilan-Adassi’, in MØmoires de l’AcadØmie
ImpØriale des Sciences de St. Petersburg, Tome X, pp.531-534, 556-559, 562-564, 566-571, 599-615.
436 UCM, Annex 50, E. D. Clarke, Travels in Various Countries of Europe, Asia and Africa, Third
Edition, Printed for T. Cadell and W. Davies Strand, London, 1813, pp. 648-650.
437 UCM, Annex 50, E. D. Clarke, Travels in Various Countries of Europe, Asia and Africa, Third

438tion, Printed for T. Cadell and W. Davies Strand, London, 1813, pp. 648-650.
UCM, Annex 57, I. V. Tunkina, Russian Science on Classical Antiquities of the Russian South
(XVIII-Mid XIX Centuries), St. Petersburg, 2002, pp.401-450.
439 UCM, Annex 52, S.B.Ochotnikov, Odesa National University, Notes of the Faculty of History,

Odesa, 1996, pp. 46-59.
1765.150. Annex 57 reminds the “testimonies of antique authors who wrote about

forbiddance to settle on Leuke” and the opinions of certain historians who “believe that

the island could not be inhabited by ‘somewhat stable population in the true sense of the
440
word’”.

5.151. The situation nowadays is not different: according to the Ukrainian press: “At

present, it is inhabited by border guards, meteorologists and ecologists.” 441 Another piece

of information published on the 10 February 2006 by the Ukrainian Press Agency

“ProUA” mentions that”[o]n the island there are border guards’ detachments,
442
representatives of the Ministry of Interior and of the anti-aircraft intervention troops”.

Another article, entitled “There, where the wind always blows from the sea”, published in

the newspaper “Golos Ukraini” on the 29 September 2006, confirms this reality: it

enumerates the persons temporarily working there – only “12 border guards, one

lighthouse keeper and one postal worker, two ornithologists and few scientific

443
researchers”. The border guards are hired on contractual basis, being monthly rotated,
444
and the researchers stay only 10 month per year, due to the difficult life conditions.

5.152. Annex 60 includes an article published by the Romanian journalist D.L. St▯hiescu

in the newspaper “Ac▯iunea” (“The Action”), on the 25 March 1938. The passages

regarding the conditions on Serpents’ Island, which were not included in the translation

provided by Ukraine, read as follows:

“... In the huge desert of the somber surrounding waters, the island is itself

desert, inhabited only by birds ... and by black serpents that crawl on the
stones.

440
UCM, Annex 57, I. V. Tunkina, Russian Science on Classical Antiquities of the Russian South
441III-Mid XIX Centuries), St. Petersburg, 2002, pp.401-450.
Article titled “By the year 2008, a town will be settled on Serpents’ Island”, published by the
Ukrainian Press Agency ProUA on 1 June 2006 (Annex RR 12); see also the article titled “A town will be
settled on Serpents’ Island”, published by the Ukrainian Press Agency PODROBNOSTI.UA on 1 June 2006

442nex RR 13).
Article titled “Ehanurov initiates a contest regarding the development of Serpents’ and Tuzla
Islands”, published by the Ukrainian Press Agency ProUA on 10 February 2006 (Annex RR 14).
443 Article titled “There, where the wind always blows from the sea”, published by the Ukrainian
newspaper Golos Ukraini (official newspaper of the Supreme Rada), issue of 29 September 2006; author:
Anastasia Sirotkina (Annex RR 15).
444
Ibid.
177 With its poverty and isolation, the island can be useful for us only as a
guard post in front of the mouths of the Danube. ...

We don’t know yet: will it be a cursed land, regarded by everybody with

horror, just li445the French Guyana, or it will be a place for meditation, of
repentance ...”

446
5.153. Annex 63 of Counter-Memorial includes the following description:

“The designation of Serpents’ Island as place of deportation for political

detainees of Romania is a result of its own geographical situation.

... The rocky ground, the un-favourable climate make out of this isolated
corner of land a territory unfitted for a longer stay. The only permanent
dwellers of the islet are the serpents – once upon a time in terrible

quantities, today in lesser number – serpents which gave its name.

A sea lighthouse with a few poor rooms for the guards and for the few
border guards represents the whole comfort of this kingdom of reptiles.
But neither the guards, nor the soldiers do resist much to the solitude.

They are changed on teams. Life is neurasthenic because of its
exaggerated monotony.

A small spot of vegetable garden, which the guards are trying hard to
maintain during summer, hardly survives to the climate and atmosphere

soaked with salt. The rest of the island is improper to any agriculture.

Obviously, in order to be able to host a number of political detainees,
Serpents’ Island should be radically developed.”

5.154. Another example in this respect can be found in Annex 66 of the Counter-

Memorial:

“9. COMFORTS

... 2. Water – as delivered, artesian well 3. Heating system – no 4. Gas –
no ... 6. Air conditioned – no 7. Interior telephone – no ... 9. Television –

no 12. Warmhouse – no 13. Nearest inhabited place – Valkovo, 60 km

445
446 See, for the full text of this article, accurately translated, Annex RM 42.
See UCM, Annex 63, Curentul, 21 March 1938. The article included in this Annex has no
translation whatsoever with the exception of the title (despite of the fact that its cover page mentions, in
brackets, ”English translation and Romanian original”). Romania translated and included it in Annex RR
16.
178 14. School – no ... 16. % bonus – 30% for distance 17. Food ration –
447
given”

5.155. According to the same Annex 66, because of the long distance to the nearest

civilised place, the guardians of the lighthouse receive a significant bonus.

5.156. According to Annex 67, similar incentives are also provided for the soldiers

located on duty due to this difficult environment:

“The Cabinet of Ministers of Ukraine HAS RESOLVED:

...

8. To enact an increase of official salaries for military personnel of the
Armed Forced and Frontier Troops who perform their duty on the island

by 50 percent commencing on 1 January 1996 and their annual
compulsory medical examination.

9. ... the duty period of military personnel of the Armed Forced and

Frontier Troops on Zmiinyi Island shall be added to their record for
granting pensions as follows: one month of the duty shall stand for two
months of the duty.” 448

So, far from having conditions favourable to sustain human habitation, Serpents’ Island is

such a hostile and difficult environment that the officials sent on duty, for short periods of

time, on rotational basis, are granted a supplementary half of salary and the time of stay

on it is counted double for pension rights.

5.157. This situation is confirmed by other Annexes to the Ukrainian Counter-Memorial

as well. Annex 68 also comprises official documents of the Cabinet of Ministers of

Ukraine. The first table included in this Annex mentions in item 8 the same text quoted

above, in the previous paragraph. 449

447 UCM, Annex 66, Document Evidencing the Layout of the Lighthouse.
448 UCM, Annex 67, Resolution of the Cabinet of Ministers of Ukraine No. 1009 dated 18 December

4495.
UCM, Annex 68, Documents for a Meeting of Cabinet of Ministers of Ukraine Regarding
Problems of Development of Zmiinyi Island, and Enclosure (1998 Information Concerning the
Implementation of the Cabinet of Ministers of Ukraine Regulation No. 1114 of 8 October 1997).
1795.158. A recent article in the Ukrainian press, published in September 2006, confirms the

toughness of the environment on Serpents’ Island:

“It is said that the work here is not different than that at any other border
post. In my view, it is not true! According to the statements of the head of
the border post, captain Vitalii Balchiuk, the main problems faced by the

border guards here are of medical type – the colds. The climate dictates its
own conditions. So, it is explainable why there are permanent deposits of
supplies for 3-4 months. In his turn, the physician of the island, major

Vladimir Sanoed, has to fight not only the traditional gastritis, but also the
psychological problems, especially in the autumn-winter period... At the
kitchen of the border guards, soldier Pavel Parhomenko is the chief. He

bakes the bread by himself! Each and every day he cooks six litres of
soup, 2-2.5 kilos of macaroni. The menu of the soldiers here is in
accordance with the ‘marine ration’, that is to the ordinary daily military

ration 50 grams of salami, 150 grams of juice, 20 grams of condensed
milk, biscuits and jam are added.” 450

This shows eloquently difficulties encountered on Serpents’ Island in the present times,

as in the past: the roughness of the climate, the medical problems – gastritis, which are a

result of the constant stress caused by the rough environment and of eating each and

every day the same poor food of the military rations (with no fruits or vegetables, as they

can not be cultivated on Serpents’ Island), but also psychological problems, also
451
witnessed in the inter-war period at the Romanian border guards and lighthouse

keepers – and the exclusive dependence from the mainland as far as food and other

essential supplies are concerned.

5.159. Annex 89 of the Counter-Memorial mentions, with reference to Serpents’ Island,
452
“... the specific nature of the remote and locked territory...”

5.160. Annex 91 shows how complex and difficult task is to create conditions on

Serpents’ Island, as a large number of State central institutions are involved:

“Ministry of Transport, Ministry of Ecological Resources, Ministry for

Energy and Fuel, Ministry of Agriculture, Ministry of Health, Ministry for

450 Article titled “There, where the wind always blows from the sea”, published by the Ukrainian

newspaper Golos Ukraini (official newspaper of the Supreme Rada), issue of 29 September 2006; author:
451stasia Sirotkina (Annex RR 15).
452 See, for example, RM, paras. 10.67, 10.85.
UCM, Annex 89, Decree No. 700/A-2003 of the Odessa Regional Administration dated 29 August
2003.
180 Extraordinary Situations, Ministry of culture and State Committee of

Ukraine for Communication shall determine specific arrangements for
development of economical activity on the island and creation of
conditions for permanent residence of people there ...” 453

5.161. In this context, the recent statement of Ukraine’s Prime-Minister Ehanurov, on the

occasion of a visit on Serpents’ Island in spring 2006, acknowledged its uninhabitable

status:

“There is a need for infrastructure. In order for us to settle on the island,

we need to have people living there on a permanent basis, and for them to
feel good there ....”454

D. Serpents’ Island cannot sustain any economic life of its own

5.162. Paragraphs 3.27-3.28 and Figure 3-5 of the Counter-Memorial discuss the social

and economic importance of Ukrainian centres of the Black Sea area. Serpents’ Island is

not mentioned in Figure 3-5. According to Figure 3-5, Serpents’ Island has neither

economic activity nor population.

5.163. The lack of conditions as to the easy access to Serpents’ Island is an impediment

for any economic life on Serpents’ Island. Annex 5 of the Counter-Memorial, which

includes a very recent official Ukrainian document of 2005, mentions the newly built

berth on the north-eastern part of Serpents’ Island, which proves to be unsatisfactory as to

its purposes to provide for safe access:

“But the berth is unreliable as the ground ho455 the anchor badly and it is
insufficiently protected from disturbance.”

5.164. Annex 12 of the Counter-Memorial, which includes an official Russian document

of 1903, pictures an identical problem more than 100 years before:

“Anchorage at the island is inconvenient due to the bad ground; however

vessels sailing to Sulina with the NO wind sometimes land at the southern

453 UCM, Annex 91, Direction No. 277-r of the Cabinet of Ministers of Ukraine dated 30 May 2002
and Subsequent Resolutions.
454
Article titled ”On the Ukrainian island of the Black Sea a hotel for ‘Robinsons’ was opened”,
455lished by the Press Agency NEWSru.com, on 6ay 2006 (Annex RR 28).
UCM, Annex 5, Sailing Directions of the Black Sea and the Sea of Azov for Waters of Ukraine,
Kiev, 2005.
181 side at 12 sajenes and 13 sajenes although the island due to its tiny stretch

provides little protection during the roughness... The disembarking on the
island is not allowed. ... the NW side is not deep enough for boats and
456
abundant in stones.”

5.165. Annex 13 includes an official Soviet document of the same kind of 1931 having

an identical language, after 30 years. It adds information about sunken ships in the near

vicinity of Serpents’ Island, and shows the difficulties of access to it. 457

5.166. Annex 14 of the Counter-Memorial includes another official Soviet document of

the same kind of 1954, when the Serpents’ Island was already under Soviet authority. It

depicts the same reality:

“The island is ... bordered with a ridge of submerged stones....

This anchoring site is not reliable since the ground within it holds poorly 458
the anchor and the site is inadequately protected from the roughness.”

5.167. The adverse climate is another obstacle to economic life. Annex 9 of the Counter-

Memorial reads:

“Maximum air temperature is observed from June to August and reaches

40°C. Average number of frosty days during a year is approximately 100
days. Maximum depth of frost penetration in ground is 51 cm. The sea

becomes frozen in December-January. In second half of February the ice
begins to thaw and the sea becomes ice-free in March.

Construction of the berth was finished in 2002. ... A period from May to

October is considered the most favourable one for navigation of ships to
Zmiinyi Island. During the rest of the period the island can be reached
459
only with helicopter.”

456 UCM, Annex 12, Black Sea and Sea of Azov Sailing Directions, Fourth Edition, St. Petersburg,
1903, pp. 64-65.
457 UCM, Annex 13, Black Sea Sailing Directions, Seventh Edition, Leningrad, 1931, pp. 47-48.
458 UCM, Annex 14, Black Sea Sailing Directions, 1954, p.76.
459
UCM, Annex 9, Information about research on drinking underground waters on Zmiinyi Island in
2002-2004, State Committee of the Natural Resources of Ukraine, Dnipropetrovsk, 2004.
1825.168. Annex 96 mentions the lack of frequency of transport to Serpents’ Island: from
460
June to December 2004 there were only 16 voyages.

5.169. This frequency is insufficient for the development of an economic life, for the rest

of the year transportation by sea being impossible. The Ukrainian press supports this

conclusion; an article published in September 2006 mentions:

“As in the past times, it is difficult to get here. So, when the border guards

invited me to visit border post number 1474, I had no hesitation: we
fly!” 461

5.170. According to Annex 73 of the Counter-Memorial, a transportation schedule was

established only in 2004 as part of the program to elevate the status of Serpents’ Island. 462

5.171. Annex 96 discloses the full dependence of Serpents’ Island on supplies brought

from the mainland:

“The said voyages were made with the purpose of ensuring vital activities
on the island, personnel rotation, and delivery of building materials and
food.” 463

5.172. This exclusive dependence from the mainland is also reflected in the Ukrainian

press:

“The climate dictates its own conditions. So, it is explainable why there
are permanent deposits of supplies for 3-4 months.“ 464

460
UCM, Annex 96, Letter from Ostrivne Odesa Oblast Municipal Enterprise to Oleksandr
Mychaylovych Hordiyenko, Chief of the Main Department of Economy, Oblast State Administration dated
5 March 2005, Enclosing Transport Communications with Serpents’ Island for 2004.
461 Article titled “There, where the wind always blows from the sea”, published by the Ukrainian
newspaper Golos Ukraini (official newspaper of the Supreme Rada), issue of 29 September 2006; author:

462stasia Sirotkina (Annex RR 15).
UCM, Annex 73, Information concerning the Fulfillment of the Complex Program of the Further
Development of the Infrastructure and the Realization of the Economic Activities on Zmiynyi Island and
the Continental Shelf for the Period of 2002-2004, The State Regional Administration Odesa, the Central
Administrative Board of Economy, November 2004.
463
UCM, Annex 96, Letter from Ostrivne Odesa Oblast Municipal Enterprise to Oleksandr
Mychaylovych Hordiyenko, Chief of the Main Department of Economy, Oblast State Administration dated
5 March 2005, Enclosing Transport Communications with Serpents’ Island for 2004.
464 Article titled “There, where the wind always blows from the sea”, published by the Ukrainian
newspaper Golos Ukraini (official newspaper of the Supreme Rada), issue of 29 September 2006; author:

Anastasia Sirotkina (Annex RR 15).
1835.173. The situation is no different from that on Serpents’ Island more than a century

before. The official Russian document of 1903 quoted by Ukraine in Annex 12 pictures

the same dependence: “foodstuff for those [lighthouse keepers] is delivered from

Sulina.” 465

5.174. This absolute dependence, incompatible with both human habitation and

economic life of its own, is proved also by the declared purpose of the berth (to ensure

delivery of essential supplies) in Annex 5 of the Counter-Memorial:

“The berth is equipped with engineering networks for diesel oil and water
466
delivery to the island ...”

5.175. The same purpose is described by the site of the scientific expedition (already

mentioned in this Reply) which visited Serpents’ Island in August 2003:

“In 2002 a permanent berth was built in order to ensure the delivery of
goods of all kind (potable water, food, fuel etc.)” 467

5.176. Annex 57 of the Counter-Memorial includes an interesting assessment as to the

lack of natural resources on Serpents’ Island, except for rocks, made already in 1836 by

the Russian captain P. A. Drozzhin. But neither these arid stones are good for nothing,

not even to use them as raw material for construction purposes:

“... all rocks on the island are made up, like on purpose, out of regular
block stones abundant in quantity, the stone in fact is siliceous, reddish

and bluish hard and absolutely unsuitable for the lighthouse
construction.” 468

5.177. Annex 74 provides for an explanation why Ukraine adopted a lot of measures

aimed at creating an unconvincing artificial appearance of Serpents’ Island:

465 UCM, Annex 12, Black Sea and Sea of Azov Sailing Directions, Fourth Edition, St. Petersburg,
1903, pp. 64-65.
466 UCM, Annex 5, Sailing Directions of the Black Sea and the Sea of Azov for Waters of Ukraine,
Kiev, 2005.
467
468 See the Ukrainian site http://www.astronomy.odessa.ua/zmej.htm, last visited on 17 July 2006.
UCM, Annex 57, I. V. Tunkina, Russian Science on Classical Antiquities of the Russian South
(XVIII-Mid XIX Centuries), St. Petersburg, 2002, pp.401-450.
184 “Objective and Task

The Programme objective is to create the appropriate conditions for

human habitation and conduct of the economical and other activities on
the island.

The main tasks of the Programme are:

To complete the work connected with documentary legitimation of the
Island status as an administrative and territorial formation;

To carry out measures with the purpose of: ...

Ensuring reliable connection and transport communication with the island;

Diversification of the economical and other activities; ...” 469

5.178. Annex 87 of the Counter-Memorial shows in his turn that a Regional Utility

Company called “Ostrivne” (from “ostriv” – “island” in Ukrainian) was established

“with the purpose of maintenance of life sustenance system and stirring up
the process of integration of the island with the economy of the region
470
(province) and settling matters concerning its adjustment”

5.179. Annex 88 is in the same line and reveals that, in order to create conditions for

economic life on Serpents’ Island, the objectives of the Ukrainian authorities are to

promote the “adjustment of administrative, economic and residential sectors of the

island” 471and the “integration of the island with the regional economy”.

5.180. Annex 95 comprises documents relating to the opening of the branch of the Aval

Bank on Serpents’ Island. The decision of the board of the Bank shows which operations

are allowed to the branch:

469 UCM, Annex 74, Resolution No. 713 of the Cabinet of Ministers of Ukraine dated 31 May 2002
regarding the approval of the”Comprehensive Programme of the Further Development of the Infrastructure
and the Conduct of the Economic Activities on Zmiynyi Island and the Continental Shelf”.
470 UCM, Annex 87, Resolution No. 111-XXIV of the Odessa Regional Council dated 10 January

4713.
UCM, Annex 88, Articles of Association of Ostrivne, Odesa Regional State Administration dated
21 March 2003.
185 “1. Receipt of deposits from legal entities and individuals, in particular -

draw deposits from individuals...

5. Transactions with foreign values:

472
- non-commercial transactions with foreign values;”

So, even if this branch would be operational (which is not the case), 473according to the

evidence included by Ukraine in Annex 95 it has no competence to perform commercial

operations.

5.181. Annex 67 of the Counter-Memorial, also an official document (a normative act) of

the Cabinet of Ministers of Ukraine of 1995, discloses the true intentions of Ukraine:

“5. The Ministry of Foreign Affairs shall proceed with a package of
measures aimed at legitimation of the state boundary with the Republic of

Romania proceeding from that that in conformity with the United Nations
Convention on the Law of the Sea the island has the exclusive economic

zone and continental shelf of its own...

7. In view of fulfillment of tasks and protection of Ukrainian sovereign
rights in the exclusive economic zone in the vicinity of the island the
Ministry of Finance shall appropriate 512 milliard Karbovanets for the

purchase of fuel for ships and aircrafts to the State Committee for the
National Boundary Protection.” 474

5.182. Annex 68 of the Counter-Memorial mentions the same intentions of legitimating

Ukraine’s claims to continental shelf and exclusive economic zone (see item 5 of the first

table included in this Annex).

5.183. In its turn, the Analytical Brief on the implementation of the Comprehensive

Programme of the Further Development of the Infrastructure and Conduct of the

Economical Activity on Serpents’ Island and on the Continental Shelf, included by

Ukraine in its Annex 84, also express with clarity that the said program was approved

472 UCM, Annex 95, Documents Relating to the Opening of a Branch of Aval Bank on Serpents’

473and.
474 See above, para. 5.91.
UCM, Annex 67, Resolution of the Cabinet of Ministers of Ukraine No. 1009 dated 18 December
1995.
186 “with the purpose of implementation of a package of measures aimed

at ensuring proper conditions for human habitation a475for conduct
of economic and other activities on the island”.

5.184. Annex 68 to the Counter-Memorial is also illustrative for the political target of the

measures taken by Ukraine. This Annex is composed by official documents prepared for

a meeting of the Cabinet of Ministers of Ukraine regarding problems of development of

Serpents’ Island. After the first table included in this Annex, a page entitled “Proposals to

the procedure of holding a conference presided over by the First Vice Prime Minister of

Ukraine Dubnaya O.V. on problem issues of Zmiinyi Island” is inserted. It contains

several notes in italics, probably a hand-written instruction by the First Vice Prime

Minister Dubnaya. The first words in Italics on this page are eloquent:

“O.V. Dubnaya.

1. A political issue.

Request

2. Development of Zmiinyi Island” 476

475 Bold in the original text. UCM, Annex 84, Analytical Brief on the Implementation of the
Comprehensive Program for Further Development of the Infrastructure and Conduct of Economic

476ivities on Zmiinyi Island and on the Continental Shelf.
UCM, Annex 68, Documents for a Meeting of Cabinet of Ministers of Ukraine Regarding
Problems of Development of Zmiinyi Island, and Enclosure (1998 Information Concerning the
Implementation of the Cabinet of Ministers of Ukraine Regulation No. 1114 of 8 October 1997).

187 CHAPTER 6

THE RELEVANT CIRCUMSTANCES

A. THE ROLE OF RELEVANT CIRCUMSTANCES IN MARITIME

DELIMITATION

6.1. The 1997 Additional Agreement lists the principles on which the Parties – thus
477
the Court – are to base themselves in order to carry out the delimitation of the

continental shelf and the exclusive economic zones in the Black Sea. Among these

principles, paragraph 4(e) mentions:

“The principle of taking into consideration the special circumstances of
the zone submitted to delimitation.”

In so doing, the Additional Agreement refers to a well-accepted principle applicable in

maritime delimitations even without a text.8

6.2. Ukraine correctly states that “‘relevant circumstances’ and ‘special

circumstances’ in the context of maritime delimitation are terms of art; they connote

circumstances which relate to the delimitation being undertaken in such a way as to affect

the outcome of that delimitation”.479 But in its Counter-Memorial it mixes up special

circumstances and other matters. It includes under the general heading “Relevant

Circumstances” a lengthy discussion of “The Geographical Factors” 480 and “State

Activities in the Relevant Area”,81which is clearly misplaced. Moreover it ignores the

most important special circumstance in the present case, i.e., the presence of Serpents’

Island, and it misrepresents the impact of the Black Sea as an enclosed sea and of the

existing delimitation agreements in the Black Sea.

477 See above paras. 2.7-2.10, responding to UCM, paras. 2.13-2.20.
478 See RM, para. 8.75-8.85.
479 UCM, para. 4.42.
480 UCM, paras. 8.1-8.38.
481
UCM, para. 8.39-8.67.
1886.3. As the Court has clearly explained – and as Ukraine purports to accept: 482

“… special circumstances are those circumstances which might modify the

result produced by an unqualified application of the equidistance principle.
General international law, as it has developed through the case-law of the
Court and arbitral jurisprudence, and through the work of the Third United

Nations Conference on the Law of the Sea, has employed the concept of
‘relevant circumstances’. This concept can be described as a fact necessary
483
to be taken into account in the delimitation process.”

6.4. This does not mean that any element, whether geographical, historical or political,

which may be taken into consideration in order to establish a maritime border, is a

relevant circumstance for the purposes of maritime delimitation. Only those facts “which

might modify the result produced by an unqualified application of the equidistance

principle” are included. Otherwise the term would lack content.

6.5. The delimitation of the maritime zones beyond the territorial sea is a two stage

process: the now well-established method “involves first drawing an equidistance line,

then considering whether there are factors calling for the adjustment or shifting of that
484
line in order to achieve an ‘equitable result’.” In other words, the relevant

circumstances come after a line has first been drawn according to the principle of

equidistance. In doing so, relevant geographical factors will necessarily have been taken

into account, in particular, what Ukraine refers to as “the physical geographical

framework of the area”. The line is provisionally drawn in view of these factors; if

necessary, the line is then adjusted in view of any special circumstances which may be

demonstrated to exist.

6.6. The 1997 Additional Agreement confirms the Parties’ adherence to this method

and to the distinction to be made between the two phases of the delimitation process.

Paragraph 4(b) refers to the principle of equidistance; the “taking into consideration [of]

482
483 See para. 6.2 above.
Maritime Delimitation in the Area between Greenland and Jan Mayen (Denmark v. Norway), ICJ
Reports 1993, p. 38, at p. 62 (para. 55); Land and Maritime Boundary between Cameroon and Nigeria
(Cameroon v. Nigeria: Equatorial Guinea intervening), ICJ Reports 2002, p. 303, at p. 442 (para. 289).
484 Ibid., p. 441 (para. 288); see also p. 443 (para. 293), also Maritime Delimitation and Maritime
Questions between Qatar and Bahrain (Qatar v. Bahrain), ICJ Reports 2001, p. 40, at p. 104 (para. 217);
Barbados v The Republic of Trinidad and Tobago, Arbitral Award of 11 April 2006, 45 ILM, p. 800, at p.
839 (para. 242).

189special circumstances” is referred to in paragraph 4(e). Both apply; but they must be

applied in turn – which implies that the relevant factors too are distinct and cannot be

used for a multipurpose function as Ukraine would like.

6.7. Having adjusted the provisional equidistance line to take account of any special

circumstances, there is then a third step, which takes into account proportionality of

coastal lengths not as a criterion for drawing the initial line nor as a special circumstance

but in order to check the equitable character of the result thus provisionally achieved. As

the Chamber put it in the Gulf of Maine case:

“… the fact that to take into account the extent of the respective coasts of
the Parties concerned does not in itself constitute either a criterion serving

as a direct basis for a delimitation, or a method that can be used to
implement such delimitation. The Chamber recognizes that this concept is
put forward mainly as a means of checking whether a provisional

delimitation established initially on the basis of other criteria, and by the
use of a method which has nothing to do with that concept, can or cannot
be considered satisfactory in relation to certain geographical features of

the specific case, and whether it is reasonable or otherwise to correct it
accordingly”. 485

6.8. Here again, paragraph 4 of the 1997 Additional Agreement distinguishes between

“the principle of equity and the method of proportionality, as they are applied in the

practice of states and in the decisions of international courts regarding the delimitation of

continental shelf and exclusive economic zones”, which is embodied in sub-paragraph (c)

485
Delimitation of the Maritime Boundary in the Area of the Gulf of Maine (Canada/United States of
America), ICJ Reports 1984, p. 246, at p. 323 (para. 185). See also e.g.: Continental Shelf Case (Libyan
Arab Jamahiriya/Malta), ICJ Reports 1985, p. 13, at pp. 44-46 (paras 57-58): “… to use the ratio of coastal
lengths as of itself determinative of the seaward reach and area of continental shelf proper to each Party, is
to go far beyond the use of proportionality as a test of equity, and as a corrective of the unjustifiable

difference of treatment resulting from some method of drawing the boundary line. If such a use of
proportionality were right, it is difficult indeed to see what room would be left for any other consideration”
(p. 45, para. 58). Cf Delimitation of the Continental Shelf (United Kingdom of Great Britain and Northern
Ireland and the French Republic), Decision of 30 June 1977, 54 ILR, p. 6, at p. 68 (para. 101)
(“Proportionality, therefore is to be used as a criterion or factor relevant in evaluating the equities of certain
geographical situations, not as a general principle providing an independent source of rights to areas of
continental shelf.”); Case of the Maritime Boundary Delimitation between Eritrea and Yemen (Eritrea v.

Republic of Yemen), Award of 17 December 1999, 119 ILR, p. 417, at p. 465 (para. 165); Barbados v.
Republic of Trinidad and Tobago Arbitral Award of 11 April 2006, 45 ILM, p. 800, at p. 839 (para. 240)
(“In this context, proportionality becomes the last stage of the test of the equity of a delimitation. It serves
to check the line of delimitation that might have been arrived at in consideration of various other factors, so
as to ensure that the end result is equitable and thus in accordance with the applicable law under
UNCLOS.”).

190on the one hand and on the other, the principles of equidistance and of taking into

consideration the special circumstances of the zone, which are mentioned in sub-
paragraphs (b) and (e), respectively.

6.9. Consequently, in contrast with Ukraine’s catch-all notion of special or relevant

circumstances, it is appropriate to analyze the factors which “might modify the result

produced by an unqualified application of the equidistance principle”, without confusing

them with those which are to be taken into consideration in order to draw the provisional

equidistance line or which might demonstrate that the outcome is markedly

disproportionate, one way or the other.

6.10. In the present case, neither the “geographical factors” mentioned by Ukraine as

“relevant circumstances” – which are relevant in drawing the provisional line – nor the

“State activities in the relevant areas” – which are, at best, relevant as showing the

Parties’ practice in the absence of an agreed delimitation – can be used as “special

circumstances” within the meaning of paragraph 4(e) of the Additional Agreement. Only

two such circumstances are of relevance in the present case: first, the presence of

Serpents’ Island with its already agreed belt of 12 nm territorial sea, and second, the fact
that the Black Sea is a small enclosed sea which has already been partially delimited

between the coastal States.

B. IRRELEVANCE OF “GEOGRAPHICAL FACTORS” AND “STATE

ACTIVITIES” AS SPECIAL CIRCUMSTANCES

(1) Geographical Factors

6.11. In its Counter-Memorial, Ukraine lays great emphasis on the alleged

“geographical predominance of Ukraine in the area” and “the disparity between coastal
486
lengths” in the western basin of the Black Sea, which, in accordance with its approach,

represent relevant circumstances making necessary a dramatic shift of the provisional
equidistance line in favour of Ukraine.

486
UCM, paras. 8.10-8.34.
1916.12. As already shown in Chapter 3, the way in which Ukraine has determined what it

considers to constitute the “relevant coasts” and the “relevant delimitation area” is

erroneous, artificial and contradicts the clearly-established case-law. Nor do the

respective lengths of the relevant coasts of the Parties constitute “relevant
487
circumstances”. As explained above, proportionality can be taken into account if the

line resulting from the application of the equidistance/special circumstances principle

results in a situation “in which the relationship between the length of the relevant coasts

and the maritime areas generated by them by application of the equidistance method, is so

disproportionate that it has been found necessary to take this circumstance into account to

ensure an equitable solution”. 488 It is appropriate to deal with proportionality only after

having identified the line resulting from the application of the equitable principles/special

circumstances approach. 489

(2) State Activities in the Relevant Area

6.13. Under the rubric of “The Relevant Circumstances”, Ukraine deals with “State

Activities in the Relevant Area”, that is, licenses for the exploration and exploitation of
490
oil and gas and fishing practices in the area. But these activities, even if they can be

demonstrated, do not constitute “relevant circumstances” properly so-called. As the

Court explained in its Judgment in Land and Maritime Boundary between Cameroon and

Nigeria:

“Overall, it follows from the jurisprudence that, although the existence of
an express or tacit agreement between the parties on the siting of their
respective oil concessions may indicate a consensus on the maritime areas

to which they are entitled, oil concessions and oil wells are not in

487 See above para 6.7.
488 Maritime Delimitation in the Area between Greenland and Jan Mayen, ICJ Reports 1993, p. 38, at

p. 67 (para. 65). See also: Delimitation of the Maritime Boundary in the Gulf of Maine Area), ICJ Reports
1984, p. 246, at p. 323 (para. 185); Delimitation of the Continental Shelf (United Kingdom of Great Britain
and Northern Ireland and the French Republic), Decision of 30 June 1977, 54 ILR, p. 6, at p. 68 (para.
101) (“it is rather a question of remedying the disproportionality and inequitable effects produced by
particular geographical configurations or features in situations where otherwise the appurtenance of roughly
comparable attributions of continental shelf to each State would be indicated by the geographical facts.”);
Barbados v. The Republic of Trinidad and Tobago, Arbitral Award of 11 April 2006, 45 ILM, p. 800, at p.

839 (para. 240) (“Thus the real role of proportionality is one in which the presence of different lengths of
coastlines needs to be taken into account so as to prevent an end result that might be ‘disproportionate’ and
hence inequitable.”).
489 See below, Chapter 9.
490 See UCM, paras. 8.39-8.65.

192 themselves to be considered as relevant circumstances justifying the
adjustment or shifting of the provisional delimitation line. Only if they are

based on expre491or tacit agreement between the parties may they be taken
into account.”

6.14. Two aspects must thus be kept in mind:

(1) Such State activities have no bearing on the drawing of the line; they can

only intervene ex post, once the line has been drawn in application of the

equidistance/special circumstances principle;

(2) In any case, these practices can only be taken into consideration if they are

based on or necessarily imply an agreement between the States

concerned. 492

6.15. As will be shown in Chapter 7, neither Ukrainian activities concerning oil and gas

concessions nor those concerning fishing activities have been carried out pursuant to any

agreement between the Parties or provide any basis whatever for implying such an

agreement. Indeed most of the activities invoked by Ukraine occurred after the

conclusion of the 1997 Additional Agreement, i.e. after the present dispute had arisen.

C. SERPENTS’ ISLAND AND ITS POTENTIAL EFFECT AS A RELEVANT

CIRCUMSTANCE

6.16. As noted in paragraph 6.10, there are two special circumstances in the present

case. The first is the presence, off the coasts of the Parties, of Serpents’ Island. The

second is represented by the enclosed character of the Black Sea, seen together with the

geographical configuration of its north-western part and the existing delimitation

agreements. These issues will be dealt with in turn.

491 Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial
Guinea intervening), ICJ Reports 2002, p. 303, at pp. 447-448 (para. 304). See also Delimitation of
Maritime Areas between Canada and France, Decision of 10 June 1992, 21 RIAA, p. 267, at p. 296 (paras.
90-1).
492
See below, paras. 7.3 -7.7.
193 (1) Serpents’ Island as a relevant circumstance

6.17. As the Court accepted in the case concerning the Land and Maritime Boundary

between Cameroon and Nigeria, “islands have sometimes been taken into account as a

relevant circumstance in delimitation when such islands lay within the zone to be

delimited and fell under the sovereignty of one of the parties.” 493

6.18. It has been demonstrated that Serpents’ Island is entitled to nothing more than a

belt of 12 nm of maritime area. This is so for two reasons: first, the specific agreements

between the Parties, starting with the 1949 ProcŁs-Verbaux; secondly, their general

agreement, in the light of Article 121(3) of the 1982 UNCLOS and paragraph 4(a) of the

1997 Additional Agreement, that Serpents’ Island is a rock which cannot generate

maritime zones beyond 12 nm. But even if Serpents’ Island could generate such zones –

that it is to say, independently of the existing agreements between the Parties – it would

undoubtedly qualify as a special circumstance and would have to be given nil effect

beyond 12 nm. 494

6.19. Furthermore (even on these assumptions) Serpents’ Island, together with its

already-attributed belt of 12 nm of maritime areas, can play a part as a special

circumstance only in Sector 1 of the delimitation area, where the coasts of the Parties are

adjacent. Due to its remoteness from the Ukrainian coast of Crimea, as well to the

presence of other notable geographical features of this coast (e.g. Capes Tarkhankut and

Kubansky), Serpents’ Island plays no role in the delimitation in Sector 2 of the

delimitation area, characterised by the coastal relation of oppositeness.

493
Land and Maritime Boundary between Cameroon and Nigeria, ICJ Reports 2002, p. 303, at p. 446
(para. 299). See also: Maritime Delimitation and Maritime Questions between Qatar and Bahrain, ICJ
Reports 2001, p. 40, at p. 109 (para. 219); Delimitation of continental shelf (United Kingdom of Great
Britain and Northern Ireland and the French Republic), Decision of 30 June 1977, 54 ILR, p. 6, at p. 83
(para. 145); Government of the State of Eritrea and Government of the Republic of Yemen (Phase Two:
Maritime Delimitation), Award of 17 December 1999, 119 ILR, p. 417, at pp. 457-8 (para. 132).
494
See RM, paras. 8.85-8, 123, 8.126 (d)-(f).
194 (a) Serpents’ Island: overview of the arguments

6.20. As mentioned in Chapter 5 above, there are three premises which must be

simultaneously taken into account when analysing the status of Serpents’ Island as a
special circumstance in the area to be delimited:

(i) the legal entitlements of Serpents’ Island pursuant to the 1949 ProcŁs-

Verbaux;

(ii) Serpents’ Island constitutes a rock unable to sustain human habitation or

an economic life of its own, within the meaning of Article 121(3) of the

1982 UNCLOS;

(iii) the position of this rocky formation, together with its already recognised
surrounding maritime area, vis-à-vis the Parties’ coasts.

These points will be mentioned in turn.

(i) The legal entitlements of Serpents’ Island pursuant to the 1949
ProcŁs-Verbaux

6.21. First, by the 1949 ProcŁs-Verbaux, Serpents’ Island was already given a

considerable effect in terms of maritime areas: the two Parties agreed upon conferring

upon it a 12 nm belt of maritime areas, beyond which were maritime areas appertaining
to Romania. Indeed, as compared to the Romanian coast of the Danube Delta, Serpents’

Island, in spite of its natural characteristics, was given a maritime zone of 12 nm, while

the Romanian coast was given a reduced effect, its maritime zone (i.e., its territorial sea)

being reduced to less than 10 nautical miles. 495 As demonstrated in Chapter 5, those

agreements dispose of the effect of Serpents’ Island beyond the 12 nm limit.

(ii) A rock unable to sustain human habitation or an economic life of
its own

6.22. Secondly, Serpents’ Island, due to its natural characteristics and in particular its

inability to sustain human habitation or economic life of its own, falls under the

provisions of Article 121(3) of the 1982 UNCLOS and thus is not entitled to extended

495
See RM, paras. 5.4, 5.11,12.18.
195maritime areas. As far as human habitation is concerned, the group of persons currently

residing, on a temporary and rotating basis, on this maritime feature cannot be considered

even as a “temporary resident population”. As to the economic life of Serpents’ Island,

this is non-existent, since the natural conditions cannot support the development of
economic activities. 496 To this should be added the dimensions and geological specificity

of this formation, a small (0.17 km ) rocky feature with a harsh environment, lacking in

natural sources of water, unfit for human habitation or for any kind of economic activity.

Consequently, Serpents’ Island is not entitled to its own continental shelf or exclusive

economic zone.

(iii) The position of Serpents’ Island vis-à-vis the Parties’ coasts

6.23. Thirdly, considering the position of Serpents’ Island, off the Romanian and

Ukrainian adjacent coasts, this rocky formation and its belt of 12 nm of territorial sea

would distort in a significant manner the delimitation line in Sector 1 if, contrary to the

position set out above, it was to be held to generate entitlements to anything more than a

12 nm belt of territorial sea and if it was to be given full effect. Even if Serpents’ Island

itself lies only in front of the Ukrainian coast, its belt of territorial sea accorded by the

1949 ProcŁs-Verbaux lies also in front of the Romanian coast.

6.24. To summarise Serpents’ Island, together with its maritime entitlements recognised

by the 1949 ProcŁs-Verbaux, qualifies as a relevant circumstance in Sector 1 of the

delimitation area, which results in the provisional equidistance line being shifted as to

take into consideration the maritime boundary on the 12 nm arc surrounding Serpents’

Island established in 1949, between Point F and Point X. The effect of Serpents’ Island

upon the delimitation line stops at Point X, whence the delimitation line reverts to an

equidistance line. The course of the maritime boundary beyond Point X is re-presented in

Chapter 8.

496
See above, paras. 5.80-5.93
196 (b) Serpents’ Island as a special circumstance in any event

6.25. This section will discuss, for the sake of argument, the position put forward by

Ukraine in its Counter-Memorial, according to which the 1949 ProcŁs-Verbaux did not

establish a maritime boundary on the 12 nm arc surrounding Serpents’ Island up to Point

X. According to Ukraine, the most advanced point established by the 1949 ProcŁs-

Verbaux corresponds to Point F – the final point of the boundary delimiting the territorial

seas of the two Parties. Beyond Point F, according to Ukraine, the Court must rule de

novo on the delimitation of the maritime areas claimed by the Parties.

6.26. In its Memorial, Romania emphasised that, in such a case, due to the

characteristics and distorting influence of Serpents’ Island on the equidistance line, and in

conformity with the well-established body of State practice and judicial and arbitral
497
decisions, this maritime formation should not be given any effect at all in the

delimitation process beyond 12 nm. In such a case, the solution would be that the

maritime boundary would follow the 12 nm arc of circle around Serpents’ Island. 498

499
6.27. As demonstrated in Romania’s Memorial, and as will be demonstrated further
500
below, small islands in locations comparable to Serpents’ Island (straddling the

equidistance or median line and therefore having a strong potential to distort it) have been

discounted, substantially and in some cases entirely, in terms of their entitlements to

continental shelf or exclusive economic zone.

6.28. Ukraine criticises Romania’s reliance on the State practice in relation to whether

islands constitute a special circumstance. But it does not contest Romania’s presentation

of the judicial and arbitral decisions. 501 The eleven decisions identified by Romania as

relevant 502remain unrebutted and Ukraine does not introduce any example to support its

approach. The conclusion to be drawn from those cases is that limited or no effect is

497
498 See RM, paras. 8.86-8.123 and the Addendum to the present Chapter.
499 See RM, paras. 11.46-11.50.
See RM, paras. 8.85-8.123, 8.126 (d)-(f).
500 See the Addendum to the present Chapter.
501 See UCM, Chapter 4, Section 4.
502 See RM, paras. 8.88-8.103.

197given to islands which have the potential to distort the delimitation line and,

consequently, to preclude an equitable overall result. 503

6.29. As far as State practice is concerned, this has to be approached having regard to

the provisions of paragraph 4(e) of the Additional Agreement, which refers to “the

principle of taking into account the special circumstances of the area submitted to

delimitation”, in conjunction with the provisions of paragraph 4(a) of the Additional

Agreement, which refers to the applicability of Article 121 of the 1982 UNCLOS “as

applied in the practice of States”. Romania’s reliance on State practice would be justified

in any event, but the Additional Agreement confers special significance on State practice
504
in this case.

6.30. Without having regard to the provisions of the Additional Agreement, Ukraine

criticises Romania’s reliance on State practice on essentially three grounds.

6.31. First, it argues that each agreed delimitation depends on its particular facts. 505 In

some elemental sense this is true: maritime delimitation ultimately comes down to an

appreciation of the particular geographical situation in light of any relevant agreements of

the parties and any other relevant circumstances. But to say that therefore no guidance is

to be gained from State practice is to deny the stipulation of the Additional Agreement,

the practice of the Court and indeed the very idea that there is a law of maritime

delimitation, as distinct from an ocean of single instances.

6.32. Secondly, Ukraine argues that extrinsic (often unrecorded) considerations may

have affected the weight given to a small island. In other words, it seems to suggest,

whenever islands are ignored or not given full effect in delimitation agreements, this is

the result of political trade-offs, the party that admits nil or reduced effect for islands
506
under its sovereignty being compensated elsewhere. But the fact that the agreements

503 See RM, paras. 8.122-8.123.
504
In this context Sub-section A of Section 4 of Chapter 4 of UCM (paras. 4.53-4.57) is erroneous, as
it disregards the special significance given to State practice by the 1997 Additional Agreement and, at the
same time, tries to equal State practice with custom (by referring, for instance, to the opinio juris – see
UCM para. 4.54)
505 UCM, paras. 4.53, 4.57.
506 See UCM, paras. 4.59-4.60.

198giving reduced effect to islands are the product of interstate negotiations does not mean

that they have no relevance in other cases. In practice, States engaged in maritime
boundary negotiations take careful account of the solutions arrived at in comparable

cases. Any delimitation agreement – for that matter, any agreement concluded between

two States – is to some extent the result of reciprocal concessions, and political

considerations may have a role to play. But the influence of extraneous political

considerations is not to be presumed or merely asserted, as Ukraine tends to do.

Moreover the give-and-take inherent in bilateral negotiations by no means invalidates

Romania’s conclusion. It is a fact that in numerous maritime delimitation agreements

States have given limited or no effect to small islands which due to their relative

insignificance or their geographical position relative to other coasts would otherwise have

a disproportionate effect on the delimitation.

6.33. Thirdly, Ukraine criticises the examples relied upon by Romania as being…

“misleading and, in important respects, incomplete. In particular, none of
these examples cited by Romania involved delimitations where the
mainland coasts of one of the parties bordering the area to be delimited
was significantly longer than those of the other party, as is the case
507
here.”

This confuses two distinct issues: (1) whether an island constitutes a special circumstance

justifying adjustment of the equidistance line and (2) the influence of comparative coastal
lengths, which constitutes, as explained, a means of testing the equity and proportionality

of a delimitation already provisionally determined.

6.34. Though casting doubt on any and all State practice, Ukraine seeks to oppose its

own catalogue of practice to the examples cited by Romania. Many examples included in

Ukraine’s list are of questionable relevance on the basis that the location of the islands in

question was not such as to generate a disproportionate effect. Further, a number of the

islands referred to by Ukraine serve as turning points for systems of straight baselines,
and again are of no relevance for the present case.

507
UCM, para. 4.58.
1996.35. The examples most relevant for this case concern small islands, with no or limited

population, situated geographically in such a way as to produce significant distortions of

the equidistant or median line. From the examples quoted in its Memorial, as well as in
508
Ukraine’s Counter-Memorial, 8 such cases may be identified. In all of these instances,

the islands in question were given limited or nil effect in the delimitation of the exclusive

economic zones and/or continental shelf. Furthermore, in all of these examples the

islands in question were more significant (in terms of size and/or population) than

Serpents’ Island.

6.36. The Addendum to this Chapter analyses in detail each relevant example given in

Ukraine’s Counter-Memorial, and deals with Ukraine’s comments regarding the

examples provided in the Memorial. From this broader examination of approximately 40

cases, the conclusion reached reinforces the fact that, as a general matter, small islands

with little or no population located close to the equidistance or median line are given little

or nil effect in delimitation.

D. THE BLACK SEA AS AN ENCLOSED SEA AND THE EXISTING

DELIMITATION AGREEMENTS IN THE BLACK SEA

6.37. In its Memorial, Romania made a detailed presentation of the other existing
509
maritime delimitation agreements in force between the riparian Black Sea States. It

508 These are: (1) the Agreement between the Government of the Republic of Indonesia and the

Government of Malaysia relating to the delimitation of the continental shelf between the two countries, 27
October 1969, entered into force 7 November 1969, cited in Case Concerning Sovereignty over Pulau
Ligitan and Pulau Sipadan (Indonesia/Malaysia), ICJ Rep 2002 p. 625, at p. 642 (para. 31) (RM, para.
8.107); (2) the Agreement between India and Sri Lanka on the maritime boundary between the two
countries in the Gulf of Mannar and the Bay of Bengal and related matters, 23 March 1976, 1049 UNTS 43
(RM, para. 8.108); (3) the Agreement between Italy and Yugoslavia concerning the delimitation of the

continental shelf between the two countries, 8 January 1968 (RM, para. 8.110); (4) the Agreement
concerning the delimitation of the continental shelf between the two countries (Tunisia and Italy), 20
August 1971, 1129 UNTS 254 (RM, para. 8.111); (5) the Agreement between Qatar and Abu Dhabi on
settlement of maritime boundary lines and sovereign rights over islands, cited in Case Concerning
Kasikili/Sedudu Island (Botswana/Namibia), ICJ Rep 1999 p. 1045, at p. 1192 n 89 (para. 110) (RM, para.
8.114); (6) the Offshore boundary agreement between Iran and Dubai, 31 August 1974 (RM, para. 8.115);
(7) the Agreement between Finland and Sweden on delimitation in the ¯land Sea and Northern Baltic Sea,

2 June 1994, 1887 UNTS 237 (RM, para. 8.116) and (8) the Agreement between the People’s Republic of
China and the Socialist Republic of Viet Nam on the delimitation of the territorial sea, the exclusive
economic zone and continental shelf in Beibu Bay/Gulf of Tonkin of 2000, 25 December 2000, UNTS reg.
no. 41860 (RM, paras. 8.118-8.119).
509 See RM, paras. 6.1-6.20.

200demonstrated that the enclosed character of the Black Sea, taken together with the

maritime boundaries established by the existing Black Sea delimitation agreements,

amount to a relevant circumstance to be taken into account in the present proceedings,

according to which any new Black Sea delimitation should not dramatically depart from

the existing delimitation methods. 510

6.38. In its Counter-Memorial, Ukraine questions the existing Black Sea delimitation

agreements 511and the closed character of the Black Sea as circumstances relevant for the

present proceedings. 512

6.39. Two points call for clarification: first, Ukraine’s analysis of the Black Sea

delimitation agreements is defective in several respects; secondly, while apparently

recognizing that “third State delimitations” may be taken into account, Ukraine questions

whether the enclosed character of the Black Sea constitutes a special circumstance.

(a) The existing delimitation agreements

6.40. Ukraine criticises Romania for referring to the Black Sea delimitation agreements

“in such a way as to imply that they were several”. 513 In fact there are six Black Sea

riparians and there are pairs of delimitation agreements in force between

Bulgaria/Turkey, Ukraine/Turkey, Russia/Turkey and Georgia/Turkey, while

delimitations have yet to be established between Romania/Ukraine, Romania/Bulgaria,

Romania/Turkey, Georgia/Russia and Ukraine/Russia (although in the latter case a

provisional agreement has been concluded). 514 Thus Ukraine’s statement that “[o]ut of 7

possible bilateral delimitation agreements in the Black Sea only 2 have been

concluded” 515is incorrect. Ukraine counts the agreement concluded by the former USSR

as only one agreement notwithstanding the fact that, after the USSR dissolution, it is

510
511 See RM, paras. 6.21-6.34.
512 See UCM, paras. 8.66-8.99.
See UCM, paras. 4.33-4.50.
513 UCM, paras. 8.77-8.79.
514 Contrary to what Ukraine asserts in UCM, para. 8.78, there is no possible delimitation between
Bulgaria and Ukraine, their coasts having no relation (of adjacency or oppositeness) whatsoever.
515 See UCM, para. 8.79.

201applicable to three pairs of States. What matters is the continuation of a maritime

delimitation agreement in force, not who first made it.

6.41. Second, Ukraine seems to question that the USSR/Turkey and the

Bulgaria/Turkey delimitations represent simplified equidistance lines. 516 But they are so

regarded by well-known authorities, as indicated in the Memorial. 517 Moreover, that the

agreements were based on equidistance (despite the disproportion between the lengths of

the Soviet and Turkish and respectively Bulgarian and Turkish coasts) can readily be seen

from maps. 518

(b) The relevance of the enclosed character of the Black Sea in

conjunction with the geographical configuration of the delimitation area and

the existing delimitation agreements

6.42. As to the second issue, while apparently acknowledging that “third State

delimitations in the Black Sea” are a relevant circumstance, 519 Ukraine refuses to accept

the simple argument that the enclosed character of the Black Sea, seen together with the

maritime boundaries established by the existing Black Sea delimitation agreements is –

on the assumption set out in paragraph 6.25 above – a relevant circumstance in the

present case.

6.43. Ukraine refers to the provisions regarding enclosed or semi-enclosed seas of the

1982 UNCLOS, as well as to two cases of delimitation located in the Mediterranean Sea,

Tunisia/Libya and Libya/Malta. 520 It is true that articles 122 and 123 of the 1982

UNCLOS do not specify any special rules of delimitation for enclosed or semi-enclosed

seas. But Romania does not consider these provisions as the most relevant. Romania

relies mainly on Articles 74 and 83 of the UNCLOS referring to the delimitation of the

continental shelf and the exclusive economic zones as well as on the case-law of this

516
See UCM, para. 8.82.
517 See, e.g., RM, footnotes 80, 81, 84.
518 See, e.g., UCM, Figures 8-10, 8-11.
519 UCM, paras. 6.43, 8.67; but cf. para. 8.81.
520 See UCM, paras. 4.33-4.50.

202Court or of international arbitral tribunals, going back to the North Sea Continental Shelf

Cases, which emphasise the need for delimitation to reach an equitable solution.

6.44. This being said, there can be no doubt that “[i]n a semi-enclosed sea like the

Mediterranean”, or an enclosed one like the Black Sea, the “reference to neighbouring
521
States is particularly apposite…” In such a situation, the delimitation agreements

which have been concluded between these neighbouring States deserve particular

attention.

6.45. This conclusion might not be applicable to all delimitations to be effected in

enclosed or semi-enclosed seas. But it is the result of the specific characteristics of the

Black Sea and of the constraining presence of Serpents’ Island immediately offshore,

when combined with the solutions arrived at in the existing Black Sea delimitations. In

this context, the North Sea Continental Shelf and Guinea/Guinea-Bissau cases are

relevant.

6.46. In the North Sea Continental Shelf cases, the semi-enclosed character of the North

Sea, together with the network of delimitation agreements already concluded, were
considered by the International Court of Justice. Indeed they helped to form the basis of

the Court’s decision. The issues of delimitation as between the Netherlands and the

Federal Republic of Germany (FRG), on the one hand, and the FRG and Denmark, on the

other, were considered in conjunction precisely so that the Court could view the

geographical situation as a whole. Thus, the Court analysed the character of the North

Sea, together with the “effects, actual and prospective, of any other continental shelf

delimitation between adjacent States in the same region” 522 in order to arrive at the

conclusion that a particular method of delimitation was not prone to bring about equitable

results.

521 Continental Shelf Case (Libyan Arab Jamahiriya/Malta), ICJ Reports 1985, p. 13, at p. 40 (para.
47).
522 North Sea Continental Shelf Cases (Federal Republic of Germany/Denmark; Federal Republic of

Germany/Netherlands), ICJ Reports 1969, p. 3, at p. 54 (para. 101(D)(3)).
2036.47. A similar approach was followed by the Arbitral Tribunal in the Guinea/Guinea-

Bissau case, where the Tribunal considered not only the delimitation situation between

the two parties, but also its implications for delimitation between the parties and the

neighbouring States (Senegal and Sierra Leone). Taking into account the “amputation”

effect that the use of equidistance might have, due to the general coastal characteristics of

the Gulf of Guinea (and not only to the two parties’ coastal situation), the Tribunal

rejected the use of a certain method and turned to other delimitation methods. 523

6.48. In the Libya/Malta case, the Court first established a provisional median line, then

adjusted that line on the basis of two relevant circumstances, the first of which was the

“general geographical context in which the islands of Malta appear as a relatively small

feature in a semi-enclosed sea”. 524 The character of the relevant sector of the

Mediterranean was thus a factor to be taken into account in order to achieve an equitable

result, influencing the scope of the adjustment of the boundary northwards. 525

6.49. It may be objected that the easterly-facing coastlines of Bulgaria, Romania and

the segment of Ukraine from the land/river boundary north to Odessa are not concave in

the same way as the North Sea coastlines of Denmark, the FRG and the Netherlands.

That is true if (but only if) Serpents’ Island is ignored for the purposes of delimitation

beyond 12 nm. However, if Serpents’ Island is treated as a basepoint for those purposes,

the situation is radically transformed. On that basis (as seen in Figure RR25 on page

206) Ukraine acquires what amounts to a south-westerly facing “coast”, in the form of a

baseline joining Serpents’ Island to the mainland, and this “coast” dominates the

delimitation virtually to the exclusion of all else. In Ukraine’s submission, Serpents’

Island generates a coastal projection without a coast in support, and the consequence is

that the substantial east-facing coastline of Romania is caught in a virtually concave

position, to its great disadvantage. If there was a real coastline between, say, Ukraine’s

523
Delimitation of the Maritime Boundary between Guinea and Guinea-Bissau (Republic of
Guinea/Guinea-Bissau), Arbitral Award of 14 February 1985, 25 ILM, p. 252, at p. 295 (para. 104), pp.
297-298 (paras. 108-111).
524 Continental Shelf Case (Libyan Arab Jamahiriya/Malta), ICJ Reports 1985, p. 13, at p. 52 (para.
73).
525 See also ibid., p. 42 (para. 53).

204mainland and Serpents’ Island, 526the lesson of the North Sea Continental Shelf Cases is

that the principle of equidistance would be modified so as to avoid an inequitable

situation for the coastal frontage caught by the concavity – in this case, Romania’s coastal

frontage. The confined area of the waters of the Black Sea (as with the North Sea) would

call for such a solution. Yet there is no coast, only a rock.

6.50. The position if Serpents’ Island generates no maritime zones beyond 12 nm is

quite different. 527 On that basis the Romanian/Ukrainian maritime boundary may be

described is a mainland equidistance/median line, in principle like all others in force in

the Black Sea. It must be noted that the delimitation agreements between the USSR and

Turkey are based on the assumption that the equidistance line leads to an equitable result,

despite disparities in the coastal lengths of the parties. Equally, in both the

Soviet/Turkish and the Bulgarian/Turkish agreements, the delimitation line was drawn

until a certain point where the interests of third parties (i.e., Romania) were considered to

come into play. Both these final points are equidistant from the relevant mainland coasts.

Under these circumstances, a dramatic shift of the Romanian/Ukrainian boundary from

equidistance is prone to bring about inequitable results.

526 See RM, Map Atlas, Map RM A 24.
527 2
Even on that basis Serpents’ Island attracts substantial maritime areas, some 525 km to the south
of the mainland equidistance line.

205 6
2

s
n
p
e
S
i t
i a
u c
d n
5 ç l
R f i
R t m
r a e
g c t
F al i
t w
h d
o l
y I
s
e
i
r
UE. CONCLUSION

528
6.51. For these reasons, in addition to those explained in the Memorial, the only

relevant circumstances in the delimitation of the maritime areas between Romania and

Ukraine are the presence of Serpents’ Island with its maritime belt of 12 nm already

established by the 1949 Romanian-Soviet agreements, and the enclosed nature of the

Black Sea seen in conjunction with its existing delimitation agreements. These factors

would combine – even in the absence of agreements to the contrary between Romania

and Ukraine – to give nil effect to Serpents’ Island beyond 12 nm.

6.52. On the other hand, “State activities” in the delimitation area, and the alleged

disparity between the Romanian and Ukrainian coastal lengths do not constitute relevant

or special circumstances as that concept is normally understood in the international law of

the sea. These elements will be analysed instead in subsequent chapters of this Reply. 529

528
529 See RM, paras. 6.30-6.34.
See below, Chapter 7 “The Oil and Fisheries Activities”; Chapter 9 “The Equitable Character of
the Delimitation”.

207 Addendum to Chapter 6

State practice cited in Romania’s Memorial and criticised

in Ukraine’s Counter-Memorial

6.53. In its Memorial, Romania made a presentation of maritime delimitation practice

concerning small islands and rocks. 530 This was the subject of a lengthy critique in the

Counter-Memorial. 531 The various individual items of State practice mentioned in this

exchange will be dealt with seriatim.

A. State practice relied upon in the Memorial

(1) Iran-Qatar, Continental Shelf Agreement, 20 September 1969

6.54. Ukraine observes that the agreement was “largely motivated by economic

considerations”, and that both Iranian and Qatari islands were ignored “as part of a trade

off: the boundary was thus delimited so as to be equidistant from the nearest points on the

coasts of the parties’ mainland territories”, citing Charney & Alexander. 532

6.55. In Limits in the Seas #25, the Office of the Geographer of the United States of

America Department of State simply states that the delimitation agreement

“is based on the equidistance principle with the exception that the presence
of all islands in the Persian Gulf was disregarded. The turning points on

the CSB are all equidistant from the mainland of the two countries.”

6.56. Similarly, Charney & Alexander state:

“Islands, rocks, reefs and low-tide elevations were ignored in the
delimitation of the equidistant boundary. The boundary was delimited so

as to be equidistant from the nearest points on the coasts of the parties’
mainland territories”

and

530
531 RM, paras. 8.104-8.121.
UCM, paras. 4.51-4.63.
532 See UCM, para. 4.60(1) quoting Charney & Alexander, vol. II, p. 1513.
208 “The boundary was delimited using the equidistance method, disregarding
all islands.”33

Although it may be the case that the fact that all islands were disregarded “as part of a

trade off”, it nevertheless remains the fact that the islands on both sides were attributed

no effect on the equidistance line drawn from the mainland coasts. This appears to have

been particularly the case for the relatively large Iranian island of Lavan Island/Sheykh

Sho’eyb, which is situated close to the Iranian coast. There are also various small insular

features situated towards the centre of the Gulf on the Iranian side.

6.57. On the Qatari side, the principal island disregarded was the relatively small island

of Jazirat Halul which is located outside Qatar’s territorial sea, some distance off Qatar

towards the middle of the Gulf. Jazirat Halul, is situated roughly two to three time

further from Qatar than Lavan Island is from Iran; it lies roughly due east of the Qatar

peninsula, while Lavan Island lies in front of the Iranian mainland. Given the relative

positions of the two major islands, it is clear that giving Jazirat Halul full or even partial

effect would have resulted in a disproportionate result.

(2) Indonesia-Malaysia, Agreement on Delimitation of the Continental Shelf, 27
October 1969

6.58. In relation to the portion of the delimitation between Borneo and Sarawak,

Ukraine, although conceding that the Indonesian islands were not given full effect,

observes (relying on the report in Charney and Alexander) that this “is believed” to have

been part of a trade-off by Indonesia in return for Malaysian acceptance of Indonesian

claims to archipelagic status, while observing that Romania omits to mention that islands
534
of both parties were given full effect in other parts of the boundary.

6.59. Charney & Alexander state that “the Indonesian islands situated northwest of

Tanjong Datu of Borneo were not given full effect”, and note that “[t]he departure from

equidistance in the present case was [a] compromise in the interest of agreement”. 535 The

comments in relation to a trade-off are essentially supposition as to the motivations

533
534 Charney & Alexander, vol. II, p. 1513.
See UCM, para. 4.60(2).
535 Charney & Alexander, vol. I, p. 1021.
209behind the undisputed fact that the Indonesian islands of Pulau Natunabesar and Pulau

Subisketjil were given much less than full effect. On the other hand, the relevant
Indonesian islands lie to the north-west of the coast of Indonesia territory on the island of

Borneo, which is essentially west-facing and they accordingly do not project onto the

area in which the median line is drawn, given the projection of Malaysian territory at the

terminus of the land boundary between the two States (Tanjong Datu). Further, the

islands effectively lie between the Malaysian territory of Sarawak on Borneo and

peninsular Malaysia. If they had been given full effect the result would arguably have

been wholly disproportionate.

(3) India-Sri Lanka, Agreement on Boundary in the Historic Waters, 26-28 June

1974

6.60. Ukraine observes that the boundary separates the historic waters between the two

countries, and although conceding that the island of Kachchativu is not used as a
basepoint, explains this on the basis that there had been a dispute as to sovereignty over

the island. It goes on to note that the small islands in the Adam’s Bridge chain appear to

have been accorded full effect.536

6.61. Despite the fact that the boundary is one drawn between the historic waters of the

two countries and that there was a dispute as to sovereignty over Kachchativu Island, the

fact remains that Kachchativu Island is accorded less than full effect; in particular,

although the island was attributed to Sri Lanka, the independent analysis contained in
Limits in the Seas #66 reveals that turning point 4 (the point closest to the island) is only

1.187 nm from the island, while it is 10.87 nm from the nearest Indian territory; on the

other hand point 4 is 12.279 nm distant from the nearest point on the Sri Lankan Delft

Island. In these circumstances, it appears that Kachchativu Island was given a

substantially reduced effect due to the disproportionate effect it would otherwise have

had.

6.62. In relation to the claim that the islands in Adam’s Bridge appear to have been
accorded full effect, Limits in the Seas #66 notes that point 6 of the boundary is

536 See UCM, para. 4.60(3).
210equidistant from Pamban Island (India) and Mannar Island (Sri Lanka). Point 6

apparently falls on a low-tide elevation between two islands in the Adam’s Bridge chain,

and appears to be approximately equidistant from those two islands. But there would be

little difference between the median line between Pamban and Mannar Islands and the

median line between the Indian and Sri Lankan islands at the centre of Adam’s Bridge.

The coastal geography in this sector is balanced – unlike in the sector in which

Kachchativu is located.

6.63. In discussing the use of reefs and rocks as basepoints, Charney & Alexander state

that, apart from the point located on the low-tide elevation on Adam’s Bridge “[t]he scale

of available charts prevents any certain statement about the use of these features along

other sections of the boundary, except to confirm that Kachchativu appears to have been
537
ignored in fixing this boundary.”

6.64. However, the subsequent 1976 Agreement between India and Sri Lanka on the

Maritime Boundary between the two countries in the Gulf of Mannar and the Bay of

Bengal and related matters (New Delhi, 23 March 1976), 538which extended the maritime

boundary out from the final point of the 1974 Agreement located on Adam’s Bridge into

the Gulf of Mannar, apparently did not use the two islands in the Adam’s Bridge chain as

basepoints. The analysis contained in Limits in the Sea #77 in relation to that Agreement

notes that point 1M, which corresponds to the southernmost point under the 1974

Agreement (Point 6), is very close to equidistant from points on Pamban Island and

Mannar Island. It further observes:

“Apparently the two countries decided to disregard as basepoints the small

Adams Bridge islands. The basepoints equidistant to point 2M are Pamban
Island (India) and an unnamed island immediately to the west of Manaar
Island (Sri Lanka).”

537
538 Charney & Alexander, vol. II, p. 1412.
1049 UNTS 43; this is the agreement which is referred to by mistake in the relevant footnote in the
Memorial, in place of the correct reference to the 1974 Agreement.
211(4) Greece-Italy, Agreement of 24 May 1977.

6.65. Ukraine criticises Romania’s reliance on the Greek-Italian agreement on the basis
that the delimitation line “in most parts does not mirror an equidistance line”. While

observing that the Greek islands of Corfu, Kefallinia and Zakynthos were given full

effect, it accepts that the Otranto Channel Islands (Fanos/Othonoi Island) were given 75%

effect, while Stamphani in the Strofades group was given half-effect, but argues that this

was “compensated for elsewhere”. A footnote observes, referring to the Limits in the

Seas study that “most points are approximately equidistant from the small Greek islands
539
and the Italian mainland”.

6.66. The Ukrainian criticism is unconvincing. This is a straightforward case of small,

remote islands being given reduced effect in order to avoid a disproportionate result. The

islands of Corfu (Kerkyra), Kefallinia and Zakynthos are all large, while Othonoi, which

lies northwest of Corfu in front of the coast of Albania towards the middle of the channel,

is much smaller. Similarly, the Strofades, a group of small islands, lie a considerable

distance west off the coast of the Peloponnesian Peninsula, and would have deflected the

line westwards if given full effect as against points on the coasts of Calabria and Sicily as

the line progressed southwards.

6.67. The report in Charney & Alexander recounts the negotiating history as follows:

“Italy was ready to give full effect to Corfu (641 square kilometres (sq.

km.)), Kefallinia (935 sq km) and Zakynthos (406 sq. km.), which are of
considerable size and in the vicinity of the coast. It proposed that no effect
should be given to the islands of the Channel of Otranto (Fanos and
Samothrake), which are of small size, and to Stamphani, at the south of the

area to be delimited, which is small and quite far from the Greek
mainland. Greece was, in principle, in favor of a full effect for every
island.

The parties finally agreed to give a three-quarters effect to the islands of
the Channel of Otranto, attributing to Greece 75 percent of the disputed
seabed. The concession of about 75 sq. km. thus made to Italy relative to
the equidistant line was, however, compensated by an equal concession to

Greece in the central part of the area of delimitation. As regards the

539 UCM, para. 4.60(4).
212 Strofades, they were given half-effect, by giving each party 50 percent of
the disputed area.” 540

It thus appears that no compensation was given for the half effect given to the Strofades.

(5) Italy-Yugoslavia, Agreement Concerning the Delimitation of the Continental
Shelf, 8 January 1968

6.68. Ukraine accepts that reduced effect was given to certain islands in the

delimitation.541 However, it attempts to distinguish the example on the basis that, first,

the example concerns opposite coastlines, and that unlike Serpents’ Island, “these islands

were located in the middle of the maritime area to be delimited between the two States’

opposing coasts, and the Yugoslav islands were relatively distant from the Yugoslav

mainland”.

6.69. This is a weak attempt to distinguish what appears a clearly relevant example of

State practice. Although the two islands of Pelagruz and Kajola/Caiola are located near

the centre of the Adriatic Sea, the small island of Jabuka to the north is less central, and
its relative effect would in any case have been diminished by the presence of the larger

island of Andrija to the southeast .

6.70. The solution adopted by the Parties was in two parts; Jabuka appears to have been

attributed reduced effect, with some smoothing as the line shifts to be governed by points

on Andrija, where the line is based on equidistance. Pelagruz and Caiola are limited to a

12 nm band facing the Italian mainland and it appears that they were given a “tapering”

effect as the governing point shifts back to the larger Yugoslav islands lying closer to the

Yugoslav mainland. On the other hand, the small Italian island of Pianosa, which lies to

the north of the Gargano peninsula between the Italian mainland and the island of

Pelagruz was given markedly less than full effect as against the Yugoslav mainland and

apparently a tapering effect as the governing point shifts back to the mainland. Another

way of looking at this is to treat the delimitation line having been based on equidistance

ignoring these small islands, with a 12 nm band around the Yugoslav island, and then

540 Charney & Alexander, vol. II, p. 1594.
541 UCM, para. 4.60(5).
213some “smoothing” of the boundary on either side of the 12 nm zone where the line

reverts to equidistance from the Italian mainland and basepoints on the larger Yugoslav
islands close to the mainland. On any analysis, the agreed delimitation line clearly

involved giving reduced effect to the various islands due to the disproportionate effect

they would otherwise have had on the delimitation.

(6) Italy-Tunisia, Agreement relating to Delimitation of the Continental Shelf, 20
August 1971

6.71. As with the Italian-Yugoslav Agreement, Ukraine attempts to undermine

Romania’s reliance on the Italian-Tunisian Agreement on the basis that it concerns

opposite coastlines, where the islands were located in the middle of the maritime area,
542
“and, for the most part, were closer to Tunisia than to Italy”. Ukraine also cites the

commentary in Charney & Alexander which reports that concessions were made in return

for a “wider package deal”, including a fishing agreement. Finally, Ukraine cites Article
1 of the Agreement which states that delimitation was accomplished on the basis of “the

median line… taking into account islands, islets, and low-tide elevations”, and that

various other small islands and low-tide elevations on both sides were taken into account

for these purposes.

6.72. As an initial point, it may be noted that Ukraine incompletely cites Article 1 of the

Agreement, which reads in full as follows:

“The boundary of the continental shelf between the two countries shall be
the median line, every point of which is equidistant from the nearest points
of the baselines from which the breadths of the Italian and Tunisian
territorial seas are measured, taking into account islands, islets and low-

tide elevations with exception of Lampione, Lampedusa, Linosa and
Pantelleria.”

Given the full text of the provision, and in particular the stipulation that, apart from the

four named Italian islands, the median line was otherwise to be used, it is clear that there

was recognition between the Parties that if a median line were based on all islands, the

four Italian islands, if given full effect, would produce a disproportionate result due to

542 UCM, para. 4.60(6).
214 543
their position. The other islands and low-tide elevations referred to by Ukraine are

situated very differently from the islands of Lampione, Lampedusa, Linosa and

Pantelleria, being much closer to the respective coasts of the Parties. In any case, the fact

that full effect was given to certain islands for the purposes of calculating turning points

is irrelevant given that the Parties clearly recognized the disproportionate effect of the

four islands located in the centre of the zone.

6.73. Again, this is a clear example of recognition by the two States that islands

constitute a special circumstance due to their position. Limits in the Seas #89 observes:

“Italy and Tunisia have agreed that the Italian islands of Pantelleria,

Linosa, Lampedusa, and Lampione, all situated near the Tunisian
mainland, constitute a special circumstance. This causes a true equidistant
line, developed from all Tunisian and Italian basepoints, to follow a course

much to Tunisia’s disadvantage. An equidistant line developed between
Tunisia and Pantelleria, for example, would deflect the boundary to within
20 miles of the Tunisian mainland. Therefore, in the equidistant line
calculation the two states agreed to disregard these four Italian islands. In

the area of the islands the boundary deviates from equidistance, as
developed from basepoints on Sicily and the Tunisian mainland, and
follows the outer limits of maritime zones drawn around each of the four

islands.”

(7) Sweden-USSR, Agreement on Delimitation of the Continental Shelf, 18 April
1988

6.74. Ukraine observes that the fact that less than full effect (roughly 75%) was given to

the islands of Götland and Gotska Sandön “cannot be separated from extraneous political
544
issues”, and cites Charney and Alexander who refer to a “political compromise”.

Götland and Gotska Sandön lie off the west coast of mainland Sweden, roughly a third of

the distance between the coasts of Sweden and the former USSR.

6.75. Ukraine does not specify the “political factors” which might have motivated a

settlement giving less than full effect to the islands, and the citation from Charney and
Alexander does not support the contention that the agreement was based on “political”

543
See also the discussion of the history of negotiations in Charney & Alexander, vol. II, pp. 1612-
1614.
544 UCM, para. 4.60(7).
215factors rather than the fact that these islands constitute a “special circumstance”. Read in

context, following a discussion as to the respective positions of the parties (which reports

that Sweden took the position that a strict equidistance line from the coast of the islands

should be used, while the USSR took the position that they should be disregarded entirely

as a special circumstance and the equidistance line should be drawn from the mainland

coast), it seems relatively clear that the “political compromise” was precisely to give the

islands some, although reduced, effect: in other words, they were considered as

constituting a special circumstance. 545

6.76. Other accounts confirm that the solution was a negotiated one in relation to the

opposing positions of the parties in relation to whether the islands constituted a special
546
circumstance, with Sweden receiving 75% of the disputed area.

6.77. In the absence of clear evidence as to other “political” reasons for the final

delimitation line, it appears that the parties reached agreement that the islands should be

regarded as a special circumstance, and therefore accorded less than full effect.

(8) Saudi Arabia-Iran, Agreement concerning Sovereignty over the Islands of

Al-’Arabiyah and Farsi, 24 October 1968

6.78. Ukraine observes that the agreement settled a dispute over sovereignty over the

two islands of Al-’Arabiyah and Farsi and that “equal effect” was given to them.

Although accepting that Kharg Island (Iran) was reportedly given half-effect, Ukraine

observes that this has to be viewed in light of the potential oil deposits in the area and the

fact that both sides had awarded concessions in that area. Ukraine further observes that

the delimitation line was equidistant between the respective claim lines, and was later
547
modified to apportion the oil fields equitably”.

6.79. Again, the Ukrainian criticism is weak and resorts to speculation. Further, it is

not accurate to state that Al-’Arabiyah and Farsi were given full effect. Rather, the two

545
546 Charney & Alexander, vol. II, pp. 2061-2.
See e.g. H Tiberg “New Sea Boundaries in a Swedish Perspective”, (1989) 10 Michigan Journal
of International Law, p. 686, at pp. 692-5.
547 UCM, para. 4.60(8).
216islands, which lie in the middle of the maritime area between the two States were each

given a 12 nm territorial sea and an equidistance line was drawn between them where

they are opposite each other. For the rest of the delimitation line, the two islands were
548
given no effect. Limits in the Seas #24 in this context observes:

“The CSB [Continental Shelf Boundary] around the islands of Arabi and
Farsi was delimited by giving effect to the 12-mile territorial seas of the
respective islands, this is in conjunction with a “local” median line

between the two islands.”

Ukraine further omits to refer to the report in Charney & Alexander, which states as

follows:

“In the vicinity of Al-’Arabiyah and Farsi, the continental shelf boundary
is defined by the 12-mile territorial sea of the two islands; and, where
these territorial seas overlap, an equidistant line is drawn between the

islands. In the northern sector, the boundary is a modification of an
equidistant line that gives half-effect to the island of Kharg. Small islets
situated close to the mainland were disregarded in the delimitation.”

Limits in the Seas #24 observes that Kharg Island,

“a rather large island lying 17 n.m. off the coast, was given half-effect in

delimiting portions of the CSB. The half-effect principle is a trade off
between considering the island as part of the mainland (full-effect) or
completely ignoring the island (no-effect).”

6.80. Further, Ukraine’s observation in relation to the adjustments to the line in relation

to petroleum structures is inaccurate; the equidistance line on the basis of giving half-

effect to Kharg Island was originally used in a 1965 Agreement which was never ratified

due to the discovery of a petroleum structure by the Saudi concessionaire. Accordingly

adjustments were made to the line in that area in order to proportion the structure, and it

is this modified version of the equidistance line, giving half effect to Kharg Island which

is embodied in the 1968 Agreement. 549 This change is explained in the following manner

in Charney & Alexander:

548
In addition, several other Saudi islands located a significant distance offshore also appear to have
been given little or no weight, e.g. Al Qiran, al Qurayyin, Janah and al Jurayd.
549 Charney & Alexander, vol. II, p. 1520, 1522.
217 “In the area north of Farsi, the parties had granted overlapping
concessions. The Saudi concession was bounded by an equidistant line

constructed from the mainland coasts of the two states, whereas the
Iranian concession was bounded by an equidistant line constructed from
the mainland coast of Saudi Arabia and Iran’s island of Kharg, some 27

km (about 17 nm) offshore and 5 by 6.5 km (3 by 4 n.m.) in area. The
1965 agreement, which never entered into force, delimited a boundary that
was equidistant between these two lines, i.e. a boundary that gave half-

effect to Kharg. This half-effect line was subsequently modified in the
1968 agreement to equitably apportion a petroleum structure in the
vicinity of the boundary. The result is a series of straight line segments
550
that zigzag back and forth across the half-effect line.”

(9) Qatar-Abu Dhabi (UAE), Agreement Settlement of Maritime Boundary
Lines and Sovereign Rights over Islands, 20 March 1969

6.81. Ukraine criticises the reliance placed on the 1969 Agreement on the basis that it

settled on a global basis territorial disputes between the parties, and provided for equal
551
rights in the Al-Bunduq oil-field. Reliance is placed on the report in Charney and

Alexander, which observes that “[e]conomic considerations motivated the delimitation
552
and to a certain extent influenced the course of the boundary.”

6.82. Again, the criticism is misplaced, given that it is clear that no effect was given to

the island of Daiyina in drawing the equidistance line and the island is only accorded a 3

nm arc. Further, although drawing attention to the fact of the slight deviation of the line

from equidistance in relation to the oil field, and that sovereignty over Daiyina was

attributed to Abu Dhabi/UAE, Ukraine does not make clear that the Agreement also

attributed sovereignty over the island of “Lasahat” (Al Ashat) and “Shura’awa”
(Shara’iwah). The latter, a small island situated northwest of Daiyina, apparently

constitutes the basepoint on the Qatari side for calculation of two of the turning points.

6.83. Further, it is not at all clear that the reference to “economic considerations” refers

specifically to the limited effect given to Daiyina. In the report in Charney & Alexander,

there is little reference to the fact that Daiyina is allocated only a band of territorial sea

and otherwise given no effect, and the discussion does not go into any great detail as to

550
551 Charney & Alexander, vol. II, p. 1522.
UCM, para. 4.60(9).
552 Charney & Alexander, vol. II, p. 1544.
218the method for calculation of the various turning points. The comment rather appears to

be directed at the dispute relating to the Al-Bunduq oil-field, as does the comment

relating to the course of the boundary. In this context, the report comments that “[t]he

delimitation used a selective use of equidistance modified to take account of special

circumstances (specifically, the location of an existing oil well and an island located
553
approximately midway between the parties’ coasts).”

(10) Iran-Dubai (UAE), Offshore Boundary Agreement, 31 August 1974

6.84. Ukraine observes that the agreed boundary was not an equidistance line, and to

the extent that it does approximate to equidistance, ignores islands on both sides, so that

the location of the islands resulted in them being “offset” against each other. 554

6.85. The boundary is admittedly not based on strict equidistance. However, the fact

remains that Sirri Island is only accorded a 12 nm territorial sea.

6.86. The basis on which Ukraine states that the boundary ignored the effects of Iranian

islands (other than the 12 nm territorial sea around Sirri Island) or the UAE’s Sir Abu

Nu’ayr Island appears to be a statement to that effect in the report in Charney &

Alexander. 555 However, the report in Charney & Alexander observes variously that:

“[e]conomic considerations – specifically a desire to facilitate offshore oil
developments – played a decisive part in the decision to delimit the
boundary but did not affect the actual location of the line.” 556

and that

“It is unclear what delimitation method was utilized for the western sector
of the boundary, although the net result was a line which is approximately
equidistant from Iran and the UAE.” 557

553
Charney & Alexander, vol. II, p. 1543.
554 UCM, para. 4.60(10).
555 Charney & Alexander, vol. II, p. 1535: “Islands, rocks, reefs, and low-tide elevations did not affect
the location of the boundary, except that in its eastern sector the boundary follows the 12 nm arc about Sirri
Island.”
556
557 Ibid., p. 1534.
Ibid., p. 1535.
219Sirri Island clearly does not govern points 1 & 2 on the eastern portion of the boundary.

These seem to be roughly equidistant from Sir Abu Nu’ayr island on the UAE side and
Bani Forur island (which is much smaller than Sirri) on the Iranian side, but they are also

approximately equidistant from the nearest points on the two mainlands. Point 3 (which

is situated on the 12 nm arc around Sirri Island) appears to be governed by points on the

UAE and Iranian mainland, ignoring the closer Iranian islands. Indeed the whole

delimitation seems to be based on a mainland coasts equidistance line.

6.87. It also bears noting that Abu Musa, the sovereignty of which was disputed,

appears to have been entirely disregarded in relation to the short stretch of the line west
of the arc around Sirri Island, and that Sirri Island itself was given no effect on this

segment. The base points in that section (points 4 & 5) appear to be based on points on

the UAE and Iranian mainland, ignoring the effect of Bani Forur Island. It accordingly

may be speculated that some islands were taken into account for determination of some

points, and not for others, and that lines joining those points were then drawn.

(11) Finland-Sweden, Delimitation in the ¯land Sea and Northern Baltic Sea, 2
June 1994

6.88. In relation to this Agreement, Ukraine argues that the boundary line is not an

equidistance line but was based on the pre-existing 1972 Agreement relating to the

continental shelf. It further observes that the fact that no effect was attributed to the

Bogskär Islands “is undermined by the fact that these islands were given full effect in the
1980 Agreement between Finland and the USSR”. 558

6.89. Whether the line is based on equidistance or not, it appears that no effect was

given to the Bogskär Islands. Further, it is irrelevant whether the agreement was

substantially based on a prior agreement if that agreement likewise gave no effect to the

islands. On the other hand, it appears that the southernmost point of the boundary under

the 1972 Convention (point 17), which corresponds to point 4 of the 1994 boundary, in

558 UCM, para. 4.60(11).
220fact corresponds to one of the points fixed by the ¯land Islands Convention (see Art
559
5).

6.90. The argument that those islands were given full effect in another agreement with a

third State is irrelevant. Depending on the situation, islands belonging to State X might

constitute a special circumstance as regards State A because they have a disproportionate

effect on the division of the continental shelf, but might not produce a disproportionate

effect and therefore constitute a special circumstance as against State B. This appears to

be the case in relation to the Bogskär Islands; they are situated some distance off the

coast of Finland in the middle of the area between Finland and Sweden, south of the

straight baselines drawn around the ¯land Islands, which protrude substantially further

west, and they would arguably have a disproportionate effect on an equidistance line

drawn between Sweden and Finland. However, they are not appreciably further south as
compared to the baselines drawn around the islands fringing the mainland of Sweden and

therefore arguably had no such disproportionate effect as against the USSR.

6.91. Finally, it should be noted that although it is argued in the Memorial (and

assumed in the Counter-Memorial) that the Bogskär Islands were not used as base points,

the report in Charney & Alexander expressed the view that “the new segment between

Points 4 and 5 must have been influenced by Bogskär. Exactly to what extent it was so

influenced, is not clear since both parties stress different elements in this respect”. 560

Elsewhere in the report, it is suggested that the line was drawn so as to compensate

Finland for Swedish fishery gains elsewhere. 561

(12) United Kingdom (Turks and Caicos Islands)-Dominican Republic,
Agreement Concerning Delimitation of the Maritime Boundary, 2 August 1996

6.92. Ukraine limits itself to noting that the relevance of the Agreement is unclear,

noting the disparities between the lengths of coastlines, and noting that the British

559
560 See the discussion in Charney & Alexander, vol. III, p. 2548.
Ibid., p. 2543.
561 Ibid., p. 2547.
221negotiators had described it as a “pragmatic solution”, 562 while accepting that the

boundary was fixed some distance to the north of the median line. 563

6.93. In fact, it appears that the boundary was not based on any particular method but

was agreed from the starting point of a claim line by the Dominican Republic in relation

to its EEZ (an EEZ had not at time been claimed for the Turks and Caicos), which was
564
then slightly adjusted and agreed as an all-purpose maritime boundary.

(13) People’s Republic of China-Vietnam, Agreement on Delimitation of the
Territorial Sea, EEZ and Continental Shelf in the Gulf of Tonkin, 25 December
2000

6.94. Ukraine observes that Romania “appears to rely particularly” on this Agreement,

especially the 25% effect allegedly attributed to Bach Long Vi. It attempts to distinguish

the example on the basis that Bach Long Vi is a “mid-ocean island”, almost equidistant
565
from the parties’ coasts. In relation to the island of Can Co, although accepting that

the island was given 50% effect, Ukraine pleads that the reduced effect of the island has
566
been noted to have been “obviously a negotiated compromise”.

6.95. Zou Keyuan, in a journal article not relied upon by Ukraine, observes:

“One important issue which is not mentioned in the Sino-Vietnamese

Boundary Agreement but critical to the delimitation, is the effect of
islands on the delimitation line. The two parties finally agreed to give the
Bach Long Vi Island twenty-five percent effect, thus this mid-ocean island

enjoys 12 n.m. of territorial sea and 3 n.m. of the E.E.Z. and continental
shelf. The other small island, Con Co Island, about 13 n.m. off the coast of

Vietnam has been given fifty percent effect in the delimitation of the
E.E.Z. and continental shelves of the two countries at the closing line at
the mouth of the Gulf. Since these two islands belong to Vietnam, China

would have wished to minimise their effect on the delimitation. The final
arrangement is obviously a result of compromise.” 567

562
563 The phrase is taken from the report in Charney & Alexander, vol. III, p. 2239.
UCM, para. 4.60(12).
564 Charney & Alexander, vol. IV, p. 2239.
565 Bach Long Vi is not much more than 50 nm from the nearest point on the Vietnamese mainland,
which hardly justifies calling it “mid-ocean”.
566 UCM, para. 4.60(13).
567
Zou Keyuan, “Implementing the United Nations Convention on the Law of the Sea in East Asia:
Issues and Trends” (2005) 9 SYBIL, p. 37, 49.
2226.96. Again, the Ukrainian criticism is a weak one, and the comment that Romania

appears to place particular reliance on the example is perhaps a recognition that this is
one of the strongest examples. First, the fact that reduced effect is given to an island

through negotiation does not discount its relevance as practice. The very fact that

agreement has been reached to give an island reduced effect is evidence that the island

constituted a special circumstance.

6.97. Further, the attempt to distinguish the geographic situation is not particularly

convincing. Although Bach Long Vi is located towards the centre of the northern-most

part of the Gulf of Tonkin, it nevertheless would have had a very substantial effect as
against the northern coast of China.

(14) Australia/Indonesia, Treaty Establishing an EEZ Boundary and Certain
Seabed Boundaries, 14 March 1997

6.98. Similarly the attempt to distinguish the Agreement between Australia and

Indonesia is weak. Ukraine observes that the Ashmore Islands “are located mid-ocean,

significantly closer to western Timor than the Australian mainland”, and disputes any

analogy with Serpents’ Island “which lies close and opposite the Ukrainian coast and in a

geographic situation where the mainland Ukrainian coast is at least three times as long as
568
the corresponding Romanian coast”.

6.99. Again, Ukraine attempts to confuse the issue of the length of respective coastlines

with the separate question of whether an island would have an inequitable effect on a

provisional equidistance line.

6.100. The report in Charney & Alexander observes that there was no argument from

Australia that using small islands, rocks and low-tide elevations within the Indonesian

system of archipelagic baselines was inappropriate. The report further notes that if
Ashmore and Cartier, and Scott Reefs had been used as basepoints with full effects, “that

line would have been forced northwards to within 35 nm of Rote, Indonesia’s

southernmost island.” After observing that the 1981 fisheries line had given Indonesia

568 UCM, para. 4.60(14).
223roughly 70% of the disputed area, the report continues “[i]t is therefore unsurprising that

the EEZ boundary… lies consistently south of the line of equidistance.” 569

6.101. In relation to the boundary between Christmas Island and Java, the report notes:

“The marked difference in size between Australia’s Christmas Island and
Indonesia’s Java Island and the fact that Christmas Islands lies about

755nm from Northwest Cape, the nearest part of the Australian mainland,
underlay the agreement to delimit a single boundary separating the EEZs
and seabeds of both countries in a manner that delivered to Indonesia
570
about two-thirds of the disputed area.”

Elsewhere it is noted that the reduction of the claim based on Christmas Island was

influenced by the decision in Jan Mayen, and that it was agreed that the difference in

coast length justified discounting the claim, although the actual ratio did not determine
571
the proportions in which the disputed area was allocated.

6.102. The Agreement, and in particular the enclaving of the Ashmore Islands, represent

another strong example of islands being agreed to constitute a special circumstance

because of the disproportionate result they would produce if given full effect. The

geography of the situation is admittedly different, but this does not affect the validity of

the basic proposition that small islands may be given reduced effect if they produce

disproportionate results; rather, at most it makes the example unhelpful for the purpose of

any attempt to place reliance on the Agreement as an analogy to provide a concrete

solution in relation to Serpents’ Island.

(15) Saudi Arabia-Bahrain, Agreement dividing the continental shelf in the

Persian Gulf, 22 February 1958

6.103. Ukraine observes that the effects of the two islands of Lubainah Al-Saghirah and

Lubainah Al-Kabirah have to be regarded in the light of the dispute over sovereignty

between the parties. It further observes that in any case the islands were not discounted,

569
570 Charney & Alexander, vol. IV, pp. 2705-2706.
Ibid., p. 2706.
571 Ibid., p. 2703.
224but would appear to have been given equal importance and used as boundary turning

points.572

6.104. The observation that the dispute relating to sovereignty over the islands was a

major factor in the Agreement is justified. The report in Charney and Alexander notes:

“Certain small islands and drying reefs belonging to both parties were also
given full effect. Other small islands and drying reefs belonging to both

parties were ignored. The allocation of Lubainah Al-Saghirah to Bahrain
and Lubainah Al-Kabirah to Saudi Arabia determined the location of part
of the boundary, since turning points are located on both of these
573
islands.”

6.105. However, as to the effect given to the islands, Ukraine’s representation of the

situation is misleading. Study of the Agreement reveals that the parties adopted a

somewhat peculiar method of delimitation, obviously dictated in part by the sovereignty

dispute, which involved placing two of the boundary turning points on a point located on

the eastern extremity of Lubainah Al-Kabirah (which was attributed to Saudi Arabia) and

respectively on another point on the western extremity of Lubainah Al-Saghirah (which

was attributed to Bahrain). The two islands were otherwise not given any effect; rather

the method for delimitation of the other points consisted of a particular method of

equidistance, based on lines joining the midpoints between specified points located on the

respective territories of the parties.

(16) Denmark-Canada, Agreement relating to delimitation of the Continental
Shelf, 17 December 1973

6.106. Ukraine criticises Romania’s reliance on the example in relation to Hans Island as

sovereignty over the island was disputed and as a result “the boundary in the area of the

island was not delimited”. 574 In relation to the treatment of Franklin Island, which

Romania observed in the Memorial that it was apparently given less than full effect,

Ukraine notes that “this must be considered in the context of the fact that its location was

572
573 UCM, para. 4.60(15).
Charney & Alexander, vol. II, p. 1491.
574 UCM, para. 4.60(16).
225 575
relatively remote from Canada’s coast”. As noted below, Ukraine appears to be

referring to a different island.

6.107. In relation to Hans Island, Limits in the Seas #72 observes:

“No boundary exists between Point Nos. 122 and 123. These two points
are situated on the north and south coasts, respectively, of Hans Island.
This island lies in the mid-channel, about 8 nautical miles north-northeast
of Franklin. Problems of sovereignty and effect of the island on a maritime

boundary led to its being discounted in the delimitation.”

The report in Charney & Alexander (which appears to have been selectively cited by

Ukraine) similarly observes:

“There is no boundary line between Points 122 and 123 in Nares Strait
because of the existence of Hans Island. The island lies in mid-channel
and measures less than a mile north-south. Sovereignty over the island is

in dispute, with neither side willing to press strongly on the issue.
Because of its status, and the distorting effect it might have on a generally
equidistant boundary, the boundary in the area of the island was not
576
delimited.”

6.108. Ukraine’s comments are misleading. However the fact that the island is mid-

channel and that turning points appear to have been placed just off its coast means that,

irrespective of under whose sovereignty the island falls, it has no effect on the course of

the equidistant line

6.109. As regards Ukraine’s comments as to [Lady] Franklin Island, there appears to be

some misunderstanding as to which island is being referred to. Franklin Island is a small

island in the region south-southwest of Hans Island in the Nares Strait, lying closer to the

coast of Greenland, which appears to have been disregarded in drawing the equidistance

line in that area. Lady Franklin Island is a Canadian island considerably to the south, off

the Hall Peninsula towards the south of the Davis Strait. Ukraine accepts that the latter

island appears to have been given reduced effect. The suggestion that Lady Franklin

Island is “relatively remote” from Canada’s coast in the overall geography of the coast is

true but irrelevant since the real issue concerned Franklin Island.

575 Ibid.
576 Charney & Alexander, vol. I, p. 372.
2266.110. Ukraine makes no reference to Crozier Island, 577which as with Franklin Island, is

located to the south west of Hans Island in the Nares Strait.

6.111. The report in Charney & Alexander further notes that “Point 114, west of Thule,

completely ignores Greenland’s Carey Islands, which are located much closer to the

boundary than is the Canadian coast.” 578

(17) Australia-Papua New Guinea, Agreement Concerning Sovereignty and
Maritime Boundaries, 1978

6.112. Ukraine comments that reliance on this example is “quite misplaced” since the

islands are located very close to the coast of Papua New Guinea. It further comments that,

apart from the small islands which were enclaved, other small islands were given full

579
effect as basepoints.

6.113. The Ukrainian criticism misses the point. The example is relied upon for the very

fact that the small Australian islands, due to their position very close to the mainland,

would have exercised a wholly disproportionate effect, effectively cutting off that portion

of the coast of Papua New Guinea, and this is quite clearly the reason why they were

enclaved with only 3 nm territorial seas, and were given no effect in relation to the

continental shelf boundary to the south. Whether small islands on the Papua New Guinea

side were used as basepoints or given full effect is not relevant, as they quite clearly did

not have a disproportionate effect, given their location; given the closeness of those

islands to the mainland, it is not surprising that they were treated as basepoints.

577
578 UCM, para. 4.60(16).
579 Charney & Alexander, vol. I, p. 376.
UCM, para. 4.60(17).
227 B. State practice relied upon in the Counter-Memorial

I. Examples in relation to adjacent coasts

(18) USA-Mexico, Exchange of Notes, 24 November 1976; Treaty on Maritime
Boundaries, 4 May 1978

6.114. Ukraine observes that according to the report in Charney & Alexander all islands

were given full effect, including Arrecife Alacran (approximately 75 nm north off the

Yucatan peninsula) in the Gulf of Mexico, as well as, in the Pacific, San Clemente and

San Nicolas, located approximately 60 nm off the US coast and Guadalupe, located

approximately 145nm off the Mexican coast. 580 Two small islands off the Mexican coast

apparently also had a minor impact on the delimitation line.

6.115. The report in Charney & Alexander observes:

“During the negotiations both sides addressed the question of what weight
certain islands should receive in calculating the equidistant lines. It would

appear that both sides might have made arguments of maximum advantage
to the effect that one island or the other ought to be discounted in the
construction of the equidistant line. The negotiators understood that if
such positions were taken it would lead to disputed boundaries on both

coasts. Accordingly, if a dispute was to be avoided, the choice fell
between using all islands in the construction of the equidistant line versus
disregarding them all. As a matter of general policy on maritime boundary

delimitations, the United States had decided that in those maritime
boundary situations where an equidistance line leads to an equitable result,
all territory on both sides, including islands, should be used in the

equidistance calculation. Consequently581 this delimitation, the
negotiators chose to use all such features.”

6.116. Two preliminary points may be made. First, it seems that the line in the Pacific

does not precisely follow the equidistance line but has been adjusted and simplified so as

to reduce the number of turning points from 8 to 3 with no consequences on the areas

received by each side. Secondly, as is well known, the position of the United States is to

take account of all islands more or less independently of considerations of equitableness.

This may be a legitimate negotiating position, and no doubt reflects an overall calculation

580 UCM, para. 4.62(i)(1).
581 Charney & Alexander, vol. I, pp. 431-432.
228of United States interests, but it departs from general State practice and the position of the

Court.

6.117. Of the various islands, only the small Arrecife Alacran could potentially be

argued to constitute a special circumstance, being isolated due north of the Yucatan

Peninsula and therefore having a relatively substantial effect on the delimitation line as

against the Louisiana coast. Despite their small size, it appears that the islands are

inhabited and support fishing and tourist activities.

6.118. On the Pacific boundary, the US island of San Clemente is relatively large and

deflects the line starting at point OP2 as against the mainland coast. San Nicolas, which

is also large and located to the northwest exerts an effect only as the line passes 120’ W.
Guadalupe, a large island, is situated substantially further south from the inner portion of

the line compared to San Clemente and appears only to exert an influence on point OP3.

From point OP3 westwards, the boundary is essentially a median line between similarly-

sized islands and is uncontroversial. It is true that San Clemente was given full effect

between points OP2 and OP3, but it is a much larger island than Serpents’ Island.

(19) Argentina/Uruguay, Delimitation of the Río de la Plata and the Maritime
Boundary, 19 November 1973

6.119. Ukraine relies on the effect given to the Isla de Lobos as a basepoint, noting that it
582
is analogous in size to Serpents’ Island.

6.120. Isla de Lobos is a small island lying relatively close in front of the coast of

Uruguay. I is substantially closer to the mainland of Uruguay (at about 4.3 nm from the
coast) than Serpents’ Island is to the mainland of Ukraine (which is about 20 nm).

6.121. Given the close location of Isla de Lobos to the Uruguayan mainland and the

relatively minor effect it has between Point E and Point F, it is only with difficulty that it

could be argued to produce a disproportionate result and constitute a special circumstance

justifying attributing it reduced or no effect.

582 UCM, para. 4.62(i)(2).
229(20) Sarawak, North Borneo, Brunei (British Orders in Council, 1958)

583
6.122. Ukraine relies on the observation in Charney & Alexander that “[s]o far as can

be determined… all islands in the vicinity of the central and eastern boundaries were
584
given full effect…”

6.123. The central boundary does not extend out to sea, but delimits the boundary as

between the two parts of Brunei from what is now that part of Malaysian territory

containing the Limbang river valley. The eastern boundary extends from the land

boundary between the Brunei and Malaysian territory to the east, passing to the west of

the large island of Pulau Labuan before proceeding out to sea. The western boundary

parts from the terminus of the land boundary between the two States to the west before

passing out to sea.

6.124. The report in Charney & Alexander observes that “[i]t appears that islands, rocks,

and low-tide elevations were only considered in respect of the central and eastern

boundary sections.” 585 In relation to the western boundary, it is observed that the only

offshore feature which might have had an effect is the Ampa Patches (a feature lying a

little less than 20 nm off the centre of the larger portion of Brunei) and “there is no

586
evidence that it was taken into account when the western boundary was selected.”

6.125. In relation to the method of delimitation, it is observed:

“It appears that two different methods were employed in delimiting these
three boundaries. The equidistance method was used to fix the eastern and

central boundaries and the first 5 n.m. of the western boundary adjacent to
the coast. The remainder of the western boundary was based on a line
which was perpendicular to the general direction of the coast” 587

583 UCM, para. 4.62(i)(3).
584 Charney & Alexander, vol. I, p. 918.
585 Ibid.
586
587 Ibid.
Charney & Alexander, vol. I, p. 919.
230These observations must be substantially qualified by the recognition that the matter

could “only be put beyond doubt by studying large-scale charts which were not

available”.588

6.126. Further, the delimitation was made by Order in Council at a time when the

relevant parts of Malaysian territory and Brunei were not yet independent. It appears

that the boundaries are not accepted by Brunei; in 2003, a dispute arose between Brunei

and Malaysia in relation to overlapping oil concessions off Sabah, implying that at least

the eastern boundary is not formally recognized.

6.127. In any event, Pulau Labuan, the island arguably having the largest effect in

relation to the outer zones is comparatively large, has long been inhabited, and does not

have a disproportionate effect on the delimitation of the eastern boundary. There is a cut-

off effect in relation to the eastern portion of Brunei, but this is due to the geography of

the area, and Brunei has continental shelf entitlement arising from the coast of the

western part; it is in relation to this that the equidistance line appears to be largely

calculated.

(21) Indonesia-Papua New Guinea, Agreement Concerning the Maritime

Boundary, 13 December 1980

589
6.128. Ukraine relies on the comments in the report in Charney & Alexander that

islands “significantly affected” the position of the delimitation line, and that all islands
590
were given full effect, including Wuvulu, which is located some 80 nm offshore.

6.129. Wuvulu is a relatively small island located some distance off the coast of Papua

New Guinea. It lies some distance to the east of the perpendicular line running from the

terminus of the land boundary. Papua New Guinea has (since 2002) purported to draw

archipelagic baselines around its offshore islands, including Wuvulu.

588
589 Charney & Alexander, vol. I, p. 918.
UCM, para. 4.62(i)(4).
590 Charney & Alexander, vol. I, p. 1041.
2316.130. In the geographical circumstances, the effect given to Wuvulu does not seem

disproportionate. Wuvulu has a substantial permanent population; it is the westermost

island of the Ninigo Atolls, which comprise 7 atolls with ca. 50 islets. Furthermore, the

Liki Islands (Pulau Liki and Isjuma), located relatively close to the Indonesian coast

some distance to the west, are apparently also given effect, although the impact of that

effect is relatively minor due to their position.

(22) German Democratic Republic-Poland, Agreement on Delimitation of the

Continental Shelf in the Baltic Sea, 29 October 1968

6.131. Ukraine relies on the fact that full effect was given to the island of Greifswalder

Oie in the 1968 Agreement, and attempts to suggest that it also played a role in the
591
drawing of the single maritime boundary agreed in 1989.

6.132. Greifswalder Oie is located relatively close to the shore and to other islands lying

off the German coast. The coast itself is markedly concave. Although the island was

given at least some effect in the 1968 Agreement, the effect is not sufficiently

disproportionate to constitute a special circumstance given the position of the island. In

any case, it bears noting that as from at least 1969, the GDR purported to draw straight
592
baselines around the island.

6.133. As to the 1989 Agreement the Ukrainian statement is misleading. The 1989

Agreement was concerned with fixing the outer limits of the Parties’ respective territorial

seas, as well as the adjustment of the continental shelf boundary from that established in

1968 and the creation of a fishing zone as a single maritime boundary. The report in
Charney & Alexander observes that although full effect was given to the German islands

in the 1968 Agreement, effect was not given to them in fixing the outer limits of the

territorial sea and adjustment of the continental shelf boundary:

“The [1989] agreement, on the other hand, disregards these features in
order to secure Polish navigational interests. Since the 1968 boundary was
relied upon to balance areas, it can of course be argued that these islands

591 UCM, para. 4.62(i)(5).
592 Cf. Limits in the Seas, #52 (1973).
232 did play a secondary role when the single maritime boundary was
drawn.” 593

II. Examples in Relation to Opposite Coasts

(23) USA-Cuba, Maritime Boundary Agreement, 16 December 1977

6.134. Ukraine stresses that under the Agreement, effect was given to all islands of both

parties, including the small islands forming the Florida Keys, which extend some 50 nm

off the mainland of Florida. 594

6.135. The Florida Keys form a near continuous chain of islands of various sizes

stretching off the southern tip of Florida, the largest of which are connected by road

bridges, and a large number of which are inhabited. Given their size, geographic

situation and character, their effect can only with difficulty be said to constitute a special

circumstance or to be disproportionate.

(24) Columbia-Costa Rica, Delimitation of Marine and Submarine Areas, 17
March 1977

6.136. The Albuquerque Cays form part of the San AndrØs and Providencia possessions

of Colombia located in the eastern Caribbean. Ukraine notes that the archipelago was

accorded equal weight as against the Costa Rica mainland. 595

6.137. As the Court is aware, the islands in the San AndrØs archipelago are the subject of

a dispute with Nicaragua which claims sovereignty over the islands of Providencia, San

AndrØs and Santa Catalina “and all the appurtenant islands and keys” capable of

appropriation.

6.138. The boundary consists of three points and two lines, which run east-west and

north-south, with the right-angle junction being on the Costa Rican side. The report in

Charney & Alexander notes that:

593
594 Charney & Alexander, vol. II, p. 2012.
UCM, para. 4.62(ii)(1).
595 UCM, para. 4.62(ii)(2).
233 “The agreement having been negotiated on the basis of equitable

principles, equal ‘weight’ was basically accorded to the Columbian small
archipelago of Albuquerque, in front of the coast of a country inhabited by
2,250,000 people. It seems that the islands obtained full ‘weight’ or

effect, though, with regard to hypothetical equidistance; Point A... slightly
favored Costa Rica; (turning) Point B favoured Columbia, thus producing
a compensatory effect which both governments deemed satisfactory.” 596

They also note that in the Agreement

“No reference to the method employed is stated and it is definitely not
equidistance.” 597

6.139. This is one of the rare examples of a very small island or formation apparently

being given full effect, even though the boundary is not an equidistance line, and is

clearly a negotiated line which was acceptable to both parties and was adopted for the

purposes of simplicity. However, the wider political context and the various disputes

between the coastal States may mean that there were other factors involved.

(25) Dominican Republic-Venezuela, Treaty on delimitation of marine and

submarine areas, 3 March 1979

6.140. Ukraine relies on the effect given to islands, in particular the Venezuelan Los
598
Monjes archipelago. The report in Charney and Alexander notes that the Dominican
599
islands utilised (apparently Saona and Beata) are all within 12 nm from the coast. The

Los Monjes archipelago lies at the entrance of the Gulf of Venezuela, lying roughly

south-southwest of Aruba.

6.141. The agreed boundary is in two sectors, a sector to the west, which appears to be

governed by the Monjes Islands on the Venezuelan side, and another sector in the east,

governed by other Venezuelan islands lying off the coast. The gap in the middle results

from the presence of Aruba and the Netherlands Antilles which effectively cut-off

Venezuela from the continental shelf in that sector.

596
597 Charney & Alexander, vol. I, p. 468.
598 Charney & Alexander, vol. I, p. 464.
599 UCM, para. 4.62(ii)(3).
Charney & Alexander, vol. I, p. 581.
2346.142. As with the previous example, there is a dispute with a third State over

sovereignty over the islands, which are also claimed by Colombia. Further, the

geography of the area is important, in particular the fact that the Netherlands Antilles lie

in front of the long Venezuelan coast in the middle sector.

6.143. In that context, it should be noted that in the Boundary Treaty concluded with the

Netherlands, 600 there is a substantial departure from equidistance in drawing almost all of

the lines, and in particular the direction of the lines moving out to sea to the west of

Aruba and to the east of Bonaire, resulting in significant gains for Venezuela in these

sectors.601

(26) Venezuela-United States of America, Maritime Boundary Treaty, 28 March
1978; Venezuela-Netherlands, Boundary Treaty, 31 March 1978; Venezuela-France,

Delimitation Treaty, 17 July 1980

6.144. Ukraine relies on the fact effect is given to Aves Island in agreements with the

United States of America, the Netherlands, and France. Particular reliance is placed on

the agreement with France, in relation to which Ukraine argues that Aves Island “was
602
given the same weight as Martinique”.

6.145. Ukraine’s reliance on Aves Island is controversial. Although the US, France and

the Netherlands have all been willing to accord it weight in concluding delimitation

agreements with their island territories, States in the region (including Saint Kitts and

Nevis, Saint Vincent and the Grenadines, and Saint Lucia), have protested against these

agreements, arguing that the island constitutes a rock within the meaning of Article

121(3) in that it does not sustain human habitation and therefore does not generate any

continental shelf or EEZ entitlement. All are States whose continental shelf entitlement

would be dramatically affected by any entitlement flowing from Aves Island. It should

be also noted that Dominica disputes Venezuelan sovereignty over the island.

600 Boundary Treaty between the Kingdom of the Netherlands and the Republic of Venezuela, 31
March 1978: 1140 UNTS 323.
601
602 Cf the discussion in Limits in the Sea #105 (1986).
See UCM, para. 4.62(ii)(4). Aves Island should not be confused with the two islands known as
Islas Las Aves in the Las Roques group, which belong to Venezuela and are located relatively close to the
north of its coast, to the east of Bonaire.
2356.146. Even if it is accepted that Aves Island is capable of generating continental shelf

and EEZ entitlement, the agreements with France, the US and the Netherlands (as it

relates to the line between Aves Island and Saba and Sint Eustatius) are delimitations

essentially as between opposite islands.

6.147. As regards Ukraine’s comment that Aves Island was given the “same weight as

Martinique” in the Agreement with France, it should be noted that neither of the two

delimitation lines as against Martinique and Guadaloupe are equidistance lines: they

rather appear to be segments of the same line. Further, although more or less “full”

weight was given as against Martinique, the report in Charney & Alexander observes:

“It seems that the decision to use one single line for both boundaries

prevailed, thus making the delimitation in the Guadeloupe sector appear to
be the result of a proportionality criterion”.603

(27) Ecuador-Costa Rica, Agreement on Delimitation of Maritime Areas, 12
March 1985

6.148. Ukraine relies on the Agreement between Costa Rica and Ecuador which delimits

the boundary between Isla del Coco (Costa Rica) and the GalÆpagos Islands (Ecuador) in

the eastern Pacific,604 noting that Isla del Coco was given full effect as against the larger

GalÆpagos group. 605

6.149. The obvious observations are that Isla del Coco lies in mid-ocean some 330 nm

southwest off mainland Costa Rica, and that the delimitation was carried out as against

another group of islands lying even further southwest out into the Pacific.

6.150. It is also noticeable that, given the distance between the islands, the equidistance

line in neither case does more than slightly flatten the 200 nm arc of each. In these

circumstances, it would be extremely difficult to suggest that giving full-effect to Isla del

Coco, despite the relative difference in size compared to the GalÆpagos Islands, would

603
604 Charney & Alexander, vol. I, p. 609.
Charney & Alexander, vol. I, p. 819.
605 UCM, para. 4.62(ii)(5) (it is assumed that the reference to Colombia is a mistake).
236have had a disproportionate effect. In addition, in any case it appears that the line is not a

true equidistance line, but deflects towards Isla del Coco in its western portion.

6.151. On the other hand, it appears that Isla del Coco was not given effect in the 1980

Costa Rica-Panama delimitation, where the line appears to have been drawn solely from

points on the mainland of each coast, even though Isla del Coco lies some 165 nm from

the agreed boundary and therefore would have deflected the line if strict equidistance

were used. 606 As against Colombia in the south, although Isla del Coco was obviously

given substantial weight, the boundary line is a simplified one running north-south. 607

(28) Australia-France, Agreement on Marine Delimitation, 4 January 1982

6.152. Ukraine relies on the fact that both Middleton Reef and Norfolk Island were given

full effect as against New-Caledonia. 608

6.153. In relation to Middleton Reef, which is situated 125 nm off the Australian coast,

the obvious observation is that the delimitation in that specific area was not so much a

delimitation between the Australian coast and New Caledonia (due to its remoteness from

the coast, Middleton Reef could not have qualified as a basepoint on the Australian

coast), but a delimitation between the reef itself and New Caledonia – again a mid-ocean

delimitation between islands. It also bears noting that the portion of the boundary

apparently governed by Middleton Reef is more than 200 nm from the formation, and

even further from the nearest point on New Caledonia. It therefore divides the claimed

extended continental shelves of France and Australia. Australia made its submission to

the Commission on the Limits of the Continental Shelf in relation to claims beyond 200

nm on 15 November 2004; the Commission is still considering the submission. This case

has no relevance for the Black Sea delimitation.

6.154. Charney and Alexander remark that:

606 Limits in the Seas, #97 (1982).
607
As to the Costa Rica-Colombia delimitation (ibid., fn 56), it is no surprise that Isla del Coco was
given full weight vis a vis Colombia. The relevant Colombian basepoint for determining the boundary,
608pelo Island, was almost as remote and even smaller than it.
UCM, para. 4.62(ii)(6).
237 “The absence of confirmed economic value in the maritime area in
question may have been one of the reasons for the French concession
609
[which accepted a full effect for Middleton Reef].”

6.155. Norfolk Island is a self-governing island of substantial size whose capacity to

generate a continental shelf and an exclusive economic zone of its own is not in doubt.

Again this is a case of a mid-ocean delimitation as between islands which are distant (a

little under 400 nm) from each other. It is difficult to envisage an equidistance line giving

rise to a disproportionate result in this situation.

(29) Japan and Republic of Korea, Agreement Concerning the Establishment of a
Boundary in the Northern Part of the Continental Shelf Adjacent to the Two
Countries, 30 January 1974

6.156. Ukraine observes that “numerous” small islands were given full effect. 610 It

appears that the delimitation line is not a true equidistance line, but has been modified in

some respects. 611 Further, the language of the Counter-Memorial makes clear that some

small islands were not taken into effect. On the Japanese side, the principal islands taken

into account are the very substantial island of Tsushima, the smaller island of Mi Shima,

which lies relatively close in front of the large island of Honshu, and, in the southern

sector, various small islands lying off the coast of Honshu. On the Korean side, the

Korean coast has a large number of islands of varying sizes lying in front of the mainland

coast, and many of these were apparently taken into account.

6.157. It is difficult to ascertain precisely which islands were used as basepoints, and

whether there were smaller islands which were disregarded. However, the general

configuration of the area is such that no particular small island was capable of having a

disproportionate effect on the course of the line, given that there are various larger islands

in the vicinity.

609
610 Charney & Alexander, vol. I, p. 907.
UCM, para. 4.62(ii)(7).
611 Limits in the Seas, #75 (1977).
238(30) Australia-France, Agreement on Marine Delimitation, 4 January 1982

6.158. Ukraine relies on the fact that Heard and McDonald Islands, although smaller than
612
the Kergeulen Islands, were given full effect.

6.159. Again, this is a mid-ocean delimitation between islands which are far apart and

uninhabited; the difference in size between the islands is not such that giving full effect to

the Australian islands would have resulted in a disproportionate result.

(31) Australia-Indonesia, Agreement Establishing Certain Seabed Boundaries, 18
May 1971

6.160. Ukraine relies on the fact that islands belonging to both sides were given full

effect, referring in particular to Pulau Enu, New Year Islet and Crocodile Island. 613

6.161. An examination of the geography reveals that New Year Islet and Crocodile

Island are located relatively close in front of the Australian mainland. Pulau Enu,

although small, is not in such a position as to cause a disproportionate effect, given that it

lies only slightly south of a compact group of much larger islands (including Pulau
614
Trangan).

6.162. The report in Charney & Alexander observes that:

“the offshore islands and peninsulas on the coasts of Indonesia and
Australia seem to offset each other so that no side gains a marked

advantage when an equidistant line is used… New 615r Islet, off
Australia’s Cobourg Peninsula, is offset by Pulau Enu…”

6.163. According to Limits in the Seas, although part of the boundary is based on

equidistance, one portion of the boundary (Points 13-16) was in any case “negotiated on

the basis of equitable principles relating to the geomorphology of the seafloor […]” 616

612 UCM, para. 4.62(ii)(8).
613 UCM, para. 4.62(ii)(9).
614 See the map attached to Limits in the Seas, #79 (1978).
615
616 Charney & Alexander, vol. II, p. 1197.
Limits in the Seas, #79 (1978).
239(32) India-Indonesia, Agreement relating to Delimitation of the Continental Shelf,
8 August 1974

6.164. Ukraine observes that the small islands of Pulau Rondo and Pulau Benggala were

given full effect.617

6.165. A study of the geographic situation reveals that neither of these formations are

situated far off the coast of the peninsula of Sumatra and that there are other larger

islands lying between them and the coast of Sumatra. It may also be of relevance that

both islands are apparently included within and form turning points for the Indonesian

archipelagic baselines system, although equidistance was apparently not measured from

the Indonesian baselines.

6.166. Further, on both sides there are narrow “fingers of land” whose tips face each

other across the strait (on the Indian side, the point of the island of Great Nicobar). In

this regard the report in Charney & Alexander notes that

“In view of the short length of coast involved on each side and in view of
the fact that all the basepoints were on or close to the respective mainlands
the equidistant method seems very suitable.”

6.167. It seems clear that it was not even arguable that use of the formations was such as

to have a disproportionate effect. The date of the Agreement is also important. It may be

questionable whether both of the formations would today qualify for continental shelf

entitlement under Article 121.

6.168. Finally, it should be noted that the 1977 agreement extending the boundary relies

on the same islands, as well as various small islands close to the west coast of Sumatra in
618
calculating the equidistance line in the western extension.

617
UCM, para. 4.62(ii)(10).
618 See Limits in the Seas, #93 (1981).
240(33) Indonesia-Thailand, Agreement relating to delimitation of the continental
shelf in the northern part of the strait of Malacca and in the Andaman Sea, 17

December 1971

6.169. Ukraine comments that the islands of Pulah Weh and Koh Racha Noi, which are
619
not known to have any permanent residents, were apparently given full effect.

6.170. The island of Pulah Weh is relatively large and lies between Sumatra and Pulau

Rondo; it is located within the Indonesian system of archipelagic baselines. Ko Racha

Noi, by comparison, lies more distant from other islands off Thailand. Although

Thailand purports to draw straight baselines, they do not encompass Ko Racha Noi and

pass landwards of the island.

6.171. Ukraine’s reliance on this example is misplaced, given that it appears to have

been the result, at least to some extent, of extraneous considerations. The report in

Charney & Alexander having observed that the delimitation is apparently based on

equidistance, states:

“It is also possible that Indonesia was prepared to allow Ko Racha Noi to
be used because Indonesia’s baseline system surrounded all Indonesian

islands and the Thai straight baselines in this area are quite modest in
comparison to some baselines claimed by other countries.” 620

(34) Iran-Bahrain, Agreement concerning delimitation of the Continental Shelf,
17 June 1971

6.172. Ukraine relies on the fact that the Iranian islands of Nakhilu and Jabrin and the

Bahraini island of Jazirat Al Muharraq were given full effect. 621 On the Iranian side,

Jazirat Al Muharraq lies very close to the major island of Bahrain, and is connected to it

by a causeway. The Iranian islands taken into account (Nakhilu and Jabrin) are located

very close (within 3 miles) of the Iranian mainland coast. In the circumstances, granting

full effect to any of the islands could not conceivably result in a disproportionate effect.

In any case, the effect of taking into account all of the islands results in their effects

being, to a large extent, offset against each other.

619
620 UCM, para. 4.62(ii)(11).
Charney & Alexander, vol. II, p. 1458.
621 UCM, para. 4.62(ii)(12).
241(35) Finland-USSR, Agreement Concerning Boundaries of Sea Areas and the

Continental Shelf in the Gulf of Finland, 20 May 1965; Agreement Concerning the
Boundary of the Continental Shelf in the North-eastern Part of the Baltic Sea, 5

May 1967

6.173. The 1965 Agreement is relied upon for the fact that many islands and islets appear
622
to have been given full effect. The 1967 Agreement is similarly relied upon for the

effect which appears to have been given to various islands despite the fact that the islands

were smaller than those in the Gulf of Finland, and were more numerous on the Finnish

side.623

6.174. The obvious point is that islands on both sides appear to have been taken into

account in both Agreements. It may also be of relevance that Finland draws straight

baselines around the large number of the islands off its coast; all of the points apparently

relied upon appear to fall upon islands included within these straight baselines. 624 On the

USSR side, the islands used are larger, and lie in front of the coast. The comment that the

islands outside the Gulf of Finland were smaller appears to be overly sweeping as the

islands relevant to the 1967 Agreement appear to be of a range of sizes.

6.175. Overall, the particular geographic circumstances made the use of small islands for

the application of equidistance inevitable. None of the islands used as basepoints appear

to be in such a position as to create a disproportionate result if an equidistance line is

used.

6.176. It also appears that 1967 extension was not a true equidistance line, apparently
625
being consistently 1 nm closer to Finland.

6.177. Subsequently, part of the boundary formed the subject of an Agreement between

626
Finland and Estonia in 1996. It appears that many of the points are very close to those

under the 1967 Agreement and that the line has been simplified.

622
UCM, para. 4.62(ii)(13).
623 Ibid.
624 Cf. Limits in the Seas #48 (1972), #56 (1973).
625 See the calculations in Limits in the Seas, #56 (1973).
626 Agreement between the Government of the Republic of Finland and the Government of the

Republic of Estonia on the Boundary of the Maritime Zones in the Gulf of Finland and in the Northern part
of the Baltic Sea, Helsinki,18 October 1996, 1964 UNTS 380.
242(36) Sweden-Poland, Agreement on Delimitation of the Continental Shelf and the
Fishery Zones, 10 February 1989

6.178. Ukraine relies on the agreement for the fact that the small island of Utklippan,

627
situated at approximately 9.5 nm from the mainland was given full effect.

6.179. In this regard, it bears noting that as with the other Swedish island, Utklippan is

comparatively close to the mainland, and that Sweden purports to draw straight baselines

around various of its islands, including a straight baseline between points on Utklippan

and Utlangen.

6.180. It also bears noting that the boundary is divided into two sectors, and although in

the first sector, full effect is given to all islands, in the second sector, as noted by

Ukraine, 628 it was apparently agreed that Gotland should be accorded reduced effect

(75%) in favour of Sweden. Poland had apparently made the argument during

negotiations that it should not be accorded any effect at all (cf. Sweden’s Agreement with

the USSR, Item 7 above).

(37) Denmark-German Democratic Republic, Treaty on Delimitation of the
Continental Shelf and Fishery Zones, 14 September 1988

6.181. Ukraine relies on the fact that, as in the Agreement with Poland, the small island

of Greifswalder Oie was accorded full effect. 629

6.182. It does not appear, however, that Greifswalder Oie was in fact attributed full

effect. It appears that there was some discussion as to whether the Danish island of

Bornholm should be given full effect as against the island given its remoteness from the

Danish mainland. As a result, it appears that the delimitation line is a negotiated one,

albeit admittedly one deriving ultimately from a line based on equidistance. The report in

Charney & Alexander notes that the

“… negotiated solution here used a proportionality calculation based on

the lengths of the opposite lying coastlines to adjust the line…”

627
628 UCM, para. 4.62(ii)(14).
See UCM, fn. 40 (accompanying the comments on the Sweden-USSR agreement).
629 UCM, para. 4.62(ii)(15).
243This is manifested in the form of a triangle taken out of the delimitation line in favour of

the GDR in the middle of the area where Bornholm exerts an effect on the equidistance

line.

(38) Oman-Pakistan, Muscat Agreement on Delimitation of the Maritime

Boundary, 12 June 2000

6.183. Ukraine places reliance on the fact that Sail Rock (Gurab), located just off Astola

Island was used as a basepoint. 630 In this regard, it relies on the commentary in Charney

and Alexander to the effect that Astola Island is clearly an island for the purposes of

Article 121(1), and that Sail Rock is clearly a feature from which extended maritime

claims can be made. 631

6.184. Astola is located roughly 25 km south of the Pakistan coast; Ukraine appears to

understate its dimensions. Sail Rock is a stone column which has apparently been

separated from Astola Island, lying roughly 7 cables (0.7 nm) off Astola Island. It

appears that Pakistan purports to draw straight baselines using Sail Rock as one of the
632
turning points.

6.185. It is clear from the general layout of the coast that the use of Astola Island and

Sail Rock had only a minor effect on the equidistance line with Oman; there are

promontories to the west and the east. In those circumstances, the effect given to Sail

Rock did not have a disproportionate effect.

(39) Denmark-United Kingdom, Agreement relating to Maritime Delimitation
between the Faroe Islands and the United Kingdom, 18 May 1999

6.186. Ukraine relies on the effect apparently given to Sule Skerry, Rona, Sula Sgeir and

the Flannan Islands, all of which are small and relatively remote from the mainland of the

633
United Kingdom.

630 UCM, para. 4.62(ii)(6).
631 Charney & Alexander, vol. IV, pp. 2812-3.
632 IBRU Boundary and Security Bulletin Autumn 1996, 87, available at

633p://www.dur.ac.uk/ibru.www/publications/bsbfirstpages/bsb4-3_prescott_p1.pdf
UCM, para. 4.62(ii)(17).
2446.187. The United Kingdom claims territorial sea from these various islands, several of

which have a substantial history of settlement. Indeed all but one of them have in the

past been inhabited either for sheepherding or seasonally for fisheries. In any case, the

Agreement is a complex one, involving the sharing of a “Shared Zone” in relation to
634
fisheries jurisdiction.

634
It may be noted that the southern limit of the shared zone was defined by a median line giving no
effect to these islands.
245 CHAPTER 7

OIL ANDFISHERIESACTIVITY

A. Introduction

7.1. In its Counter-Memorial Ukraine relies on a series of de facto lines allegedly
635
endorsed by Romania. It seeks to invoke a pattern of Ukrainian conduct in the

disputed area, allegedly coinciding with the Ukrainian claim. Specifically it refers to the

grant of oil and gas exploration and exploitation licenses as well as to fishing and

patrolling practices allegedly carried out. It considers these State activities as a “relevant

circumstance which operates in favour of the continental shelf/EEZ claim line proposed
636
by Ukraine”.

7.2. This chapter analyses these Ukrainian contentions, correcting certain erroneous

statements and casting light on others, in order to demonstrate, first, that Ukraine’s

assertions do not have the significance it gives them and, second, that the “State

activities” mentioned by Ukraine do not constitute relevant circumstances in the present

case, whether operating in favour of Ukraine or of Romania.

B. Oil and Gas Concessions in the Delimitation Area

(1) General Considerations

7.3. The Continental Shelf (Tunisia/Libya) case 637is the only one where the Court

considered oil concessions as actually having an effect in terms of maritime delimitation.

While it did not accept that the concession line was, as such, opposable to the Parties, the

Court found that close to the coasts the concessions of the parties confirmed the existence

of a modus vivendi:

“the Court could not fail to note the existence of a de facto line from Ras
Ajdir at an angle of some 26° east of north, which was the result of the
manner in which both Parties initially granted concessions for offshore

635
See UCM, paras. 8.39-8.65.
636 UCM, para. 8.41.
637 Continental Shelf Case (Tunisia/Libya), Merits, ICJ Reports 1982, p. 18.
246 exploration and exploitation of oil and gas. This line of adjoining
concessions, which was tacitly respected for a number of years, and which

approximately corresponds furthermore to the line perpendicular to the
coast at the frontier point which had in the past been observed as a de facto
maritime limit, does appear to the Court to constitute a circumstance of
638
great relevance for the delimitation.”

7.4. By contrast in the Gulf of Maine case the Chamber considered that the practice of

the parties in areas of overlapping claims could be taken into account only if it was

uniform, and continuous over a significant period:

“…even supposing that there was a de facto demarcation between the
areas for each of the Parties issue permits (Canada from 1964 and the

United States from 1965 onwards), this cannot be recognized as a situation
comparable to that on which the Court based its conclusions in the
Tunisia/Libya case. It is true that the Court relied upon the fact of the

division between the petroleum concessions issued by the two States
concerned. But it took special account of the conduct of the Powers
formally responsible for the external affairs of Tunisia – France – and of

Tripolitania – Italy – which it found amounted to a modus vivendi, and
which the two States continued to respect when, after becoming
independent, they began to grant petroleum concessions.” 639

7.5. In the Gulf of Maine case, the Chamber of the Court considered that:

“the period from 1965 to 1972, ‘at least’, which, according to Canada, is

the one in which the modus vivendi was instituted, is too brief to have
produced a legal effect of this kind, even supposing that the facts are as
claimed.” 640

7.6. In Cameroon v. Nigeria, the Court, after considering its own jurisprudence in the

field and the findings of the arbitral tribunals in Guinea/Guinea Bissau and St. Pierre et

Miquelon, said:

638
639 Ibid., p. 71 (para. 96).
Delimitation of the Maritime Boundary in the Area of the Gulf of Maine, ICJ Reports 1984, p. 246,
at p. 310 (para. 150).
640 Ibid., pp. 310-311 (para. 151).
247 “Only if [the oil concessions] are based on express or tacit agreement
641
between the parties may they be taken into account.”

And it concluded that:

“…although the existence of an express or tacit agreement between the
Parties on the sitting of their respective oil concessions may indicate a
consensus on the maritime areas to which they are entitled, oil concessions

and oil wells are not in themselves to be considered as relevant
circumstances justifying the adjustment or shifting of the provisional
642
delimitation line.”

7.7. Thus oil concessions can only be taken into account in the process of maritime

delimitation if they demonstrate a uniform, continuous and constant behaviour manifested

within a considerable period of time and demonstrating an agreement, tacit or expressed,

between the parties. Neither of these elements can be found in the present case.

(2) Paragraph 4(f) of the Additional Agreement

7.8. A relevant feature in evaluating the significance of oil concessions is the legal

status of the area in dispute between Romania and Ukraine. According to Paragraph 4(f)

of the 1997 Additional Agreement:

“Until reaching a solution concerning the delimitation of the continental

shelf, the Contracting Parties shall refrain from exploitation of the mineral
resources of the zone submitted to delimitation, the coordinates of which
shall be established at the beginning of the negotiations on the basis of the
643
above-mentioned principles…”

The 1997 Additional Agreement clearly admits the existence of a disputed area, which is

in fact the area defined by the Parties’ claims before the Court. This is in striking

contrast to Ukraine’s claim that by its Note Verbale of 7 November 1995 to the

Romanian Ministry of Foreign Affairs 644or by its Note Verbale of 19 November 2002 to

641 Land and Maritime Boundary between Cameroon and Nigeria, Merits, ICJ Reports 2002, p. 448

642ra. 304).
643 Ibid., p. 447 (para. 304).
644 See RM, Annex 2; UCM, Annex 1.
See UCM, para. 8.60.
248 645
the Bulgarian Embassy in Kiev it affirmed its sovereign rights over large segments of
646
the disputed area, considered as “unquestionably to appertain to Ukraine”.

7.9. Paragraph 4(f) of the 1997 Additional Agreement excludes any potential effect of

exploration and exploitation activities in the relevant area. Thus, any practice occurring

after the conclusion of the 1997 Agreement is irrelevant in the present proceedings. On

the other hand, any practice that occurred before the 1997 Agreement was superseded by

its provisions that referred to the establishment of a disputed area.

(3) Ukraine' s Concession Practice is Without Influence on the
Delimitation

7.10. In any event, Ukraine’s concession practice offers no support to its claimed
delimitation.

7.11. Ukraine invokes three areas it licensed in the disputed area after 1991:

- the Delphin block, licensed in 1993, located approximately in the north-eastern

part of the delimitation area, straddling the median line between the mainland

opposite coasts of the two countries;

- the Olympiiska block, licensed in 2001, roughly situated in the north-western area

of the maritime perimeter to be delimited, close to the Ukrainian maximal claim line;

and

- the Gubkina block, licensed in 2003, positioned in the northern part of the

delimitation area, south of the territorial sea of Serpents’ Island.

647
7.12. An initial point is that that the area covered by the Ukrainian concessions does

not even roughly correspond to its claim in the present proceedings. If lines are drawn

arbitrarily to connect the exterior main points of the Ukrainian concession blocks, the

area resulting covers far less than the area claimed by Ukraine, and even less than the

area covered by the Romanian concessions as reflected by the UCM.

645
See UCM, para. 8.61.
646 See UCM, para. 8.60.
647 See UCM, Figure 8-7 (facing p. 214).
2497.13. Further, the Ukrainian licensing practice has developed only very recently, well

after the critical date - 1997. Two of the three licenses, the Olympiiska block (2001) and

the Gubkina block (2003), were subsequent to the 1997 Additional Agreement. Taken in

its entirety, Ukraine’s practice does not even remotely meet the criteria of time,

consistency and express or implied agreement as laid down by the Court. Ukraine’s
648 649
conduct was limited in time and in extent, both territorially and materially.

According to the Chamber of the Court in the Gulf of Maine case, such a limited practice

is to be regarded as too brief to have produced a legal effect of this kind, “even supposing
650
that the facts are as claimed”.

(a) The Olympiiska and the Gubkina blocks

7.14. With regard to the Olympiiska and the Gubkina blocks, the latter being only

partially in the area in dispute, both were licensed after the entry into force of the 1997
651
Additional Agreement, in 2001 and 2003, respectively.

7.15. Insofar as the exploration of such deposits is concerned, these licenses cannot

have any relevance for the delimitation. Under the Additional Agreement, neither Party

is entitled unilaterally to regulate or undertake exploitation activities in the area in

dispute. By contrast exploration activity is allowed to be carried out by each Party. Thus

exploration activities undertaken by either Party in the disputed zone since 1997 could

not give rise to any basis of title in any event: they would merely involve the exercise of

provisional rights, mutually tolerated.

7.16. Insofar as the permits given for these two blocks refer to exploitation, this is in

clear breach of the 1997 Additional Agreement, which expressly prohibits the

exploitation by one Party of the mineral resources in the delimitation area. Moreover

after 1997 Romania constantly drew attention to the provisions of the Additional

648 The concessions cover only a limited part of the area in dispute defined by the claims of the two
Parties, and do not coincide, even partially, with the Ukrainian claim.
649 The concessions only cover a limited range of activities (mainly excluding exploitation of mineral

650ources) – see reference 47 at page 214 of Ukraine’s Counter-Memorial.
651 ICJ Reports 1984, p. 246, at pp. 310-311 (para. 151).
See UCM, paras. 8.48, 8.51.
250Agreement which prohibit the exploitation of the mineral resources of the area in

652
dispute.

7.17. In fact when drilling the two exploratory wells in the Olympiiska block:

“Chornomornaftogaz [the license holder] has been directed by the
Ukrainian Government that it was not authorised to carry out any
exploitation (as opposed to exploration) activities in the areas in respect of

which Romania disputes Ukraine’s entitlement pending settlement of the
maritime boundary dispute with Romania”. 653

7.18. Moreover, both permits were “granted by the Ministry of Ecology and Natural

Resources of Ukraine” 654 to the State-owned Ukrainian company Chornomornaftogaz.

As the Court said in a similar situation in the Continental Shelf (Tunisia/Libya) case:

“Both the Law and the Regulation which followed it are purely internal

legislative acts, intended to identify domestic zones for the petroleum
exploration and exploitation activities of Libya and could… hardly be

considered655en as a unilateral claim for maritime lateral boundaries with
Tunisia” .

(b) The Delphin block

7.19. The Delphin block was licensed before the entry into force of the 1997 Additional
656
Agreement. But contrary to what Ukraine asserts, the Delphin block is only partially,

i.e. less than 50%, in the area in dispute. The same is true for the well drilled in this
657
block in 1995. Moreover, the Delphin Area Licence Agreement clearly states that

“in the event that the territorial issue is resolved in a manner resulting in
all or part of the adjacent Alternate Area (as described in Exhibit C) being

in Ukrainian territorial waters, the State Committee shall issue a new
licence to the Licence Participant for all or that part of the adjacent
Alternate Area which lies within Ukrainian territorial waters…” 658

652 See below, paras. 7.21-7.23, 7.27, 7.30.
653 UCM, p. 214, fn. 47.
654
655 See UCM, p. 214, para. 8.48 and 8.51, p. 214
Continental Shelf Case (Tunisia/Libya), ICJ Reports 1982, p. 69, para 92.
656 UCM, para. 8.46.
657 As referred to in UCM, para. 8.47. Figure 8.7 of Ukraine’s Counter-Memorial depicts two wells
in the Delphin block; one of these is situated outside the area in dispute, while the second one seems to be

658Romania’s claimed delimitation line. So, neither falls within the area of Romania’s claim.
UCM, volume 5, Annex 97.
2517.20. This clause shows Ukraine’s awareness that the area is in dispute and contradicts

the assertions made in its Counter-Memorial. 659 The coordinates mentioned as defining

the “licence area covered by the Licence Agreement” (described as Points A, B, 660C, D)

661
seek to establish that the license area extends farther to the west. In fact, the License

Agreement sets the limits of the actual licensed area, restricted to a much smaller

perimeter, pending a final delimitation with Romania. Article III, para.3, of the Licence

Agreement, read in conjunction with Exhibit C of the same document, reveals the

existence of an adjacent Alternate Area for which a new license will be needed “in the

event that the territorial issue is resolved in a manner resulting in all or part of the

adjacent Alternate Area… being in Ukrainian territorial waters”. Whether the “Alternate

Area” will actually be licensed is thus dependent on the outcome of the delimitation.

(4) Romania consistently objected to Ukrainian hydrocarbon activity

7.21. Notwithstanding the “safeguard clause” of the Additional Agreement, Romania

has been vigilant with respect to Ukrainian conduct within the area. Further to Ukrainian

drilling operations in the Olympiiska perimeter in 2001, without any prior notification to

Romania, the Romanian Ministry of Foreign Affairs sent several Notes Verbales drawing

attention to the necessity to respect the provisions of the Additional Agreement.

7.22. By Note Verbale C23/2533/6 June 2001, Romania declared that:

“as the respective exploitation point is located in the northern area of the

contiguous zone of Romania, about 12 kilometres away from its territorial
sea, in a maritime area still subject to the delimitation process, the

Romanian Party considers that undertaking this kind of activities in that
perimeter represents a violation of the principle provided by art.4(f) of the
Additional Agreement…” 662

659 UCM, para. 8.45.
660 Point B has been described in the Counter-Memorial by the coordinates 400”N, 30°00'00”E
(UCM, para. 8,45). However, according to the License Agreement this point has the coordinates
44°30'00”N, 31°0000”E (UCM, Vol. 5, Annex 97).
661
662 See UCM, Annex 97.
Note Verbale no. C23/2533 of 6 June 2001 of the Romanian Ministry of Foreign Affairs addressed
to the Ukrainian Embassy in Bucharest (Annex RR 29).
2527.23. On 2 July 2001, by Note Verbale no. C23/3342, the Romanian Ministry of

Foreign Affairs sought clarification as to the “situation created by the carrying out, by the

Ukrainian gas enterprise Cernomornaftogaz, of drillings in the maritime area adjacent to

Serpents’ Island”. 663

7.24. Later on, in 2003, referring to the geophysical exploration activities carried out in

the Black Sea by the Ukrainian ship “Iskatel”, Romania noted, by Note Verbale no. DVI-

1/6784/15 December 2003, that

“taking into account the fact that the above mentioned ‘Iskatel’ ship is
carrying out scientific research activities in an area which partially

overlaps the area subject to delimitation between Romania and Ukraine,
the Romanian Party deems that it would have been advisable for the

Ukrainian party to have notified, through664plomatic channels, the
undertaking of the research activities”.

7.25. In respect of the same issue, in its Note Verbale no. EVI-1/2850/ 11 May 2004,

Romania further expressed

“its hope that the Ukrainian party should refrain, in the future, from such

unilateral actions, which are likely to influence the carrying out of the
negotiations regarding the delimitation of the maritime spaces of the two
countries in the Black Sea”. 665

7.26 In 2004, additional Notes Verbales were transmitted to similar effect. In particular

by Note Verbale no. EVI-1/6352/21 October 2004, Romania reacted as to a piece of news

referring to “the Ukrainian party’s decision to licence, for exploitation purposes, maritime

perimeters located South of Serpents’ Island, in the area subject to the delimitation

process between Romania and Ukraine”. In this respect, Romania emphasized that

“should the mentioned public information be confirmed, such licensing

activities carried out by Ukraine are in no way legally opposable to

663 Note Verbale no. C23/3342 of 2 July 2001 of the Romanian Ministry of Foreign Affairs addressed

664the Ukrainian Embassy in Bucharest (Annex RR 30).
Note Verbale no. DVI-1/6784 of 15 December 2003 of the Romanian Ministry of Foreign Affairs
665ressed to the Ukrainian Embassy in Bucharest (Annex RR 31).
Note Verbale no. EVI-1/2850 of 11 May 2004 of the Romanian Ministry of Foreign Affairs
addressed to the Ukrainian Embassy in Bucharest (Annex RR 32).
253 Romania, being unable to affect the process of delimitation of the
666
maritime spaces of the two States in the Black Sea”.

7.27. As this Note was not replied to by Ukraine, the position was reiterated in Note

Verbale no. EVI-1/6503/ 2 November 2004, by which Romania underlined

“the importance for this situation to be clarified as soon as possible, taking
into account the fact that, in case that the mentioned information confirms,

the Ukrainian party’s actions would constitute a breach of the provisions
of art. 4 (f) of the Agreement Additional to the Treaty on the Relations of
Good Neighbourliness and Co-Operation between Romania and Ukraine
667
(Constanta, 2 June 1997).”

7.28. Following the eventual receipt of a Ukrainian Note Verbale, Romania stated in its

Note Verbale no. EVI-1/7242/27 December 2004, that:

“the qualification that the Ukrainian party gives… to the North-Western
area of the Black Sea, characterising it as an area ‘with respect to which

Ukraine has sovereign and jurisdictional rights’ is not justified, being 668
likely to prejudge the solution offered by the International Court”.

7.29. In 2006, further Notes Verbales 669 were sent, following indications in the

Ukrainian media with respect to the Ukrainian authorities’ intention to issue licenses in

the Black Sea. The Romanian side clearly expressed its view that

“such licensing activities are not of such a character as to affect the 670
process of delimitation of the maritime spaces of the two States”.

7.30. It also made it very clear that

666
Note Verbale no. EVI-1/6352 of 21 October 2004 of the Romanian Ministry of Foreign Affairs
667ressed to the Ukrainian Embassy in Bucharest (Annex RR 33).
Note Verbale no. EVI-1/6503 of 2 November 2004 of the Romanian Ministry of Foreign Affairs
addressed to the Ukrainian Embassy in Bucharest (Annex RR 34).
668 Note Verbale no. EVI-1/7242 of 27 December 2004 of the Romanian Ministry of Foreign Affairs
addressed to the Ukrainian Embassy in Bucharest (Annex RR 35).
669
See Note Verbale no. H2/1607 of 22 May 2006 of the Romanian Ministry of Foreign Affairs
addressed to the Ukrainian Embassy in Bucharest (Annex RR 36), Note Verbale no. H2/1749 of 4 July
2006 of the Romanian Ministry of Foreign Affairs addressed to the Ukrainian Embassy in Bucharest
(Annex RR 37), Note Verbale no. H2/2457 of 14 August 2006 of the Romanian Ministry of Foreign
Affairs addressed to the Ukrainian Embassy in Bucharest (Annex RR 38).
670
Note Verbale no. H2/1607 of 22 May 2006 of the Romanian Ministry of Foreign Affairs addressed
to the Ukrainian Embassy in Bucharest (Annex RR 36).
254 “the disputed zone is defined by the maximum claims of each of the

parties and, therefore, until a final solution is settled in this regard, a
delimitation line per se is not to be conceived; consequently, Ukraine’s
unilateral fixing the coordinates of such a line is devoid of any legal
671
effects as to the Romanian party”.

At the same time, Romania reiterated the idea that

“to the extent that… the mentioned activities are taking place in the
maritime area where the claims of Romania, respectively, Ukraine

overlap, such initiatives represent an infringement of the provisions of the
Additional Agreement”. 672

7.31. For these reasons the Ukrainian licenses cannot constitute a relevant circumstance

for the delimitation, nor do they support the Ukrainian claimed maritime boundary in any

way.

(5) Romania’s Practice Regarding Oil and Gas Concessions

7.32. This conclusion is supported by an analysis of Romania’s own conduct with

respect to the exploration and exploitation of gas and oil deposits in the delimitation area.

In its Counter-Memorial, Ukraine gave a brief presentation of Romanian conduct,

concluding that “Romania respected a line which is entirely inconsistent with the

exaggerated claim line which it has set out in its Memorial”. 673 The Ukrainian

presentation is incomplete and misleading.

7.33. First, the concessions depicted in Figure 8-7 of the Ukrainian Counter-Memorial

represent only a minor part of the activities performed by Romania in the delimitation

area.

671 Note Verbale no. H2/1749 of 4 July 2006 of the Romanian Ministry of Foreign Affairs addressed

672the Ukrainian Embassy in Bucharest (Annex RR 37).
Note Verbale no. H2/2457 of 14 August 2006 of the Romanian Ministry of Foreign Affairs
673ressed to the Ukrainian Embassy in Bucharest (Annex RR 38).
UCM, para. 8.54, and for the presentation see UCM, paras. 8.55-8.57.
2557.34. Romanian activities in the area date since the 1960s when the first exploration

took place. Ever since, extended seismic profiles have been carried out in an area whose

outer limit coincides almost exactly with the maritime boundary claimed by Romania in

the present proceedings. These exploration activities are shown on the maps reproduced

as Figures RR26 and RR27 at pages 258 and 259 of this Reply.

7.35. In the 1990s, in an effort to attract foreign investment, the Romanian authorities

licensed foreign companies to explore and exploit mineral resources in certain areas of

the Black Sea. The Pelican, Istria, Midia and Neptun blocks are depicted on Figure 8-7

facing page 214 of the Ukrainian Counter-Memorial, and were granted to British

companies such as Enterprise Oil and Palladin, and French ones, such as Total/Elf. At

the time these concessions were granted Romania was aware of possible Ukrainian

claims (even though the full extent of Ukrainian claims was not known). Accordingly

Romania designated the blocks in such a way as not to overlap on what it thought could

be the maximum Ukrainian claim.

7.36. These concessions were granted even before the November 1995 Ukrainian Note

Verbale informing Romania of the continental shelf and the exclusive economic zone
674
limits claimed by Ukraine. By fixing the limits of the concessions as it did, Romania

could not possibly have respected a line which it was not aware of at the time.

7.37. Contrary to what can be said about the Ukrainian activities, Romania’s conduct

regarding oil and gas activities in the delimitation area was performed during a

reasonably extended period of time (more than 40 years) and is characterised by

uniformity, continuity and constancy. Further it covered an area which substantially

coincides with the maritime area claimed by Romania in the present proceedings – see
675
Figures RR26 and RR27 below .

674
675 UCM, para. 8.60. For this Note and the Romanian reaction to it, see further paras 7.41-7.43.
Examples of seismic profiles of some structures explored by Romania are enclosed as Annex RR
39.
2567.38. Still, Romania does not claim that its activities in the delimitation area amount to
a relevant circumstance to be taken into account by the Court in deciding about the

maritime boundary:

- first, such practices are not “relevant circumstances” in the law of

maritime delimitation;

- second, taking into account of said practices and activities would not be in

line with a good faith interpretation of the 1997 Additional Agreement.

7.39. Nevertheless, were the Court inclined to take these practices into consideration,
the Romanian activities are more significant than the limited activities Ukraine has

performed in the delimitation area.

257 5
2

e
m
s
h
t
; i
i a
a o
m R
R b
r b d
a d o
o m p
a a e
i c e
i e a
e a e
e e t
t t i
i a n
u m p
o e e
2 e t d
R r i u
e c w c
u s i ie
i ti n l
F i o i
c y ta
n c i
i f l
a e de
l t e
x o o
e l o
a s p
n t a
a i n
o c ai
R e k
e U
t ht
n e
s i
l t
d
e
s
A 9
2

t
p
d
n
a
r
e
t
i s
t i
o c
7 e g
R r i
R c p
r e r
g i v
F t f
a o
n e
t (
r
l
x
s
i
a
m
RC. FISHERIES PRACTICE IN THE DELIMITATION AREA

7.40. Ukraine contends that fisheries practice should be taken into account as relevant

circumstances, and that these activities (or their absence) “prove” that the boundary
676
claimed by Ukraine corresponds to an “interim” line respected by both Parties. But in

fact Romania has never recognized any such “interim” line (as Ukraine in fact admits 677).

Since the beginning of the bilateral negotiations on maritime delimitation, Romania’s

position has been clear and consistent.

(1) Ukrainian Claims have never been Recognised

7.41. In 1995 Ukraine communicated to Romania the coordinates of its claimed

exclusive economic zone area. 678 From the beginning, Romania contested through

diplomatic channels the validity of this so-called “interim line”. 679

7.42 The irrelevance of such a unilaterally established line was clearly pointed out by

the Court in the Tunisia/Libya case:

“The Court would therefore observe at the outset that an attempt by a

unilateral act to establish international maritime boundary lines regardless
of the legal position of other States is contrary to the recognized principles
of international law, as laid down, inter alia, in the Geneva Conventions of

1958 on the Law of the Sea, especially the Convention on the Continental
Shelf, which provide that maritime boundaries should be determined by
680
agreement between the Parties.”

7.43. Romania has constantly rejected any line unilaterally asserted by Ukraine,

681
whether for fisheries or any other purposes. Moreover, the so called Ukrainian

“interim” line has not been accepted by third States, contrary to what Ukraine seems to
682
suggest. Note Verbale ▯▯-12-723 of 3 October 2002 of the Bulgarian Embassy in

676 UCM, paras. 8.58-8.65.
677
678 UCM, para. 8.62.
679 See UCM, Vol. 3, Annex 26.
See UCM, para. 8.62.
680 Continental Shelf Case (Tunisia/Libya), Merits, ICJ Reports 1982, pp. 66-67 (para. 87).
681 See above, paras.7.21-7.31, as well as UCM, para. 8.62
682 UCM, para. 8.61.

260Kiev, invoked by Ukraine, refers to the “conditions upon which foreign vessels can

conduct fishing activities in the exclusive economic zones of heightened concern of

Ukraine and Romania”. 683 Even if this Note was prompted by “several incidents of

Bulgarian fishermen being arrested for illegal fishing” (which is not apparent from the

text), it contains simply a request for information as to the legislation and does not

express or imply any legal position. In particular it did not imply any recognition of a

Ukrainian claim.

7.44. Indeed, Ukraine itself did not suggest otherwise. In its reply to the Bulgarian note,

the Ukrainian Ministry of Foreign Affairs made clear that the line it claimed as

representing the limit of the Ukrainian exclusive economic zone was to be considered as

valid only pending the “conclusion of the Treaty on the Delimitation of the Continental
684
Shelf and Exclusive Economic Zones of Ukraine and Romania in the Black Sea”.

7.45. Note Verbale no 145/97 of 12 March 1997 of the Embassy of the Republic of
685 686
Turkey, also relied upon by Ukraine, equally does not involve any recognition of the

validity of the Ukrainian claim in the area. Even if it seems to imply a certain admission

of wrongdoing in respect to Turkish fishermen, it uses reserved language, simply stating

that “a thorough investigation has been made” and that “the Turkish authorities will take

the necessary measures to prevent the occurrence of such acts”.

7.46. Nor do the other fishing incidents relied upon by Ukraine sustain its claim. Not a

single response to the Ukrainian diplomatic notes has been produced. The incidents

imply nothing more than a unilateral claim which cannot have any impact in the

resolution of the present dispute. Moreover, the legal positions expressed in these notes

verbales is hardly in conformity with the provisions of the 1997 Additional Agreement

recognizing the overlapping claims.

683 See UCM, Annex 102.
684 Note Verbale No. 612/23-400-5121, 19 November 2002, UCM, Vol. 5, Annex 103.
685
686 UCM, Annex 105.
UCM, para. 8.64 (1).

261 ▯ The communiquØ of 15 January 1998 of the Ministry of Foreign Affairs of

Ukraine and the further notes verbales sent to the Ministry of Foreign Affairs of

687
Turkey placed the incident involving Turkish fishing boats that occurred that

day in the “exclusive economic zone” of Ukraine, failing to indicate that the area

concerned was actually in dispute between Romania and Ukraine.
688
▯ Equally, Ukraine’s Notes Verbales 9781/AE/077-00 of 7 February 2000 and
689
211/23-171 of 5 April 2003 refer, inaccurately, to incidents that had happened

in the area under dispute as having taken place in the exclusive economic zone of

Ukraine.

▯ Note Verbale No. 72/22-446-1134 of 31 May 2002 690 also suggests that the

relevant fishing incident occurred “in the exclusive (maritime) economic zone of

Ukraine”. It is surprising that Ukraine invokes this note as it seems never have

been sent to the Ministry of Foreign Affairs of the Republic of Turkey. As

indicated in handwriting on the note annexed to the Ukrainian Counter-Memorial

“this Note was not delivered to the Ministry of Foreign Affairs of Turkey since

[it] occurred against the very positive background created by the visits of the

Minister of Foreign Affairs of Turkey to Ukraine and of the President to the

Republic of Turkey and so this Note could have been an unpleasant dissonance

for the political climate of relations with the Republic of Turkey”. Under these

circumstances, the Note has no legal significance. 691

7.47. Furthermore, responding to a Note Verbale of the Romanian Embassy in Ankara

to the Turkish Ministry of Foreign Affairs by which the former informed the latter that

“the issue of the delimitation of the exclusive economic zones and of the

continental shelf of Romania and Ukraine in the Black Sea has not been

687
688 UCM, Annexes 106 & 107.
UCM, Annex 108.
689 UCM, Annex 110.
690 UCM, Annex 109.
691 The English translation used for both the Note Verbale and the handwritten caption is the one
provided by the Ukrainian party in UCM, vol. 5, Annex 109.

262 settled yet and over a certain area of the Black Sea the claims of the two
692
countries in respect of their maritime spaces overlap”,

the Turkish Ministry of Foreign Affairs communicated to the Romanian side that

“With regard to the delimitation of the exclusive economic zone and the
continental shelf in the Black Sea between Romania and Ukraine, the

Ministry would like to confirm that the diplomatic correspondence of
relevant Turkish authorities on incidents concerning Turkish fishing boats
in the Black Sea cannot be construed as an endorsement by Turkey of the
693
delimitation claims of neither Ukraine nor Romania.”

7.48. In conclusion, the Ukrainian maritime claim with respect to the delimitation of the

maritime areas in the Black Sea has been recognised neither by Romania nor by any third

State. In any event, the Ukrainian Notes Verbales and communiquØs regarding certain

fishing incidents occurred in the Black Sea are not opposable to Romania.

(2) Ukraine’s limited practice does not sustain any de facto line

7.49. Even had Romania respected the Ukrainian de facto line, this would not make it a

de facto delimitation line. As Kolb notes:

“a de facto line can be tolerated as such, i.e., as a de facto line in relation
to the exploitation of sponge banks, oil concessions, fishery surveillance
etc. It can be tolerated in the knowledge that it is only a provisional line
694
until such time as a maritime boundary is established.”

7.50. Ukraine does not say how the five alleged cases of third-State fishing boats

intercepted by Ukrainian authorities should be taken into account in the present case. The

five incidents invoked by Ukraine seem to have taken place in a narrow areas in the

eastern part of the delimitation area, closer to the delimitation line claimed by Romania

692 Note Verbale no. 2329 of 16 November 2006 of the Embassy of Romania in Ankara addressed to
the Ministry of Foreign Affairs of Turkey (Annex RR 40).
693 Note Verbale no. 2006/DHGY/480618 of the Ministry of Foreign Affairs of the Republic of
Turkey addressed to the Embassy of Romania in Ankara (Annex RR 41).
694
R. Kolb, Case Law on Equitable Maritime Delimitation, Martinus Nijhoff, The Hague, 2003, p.
182.

263than to the one claimed by Ukraine. This contradicts the Ukrainian assertion that “the

EEZ/continental shelf boundary claimed by Ukraine furthermore corresponds generally to

the limit of the Parties’ exclusive fishing zone as respected by both Romania and

Ukraine”. 695 The incidents were concentrated in a small sector of the area under

dispute 696and cannot support Ukraine’s claim, which is situated far to the west. A line

defining the area where all these incidents took place would enclose less than 15% of the

area of overlapping claims.

7.51. Ukraine cannot be considered to have undertaken acts à titre de souverain in the

disputed area, despite its suggestion to this effect. 697 It was fully aware that this area was

claimed by Romania. The contention that Romania “has not considered that it possesses

sovereign rights in this area” because it has not carried out activities “customarily
698
undertaken by a State in areas where it claims sovereign rights” is groundless since

Romania had expressly notified Ukraine of its claims.

(3) Judicial Consideration of Fisheries Issues

7.52. Ukraine assumes that monitoring fishing activities constitutes a relevant

circumstance for maritime delimitation purposes. The jurisprudence of this Court and

arbitral tribunals offers no support for this thesis. When activities related to fishing were

found to be relevant for delimitation purposes, it was for entirely different reasons.

7.53. In the Gulf of Maine case, fisheries-related matters were addressed at length. The

Chamber considered whether the delimitation line it had established had “catastrophic

repercussions” for the livelihood of the coastal population of the two countries concerned

– which were economically dependent on fisheries resources – and concluded that it had

not. It observed:

“It is… evident that the respective scale of activities connected with
fishing – or navigation, defence, or, for that matter, petroleum exploration

695 UCM, p. 216, para. 8.58.
696 UCM, Figure 8-9, facing p. 218.
697
698 UCM, para. 8.65.
Ibid.

264 and exploitation – cannot be taken into account as a relevant circumstance,
or if the term is preferred, as an equitable criterion to be applied in
determining the delimitation line. What the Chamber would regard as a

legitimate scruple lies rather in the concern that lest the overall result, even
though achieved through the application of equitable criteria and the use of
appropriate methods for giving them concrete effect, should unexpectedly

be revealed as radically inequitable, that is to say, as likely to entail
catastrophic repercussion for the livelihood and economic well-being of
the population in the countries concerned”. 699

The Chamber thus considered that such significant fisheries resources and activities as

those of the Grand Banks were not relevant circumstances for delimitation purposes; they

were taken into account only in assessing the equity of a line established on other grounds.

7.54. This approach was followed in the St Pierre and Miquelon case, where the

Arbitral Tribunal said that it “has an obligation to assure itself that the solution reached is
not ‘radically inequitable’, the term used by the Chamber of the International Court of

Justice in the Gulf of Maine case…”, and went on to conclude that “the facts submitted to

the Court indicate that the proposed demarcation will not have a radical impact on the

existing pattern of fishing in the area.”00

7.55. In the Eritrea/Yemen case, the Tribunal was again very cautious in respect to

fisheries, giving this factor no significance in terms of the delimitation:

“It is not possible or necessary for the Tribunal to reach a conclusion that
either Eritrea or Yemen is economically dependent on fishing to such an
extent as to suggest any particular line of delimitation… The evidence

concerning fish consumption advanced by each Party was presumably
aimed at establishing that the Tribunal' s adoption of the line of
delimitation proposed by the other Party would constitute a serious dietary

or health threat to the population of the first Party. However, the evidence
on this matter is conflicting and uncertain. It is difficult if not impossible
to draw any generalized conclusions from the welter of alleged facts

advanced by the parties in this connection…
Based on the foregoing, the Tribunal finds no significant reason on any

other grounds concerning fishing –whether related to the historical

699
Delimitation of the Maritime Boundary in the Area of the Gulf of Maine, I.C.J. Reports 1984, p.
700 (para. 237).
Delimitation of Maritime Spaces between Canada and France (St. Pierre et Miquelon), (1992) 21
RIIA, p. 294 (paras. 84-85).

265 practice of fishing in general, to matters of asserted economic dependency
on fishing, to the location of fishing grounds, or to the patterns of fish

consumption by the population – for accepting, or rejecting, the arguments
of either Party on the line of delimitation proposed by itself or by the other
Party. Neither party has succeeded in demonstrating that the line of

delimitation would produce a catastrophic or inequitable effect on the
fishing activity of its national or detrimental effects on fishing
communities and economic dislocation of its nationals.

For these reasons, it is not possible for the Tribunal to accept or reject the

line of delimitation proposed by either Party on fisheries grounds. Nor can
the Tribunal find any relevant effect on the legal reasons supporting its
own selection of a delimitation line arising from its consideration of the

general part fishing practice of either Party or the potential deprivation of
fishing area or access to fishing resources, or arising from nutritional or
other grounds.” 701

7.56. In the Barbados/Trinidad and Tobago case, the Arbitral Tribunal noted that:

“resource-related criteria have been treated more cautiously by decisions
of international courts and tribunals, which have not generally applied this
factor as a relevant circumstance.” 702

The Tribunal further commented on the effect of the equidistance line on Barbadian

fishermen, noting that:

“this deprivation is profoundly significant for them, their families, and

their livelihoods, and… its deleterious effects are felt in the economy of
Barbados. But injury does not equate with catastrophe. Nor is injury in the
course of international economic relations treated as sufficient legal
703
ground for border adjustment.”

7.57. In conclusion, fishing practices and State activities related to them have not

generally been considered relevant circumstances in maritime delimitation. The only

exception is the Jan Mayen decision, where the Court noted that:

“It appears… that the median line is too far to the west for Denmark to be
assured of an equitable access to the capelin stock, since it would attribute

701 Maritime Boundary Delimitation between Eritreea and Yemen (Eritrea v. Republic of Yemen),
Second Stage: Maritime Delimitation, (2001) 22 RIAA, p. 335, at pp. 350-352 (paras. 64, 70, 72-73).
702
Delimitation of the Exclusive Economic Zone and the Continental Shelf between Barbados and the
703ublic of Trinidad (2006) 45 ILM, p. 800, at p. 839 (para. 241).
Ibid., p. 845 (para. 267).

266 to Norway the whole of the area of overlapping claims. For this reason
also the median line thus requires to be adjusted or shifted eastwards.” 704

7.58. As the Tribunal noted in the Barbados/Trinidad and Tobago case:

“the Jan Mayen decision is most exceptional in having determined the line
of delimitation in connection with the fisheries conducted by the parties in
dispute. However, as the question of fisheries might underlie a number of

delimitation disputes, courts and tribunals have not altogether excluded the
role of this factor but, as in the Gulf of Maine, have restricted its
application to circumstances in which catastrophic results might follow
705
from the adoption of a particular delimitation line.”

The position taken by the Court in Jan Mayen can be explained by circumstances

particular to that case: while the Greenland population was heavily depending on fishing,

which constituted the mainstay of its economy, Jan Mayen had no settled population. As

the Court noted: “Norwegian fishing interests in the waters surrounding Jan Mayen are

however the interests of mainland Norway, not of Jan Mayen as such.” 706 This is not at

all the situation in the present case.

7.59. Thus, in accordance with the consistent jurisprudence of international courts and

tribunals, Ukraine could rely on its own fisheries practice only if it had established that

the line proposed by Romania would entail “catastrophic repercussions” for the Ukrainian

population, or, at the very least, that it would prevent “equitable access” of its nationals to

fisheries resources, and that its own line would avoid such results. Ukraine does not

make such a claim and certainly does not (and cannot) advance any proof in this respect.

(4) Limited Dependence of the Parties on Black Sea Fisheries

7.60. The Ukrainian claim has to be assessed taking into account the characteristics of

the fisheries sector in the Black Sea. In accordance with the FAO Fishery Country

704 Maritime Delimitation in the Area between Greenland and Jan Mayen (Denmark v. Norway),
I.C.J. Reports 1993, p. 72 (para 76).
705 Delimitation of the Exclusive Economic Zone and the Continental Shelf between Barbados and the

706ublic of Trinidad, (2006) 45 ILM, p. 800, at p. 839 (para. 241).
Maritime Delimitation in the Area between Greenland and Jan Mayen (Denmark v. Norway),
Judgment, I.C.J. Reports 1993, p. 73 (para. 79).

267 707
Profile for Ukraine, distant-water fisheries account for most of the Ukraine catch

(about 70%), and the Black Sea, Azov Sea and inland fisheries only for 30%. The value

of harvested and reared fish products in 2001 represented only 0.005 % of the Ukrainian

GDP, while the number of professional staff, as well as people connected with fisheries,
fish products processing, transporting, sale, shipbuilding and ship repairing was about

0.3% of the economically active population. Given these figures, neither the Romanian

proposed line nor any other line could seriously be considered to entail catastrophic

repercussions for the Ukrainian population. The same is true of Romania as well. In

these circumstances the overall situation of the fisheries sector in the Black Sea is without

any relevance to the present delimitation.

D. CONCLUSIONS

7.61. For these reasons neither the oil and gas concessions in the delimitation area, nor

the fisheries practice qualifies as relevant circumstances in the delimitation of the

maritime areas of Romania and Ukraine in the Black Sea.

7.62. In any event, so far as the oil and gas concessions are concerned:

- the Ukrainian practice is recent and non-consistent with the Ukrainian claim in

delimitation;

- Romania constantly and clearly made known its views on the lack of relevance of

such Ukrainian activities;
- its invocation by Ukraine is inconsistent with paragraph 4(f) of the 1997 Additional

Agreement.

By contrast:

- the Romanian conduct concerning the gas and oil exploration and exploitation

activities in this area was performed on a constant basis for more than 40 years and

coincides closely with Romania’s claimed line.

707 See http://www.fao.org/fi/fcp/en/UKR/profile.htm (last visited: 15 October 2006)

2687.63. As to fisheries:
- the practice of the Parties can have no bearing on the maritime delimitation in the

present case since neither Party economically depends on fisheries activities in an

area in which pelagic fish stocks are limited;

- the practice invoked by Ukraine is recent and only covers a small part of the area in

dispute;

- it has always been challenged by Romania and has never been recognised by third

States.

269 CHAPTER 8

THE DELIMITATION LINE

A. THE POSITIONS OF THE PARTIES

8.1. For the reasons given in Chapter 3, the delimitation the Court is called upon to

effect is to be approached in two sectors, characterised by relationships of, respectively,

adjacency and oppositeness of the relevant coasts of the Parties.

8.2. As to Sector 1, the mainland basepoints which would determine the equidistance
line are not in dispute. What is in dispute is the use of Serpents’ Island as a basepoint for

the drawing of an equidistance line, even on a provisional basis.

8.3. Romania’s position is that Serpents’ Island generates a 12 nm band of territorial

sea, but no continental shelf or exclusive economic zone, and this for three distinct

reasons:

(a) The first segment of the maritime boundary delimiting the maritime areas of the

two States situated beyond their territorial seas was established by successive

Romanian/Soviet agreements: from the final point of the boundary separating
what now constitute the territorial seas of the two States (Point F), the maritime

boundary passes on the 12 nm arc of circle around Serpents’ Island until it reaches

a point situated on that arc due east of Serpents’ Island, as depicted on all charts

submitted by Romania (Point X).

(b) Serpents’ Island is a rock which cannot sustain human habitation or economic life

of its own and which generates no entitlement to a continental shelf or exclusive
economic zone. That being so, it cannot constitute a basepoint for delimitation of

these maritime zones.

(c) If, however, the question of maritime delimitation around Serpents’ Island were to

be considered independently of any treaties or agreements in force between the

Parties but on the assumption that Serpents’ Island was to be a basepoint,
nonetheless it would be given nil effect in the delimitation, having regard to its

size and location and to the wholly disproportionate impact it would have.

270Since Serpents’ Island is to be given nil effect beyond 12 nm, it cannot be, even

provisionally, a basepoint for the delimitation and it cannot be, even provisionally, taken
into account in constructing the equidistance line.

8.4. In its Counter-Memorial, Ukraine contests the conclusion that delimitation is to be

effected in two sectors. Rather it starts from a median/equidistance line drawn using

Serpents’ Island as a basepoint and adjusts that median line further to Romania’s

disadvantage in order to take account of what it claims to be the markedly

disproportionate coastal lengths of the two States. In its view this is a single sector of

delimitation between the whole Ukrainian coast and the whole Romanian coast. Thus the

coastal proportion between the two Parties is approximately 4:1 in Ukraine’s favour,

which qualifies as a relevant circumstance that makes it necessary the adjustment of the

equidistance line drawn dominantly from Serpents’ Island. Ukraine does this while
professing still that the case is not about Serpents’ Island. But without that basepoint, the

delimitation would – even accepting all of Ukraine’s other assumptions – be radically

different. Serpents’ Island is thus, in effect, the advance guard of coastal frontages

situated a great distance from it.

B. UKRAINE’S DELIMITATION LINE

8.5. If one follows Ukraine’s approach, the influence of Serpents’ Island even on the

delimitation between the opposite coasts in Sector 2 is enormous. By far the longest part

of the delimitation line proposed by Ukraine is controlled not by the Ukrainian mainland

coasts but by Serpents’ Island. Only its short final segment (situated south of the point
708
referred to by Ukraine as “Point C”) is controlled by the Ukrainian coast of Crimea,

south to Cape Khersones. Such an approach practically eliminates all the Crimean coast

between Cape Tarkhankut and Cape Khersones, a coast which undoubtedly generates

overlapping entitlements vis-à-vis the Romanian coast.

708 See UCM, Figure 7-1.

2718.6. In effect, Ukraine seeks to treat this tiny feature as a surrogate for coasts lying
well behind it to the north and east to which it bears no relation other than the accident of

the same sovereignty. The consequence is confusion twice confounded. First, Ukraine

confuses the issue of (a) the effect to be given to a tiny maritime feature which lies in

front of adjacent coasts and is in a position to obstruct their delimitation with (b) the

effect of longer mainland coasts on a provisional delimitation line. The circumstance of

Serpents’ Island belonging to Ukraine (as it now does) has nothing whatever to do with

the respective lengths of the mainland coasts. Secondly, Ukraine confuses (c) the

construction of a provisional delimitation line with (d) the consideration to be given to
the respective coastal frontages of the parties. As explained above, it is inadmissible to

use the ratio of relevant coastal lengths (even assuming Ukraine has got these correct) as

a criterion for the drawing of a provisional line. Yet this is what Ukraine does, seeking to

give credibility to the exercise by purportedly adjusting further an “equidistance line”

drawn from Serpents’ Island, which line would be the same whether or not the coasts

existed or belonged to Ukraine.

8.7. Of course, an island is entitled to its own maritime areas, measured from its own

baselines or basepoints. If such an island is situated in a coastal relation (whether of
adjacency or oppositeness) with the coastlines of other States, a delimitation between

their respective maritime areas may be envisaged. But the effect to be given to the

respective island is one thing; that to be given to the coastal frontage of the mainland

territory of the State to which the island belongs is quite another. In the present case the

two must be kept distinct in the interests not merely of equity but of clarity of analysis.

8.8. In effect, Ukraine performs a delimitation between the Romanian coast and the

coast of Serpents’ Island (with a short segment of the Crimean coast). But it does not
determine the relevant area between these coasts, or focuses on the total lengths of these

coasts. They would have been: for Romania, the total Romanian coast (situated in a

relation of adjacency with Serpents’ Island and of oppositeness to Crimea), and for

Ukraine the south-western coast of Serpents’ Island (facing the Romanian coast) and the

Crimean segment between Cape Khersones and Cape Sarych. On that basis the length of

272the Romanian relevant coast would have been 269.67 km while the length of the relevant

Ukrainian coast would have been of only approximately 42 km – i.e. a ratio of 6.4:1 in

Romania’s favour!

C. ARRIVING AT A DELIMITATION LINE: A PRINCIPLED APPROACH

8.9. For these reasons, as well as those given in preceding Chapters, Ukraine’s method

of delimitation is inadmissible. It is necessary to ask, first, whether Serpents’ Island can

provide a basepoint for the drawing of any line beyond 12 nm; secondly, whether, even if

it could do so, the provisional equidistance line in the present case should not be a

mainland-coasts equidistance line, and thirdly, what factors should be taken into account
709
in adjusting that provisional line with a view to achieving an equitable result.

8.10. In the interests of responsiveness, these issues will be considered on the footing

(quod non) that there is no maritime boundary agreement beyond Point F and that

Serpents’ Island is not a rock to which Article 121(3) applies. Even on those

assumptions, Ukraine’s proposed delimitation line contradicts both State practice and the

settled jurisprudence of maritime delimitation.

(1) Serpents’ Island as a basepoint

8.11. Chapter 7 of the Counter-Memorial (“The Provisional Equidistance Line”)

devotes almost all its space to Serpents’ Island and, more specifically as to why Serpents’

Island should be treated as one of the basepoints for constructing the provisional

equidistance line. Ukraine argues that, in accordance with its legislation “the low-water

mark on Serpents’ Island shore may be used, together with any other appropriate point on

Ukraine mainland baselines in order to determine the relevant basepoints for the
710
construction of the provisional equidistance line”.

709
There remains the further step of checking the equity of the result by reference to the
710portionality criterion: this is discussed in Chapter 9.
UCM, para. 7. 28.

2738.12. In considering whether – even on the assumptions set out in paragraph 8.10 above

– Serpents’ Island constitutes a basepoint opposable to Romania, it is relevant to note that

it was not mentioned for this purpose even by Ukraine’s own official acts. Ukraine has

already enacted a system of baselines, but points located on Serpents’ Island are

conspicuously absent from the points which define these baselines. The List of the

geographical coordinates of the points defining the position of the baselines for

measuring the width of the territorial waters, economic zone and continental shelf of the

Black Sea, communicated by the Permanent Mission of Ukraine to the Secretary-General
711
of the United Nations on 11 November 1992, includes no reference to Serpents’ Island.

Serpents’ Island is, in Ukraine’s view, a basepoint that does not lie on any baseline.

8.13. Ukraine relies in this regard on Article 5 of the 1991 Statute concerning the State

Frontier, which provides that the territorial sea of Ukraine is measured “from the line of

minimum low tide both on the mainland and on islands belonging to Ukraine, or from the

straight baselines joining the corresponding points”, and Article 2 of the Ukrainian law

relating to the EEZ which provides that “the exclusive (marine) economic zone of

Ukraine shall be comprised of maritime areas beyond and adjacent to the territorial sea of
712
Ukraine, including areas surrounding islands belonging to Ukraine”. However, both

the Statute concerning the State Frontier and the law relating to the EEZ are drafted in

general terms, without making any specific reference to Serpents’ Island. If the actual

delimitation is as Ukraine now proposes, this is hard to credit.

8.14. Alternatively Ukraine argues that:

“if the list of the coordinates did not include any point on Serpents’ Island,
it was because it was not necessary. As a matter of fact, the list defined

the basepoints of a system of straight baselines, and Serpents’ Island was
not part of that system”.713

This is misleading. In fact the system employed by Ukraine, as notified to the Secretary-

General, is only in part a system of straight baselines as in several sectors the baseline

711
712 Annex RM 27.
713 UCM, para. 7.27.
UCM, para 7.46.

274follows the low-water line of the coast. 714 Further, the word “straight” is not actually

used, but the list is said to include simply “the geographical coordinates of the points

defining the position of the baselines” without any qualification. If Ukraine claimed that

Serpents’ Island was used as a basepoint for measuring the breadth of its various

maritime claims, it should have included it in this notification.

8.15. The 1982 UNCLOS requires States to give due publicity to the baselines it uses
715
for the purposes of measuring the breadth of its territorial sea. Ukraine cannot claim,

after registering a list of coordinates of its basepoints with the Secretary-General of the

United Nations, that it is entitled to rely on additional basepoints for the purposes of the

delimitation of its maritime spaces with Romania.

8.16. Ukraine’s other contention, namely that the communication of the coordinates

was done shortly after Ukraine’s independence, and that Ukraine “took the system of

baselines that had been previously adopted by the USSR in the Black Sea”, 716 does not

take matters further. Even if Ukraine notified the coordinates only a short time after its

independence, it gave no hint of any intention to modify the baselines thereafter, right up

until the filing of the Counter-Memorial.

(2) Ignoring small islands in defining a provisional equidistance line

8.17. In its Counter-Memorial, Ukraine argues that a provisional equidistance line

should have taken into account all available basepoints and that in drawing a mainland

equidistance line as its provisional line Romania has failed to give effect to this

714 See the description of Ukraine baselines in Annex RM 27.
715 See Art. 16 “Charts and lists of geographical coordinates”:

“1. The baselines for measuring the breadth of the territorial sea determined in accordance with
articles 7, 9 and 10, or the limits derived therefrom, and the lines of delimitation drawn in
accordance with articles 12 and 15 shall be shown on charts of a scale or scales adequate for
ascertaining their position. Alternatively, a list of geographical coordinates of points, specifying
the geodetic datum, may be substituted.
2. The coastal State shall give due publicity to such charts or lists of geographical coordinates and
shall deposit a copy of each such chart or list with the Secretary-General of the United Nations.”
716
UCM, para. 7.46.

275established practice. 717 Again, of course, this argument only comes into play on the

assumptions set out in paragraph 8.10 above. But even on those assumptions (and even

disregarding Ukraine’s own conduct as set out in the preceding section), Ukraine’s

argument is supported neither by the jurisprudence nor by the practice.

8.18. As to the former, Ukraine notes that in the Anglo-French Continental Shelf

arbitration, the Tribunal took into account the Scilly Islands when drawing a provisional
718
equidistance line. However, the situation of the Scilly Islands bears no resemblance to

the circumstances of the present case. The Scilly Islands are of considerable size and

have a resident population. They also constitute the continuation of the coast of Great
719
Britain: the Tribunal likened them to an “exceptionally long promontory”. The Scillies

have a counterpart in the presence of the islands of Ushant (Oouessant) which

counterbalance, to a certain degree, their influence on the equidistance line. By contrast,

in the present case none of these circumstances is present: Serpents’ Island is a small,

uninhabitable feature which is in no sense integrated with the mainland coast, from which

it lies at more than 20 nm. Eloquent in this respect is the satellite picture of the area

included on page 142 of this Reply and also reproduced on page below, which shows

Serpents’ Island lying out at sea, by no means integrated with the mainland.

717 UCM, paras. 7.5-7.26.
718 UCM, paras. 7.16-7.17.
719 Case concerning the Delimitation of the Continental Shelf between the United Kingdom and
France (United Kingdom of Great Britain and Northern Ireland v. French Republic), Award of 30 June

1977, RIIA, Vol. XVIII, p. 113 (para. 224).

276Satellite picture of the Danube Delta, publicly available on the Internet site
http://earth.jsc.nasa.gov/sseop/EFS/lores.pl?PHOTO=NASA7-720-6. As clearly seen from the picture,

Serpents’ Island, which, because of its tiny size, can hardly be seen in the image, is not integrated with the
Ukrainian mainland.

720
8.19. In the Libya/Malta case (also relied on in Ukraine’s Counter-Memorial), Filfla

was ignored when establishing the provisional median line. The Court noted that:

“the equitableness of an equidistance line depends on whether the
precaution is taken of eliminating the disproportionate effect of certain

720 UCM, para. 7.18.

277 ‘islets, rocks and minor coastal projections’ to use the language of the
Court in [the North Sea Continental Shelf case]. The Court thus finds it

equitable not to take account of Filfa in the calculation of the provisional
median line between Malta and Libya. Having established such a
provisional median line, the Court still has to consider whether other

considerations, including the facto of721oportionality, should lead to an
adjustment of that line being made.”

722
8.20. The Jan Mayen case, also invoked by Ukraine, is irrelevant for present

purposes since the provisional line there was established between the coasts of two

islands, Greenland and Jan Mayen. These were not considered in conjunction with any

larger mainland; there was no other maritime formation that could be taken into account.

Also irrelevant is the Cameroon v. Nigeria case, where no islands belonging to the two

States were present in the delimitation area.

8.21. The Qatar/Bahrain case is equally irrelevant: the way the Court established the

provisional equidistance line was determined by circumstances that were peculiar to that

specific geographical context. For instance, as the Court was unable to determine

whether Fasht al Azm formed part of the island of Sitrah, it drew two equidistance lines
723
reflecting each of the hypotheses.

8.22. On the other hand, and contrary to Ukraine’s assertion, there are many instances

where minor maritime formations, especially ones that are not contiguous to the

mainland, have been ignored when plotting the provisional equidistance line.

8.23. Thus, in the Gulf of Maine case – frequently relied on by Ukraine but curiously

ignored in this context – the Chamber pointed out that:

“the potential disadvantages inherent in any method which takes tiny
islands, uninhabited rocks or low-tide elevations, sometimes lying at a

considerable distance from terra firma, as basepoints for the drawing of a
line intended to effect an equal division of a given area. If any of these

721
722 Continental Shelf (Libyan Arab Jamahiriya v Malta), ICJ Reports 1985, p. 13, at p. 48 (para. 64).
723 UCM, para. 7.19.
Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain),
ICJ Reports 2001, p. 40, at p. 104 (para. 216).

278 geographical features possess some degree of importance, there is nothing

to prevent them their subsequently being assigned whatever limited
corrective effect may be equitably ascribed to them, but this is an
altogether different operation from making a series of such minor features
724
the very basis for the geometrical construction of the entire line.”

The Chamber went on to add that, in relation to Article 6(2) of the 1958 Geneva

Convention on the Continental Shelf,

“a line drawn in accordance with the indication given by that provision
(‘equidistance from the nearest points of the baselines from which the breadth
of the territorial sea of each State is measured’) might well epitomize the

inherent defects of a certain manner of interpreting and applying the method
here considered… inasmuch as the likely result would be the adoption of a

line all of whose basepoints were located on a handful of isolated rocks, some
very distant from the coast, or on a few low-tide elevations: these are the very
type of minor geographical features which… should be discounted if it is

desired that a delimitation line should result so far as feasible in an equal
division of the areas in which the respective maritime projections of the two
countries coasts’ overlap.” 725

8.24. In the Eritrea/Yemen case, the Tribunal began with a mainland-coast equidistance

line: in its view the boundary should be “as far as practicable, the median line between
726
the opposite mainland coast”, a solution “in accord with practice and precedent in the
727
like situations”. The effect of the presence of islands was assessed at a later stage. The

Tribunal also remarked that, even after the provisional equidistance line had been

adjusted to take into account the presence of certain islands, “the final solution is that the

international maritime boundary line remains for the greater part a median line between
728
the mainland coasts of the Parties”.

8.25. As to State practice, reference may be made to the review appended to Chapter 6.

Of course those cases fall into several categories, more or less relevant to the present

724
Delimitation of the Maritime Boundary in the Gulf of Maine Area (Canada/United States of
America), ICJ Reports 1984, p. 246, at pp. 329-30 (para. 201).
725 Ibid., p. 332 (para 210).
726 Government of the State of Eritrea and Government of the Republic of Yemen (Phase Two:
Maritime Delimitation), Award of 17 December 1999, 119 ILR, p. 417, at p. 457 (para. 132).
727 Ibid., p. 457 (para. 132).
728
Ibid., p. 458 (para. 132).

279dispute. Of no relevance are delimitations between isolated mid-oceanic islands, where
729
there is by definition no mainland coast. Of limited relevance are cases where effect of

some kind was given to relatively large but isolated islands, located a substantial distance
730
from the mainland of the State to which they appertain. Rather the present dispute

concerns cases of small islands or islets lying in front of the coasts of one or both parties.

Of this latter group, a significant proportion of negotiated delimitations apparently started

from a provisional equidistance line drawn from mainland coasts. This was true, for

example, of Case 1, the 1969 Iran-Qatar Agreement and Case 6, the 1971 Italy-Tunisia

agreement as regards the four Italian islands. Such an approach also appears to have been

used in Case 5, the 1968 Italy-Yugoslavia Agreement, Case 8, the 1968 Agreement

between Saudi Arabia and Iran, Case 9, the 1969 Qatar-Abu Dhabi Agreement, Case 10,

the 1968 Iran-Dubai Agreement, Case 13, the 2000 Vietnam-China Agreement and Case

36, the 1989 Swedish-Polish Agreement. As discussed in Chapter 6, the basis for this

approach is the effect that the islands in question would otherwise have on the provisional

equidistance line.

8.26. Romania has followed precisely the same method as that adopted by the Court

and international arbitral tribunals in the cases mentioned above.

8.27. If Ukraine’s assertion that Serpents’ Island has to be considered as a basepoint

relevant for drawing the delimitation line were to be upheld, this minor maritime

formation would control the delimitation line almost completely (as can be seen from

Figure 7-1 from Ukraine’s Counter-Memorial). It would eliminate entirely the Ukrainian

coast situated behind Serpents’ Island (i.e., the adjacent Ukrainian coast) and almost

entirely its opposite Crimean coast (only the short segment between Cape Khersones and

Cape Sarych would matter). According to Ukraine, “Cape Tarkhankut [on the Crimean

coast] is undeniably the most salient projection of the Ukrainian coast in the entire

729 See e.g. Case 27 (Ecuador-Costa Rica), Case 28 (Australia-France (New Caledonia) and Case 30

(Australia-France (Kerguelen Islands)). Ukraine appears to accept the irrelevance of delimitations between
mid-ocean islands given the fact that the majority of the relevant practice in this regard is only referred to in
730ootnote (see UCM, para. 4.62, fn. 56).
See in particular Case 14 (Australia/Indonesia).

280 731
area”, yet on this approach Cape Tarkhankut would have no role in establishing the

actual delimitation.

8.28. For these reasons as well as those set out in the preceding sub-section, the
provisional equidistance line in the present case should be drawn as a mainland cost

equidistance line, ignoring Serpents’ Island as a basepoint. To use Serpents’ Island as a

basepoint for this purpose would contradict not only Ukrainian internal legislation (which

does not refer in any way to Serpents’ Island as a basepoint) and its own notification of

its basepoints, but also the jurisprudence of this Court and arbitral tribunals.

(3) Adjusting the provisional equidistance line

8.29. The mainland-coasts equidistance line so drawn is of course provisional and

subject to adjustment. The factors which might call for adjustment have been dealt with

already:

(a) Serpents’ Island is a tiny feature, well off-shore and (independently of its status as

a “rock” for the purposes of Article 121(3)) uninhabited and without economic

significance. In accordance with the considerations set out in Chapter 6, and

especially the extensive State practice with features of this kind, it should be given

zero effect.

(b) Oil and fisheries practices Ukraine relies on are not relevant circumstances in

principle, and certainly not in this case, for the reasons stated in Chapter 7.
(c) The relevant coasts of the Parties were established in Chapter 3, and Ukraine’s

objections in that regard fully rebutted. The actual coastal ratio is 1:1.4 in favour of

Ukraine, which is by no means such as to require the adjustment of a provisional

equidistance line. Whether the effect of Romania’s claimed line passes the test of

proportionality, and is otherwise equitable, is discussed in Chapter 9.

731
UCM, para. 8.17.

281D. THE DELIMITATION LINE

8.30. For these reasons, a mainland-coasts equidistance line, adjusted to leave a 12 nm

territorial sea to Serpents’ Island, is the appropriate delimitation line to be drawn in this

case. In drawing that line with precision, the two sectors need to be briefly considered.

(1) The delimitation between adjacent coasts (Sector 1)

8.31. The provisional equidistance line is drawn from the final point of the land/river

boundary between the two States taking into account the salient basepoints of the

adjacent Romanian and Ukrainian coasts. These, as noted in the Memorial, are: on the

Romanian coast, the outer end of the Sulina Dyke, and on the Ukrainian coast, Cape

Kubansky (as well as any point situated on the line uniting Cape Kubansky with Point

1438) and Cape Burnas. The resulting provisional equidistant line was depicted by

Romania in Figure 28 at page 219 of its Memorial; it is also depicted as Figure RR28 at

page 283 of this Reply.

282 8
2

t
a
l
r
n t
) c a
2 j c
e a n
8 u h n
2 i t a
R F e k
r RM w U
u g e n
F i e a
o i i
o t n
( a m
i o
i R
q
e
h
T8.32. As can be seen, the first segment of this provisional equidistance line (from
the last point of the Romanian/Ukrainian land/river boundary to the point where it

meets the eastern 12 nm arc of circle around Serpents’ Island) must be superimposed

on maritime areas already delimited by the 1949 ProcŁs-Verbaux. Further, most of

the maritime boundary on the 12 nm arc of circle around Serpents’ Island between

Point F and Point X lies south of the provisional equidistance line, allocating Ukraine

a circular zone of maritime spaces in an area situated in front of the Romanian coast.

However, close to Point X, the 1949 maritime boundary intersects the provisional

equidistance line, so that Point X is situated immediately to the north of it. Thus the

determination of the maritime boundary beyond Point X must start by joining Point X
to the provisional equidistance line.

8.33. In its Memorial, Romania argued that this should be done via Point Y, situated

practically mid-way between the 12 nm arc around Serpents’ Island and Point T, the

tripoint as between the Romanian and Ukrainian adjacent coasts and the opposite
732
Crimean coast.

733
8.34. Contrary to Ukraine’s assertion, this point is neither “mysterious” nor based
734
on “compensation” or “distributive justice”. The criteria taken into account in
determining the location of Point Y were, on the one hand, the need to respect the

choice of the Parties, who evidently accorded importance to Point X when they drawn

the maritime boundary on the 12 nm arc around Serpents’ Island, and, on the other

hand, to respect the principle that in the absence of any other relevant circumstance, a

further departure from the equidistant line would not be justified.

8.35. Another solution would have been to establish the maritime boundary beyond

Point X not on the equidistant line, but on the line going through Point X and

perpendicular to the closing line of Musura Bay. Such an approach would have been

more responsive to the bases on which Point X was established.

732 See RM, paras. 11.68-11.72.
733 See UCM, para. 4.19.
734 See UCM, paras. 4.16-4.18.

2848.36. As far as the “distributive justice” issue is concerned, this is irrelevant in

determining the line at this stage. All Romania did was to note that the maritime area

north of the provisional equidistance line allocated to Romania by a line drawn from

Point X through Point Y to Point T roughly equals the area lost by Romania because

of the unjustified departure from equidistance when delimiting the territorial seas

between Romania and the USSR in 1949, a factor to be kept in mind when
considering the overall equity of the solution adopted.735

8.37. For the reasons stated in Chapters 4-6, there are no other relevant

circumstances in Sector 1 of the delimitation area. Thus (subject to any considerations

of proportionality or equity, to be considered in Chapter 9), the maritime boundary in

this sector has the following trace: from Point F (the final point of the maritime
boundary separating the territorial seas of Romania and Ukraine) on the 12 nm arc of

circle around Serpents’ Island up to Point X, then in an easterly direction until it

reaches the line equidistant between the adjacent Romanian and Ukrainian coasts at

Point Y; then on the line equidistant to the adjacent Romanian and Ukrainian coasts to

Point T, the mainland coast tripoint. The delimitation line in Sector 1 is presented

below, as Figure RR29 (reproducing Figure 30 of Romania’s Memorial).

735 See RM, para. 11.72.

285 8
2

1
r
) c
3 S
e n
9 u y
2 i a
R F n
r R ou
u g b
F i e
o t
o r
( m
e
T (2) The delimitation between opposite coasts (Sector 2)

8.38. From Point T southwards, the delimitation is governed by the opposite
Romanian and Ukrainian coasts. The relevant coasts in this sector are:

- in case of Romania: the whole Romanian coast, between the last point of the

Romanian/Ukrainian land/river boundary to the last point of the Romanian/Bulgarian

land boundary;
- in case of Ukraine: the Ukrainian coast of Crimea between Cape Tarkhankut

and Cape Sarych.

8.39. The median line, calculated taking into account the salient basepoints on the

relevant opposite coasts of the two States (the outer ends of the Sulina Dyke and of
the Sacalin Peninsula on the Romanian coast, and capes Tarkhankut and Khersones

on the Ukrainian coast) is depicted on Figure 32 at page 228 of Romania’s Memorial.

The maritime boundary in Sector 2, which coincides with the segment of the median

line between Point T (which is also equidistant to the Ukrainian coast adjacent with
the Romanian one) to the north, and Point Z (beyond which interests of third States

could be affected) to the south is depicted in Figure 33 at page 230 of Romania’s

Memorial. It is reproduced as Figure RR30 below.

287 Figure RR30
(following RM Figure 33)

The maritime boundary in Sector 2

288E. CONCLUSION

8.40. For all these reasons, the maritime boundary between the exclusive economic

zones and the continental shelf of Romania and Ukraine is as described in paragraph

11.86, as well as in the Submission of Romania’s Memorial:

- from Point F, at 421'N, 30°0227E, on the 12 nm arc surrounding Serpents’

Island, to Point X, at20'N, 30°212'E;

- from Point X in a straight segment to Point 59'N, 30°416'E;

- then on the line equidistant between the Romanian and Ukrainian adjacent coasts,

from Point Y to Point T, at45'N, 31°040'E;

- and then on the line median between the Romanian and Ukrainian opposite coasts,

from Point T to Point Z, at50N, 31°2010'E.

8.41. This boundary is illustrated on Figure 34 at pages 232 and 241 of Romania’s

Memorial. It is also depicted in Figure RR31 at page 290 of this Reply.

289 Figure RR31

(following RM Figure 33)
The maritime boundary between the exclusive
economic zones and the continental shelf of
Romania and Ukraine in the Black Sea 290 CHAPTER 9

THE EQUITABLE CHARACTER OF THE DELIMITATION

A. INTRODUCTION

9.1. In its Memorial, Romania demonstrated the equitableness of its proposed

delimitation line, showing that it satisfies the proportionality test, leads to no cut-off of

the natural prolongation of the Ukrainian territory and respects the security interests of
736
both countries.

9.2. By contrast Ukraine argues that the delimitation advocated by Romania is

inequitable. It goes even further: even a provisional equidistant/median line calculated

between the Romanian coast and Serpents’ Island does not lead to an equitable

delimitation of the maritime areas of the two countries and should be shifted closer to the

Romanian coast in order to take into account the alleged disparity between the coastal

lengths of the two Parties. Not surprisingly (given the criterion it uses to achieve this

shift), Ukraine argues that the boundary it claims satisfies the test of proportionality.

B. THE PROPORTIONALITY TEST IN INTERNATIONAL CASE-LAW

9.3. The adjustment proposed by Ukraine is grounded on a haphazard notion of the

place of proportionality in maritime delimitation.

9.4. In the North Sea Continental Shelf cases the Court held that a factor to be taken

into account in maritime delimitations is

“the element of a reasonable degree of proportionality… between the
extent of continental shelf appertaining to the States concerned and the
737
lengths of their respective coastlines.”

736
737 See RM, paras. 12.1-12.19.
North Sea Continental Shelf Cases (Federal Republic of Germany/Denmark; Federal Republic of
Germany/Netherlands), ICJ Reports, p. 3, at p. 52 (para. 98).

2919.5. In the Anglo-French Continental Shelf case, the Court of Arbitration noted that:

“it is disproportion rather than any general principle of proportionality

which is the relevant criterion or factor. The equitable delimitation of the
continental shelf of the continental shelf is not, as this Court has already
emphasized… a question of apportioning – sharing out – the continental

shelf amongst the States abutting upon it. Nor is it a question of simply
assigning to them areas of the continental shelf in proportion to the length
of their coastline; for to do this would be to substitute for the delimitation

of boundaries a distributive apportionment of shares. Furthermore, the
fundamental principle that the continental shelf appertains to a coastal
shelf as being the natural prolongation of its territory places definite limits
on recourse to the factor of proportionality… Proportionality therefore is

to be used as a criterion or factor relevant in evaluating the equities of
certain geographical situations, not as a general principle providing an
independent source of rights to areas of continental shelf.” 738

9.6. In the Tunisia/Libya case the Court assigned to proportionality the role of a test

effected in order to check the equitableness of the delimitation line it had previously

established. Finding that the “relevant coastline of Libya stands in proportion of

approximately 31:69 to the relevant coastline of Tunisia”, that “the coastal front of Libya,

represented by a straight line drawn from Ras Tajoura to Ras Ajdir, stands in the

proportion of approximately 34:66 to the sum of the two Tunisian coastal fronts

represented by a straight line drawn from Ras Kaboudia to the most westerly point of the
Gulf of Gabes, and a second straight line from that point to Ras Ajdir”, and that “the

areas of shelf below the low-water mark within the area relevant for delimitation

appertaining to each State following the method indicated by the Court stand to each

other in approximately the proportion: Libya 40; Tunisia 60”, the Court went on to

conclude that “this result, taking into account all the relevant circumstances, seems to the

Court to meet the requirements of the test of proportionality”. 739

9.7. It should be noted that the Court found that a ratio of approximately 1: 2 in
respect to the lengths of the coastlines and coastal fronts was proportionate to a ratio of

approximately 2:3 in respect to the areas of continental shelf.

738
Delimitation of continental shelf (United Kingdom of Great Britain and Northern Ireland and the
French Republic), Decision of 30 June 1977, 54 ILR, p. 6, at pp. 67-8 (para. 101).
739 Continental Shelf (Tunisia/Libyan Arab Jamahiriya), ICJ Reports 1982, p. 18, at p. 91 (para. 131).

2929.8. In the Libya/Malta case the Court held that:

“to use the ratio of the coastal length as of itself determinative of the

seaward reach and area of continental shelf proper to each Party, is to go far
beyond the use proportionality as a test of equity, and as a corrective of the
unjustifiable difference of treatment resulting from some method of drawing

the boundary line. If such a use of proportionality were right, it is diffi740t
indeed to see what room would be left for any other consideration…”

Nevertheless, the Court found it necessary that the delimitation line between the areas of

continental shelf appertaining respectively to the Parties be adjusted so as to lie closer to

the coasts of Malta.

9.9. The Court further asserted that the proportionality test may be applied in the case

of opposite coasts, but it took care to emphasize the difficulties that occur in such

geographical situation. It noted in this respect that:

“These difficulties are particularly evident in the present case, where, in the
first place, the geographical context is such that the identification of the

relevant coasts and the relevant areas is so much at large that that virtually
any variant could be chosen, leading to widely different results; and in the
second place the area to which the Judgment will in fact apply is limited by
741
reason of the existence of third States.”

9.10. In the Gulf of Maine case, the Chamber found that, in the second sector the ratio

of the coastal lengths to be applied for the purposes of determining the location of the

corrected median line was approximately 1.32:1 in favour of the United States and it

shifted the line closer to the Canadian coast in this sector.

9.11. In the St. Pierre & Miquelon case, the parties differed markedly as to the length of

the relevant coasts and as to the size of the relevant area. The Court of Arbitration held

that the ratio between the Canadian and French coastlines was 15.3:1, while the ratio of

the actual maritime spaces appertaining to parties was 16.4:1. It concluded that “there is

certainly no disproportion to in the areas appertaining to each of the Parties.

740
Case Concerning the Continental Shelf (Libyan Arab Jamahiriya/Malta), ICJ Reports 1982, p. 13,
at p. 45 (para. 58).
741 Ibid., p. 53 (para. 74).

293Consequently, the requirements of the test of proportionality, as an aspect of equity, have

been satisfied”. 742

9.12. In the Jan Mayen case, the Court found that a marked “disparity between the

respective coastal lengths of the relevant area” had to be taken into account and

accordingly shifted the line closer to the Jan Mayen coast. In fact the ratio of relevant

coastal lengths was found by the Court to be 9 to 1.

9.13. On the role of proportionality, the Court noted that:

“Taking account of the disparity of coasts lengths does not mean a direct
and mathematical application of the relationship between the length of the
743
coastal front of eastern Greenland and that of Jan Mayen.”

Equally in respect to the proportionality test, the Court stressed that:

“The practical implementation of the principle may sometimes be

complicated, as in the Libya/Malta case, by the presence of claims of third
States, or by difficulties in defining with sufficient precision which coasts
are relevant.” 744

9.14. In the Eritrea/Yemen case, the Tribunal employed again the test of

proportionality. According to its calculations, the ratio of coastal lengths was 1 to 1.31,

while the ratio of water areas was 1:1.09 and asserted that the line of delimitation it has

decided upon resulted in no disproportion. 745

9.15. Proportionality played virtually no role in the Qatar/Bahrain and

Cameroon/Nigeria cases, where the Court concluded that the lengths of the relevant

742
Delimitation of Maritime Areas between Canada and France, Decision of 10 June 1992, 21
RIAA, p. 265, at p. 296 (para 93).
743 Maritime Delimitation in the Area Between Greenland and Jan Mayen (Denmark v. Norway),
Judgment, ICJ Reports 1993, p. 69 (para. 69).
744 Ibid., p.68 (para 67).
745
Award of the Arbitral Tribunal, 17 December 1999, RIIA, Vol. XXII (2001), Part IV, p. 335, at p.
373 (para. 168)

294coasts between the parties were roughly the same, without carrying out precise

746
calculations.

9.16. In Barbados/Trinidad and Tobago, the Tribunal stated that it

“will thus not resort to any form of ‘splitting the differences’ or other
mathematical approaches or use ratio methodologies that would entail
attributing to one Party what as a matter of law might belong to the
747
other.”

It added that:

“proportionality is not a mathematical exercise that results in the
attribution of maritime areas as a function of the length of the coasts of the
Parties or other such ratio calculations, an approach that instead of leading

to an equitable result could itself produce inequity. Proportionality is a
broader concept, it is a sense of proportionality, against which the

Tribunal can test the position resulting from the provisional application of
the line it has drawn, so as to avoid gross disproportion in the outcome of
748
the delimitation”.

The Tribunal did, in this case, effect a certain adjustment of the equidistance line in order

to take into account the ratio of the coastal frontages of the two countries abutting on the

area of delimitation. In the Tribunal’s own terms it made a “limited” adjustment to the

provisional equidistance line to take account of a substantial difference in coastal

frontage. 749 The Tribunal did not quantify the difference with precision, but it had been

presented as being of the order of 8:1 in favour of Trinidad and Tobago. 750 The lesson to

be drawn is that even substantial differences in relevant coastal frontages do not

751
necessarily translate into substantial adjustments of a provisional equidistance line.

746 Land and Maritime boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial
Guinea intervening), ICJ Reports 2002, pp. 446-7 (para. 301).
747
748 Award of the Arbitral Tribunal, 11 April 2006, 45 ILM, p. 798, at p. 853 (para. 338).
749 Ibid., p. 857 (para. 376).
Ibid., p. 855 (para. 350), and for the adjustment itself see ibid., p. 857 (paras. 372-3) and Map V,
ibid., p. 859.
750 Ibid., p. 852 (para. 326).
751 The issue of minor off-shore islands did not arise.

2959.17. Although the case law is not entirely consistent, several conclusions may be

drawn as to the role to be played by proportionality.

9.18. First, only a marked disproportion between the ratio of the lengths of coastline of

the parties on the one hand and the ratio of the water expanses allotted to them on the

other hand justifies the conclusion that an equidistant/median line does not lead to an

equitable solution.

9.19. Second (and conversely), in assessing any possible disparity, recourse is not to be

made to strict mathematical proportionality. Courts have stressed that such an approach

would be tantamount to apportioning the maritime spaces in proportion to the coast

length of each State and would thus leave virtually no room for any other factors to be

taken into account.

C. THE EQUITABLENESS OF THE LINE

(1) The Calculation of Proportionality

9.20. Against this background it is useful to compare the way in which the

proportionality test is applied by the Parties and the outcome reached in terms of such

criteria as non-encroachment.

9.21. Ukraine asserts that on the basis of a provisional equidistance line taking into

account Serpents’ Island, the area ratio would be 1:2.3 in its favour, while the ratio of

coastal lengths would be 1:3.7 or even 1:4.1 in its favour. In its view such a division
752
would not satisfy the “element of a reasonable degree of proportionality”.

9.22. Of course such a hypothesis requires one to accept Ukraine’s incorrect figures for

the relevant coasts, its flawed identification of the relevant area and its drawing of a
equidistant/median line that awards full effect to Serpents’ Island. But even so, it is far

from obvious that the resulting figures justify an adjustment of the equidistant/median

752 UCM, paras. 9.24-9.25.

296line. If a coastal ratio of 1: 2 is not disproportionate to an area ratio of 2:3 (as the Court

found in the Tunisia/Libya case), why should a coastal ratio of 1 to 3.7 be
disproportionate to an area ratio of 1:2.3?

9.23. In any case, the figures provided by Ukraine for the purposes of applying the

proportionality test are mistaken, and its criticism of Romania’s proportionality

calculations is flawed.

753
9.24. Romania has dealt with these two issues at length in this Reply. It has

demonstrated that the coast of Ukraine between Point S (the southern entrance point of

the Nistru/Dniester Firth) and Cape Tarkhankut has no geographical relationship with the

Romanian coast and consequently is irrelevant for purposes of effecting the delimitation

of the maritime spaces of Romania and Ukraine.

9.25. Equally, it has shown that Ukraine’s definition of the relevant area is at variance

with relevant norms of the law of the sea and with the practice of this Court and of

arbitral tribunals, as Ukraine purports to include in the relevant area spaces where the

entitlements of the two countries do not overlap (the entire area north of the line uniting

Point S and Cape Tarkhankut) and to exclude areas where they do overlap (the roughly

triangular area, defined by Cape Sarych, a point described by Ukraine as “a point

between the Romanian and the Ukrainian coasts where the interest of third States

potentially come into play” and the point of intersection between the meridian going

through Cape Sarych and the boundary delimiting the Ukrainian and Turkish maritime
spaces). Furthermore, Ukraine’s definition of the relevant area is prejudicial to the

interests of a third party, Bulgaria, because of the failure to identify as the southern limit

of the relevant area the equidistant line between Romania and Bulgaria, and its inclusion

of an area situated beyond this line.

9.26. Ukraine alleges that “Romania has no hesitation using its entire coast up to its

boundary with Bulgaria despite the fact that the longer segment of this coast, south to the

753 See above, paras. 3.10-3.11, 3.31-3.35, 3.39-3.46, 3.75-3.87.

297Sacalin Peninsula, has absolutely no effect on the construction of Romania’s claim line”
754
and that Romania “double counts” the northern sector of its coast. However, Ukraine

does not actually contest that the Romanian southern coast (between Sacalin and the

border with Bulgaria) is opposite to the coast of Crimea and that it is relevant for the

delimitation of the maritime spaces of the two countries. In fact Ukraine used this

segment in carrying out its own test of proportionality. Ukraine also used as its

basepoints only Serpents’ Island and Cape Khersones; consequently, if we are to follow

the same logic, Ukraine is equally to blame, as the entire coast between the Romanian-

Ukrainian border and Cape Khersones, which, to borrow Ukraine’s own turn of phrase,
“has absolutely no effect on the construction of [Ukraine]’s claim line”, is nevertheless

included by Ukraine in its calculation for the proportionality test.

9.27. Equally, the accusation of “double counting” the northern sector of the coast is

incorrect. Romania used the most advanced points of its coast for purposes of

establishing the equidistance line, in accordance with the established practice in the field

of maritime delimitation. At the same time, when establishing the total length of its

relevant coast, Romania did not take into account twice the length of the sector of its

coast relevant for both sectors of delimitation – thus performing no “double-counting”.

9.28. The flawed definition by Ukraine of the relevant area and of the relevant coasts

leads, inevitably, to distorted proportionality calculations. Ukraine contends that the

coastal length ratio is 4.1 to 1 in its favour and that the area ratio resulting from the

delimitation it proposes is of 3.1 to 1 in its favour: in Ukraine’s view, its delimitation

satisfies the proportionality test.

755
9.29. An accurate calculation of these ratios was provided in Romania’s Memorial,

and there is no need to revisit them here. It suffices to say that, if the relevant coasts and

the relevant area are properly defined, the coastal length ratio is only 1.4 to 1 in favour of

Ukraine. On this basis Romania’s proposed delimitation (which leads to an area ratio of

754
UCM, para. 10.12.
755 RM, Chapter 12.

2981.7 to 1 in favour of Ukraine) satisfies the proportionality test. On the contrary, the

maritime boundary advocated by Ukraine, leading to a ratio of the maritime areas
allocated to each of the two Parties of 3.1:1 (according to the Ukrainian calculations

based on the erroneous establishment of the delimitation area) or 3.5:1 (if the calculations

are based on the correct definition of the delimitation area) in favour of Ukraine, clearly

fails to satisfy the “proportionality test” if the relevant coasts of the Parties are properly

defined (the ratio being, as shown above, only 1.4:1 in favour of Ukraine).

9.30. Against this background the warnings issued by this Court and Tribunals as to the

application of the proportionality test should be underlined. Several difficulties are

involved, for instance:

- There is often uncertainty as to the exact definition of the relevant coasts

and of the relevant area, which makes any calculation of the ratios inherently
problematic. In the present case, the calculations of the relevant coasts and

relevant area done by the two Parties diverge widely (mainly on account on the

differences in the identification of the areas of overlapping entitlements).

- This difficulty is compounded by the existence of third parties. As the

maritime boundary between Romania and Bulgaria is not established, it is

difficult to calculate the precise area of maritime spaces that will finally come

under the jurisdiction of Romania.

9.31. The present case illustrates the difficulties that the use of proportionality can

entail and the need for caution.

(2) The Principle of Non-encroachment

9.32. As to the issue of cut-off, Ukraine asserts that the Romanian delimitation line

results in a “two-fold cut-off of Ukraine’s maritime entitlements”. 756 For one thing, “the

maritime entitlements of Serpents’ Island are dramatically truncated by allotting to that

756 UCM, para. 10.24.

299island no continental shelf and no exclusive economic zone”. 757 Secondly, Ukraine’s

south-facing mainland coast is “deprived of the area to which it is entitled”. 758

9.33. As to the first argument, the maritime entitlements of Serpents’ Island were

confined to its 12 nm arc of circle surrounding it as a result of the establishment of the

maritime area in its region, by the bilateral Soviet/Romanian agreements, concluded ever

since 1949. Moreover, even disregarding (arguendo) the provisions of the 1949 and the

subsequent Romanian/Soviet ProcŁs-Verbaux, Serpents’ Island, as a rock unable to

sustain human habitation and economic life of its own, is not entitled to continental shelf

and exclusive economic zone. But in any event a tiny maritime feature – “rock” or not –

is not considered in international practice as entitled to substantial effect in delimitation,

as shown by the review of practice appended to Chapter 6 above. In these circumstances

Ukraine’s first concern about encroachment must be rejected.

9.34. As to its second concern, that its south-facing mainland coast is “deprived of the
759
area to which is entitled”, the following comments are in order. First, the projection of

“Ukraine’s south-facing mainland coast” (properly speaking, the entire coast of Ukraine

comprised between the Nistru/Dniester Firth and Cape Tarkhankut) does not actually

abut on the area where the claims of the two Parties overlap. This projection is “cut” by

the projection of the Ukrainian coasts that are adjacent, and respectively opposite, to the

Romanian coast. In any case, the water expanses lying off this Ukrainian stretch of coast
760
are attributed to Ukraine by the Romanian delimitation proposal.

9.35. It is actually the delimitation advocated by Ukraine that leads to a cut-off of

Romania’s maritime entitlements, in particular of the northern sector of its coast between

Sulina and Sacalin, which is left with a merely residual area of maritime space.

Ukraine’s claimed line results in a dramatic curtailment of the maritime areas off the

757 Ibid.
758 Ibid.
759
760 Ibid.
See, e.g., RM Figure 34.

300Romanian coast, as if the projections of every stretch of Ukraine’s coast run unobstructed
in every direction while there is no opposing or adjacent Romanian territory.

9.36. The fundamental inequity of Ukraine’s line is obvious when comparing the areas

left to the Romanian northern sector with the spaces attributed to the opposite stretch of

the coast of Crimea. Thus, while the maritime areas facing the Romanian coast of the

Danube Delta are left by the Ukrainian claimed maritime boundary to a breadth varying

between 12 and 18 nm, the opposite Ukrainian coast of Crimea enjoys maritime areas

with a breadth of 150 nm – that is, approximately 10 times larger. This situation is
illustrated in Figure RR32 at page 302 of this Reply.

301 2
3

s
i n
n e
i l
t t
i e
l n
3 d n
R e a
e o o
u o R
i p h
F n n
n o
a e
k a
U or
h n
T e (3) The Impact on Security Interests of the Parties

9.37. The limited role that security interests play in maritime delimitation has already

been described. 761 In its Counter-Memorial, Ukraine comments that “if security interests

do have a role to play, it is clear that Ukraine’s legitimate interests around Serpents’

Island must be respected as well as any purported security interests of Romania”. 762

9.38. The line proposed by Romania does respect Ukraine’s security interests,

recognizing to Serpents’ Island a belt of maritime spaces of 12 nm – the most it can claim

under international law. It should be recalled that the security interests of Romania were

disregarded when establishing the Soviet-Romanian maritime boundary in this sector by

allotting Serpents’ Island more water areas than the opposing Romanian mainland.
763
Ukraine’s undertaking not to place offensive weapons on Serpents’ Island by no means

compensates for this situation.

9.39. Ukraine seems to interpret the term “security” in an unjustifiably narrow sense as

referring only to defence matters. This is in contradiction with the jurisprudence of this

Court, 764which understands security as relating equally to economic factors and access to

resources. In this respect, a delimitation line that runs unreasonably close to the coast (as

the line advocated by Ukraine does) cannot be considered as respecting the security

interests of the coastal State.

9.40. Ukraine also alleges that “Romania’s claim line cuts right through area which

have been traditionally exploited by Ukraine for both hydrocarbon exploration and
765
fishing”. Leaving aside the curious notion of “exploitation for exploration”, this

assertion is contradicted even by the relevant sections of Ukraine’s Counter-Memorial,

761 See RM, paras. 12.14-12.19.
762
763 UCM, para. 10.26.
Art. 3 of the 1997 Additional Agreement provides that the parties shall conclude a bilateral
agreement on confidence and security building measures (CSBMs), by which Ukraine shall commit “not to
locate offensive military devices” on Serpents’ Island – see Annex RM 2. The CSBMs agreement has not
been finalised yet.
764 RM, para. 12.16.
765 UCM, para. 10.27.

303analysed in Chapter 7 above. The sporadic attempts of Ukraine at exploration of

marginal parts of the area of overlapping claims cannot be qualified as “traditional” or

“historical”. But even if they could be, the specific characteristics of this case made the

oil and gas practices of the two Parties devoid of any relevance as concerns the
766
delimitation of their maritime areas.

9.41. These considerations apply equally to fishing, with the further qualification that

Ukraine has not actually offered any evidence of fishing activities in the relevant area: it

has alleged only that, on a few occasions, it arrested vessels of third parties engaged in
fishing.767 This is by no means sufficient – even if the facts were as alleged – to affect

the delimitation.

D. CONCLUSIONS

9.42. In conclusion, Romania’s proposed maritime boundary conforms with the

requirements of equity. It complies with the provisions of the bilateral Romanian/Soviet

ProcŁs-Verbaux. It fully satisfies the demands of equitableness, including the

“proportionality test”. It does not encroach on Ukraine’s maritime entitlements and it

does not affect its security interests.

9.43. In contrast, Ukraine’s delimitation leads to manifestly inequitable consequences,

leaving to Ukraine areas of maritime spaces that are clearly out of proportion if compared

to the length of its relevant coast, encroaching on Romania’s maritime entitlements and

failing to take into account Romania’s legitimate security interests. Moreover, Ukraine’s

proposed delimitation line is contrary to the provisions of the bilateral Romanian/Soviet

ProcŁs-Verbaux concluded since 1949.

766
See above, paras. 7.10-7.31.
767 See above, paras. 7.40-7.59.

304 CHAPTER 10

SUMMARY

1. The Court’s Jurisdiction. In accordance with the will of the Parties, expressed

in the 1997 Additional Agreement, the Court has jurisdiction in the present

case to decide upon a final delimitation line between the continental shelf and

the exclusive economic zones of Romania and the continental shelf and

exclusive economic zone of Ukraine. This jurisdiction extends, in the sector
of the agreed 12 nm arc around Serpents’ Island, to determining the maritime

boundary between the continental shelf and the exclusive economic zones of

Romania and the territorial sea of Ukraine around the Serpents’ Island.

Ukraine reaches the contrary conclusion on this latter point only by

misconstruing a dictum of the Court of Arbitration in the Anglo-French

Continental Shelf case. Ukraine’s contention is contradicted by arbitral

decisions in such cases as Dubai/Sharjah, Eritrea/Yemen and St. Pierre et
Miquelon. But even if Ukraine is correct on this issue, the Court has

jurisdiction to determine the point on the 12 nm arc around Serpents’ Island

from which the single maritime boundary shall proceed: it is not in dispute

that there is such a point and that the Court can determine its location.

2. Existing Agreements relating to the Delimitation. Under international law, in

particular the principle of the paramountcy of agreement in maritime

delimitation (reflected in Articles 74(4) and 83(4) of the 1982 UNCLOS), the

delimitation of the maritime spaces of Romania and Ukraine is to be decided

in accordance with the provisions of relevant agreements in force between the

two countries relating to delimitation. Several Romanian-Soviet agreements,
especially the ProcŁs-Verbaux concluded in 1949, 1963 and 1974, govern an

initial sector of the delimitation line. The principles regarding the delimitation

enshrined in the 1997 Additional Agreement fall to be applied by the Court in

respect to the determination of the remainder of the delimitation line. The

wording of the Additional Agreement demonstrates that these principles were

305 intended to be applied not only during the bilateral talks, but also in the event
of the adjudication of the dispute by the Court.

3. The Approach to the Delimitation. The appropriate approach in delimiting the

maritime areas of Romania and Ukraine in the Black Sea is first to draw a

provisional equidistance/median line between the relevant coasts of the
Parties, and subsequently to adjust it, if necessary, to take into account (a)

relevant agreements and (b) the relevant/special circumstances of the

delimitation area, in order to reach an equitable result. The first stage is to

asses the geographical framework of the delimitation, in order to identify the

coasts of the Parties relevant to the delimitation, as well as the area relevant to

delimitation. This method is in accordance with the provisions of the 1997

Additional Agreement.

4. The Need for a Two-Sector Approach. As recognized in the 1997 Additional

Agreement, the geography of the area where the delimitation is to be effected

imposes a two-sector approach, distinguishing between a sector where the

coasts are adjacent and a sector where the coasts are opposite. The Romanian

coastal segment situated between the last point of the land/river border

between Romania and Ukraine and the outer limit of the Sacalin Peninsula
stands in a relationship of adjacency with the segment of the Ukrainian coast

situated to the north of the boundary with Romania. This relation of

adjacency ceases when the Ukrainian coasts changes its direction at the

Nistru/Dniester Firth (Point “S”). At the same time, the entire length of the

Romanian coast faces across the western basin of the Black Sea a separate

segment of the Ukrainian coast – the western coast of the Crimea Peninsula –

with which it is in a relationship of oppositeness.

5. The Relevant Coasts. The identification of the relevant coasts of the Parties,

in accordance with the rules and principles applied by the Court and by

arbitral tribunals, leads to the conclusion that while the entire Romanian coast

is relevant to the delimitation, the coast of Ukraine between the

Nistru/Dniester Firth and Cape Tarkhankut bears no relationship of either

306 adjacency or oppositeness with any segment of the Romanian coast and does
not give rise to any areas of overlapping entitlement; it accordingly has to be

disregarded for purposes of establishing the boundary between the maritime

spaces of Romania and Ukraine. Ukraine seeks to avoid this conclusion by

constructing projections that do not reflect the actual direction of its coast and

involve Ukrainian coasts which face each other and are neither adjacent nor

opposite to any Romanian coast.

6. Romania’s Basepoints. The use of the Sulina Dyke and the Sacalin Peninsula

as basepoints on the Romanian coast is justified under international law.

7. The Relevant Area. Ukraine’s definition of the relevant area is flawed
because it purports to include spaces where the entitlements of the two

countries do not overlap (the area north of the line uniting Point S and Cape

Tarkhankut) and to exclude areas where they do overlap (the roughly

triangular area, defined by Cape Sarych, a point described by Ukraine as “a

point between the Romanian and the Ukrainian coasts where the interest of

third States potentially come into play” and the point of intersection between a

perpendicular line drawn on Cape Sarych and the boundary delimiting the

Ukrainian and Turkish maritime spaces). Furthermore, Ukraine definition of
the relevant area is prejudicial to the interests of third parties (notably

Bulgaria), because of its inclusion of an area situated south of the equidistance

line between Romania and Bulgaria.

8. The Character of Serpents’ Island. A particularity of the geographical setting

is the presence of Serpents’ Island, a very small rocky formation, devoid of
water resources other than rainfall and practically devoid of soil, vegetation

and fauna. Due to its natural characteristics, Serpents’ Island is a “rock”

unable to sustain human habitation or an economic life of its own, and

consequently falls under the provisions of Article 121(3) of the 1982

UNCLOS, not being thus entitled to continental shelf and exclusive economic

zone.

3079. Ukraine’s Attempts to Change the Character of Serpents’ Island. Ukraine’s
recent attempts at artificially creating the appearance of habitation and of

economic life on Serpents’ Island are devoid of legal effects vis-à-vis

Romania. Actually, the only effect of this conduct is its implicit recognition

of the character of Serpents’ Island as a “rock” having the characteristics

referred to in Article 121(3).

10. The Agreed Boundary around Serpents’ Island. In the first, northern sector of

the delimitation area an initial segment of the boundary was already

established by Romania and the USSR, through agreements which are binding

on Ukraine. These agreements established an all-purpose boundary between

the maritime spaces of the former USSR and Romania following an arc

surrounding Serpents’ Island up to a point to the east. The existence of a

maritime boundary on the 12 nm arc around Serpents’ Island was consistently

represented in all the demarcation documents concluded between the Parties.
Ukraine’s argument that the intention was to delimit only the territorial seas,

and then further seawards to establish the outer limit of the Soviet territorial

sea vis-à-vis the high seas is plainly contradicted by the language used, by the

graphical representation of the boundary, as well as by the new evidence

included in this Reply (Plates I and V included in volume 1 of the 1949

Catalogue of Maps). There is nothing to suggest that a change of regime was

envisaged at any point along the boundary described by the Romanian-Soviet

documents.

11. Evidence of the Agreed Boundary in Maps and Charts. Further, the boundary

established by these agreements is depicted on various charts emanating from

the Soviet Union, Ukraine, Romania or third parties. These charts constitute

“corroborative evidence” in the meaning of this Court’s jurisprudence. They

consistently employ symbols normally used to represent international
maritime boundaries, in accordance with the guidelines of the International

Hydrographical Organization. These charts reflect the common understanding

of the Romanian and Soviet authorities as well as the “informed opinion” of

308 third parties as to the course and extent of the maritime boundary around
Serpents’ Island, up to Point X to the east of it.

12. The Maritime Boundary in the First Sector. In conformity with the

jurisprudence of this Court and of arbitral tribunals, the drawing of the

provisional equidistant line is to be effected without taking into account minor
maritime formations that would unreasonably deflect the trace of the line – in

this case, Serpents’ Island. In any case, Serpents’ Island was excluded as a

basepoint by Ukraine itself through international conduct opposable to it.

Consequently, in the first sector, the provisional line follows the equidistance

line between the two coasts, taking into account the existing 12 nm boundary

around Serpents’ Island.

13. The Maritime Boundary in the Second Sector. In the second sector, the

provisional line follows the median line between the opposing Romanian and

Ukrainian coasts.

14. Relevant Circumstances which might Modify the Provisional Equidistance

Line. Only the factors which “might modify the result produced by an

unqualified application of the equidistance principle” – to use the language of

the Court in the Jan Mayen case – may be analyzed as possible relevant

circumstances. The alleged “geographical predominance of Ukraine in the

area” and “disparity between coastal lengths”, as well as the “State activities”

invoked by Ukraine do not fall within the ambit of this notion. In any event

any disparity in coastal lengths is not such as to deflect the provisional

mainland-coasts equidistance line.

(a) Serpents’ Island. One special circumstance is the presence of Serpents

Island. Irrespective of its qualification as a “rock” and independently of

any agreements binding on the Parties, Serpents Island should be given

nil effect in the delimitation beyond 12 nm, taking into account its
natural characteristics, its miniscule size, the fact that it cannot sustain

human habitation or an economic life of its own and its distorting effect

on a delimitation line. State practice and the jurisprudence of this Court

309 and of arbitral tribunals show that no or very limited effect is given to
such islands.

(b) The Enclosed Character of the Black Sea in conjunction with the

Geographical Configuration of the Delimitation Area and the Existing

Delimitation Agreements. These factors combine leading to a situation
in which any dramatic alteration of the method of delimitation used in

the Black Sea existing agreements could bring about inequitable results

for the Parties.

(c) Hydrocarbon and Fishery Activities. The State activities invoked by

Ukraine do not justify an adjustment of the provisional line. In
particular, the oil concessions granted by the two countries do not reveal

a consistent pattern developed over a significant period a time, much

less the existence of a de facto line accepted by both parties. Further,

taking into account such activities would not be in accordance with a

good faith interpretation of the relevant provision of the 1997

Additional Agreement, which forbids the unilateral exploitation of the

mineral resources in the disputed area. Equally, the few incidents in

which the Ukrainian border police intercepted fishing boats of third
parties on the margins of the area of overlapping claims are devoid of

legal significance for the delimitation. In accordance with the

jurisprudence of this Court, State activities, either relating to mineral

resources, fisheries or anything else, can be taken into account only if

they are based on or necessarily imply an agreement between the

parties, which is not remotely the case here.

(d) Coastal Lengths of the Parties. There is no disproportion of the length

of the relevant coasts of the Parties, so there is no rationale to adjust the

provisional line in accordance with this criterion.

15. Conclusion on Relevant Circumstances. Accordingly, there are no

circumstances justifying the adjustment of the provisional

equidistance/median line drawn from the mainland coasts of the two Parties,

310 apart from the existing agreed boundary surrounding Serpents’ Island and its
characteristics which would accord it nil effect beyond 12 nm in any event.

16. Application of the Proportionality Test. The delimitation line put forward by

Romania is equitable and fully satisfies the proportionality test. Ukraine

reaches a different conclusion only by misidentifying the relevant coasts of the
Parties and the relevant area to be delimited. Further, Romania’ delimitation

leads to no cut-off effect, leaving to Ukraine the expanses of maritime spaces

off its coasts, while the line claimed by Ukraine dramatically truncates the

continental shelf and exclusive economic zone of Romania, especially in

respect of the northern sector of the Romanian coast which is left with

virtually no other space than the territorial sea. Romania’s delimitation line

reflects due consideration for Ukraine’s security interests, as well as for

Serpents’ Island, which has a belt of maritime spaces of 12 nm. By contrast

Ukraine’s delimitation is unjustifiably close to the Romanian coast and thus
encroaches on the security interests of Romania.

311 SUBMISSION

For the reasons set out in the Memorial, as well as in this Reply, Romania respectfully

requests the Court to draw a single maritime boundary dividing the maritime areas of

Romania and Ukraine in the Black Sea, having the following description:

a. from Point F, at 421'N, 30°027'E, on the 12 nm arc surrounding Serpents’
Island, to Point X, a20N, 30°212'E,

b. from Point X in a straight segment to Poin59N, 30°416'E,

c. then on the line equidistant between the Romanian and Ukrainian adjacent coasts,

from Point Y to Point T, 45'N, 31°0840E,

d. and then on the line median between the Romanian and Ukrainian opposite coasts,

from Point T to Point Z, 50'N, 31°210E.

This line is depicted, for illustrative purposes, on Figure RR31 on page 290 of this

Reply, reproducing Figure 34 on page 232 of Romania’s Memorial.

Dr. Bogdan AURESCU,

Agent of Romania

Bucharest

22 December 2006

312 Appendix

The Diplomatic History relative to the Dispute

Introduction

A.1. In its Memorial, Romania presented the historical background to the dispute
768
before the Court. In particular, it was shown that the circumstances that underlay the
establishment of the delimitation line between the maritime spaces of Romania and

Ukraine (such as the appurtenance of Serpents’ Island to the USSR and the delimitation

of the final segment of the border on the River Danube between Romania and Ukraine)

resulted from transactions not in compliance with international law. Romania does not

seek to reverse these transactions. But it has equally stressed that the inequitable effects

of these transactions, unlawful ab initio as they were, should not be magnified and

extended when establishing the maritime boundary between the continental shelf and the

exclusive economic zones of the two countries in the Black Sea.

A.2. In its Counter-Memorial Ukraine advances its own version of the history, based

largely on Soviet or Ukrainian sources. Moreover it covers a much broader range of

issues, going well beyond those which are relevant to this case, including such question

as sovereignty over Bessarabia and Northern Bukovina, the circumstances of the

reunification of these provinces with Romania at the end of World War I, and Romania’s

participation to World War II. Although these issues are completely irrelevant for

present purposes, a number of misstatements in the Ukrainian presentation require at least

a brief response.

A.3. An initial point is that nowhere in the Counter-Memorial is there any refutation of

the contention actually made by Romania, i.e. that the incorporation of Serpents’ Island

and of several islets at the mouth of the Danube as Soviet territory departed from the

provisions of the Paris Peace Treaty of 1947, was inconsistent with international law and

768 See RM, Chapter 3.

313directly influenced the establishment of the maritime boundary between the two

countries. Chapter 5 of Ukraine’s Counter-Memorial (“The Diplomatic History”) says

almost nothing about the circumstances in which these territories came to belong to the

Soviet Union. Rather it makes irrelevant comments about such matters as an alleged

769
“Greater Romania” plan envisaged by the Romanian authorities.

The earlier political history

A.4. In its version of history, after recording several developments concerning
th th
Bessarabia and Bukovina in the 18 and 19 centuries and noting the recognition of

Romanian sovereignty over Serpents’ Island under the terms of the Treaty of Berlin of

1878, Ukraine abruptly passes to the assertion that Romania “illegally seized the ethnic
770
Ukrainian territories of Southern Bessarabia and Northern Bukovyna in 1918”.

A.5. In fact Bessarabia and Bukovina had been, until 1812 and 1775 respectively,

integral parts of the Romanian Principality of Moldova. Prior to the events of 1940,

neither Southern Bessarabia nor Northern Bukovina were at any stage part of Ukraine.

But in 1940, following an ultimatum of the Soviet Union, they were incorporated into the

USSR and become part of the Ukrainian Soviet Socialist Republic. 771

A.6. Bukovina was detached from Moldova and became part of the Austrian Empire in

1775. At that time its population was overwhelmingly Romanian, but an influx of

Ukrainian (Ruthenian) immigrants altered the composition of the province. In 1774 there

were 64,000 Romanians, 8,000 Ruthenians and 3,000 other nationalities; in 1810 the

Romanians still constituted 75% of the population (150,000 Romanians and 48,000 other
772
nationalities); in 1848, that percentage has fallen to 55%.

769 UCM, para. 5.29.
770 UCM, para. 5.14.
771 See RM, paras. 3.20-3.22.
772 See Keith Hitchins, Românii 1774- 1866 (Romanians, 1774-1886), Humanitas, Bucharest, 1998, p.

280.

314A.7. Bessarabia was ceded to the Russian Empire by the Ottoman Empire in 1812. It

was organised as an autonomous province until 1828, when it was united with the

773
General Government of Novorossiissk.

A.8. To characterise Southern Bessarabia as an “ethnic Ukrainian territory” is to distort

the facts. As the British historian Dennis Deletant recorded,

“In the two southern Bessarabian counties of Cetatea Alba and Ismail (the
greater part of which were added to the Ukrainian SSR) the Ukrainians,
however, represented only 14% compared with 24% for Romanians, 22%

for Russians and 20% for Bulgarians. If the figures for the other two
southern Bessarabian counties of Tighina and Cahul are entered into
calculation (and parts of them were incorporated into the Ukrainian SSR),

the Ukrainian percentage drops to a mere 8% compared with the figure for
Romanians of 37%, for Russians of 17%, for Bulgarians of 15 % for
774
Gagauz of 9% and for Jews of 3%.”

A.9. Equally, the Ukrainian contention that “at the beginning of 1918, in violation of

generally recognised rules of international law relating to State succession, and ignoring

the right of the Ukrainian population in Southern part of Bessarabia to self-determination,
775
Romania occupied the entire territory of Bessarabia” is far from the truth.

A.10. An account of the events leading to the reunification of this province with

Romania is provided by the French historian Catherine Durand 776, who records that the

Romanian ethnic majority of the population clearly expressed its free will in favour of the

unification of Bessarabia with Romania.

A.11. The reunification of Bukovina with Romania was accomplished in similar

circumstances, as shown in the account of these events given by the American historian
777
Keith Hitchins.

773
774 See Keith Hitchins, Rumania, 1866-1947, Clarendon Press, Oxford, 1994, p. 239.
D. Deletant, “The Ribbentrop-Molotov Pact and its consequences for Bessarabia”, in D Deletant &
M Pearton, Romania Observed. Studies in Contemporary Romanian History, Editura Enciclopedic▯,
Bucharest, 1998, p. 53.
775 UCM, para. 5.15.
776 See Catherine Durand, Histoire des roumains, Fayard, Paris, 1995, pp. 217-220.
777 Keith Hitchins, Rumania, 1866-1947, Clarendon Press, Oxford, 1994, pp. 279-280.

315A.12. In particular, there was no question of a Romanian military occupation of the two

provinces, as contended by Ukraine. As shown by the accounts of Durand and Hitchins,

at the request of the representative bodies that had assumed power in these territories

after the breakdown of central authority in the Russian and Austro-Hungarian Empires,

Romanian troops were sent to assure the protection of the population against marauding

groups of soldiers from the disintegrating Russian and Austro-Hungarian imperial armies.

The reasons for the sending of Romanian troops in Bessarabia were fully explained to the

Ukrainian Government in a Romanian Note Verbale of 19 June 1918. 778 It should be

noted that at that time conditions of anarchy prevailed in Ukraine, due to revolution and

779
then the civil war.

A.13. The reunification of Bukovina and Bessarabia with Romania were endorsed by

the Paris Peace Conference in 1919. Specifically the reunification of Bukovina with

Romania was recognised by article 59 of the Treaty of Peace between the Allied and

Associated Powers and Austria, signed in St. Germain-en-Laye, on 10 September 1919,

which entered into force on 16 July 1920. Bessarabia’s reunification with Romania was

recognized by the Treaty between the Principal Allied Powers and Roumania respecting

Bessarabia, signed at Paris, on 28 October 1920, the Preamble of which affirmed that

“from geographic, ethnographic, historic and economic points of view, the reunion of

Bessarabia to Roumania [wa]s fully justified” and that “the population of Bessarabia has

given proof of its desire to see Bessarabia reunited to Roumania”.

A.14. Ukraine’s further remarks to the effect that “Romania’s… obstinate refusal to

negotiate and reach a fair and equitable territorial settlement with Ukraine and Russia,

and later to Soviet Union, resulted in further tensions and stalemate in relations” 780are

both partisan and irrelevant. The history of the Romanian and Soviet diplomatic contacts

778 Note Verbale of 19 June 1918 of the Romanian Ministry of Foreign Affairs addressed to the
Government of Ukraine (Annex RR 42).
779 See Olivier de Laroussilhe, L’Ukraine, Presses Universitaires de France, Paris, 1998, p.53: “Neuf

780vernement diffØrents tenteront en vain de s’assurer la maîtrise de l’Ukraine qui glisse dans l’anarchie.”
UCM, para. 5.22.

316in the interwar period suggests that the issue of Bessarabia was intractable: to lay the

blame for the failure of these negotiations only on Romania, as Ukraine does, is
781
unjustified.

A.15. Ukraine asserts that “geopolitical changes” having taken place in Europe “in the

late 1930s and the early 1940s permitted the Soviet Union to restore Ukraine’ legitimate

rights to Northern Bukovyna and Southern Bessarabia.” 782 By “geopolitical changes”,

Ukraine appears to mean the conclusion, on 23 August 1939, of the Treaty of Non-

Aggression between Germany and the Union of Soviet Socialist Republics (the Molotov-

Ribbentrop Pact) and its effects. That Pact was supplemented by a secret Protocol, Point

3 of which provided a direct reference to Bessarabia:

“3. With regard to Southeastern Europe attention is called by the
Soviet side to its interest in Bessarabia. The German side declares its
783
disinterest in this area.”

A.16. Ukraine contends that the additional Protocol to the Molotov-Ribbentrop Pact was

“legally null and void from the moment of its signing due to the arbitral [sic] provisions

concerning Estonia, Latvia, Lithuania, Finland and Poland” 784 but implies that the

provisions concerning Bessarabia were actually legitimate. The only reason offered for

this distinction is that “the interest of the USSR regarding Bessarabia did not arise on the

785
basis of the Protocol; it has existed before the Protocol was concluded.” Ukraine’s

distinction is groundless: if one admits that the provisions concerning Estonia, Latvia,

781
Keith Hitchins, Rumania, 1866-1947, Clarendon Press, Oxford, 1994, pp. 434-435: “The two sides
negotiated intermittently until 1924, when discussions in Vienna between Rumanian and Soviet delegations
broke down. Neither side showed any inclination to resume the dialogue until 1929, when Rumania
adhered to the Moscow Protocol, an instrument proposed by the Soviet Union to bring the Kellogg- Briand
Pact into force sooner than stipulated… The resumption of direct contacts in Moscow did not lead

immediately to new negotiations, as both sides stuck to their original positions on Bessarabia. Maniu, the
Rumanian Prime Minister, made normal relations dependant upon the Soviet Union’s recognition of the
Dniester River as the boundary between the two countries, while Maxim Litvinov, the Soviet Foreign
Minister, just as firmly rejected such a condition… Rumanian-Soviet talks resumed in September 1932 in
Geneva. Within a short time the draft of a treaty had been agreed to, but the Rumanian government

782ected to ambiguous language regarding ‘existing matters in dispute’ that is, Bessarabia.”
783 UCM, para. 5.23.
Quoted in Norman Davies, Europe. A History, Oxford University Press, 1996, p.997.
784 UCM, para. 5.26.
785 Ibid.

317Lithuania, Finland and Poland were arbitrary, one cannot contend that the provision

concerning Romanian territories was somehow legitimate.

A.17. Having practically obtained carte blanche to pursue its aggressive aims, the
786
Soviet Union immediately formulated its claims to Romanian territory. Indeed it went

even further than the terms of the Molotov-Ribbentrop Pact, claiming also Northern

Bukovina, and only the opposition of Germany prevented it from laying claim to the

entire province. The annexation of Bessarabia and Northern Bukovina by the Soviet

Union followed. 787

A.18. Ukraine gives short shrift to the circumstances of the annexation of the two

provinces, noting only that: “On the demand of the Soviet Union, Romania withdrew

from the occupied territories without any armed resistance in the summer of 1940.” 788 A

better description of the events is offered by the British historian Norman Davies:

“On 27 June 1940, ten days after the Baltic states, the Romanian provinces

of Bukovina and Bessarabia were grabbed789idst fanfares of their
“reunion with the Soviet fatherland.”

As to the consequences of the Soviet annexation for the inhabitants of Bessarabia, Davies

states:

“…the Moldavian peasantry has been recently robbed of both its land and
its culture. They were forced to live and work in collectives, whose

surplus was taken away by the Soviet State. Thousands upon thousands
have been driven to their death in Gulag, or shot as so-called saboteurs.

Their language had been arbitrarily transferred to the Cyrillic alphabet, so
that Soviet-educated children could not longer read pre-war Romanian or
Moldavian literature. They were denied all contact with the western half of

their province in Romania, which they were told was a foreign country.
They were beaten, beggared, and bullied.” 790

A.19. As Deletant says:

786
Norman Davies, Europe. A History, Oxford University Press, 1996, p. 997.
787 RM, paras. 3.20-3.22.
788 UCM, para. 5.23.
789 Norman Davies, Europe. A History, Oxford University Press, 1996, p. 1009.
790 Ibid, p.1011.

318 “The Soviet ultimatum of 26 June and its annexation of Romanian

territory under the threat of the use of force merit the description of an act
of international blackmail. The threat to use force was a threat to commit

acts outlawed by the two conventions on the definition of aggression,
signed on 3 and 4 of July 1933, to which both the Soviet Union and
Romania were signatories.” 791

A.20. Despite the assertions of Ukraine to the contrary, the annexation of the two

provinces by the Soviet Union cannot be justified by invoking the ethnic compositions of

their populations.

“The frontier imposed upon Romania by the Soviet Union was by no

means the ethnic line between Romanian and Ukrainians and the claim in
the ultimatum that Bessarabia was principally peopled by Ukrainians was
wildly inaccurate. Even the census taken in 1897 while the province was

under Russian rule could not be adduced to bring the slightest support to
the ultimatum’s contention.” 792

A.21. Not even the Soviet authorities believed in this “ethnic” argument. In spite of

arguing, in its Note of 26 June 1940, that Bessarabia was “populated mainly by

Ukrainians”, the vast majority of the Bessarabian territory annexed by the Soviet Union

was included in a newly-created Soviet Socialist Republic – the Moldavian Soviet

Socialist Republic (the predecessor of the present Republic of Moldova), a Romanian-
793
speaking State .

A.22. Ukraine asserts that “there was no legally established State border between

USSR and Romania: before 1940 their territories were divided by a provisional

791
D. Deletant, “The Ribbentrop-Molotov Pact and its consequences for Bessarabia”, in D. Deletant
& M Pearton, Romania Observed. Studies in Contemporary Romanian History, Editura Enciclopedica,
Bucharest, 1998, p. 51.
792 Ibid, p. 52.
793 The constitutional name of the official language of the Republic of Moldova is “Moldovan”; this
has to be interpreted together with the provisions of the Law on functioning of the spoken language on the

territory of the Republic of Moldova of 1 September 1989, representing the framework legislation on
languages currently in force, which recognizes in its Preamble the “existing linguistic Romanian-Moldovan
identity”; see also the Declaration of Independence of the Republic of Moldova, 27 August 1991 – Law
691/1991 published in the Official Bulletin of the Republic of Moldova no. 11/27.08.1991 (Annex RR 43),
which refers to earlier decisions of the Moldovan Parliament, among which “the declaration of the
Romanian language as State language”

319 794
demarcation line”. It also claims that “the Exchange of Notes concluded in 1940

between the USSR and Romania provided a legal basis for various territorial changes
795
affecting Ukrainian territory in areas previously occupied by Romania.” The

implication seems to be that the Soviet ultimatum replaced a provisional situation with a

definitive territorial settlement. Again, Ukraine tells only half of the story, overlooking

the fact that the bilateral Romanian-Soviet negotiations on the delimitation of the border

resulting from the Soviet ultimatum of 1940 ultimately broke down because of the

796
seizure by Soviet troops of several Danube islands. The Romanian Government made

very clear to the Soviet authorities that it did not recognize this de facto situation
797 798
imposed by force. Eventually, armed hostilities started between the two countries.

A.23. The Peace Treaty of 1947 established that “the Soviet Roumanian frontier is thus

fixed in accordance with the Soviet Roumanian Agreement [sic] of June 28, 1940…”

This provision referred in fact to the 1940 Soviet ultimatum which was not an agreement.

As this ultimatum contained no reference to Serpents’ Island or to the Danubian islets

that were seized by Soviet troops in autumn 1940 the said provision cannot be understood

as referring to these territories which were recognised by the Peace Treaty as appurtenant

to Romania.

A.24. Ukraine fails to mention that the “transfer” to the Soviet Union of these territories

lacked any basis in the 1947 Peace Treaty and that they were ceded to the Soviet Union

by means of the Protocol to Specify the Line of the State Boundary between the People’s

794
UCM, para. 5.22.
795 UCM, para. 5.31.
796 See RM, para. 3.24.
797 See Telegram no. 1491 of 30 April 1941 of the Romanian Legation in Moscow addressed to the

798anian Minister of Foreign Affairs (Annex RR 44).
The reasons for Romania’s entering the hostilities were stated by Grigore Gafencu, the Romanian
Minister to Moscow, in a meeting with Viacheslav Molotov, the People’s Commissar for Foreign Affairs:
“By its brusque ultimatum of last year, when it demanded not only Bessarabia, but also the Bukovina and a
corner of ancient Moldavia; by its subsequent violation of our territory, by the seizures made in the Lower

Danube at the very moment when negotiations were in progress to fix the new line of demarcation the
USSR destroyed all feelings of confidence and security in Romania and aroused the justified fear that the
very existence of the Romanian state was in danger. Romania therefore sought support from another
quarter.” See D. Deletant, “The Ribbentrop-Molotov Pact and its consequences for Bessarabia”, in D.
Deletant & M Pearton, Romania Observed. Studies in Contemporary Romanian History, Editura

320Republic of Romania and the Union of Soviet Socialist Republics of 4 February 1948 and

of a further ProcŁs-Verbal of 23 May 1948.

A.25. Ukraine’s assertion to the effect that “the facts are that the [Romanian-Soviet
799
boundary] line was part of the immediate post-war settlements” is inaccurate. The

Romanian-Soviet boundary line established in 1948-1949 was not “part of the immediate

post-war settlements”, but was subsequent to and in contradiction with the provisions of

the Peace Treaty of Paris. As Deletant notes:

“What was not consented to by Romania was the Soviet Union’s
annexation of the district of Her▯a in northern Moldavia, for it was not
mentioned in the text of the ultimatum, not her occupation of four islands
800
at the mouth of the Danube in autumn 1940.”

Romania was in no position to oppose the cession of Serpents’ Island to the Soviet

Union, regardless of the fact that this lacked any basis in the provisions of the Paris Peace

Treaty. The 1948 Protocol is indeed an unequal treaty.

A.26. No doubt “it was for the Powers negotiating the Paris Peace Treaty of 1947 to

decide what reparations Romania… should make after the war”, provided of course that
801
any such decision was incorporated in the Treaty of Peace. But the fact is that the

Powers decided that Serpents’ Island should belong to Romania. Ukraine notes that “a

permanent settlement of the USSR-Romania State border was agreed in the Paris Peace

Treaty of 1947”, 802but conceals the fact that the incorporation of Serpents’ Island in the

Soviet territory in 1948 was in clear departure from the settlement established by the

Peace Treaty.

Enciclopedica, Bucharest, 1998, pp. 45-46. Also, the references by Ukraine (UCM, para. 5.28) to alleged

799 crimes committed by Romanian troops are groundless and irrelevant.
800 UCM, para. 5.45.
D. Deletant, “The Ribbentrop-Molotov Pact and its consequences for Bessarabia”, in D. Deletant
& M. Pearton, Romania Observed. Studies in Contemporary Romanian History, Editura Enciclopedica,
Bucharest, 1998, pp.51-52.
801 UCM, para. 5.30.
802 UCM, para. 5.31.

321Romania’s alleged “denunciation” of the 1948 Protocol and the 1961 Border Regime

Treaty

A.27. In its Counter-Memorial, Ukraine refers to Romania’s “denunciation” of the 1948

Protocol to Specify the Line of the State boundary between Romania and the USSR and

of the 1961 Treaty on the Romanian/Soviet State Border Regime. 803 Although Ukraine

does not seem to draw from this episode any conclusions relevant to the present

proceedings, some brief response is called for.

A.28. The new political circumstances of the 1980s and 1990s allowed for formerly

“delicate” issues to be openly approached. One of these issues was the German-Soviet

Non-Aggression Treaty of 23 August 1939, together with its secret Additional Protocol,

commonly known as the “Molotov-Ribbentrop Pact”. Although known especially for its

provisions regarding Poland or the Baltic States, the Pact referred to Romania as well: the

USSR expressed its interest regarding Bessarabia while Germany expressed its disinterest
in this respect. The Pact was a basis for actions of the Soviet Union leading to the

annexation of certain Romanian territories. 804To condemn the Molotov-Ribbentrop Pact

as arbitrary or to declare it null and void did not entail questioning the provisions of the

Peace Treaties concluded after Word War II or the principles of the Helsinki Declaration

of 1975 regarding the inviolability of borders. The USSR itself did so: a Decision of the

Congress of the People’s Deputies regarding the legal and political aspects of the 1939

Russo-German Non-Aggression Pact was adopted on 24 December 1989. Other European

States, including the Baltic States, did the same.

A.29. Romania’s position as to the unlawfulness of the Molotov-Ribbentrop Pact was

clearly expressed in its bilateral relations with the USSR and, following the latter’s

dissolution, with its successor States. This did not mean that Romania denied its

obligations as a State party to other international agreements (in particular the 1947 Paris

Peace Treaty) or that it disregarded accepted international law principles regarding the

stability of borders. In fact Romania was the first State in the world to recognise the

803
UCM, paras. 5.39, 5.93.

322independence of the Republic of Moldova, from the date of its proclamation (27 August

1991). 805 Even the Declaration of Independence of the Republic of Moldova, while

proclaiming a new State on territories included in the USSR as a result of the Ribbentrop-

806
Molotov Pact, condemned that Pact in clear terms, without any sense of contradiction.

A.30. The political debates in Romania concerned not only the Molotov-Ribbentrop

Pact, but also the two other documents by which, in 1948, the USSR detached additional

territories from Romania (e.g., Her▯a – a Moldovan county that had never been under

Russian rule and had not been mentioned in the 1940 Soviet ultimatum, the five

Danubian islets, as well as Serpents’ Island). These were the Protocol to Specify the Line

of the State boundary between Romania and the USSR, concluded on 4 February 1948, 807

and the ProcŁs-Verbal of Delivery-Reception of Serpents’ Island, signed on 23 May

808
1948. Contrary to Ukraine’s assertion that these Protocols “did not exceed the scope of

the 1947 Peace Treaty, but simply clarified the border as imprecisely determined by that
809
Treaty”, they clearly departed from the provisions of the Treaty, in particular Article 1

fixing Romania’s boundaries, which were marked on the attached map in green, which

was neither unclear nor imprecise (see UCM, Figure 5-3).

A.31. The two documents concluded in 1948 departed from the provisions of the 1947

Paris Peace Treaty, were signed by Romania under duress and were never ratified or

confirmed by the Romanian Parliament. 810

804 See RM, paras. 3.20-3.23.
805
See the Declaration of the Government of Romania on the occasion of the proclamation of the
independence of the Republic of Moldova, 27 August 1991 (Annex RR 45): “The proclamation of a
Romanian independent state on territories forcibly annexed following the secret understandings of the
Molotov-Ribbentrop Pact represents a decisive step towards the peaceful liquidation of its ruthless
consequences, that were targeted against the rights and interests of the Romanian people. Recognising the

independence of the Republic of Moldova, the Government of Romania expresses its readiness to establish
diplomatic relations, to allow all necessary support to the authorities of the Republic of Moldova in order to
consolidate its independence and to act for the development of the brotherly cooperation between the
Romanians situated on the two banks of the Prut.”
806
See the Declaration of Independence of the Republic of Moldova, 27 August 1991– Law 691/1991
published in the Official Bulletin of the Republic of Moldova no. 11/27.08.1991 (Annex RR 43).
807 Annex RM 11.
808 Annex RM 12.
809
810 See UCM, para. 4.39.
For further details see Jean-Baptiste Duroselle, Histoire diplomatique de 1919 à nos jours,
Armand Collin, Paris, 12 ed., 2001, pp. 27-28; Catherine Durandin, Histoire des Roumains, Fayard, Paris,

323A.32. It is in this context that Romania made known to Ukraine its position that these

documents were to be considered as null and void. Note Verbale H(01)/2805 of 28 July

1995 of the Ministry of Foreign Affairs of Romania to the Embassy of Ukraine in
811
Bucharest was to this effect.

A.33. As far as the 1961 Romanian/Soviet Border Regime Treaty is concerned, 812

Romania informed Ukraine in 1993 that it considered it as “terminated” as a consequence

of the disappearance of the USSR. This clearly conveyed the intention of the Romanian

Party not to prolong the Treaty, Article 59 of which provided for its termination should a

Contracting Party express its wish to do so at least six months prior to the expiry of its

current term. Consequently, the 1961 Treaty terminated, if not in 1991 following the

disappearance of the USSR, at least on 27 June 1996 following the Romanian notification

and in accordance with Article 59. 813

A.34. The termination of the 1961 Border Regime Treaty, as well as the

acknowledgement of the invalidity of the 1948 Protocol, had no consequences on the

Romanian/Ukrainian border. It is a basic rule of international law that boundaries do not
814
disappear as a consequence of the termination of the treaty establishing them. Indeed

the Romanian/Soviet border was established by a succession of treaties. The termination

of the 1961 Treaty affected only the technical provisions regulating the border regime,

which were considered by Romania as outdated and not in conformity with the state of

bilateral relations between Romania and Ukraine.

1995, pp. 356-359; Georges Castellan, Histoire du peuple roumain, Armeline, Crozon, 2002, pp. 233-234,
Keith Hitchins, Rumania 1866-1947, Clarendon Press Oxford, 1994, pp.514-534, Dennis Deletant,
Romania under Communist Rule, Center for Romania Studies, Ia▯i-Oxford-Portland, 1999. In UCM, para.
5.29, Ukraine tries to assert that the occupation of Romania by the USSR and its loss of territory were
justified. In fact, the occupation by the Red Army of the Romanian territory was even more unjust as after

23 August 1944, Romania joined the Allied Powers against Germany and contributed decisively to
811rtening the World War II with more than 6 months.
See UCM, para. 5.39.
812 See UCM, para. 5.93.
813 Article 59 of the Border Regime Treaty provided: “This Treaty shall remain in force for a period

of ten years from the day of its entrance into force. If none of the Contracting Parties notifies with at least 6
months before the expiry of the validity of this Treaty about its intention to denounce the Treaty or to
modify it, the validity of the Treaty shall be automatically prolonged for a new period of five years.” The
Treaty entered in force on 27 June 1961.
814 See e.g., Territorial Dispute (Libyan Arab Jamahiriya/Chad), ICJ Reports 1994, p. 37 (para. 72).

324A.35. Romania acknowledged the legal situation on several occasions. Thus, the 1997

Treaty on Relations and Additional Agreement, as well as the 2003 Border Regime

Treaty, explicitly recognised that the Romanian/Ukrainian State border follows the trace

described by the 1961 Border Regime Treaty and the corresponding border documents.

The provisions of the 1961 Border Regime Treaty as to the description of the State border

never ceased to have legal effects, as acknowledged by all relevant Romanian/Ukrainian

agreements.

A.36. The termination of the 1961 Border Regime Treaty, as well as the

acknowledgement of the 1948 Protocol as null and void, did not affect in any way the

legal status of the 1949, 1963 and 1974 ProcŁs-Verbaux. These agreements, providing a

detailed description of the State border between Romania and the Soviet Union, were

duly approved by the Romanian authorities in accordance with the relevant legal

provisions then in force and could not have been unilaterally terminated. Neither

Romania nor Ukraine attempted to do so. Their validity was expressly recognised by

Romania and Ukraine in 1997, by the Treaty on Relations and the Additional Agreement,

and again in 2003, by the Border Regime Treaty.

325 LIST OF ANNEXES

Annex RR 1 Letter of the Agent of Romania to the Registrar of the International

Court of Justice, dated 25 September 2006;

Annex RR 2 ProcŁs-Verbal no. 10 of the Romanian-Soviet Commission for the
Demarcation of the State Border between the People’s Republic of

Romania and the Union of the Soviet Socialist Republics, signed in
Ismail, on 5 December 1948;

Annex RR 3 Note Verbale of the Permanent Mission of Romania in New York,

dated 18 June 1997;

Annex RR 4 Decree No. 142 of 25 April 1986 of the Council of State
concerning the establishment of the Exclusive Economic Zone of

the Socialist Republic of Romania in the Black Sea;

Annex RR 5 Maps of the Black Sea without a representation of Serpents’
Island;

Annex RR 6 Promotional materials on Serpents’ Island, publicly available at the
exhibition organised on the occasion of the 3rdInvestment and
Innovations Forum (Odessa, 1 – 3 June 2006);

Annex RR 7 Article titled “Recycling the garbage – one of the biggest problems
on Serpents’ Island”, published by the Ukrainian Press Agency

“Reporter”, on 3 May 2006;

Annex RR 8 Article titled “You win Serpents’ Island or you will be defeated in
the fight for it”, published by the Ukrainian weekly newspaper

“Argument. Vlada”, the issue of the week 10 – 16ebruary 2006;
author: Stella Maks;

Annex RR 9 Document No. 452, sent by the captain of the Sulina Port to the

European Danube Commission, 7 May 1921, General Department
of the National Archives, Galatzi, Romania, European Danube
Commission, 726/1921-1923;

Annex RR 10 Document No. 1131, sent by the captain of the Sulina Port to the
European Danube Commission, 6 December 1922, General
Department of the National Archives, Galatzi, Romania, European
Danube Commission, 726/1921-1923;

326Annex RR 11 Document No. 339, sent by the captain of the Sulina Port to the
European Danube Commission, 9 March 1923, General
Department of the National Archives, Galatzi, Romania, European
Danube Commission, 1341/1923-1940;

Annex RR 12 Article titled “By the year 2008, a town will be settled on Serpents’
Island”, published by the Ukrainian Press Agency “ProUA”, on 1
June 2006;

Annex RR 13 Article titled “A town will be settled on Serpents’ Island”,
published by the Ukrainian Press Agency “PODROBNOSTI.Ua”,
on 1 June 2006;

Annex RR 14 Article titled “Ehanurov initiates a contest regarding the
development of Serpents’ and Tuzla Islands” published by
“ProUA” Press Agency on 10 February 2006;

Annex RR 15 Article titled “There, where the wind always blows from the sea”,
published by the newspaper “Golos Ukraini” on 29 September
2006, issue no. 181 (3931); author: Anastasia Sirotkina;

Annex RR 16 Article titled “Serpents’ Island, asylum for political deportees”,
published by the Romanian newspaper “Curentul”, on 21 March
1938;

Annex RR 17 Article titled “A prison with 1500 cells for political convicts is
going to be built on Serpents’ Island”, published by the Romanian
newspaper “Informatia”, on 22 March 1938;

Annex RR 18 Article titled “Vasili Chernolutzky proposed the building of a
church on Serpents’ Island”, published by the Ukrainian on-line
publication “Reporter”, on 19October 2006;

Annex RR 19 The transcripts of phone calls attempting to contact the branch of
the Aval Bank on Serpents’ Island;

Annex RR 20 Article titled “The Settlement ‘Bilyi’ will appear on Danube”,
published in the Ukrainian newspaper “Slovo”, on 29 September
2006; author: Tamila Cernatzkaia;

Annex RR 21 Article titled “11 million hryvnyas were allotted for the
development of Serpents’ Island”, published by the Ukrainian
Press Agency “ProUA”, on 24 January 2006;

327Annex RR 22 Article titled “Serpents’ Island might become an off-shore zone”,
published by the Ukrainian branch of ITAR-TASS Press Agency,
on 18 April 2006;

Annex RR 23 Note Verbale no. H2/232 of 25 January 2006 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 24 Note Verbale no. H2/1191 of 19 April 2006 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 25 Note Verbale no. H2/1416 of 10 May 2006 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 26 Note Verbale no. H2/1724 of 2 June 2006 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 27 Article titled “Snakelike Investments”, published by the Ukrainian
newspaper “Ekonomicheskye Izvestia”, on 24October 2006;

Annex RR 28 Article titled “On the Ukrainian island of the Black Sea a hotel for

‘Robinsons’ was opened”, published by the Ukrainian Press
Agency “NEWSru.com”, on 6 May 2006;

Annex RR 29 Note Verbale C23/2533/6 June 2001 of the Romanian Ministry of

Foreign Affairs addressed to the Ukrainian Embassy in Bucharest;

Annex RR 30 Note Verbale no. C23/3342/ 2 July 2001 of the Romanian Ministry
of Foreign Affairs addressed to the Ukrainian Embassy in

Bucharest;

Annex RR 31 Note Verbale no. DVI-1/6784/15 December 2003 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in

Bucharest;

Annex RR 32 Note Verbale no. EVI-1/2850/11 May 2004 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in

Bucharest;

Annex RR 33 Note Verbale no. EVI-1/6352/21 October 2004 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in

Bucharest;

328Annex RR 34 Note Verbale no. EVI-1/6503/2 November 2004 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 35 Note Verbale no. EVI-1/7242/27 December 2004 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 36 Note Verbale no. H2/1607/22 May 2006 of the Romanian Ministry
of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 37 Note Verbale no. H2/1749/4 July 2006 of the Romanian Ministry
of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 38 Note Verbale no. H2/2457/14 August 2006 of the Romanian
Ministry of Foreign Affairs addressed to the Ukrainian Embassy in
Bucharest;

Annex RR 39 Examples of seismic profiles of geological structures from the
delimitation area explored by Romania

Annex RR 40 Note Verbale no. 2329/ 16 November 2006 of the Romanian

Embassy in Ankara addressed to the Ministry of Foreign Affairs of
the Republic of Turkey;

Annex RR 41 Note Verbale no. 2006/DHGY/480618/ 24 November 2006 of the

Ministry of Foreign Affairs of the Republic of Turkey addressed to
the Romanian Embassy in Ankara;

Annex RR 42 Note Verbale of 19 June 1918 of the Romanian Ministry of

Foreign Affairs addressed to the Government of Ukraine;

Annex RR 43 Declaration of Independence of the Republic of Moldova, 27
August 1991 – Law 691/1991 published in the Official Bulletin of

the Republic of Moldova no. 11/27.08.1991;

Annex RR 44 Telegram no. 1491 of 30 April 1941 of the Romanian Legation in
Moscow addressed to the Romanian Minister of Foreign Affairs;

Annex RR 45 Declaration of the Government of Romania on the occasion of the
proclamation of the independence of the Republic of Moldova, 27
August 1991.

329 LIST OF FIGURES

Figure RR1 The English-French maritime boundary as established by the Court

of Arbitration in the Anglo-French Continental Shelf Case;

Figure RR2 The Ukrainian Coast and its Various Segments;

Figure RR3 The relevant coasts of the two States (reproduced from RM, Figure
11);

Figure RR4 The projections of the Ukrainian coast, as presented by UCM,

Figure 3-4;

Figure RR5 The accurate projections of the Ukrainian coast;

Figure RR6 Relevant v. irrelevant coasts as defined by the Court in the Jan
Mayen case;

Figure RR7 Relevant v. irrelevant coasts as defined by the Court in the

Tunisia/Libya case;

Figure RR8 Relevant v. irrelevant coasts as defined by the Court in the Gulf of
Maine case;

Figure RR9 The Romanian coast;

Figure RR10 Satellite image of the Sulina dyke. Source: Google Earth, accessed

20.11.2006;

Figure RR11 The differences between the Parties regarding the relevant area of
delimitation;

Figure RR12 The relevant delimitation area;

Figure RR13 Extract from the sketch included in the individual ProcŁs-Verbal of

Point 1439;

Figure RR14 Reproduction of Map 134;

Figure RR15 Reproduction of Map 133 annexed to the September 1949 ProcŁs-
Verbal;

Figure RR16 The position of various points depicted by Map 134, in accordance

with the factual situation of 1949 (the year when Map 134 was
issued);

330Figure RR17 The differences between the positions of the final point of the
boundary depicted in Map 134 and Point 2, respectively Point F;

Figure RR18 Plate I annexed to the 1949 ProcŁs-Verbal;

Figure RR19 Plate V annexed to the 1949 ProcŁs-Verbal;

Figure RR20 Differences regarding the position of the final points of the
boundary as depicted on Map 134, Plate I and Plate V;

Figure RR21 The Construction of Point X;

Figure RR22 Extract from International Charts Series INT1 Symbols,
Abbreviations, Terrms Used on Charts, published by Bundesamt
fur Seeschiffahrt und Hydrographie, Hamburg/Rostock, 1996;

Figure RR23 Accurate representation of the map annexed to the 1997 Romanian
notification to the UN;

Figure RR24 The illustrative map of the Romanian Maritime Claims, as
reproduced in the Law of the Sea Bulletin, No. 19 (October 1991);

Figure RR25 Ukraine’s hypothetical coastal façade uniting Serpents’ Island with

the mainland coast;

Figure RR26 Romania’s exploration activities carried out in the delimitation
area;

Figure RR27 Romania’s exploration activities carried out in the area in dispute
(area of overlapping claims);

Figure RR28 The equidistant line between the adjacent relevant Romanian and
Ukrainian coasts (following RM Figure 28);

Figure RR29 The maritime boundary in Sector 1 (following RM Figure 30);

Figure RR30 The maritime boundary in Sector 2 (following RM Figure 33);

Figure RR31 The maritime boundary between the exclusive economic zones and

the continental shelf of Romania and Ukraine in the Black Sea
(following RM Figure 33);

Figure RR32 The Ukrainian proposed delimitation line encroaches on the

Romanian entitlements.

331

Document Long Title

Reply of Romania

Links