Jump to Content Jump to Main Navigation
Signed in as:

5 Non-Codified Means of Interpretation

From: Treaty Interpretation in Investment Arbitration

J. Romesh Weeramantry

From: Oxford Legal Research Library (http://olrl.ouplaw.com). (c) Oxford University Press, 2023. All Rights Reserved. Subscriber: null; date: 06 June 2023

Awards — Advisory opinions

(p. 115) Non-Codified Means of Interpretation

5.01  This chapter concerns techniques of treaty interpretation that are not specifically referred to in the Convention Rules. Subsequent to a general introduction in Section A, the chapter examines how the treaty interpretation process utilizes international court decisions and arbitral awards (Section B), treaties or international instruments (Section C), scholarly opinion (Section D), the principle of effectiveness (Section F), and legal maxims (Section G). Section E deals with inter-temporal aspects relevant to treaty interpretation.

A.  Introduction

5.02  All major criteria expressly referred to in Articles 31 and 32 of the Vienna Convention have been discussed in Chapter 3 and 4 supra. Any extensive study of treaty interpretation, however, cannot end there. In practice, a number of other approaches, although not codified in Articles 31 and 32, are frequently deployed by international courts and tribunals to interpret treaties. As was seen in Chapter 4 supra, Article 32 specifies two examples of ‘supplementary means of interpretation’—the use of preparatory work and circumstances surrounding the treaty’s conclusion. But Article 32 does not provide an exhaustive list. Moreover, as Sinclair has accurately remarked,

few would seek to argue that the rules embodied in Articles 31 to 33 of the [Vienna Convention] are exhaustive of the techniques which may be properly adopted by the (p. 116) interpreter in giving effect to the broad guidelines laid down in those rules. … [T]here are many other principles of logic or of common sense which might have been included if the intention had been to draw up a comprehensive catalogue of all those aids to interpretation which have from time to time, and depending upon the circumstances of the particular case, been invoked by international tribunals.1

5.03  A number of these alternate approaches to interpretation, whether they strictly fall within the scope of Article 32 or whether they constitute mere rules of logic or applications of common sense, will be discussed in this section in no particular order of importance.

B.  Use of Prior Awards or Decisions

5.04  The role prior FIAT decisions play in interpreting investment treaties has been the focus of considerable discussion and commentary.2 From an historical viewpoint, investment treaty arbitration has undergone a sea change since Christoph Schreuer observed in 2001 that there was a dearth of ICSID award references to prior ICSID awards due to ‘their relatively small number and the difficulty in gaining access to them’.3 But since then, the once gentle stream of FIAT awards has transformed into a torrent. At present, well over 300 investment arbitration decisions and awards are publicly accessible.4 The numbers of newly published FIAT awards presently increase almost on a weekly basis. The number and availability of these awards has created a paradox of precedent: the common practice of FIATs referring to past decisions and awards to confirm or support their decisions despite the absence of a system of binding precedent that compels adhesion to those past awards.5

(p. 117) 5.05  A comparison helpful in this discussion on reliance on past decisions is the practice of the ICJ.6 When making comparisons with the ICJ, it should be borne in mind that the ICJ is a permanent adjudicative body, whereas FIATs are non-permanent arbitral tribunals. Consequently, when the ICJ considers its prior decisions, it is considering its own decisions. In contrast, when FIATs refer to past FIAT decisions, they are making reference to decisions made by other arbitral tribunals, differently composed and usually operating under different international investment treaties. Article 59 of the Statute of the ICJ provides that ‘[t]he decision of the Court has no binding force except between the parties and in respect of that particular case’.7 Nonetheless, the Court has referred to its decisions as forming ‘settled jurisprudence’8 and Article 38(1)(d) classifies ‘judicial decisions’ as a ‘subsidiary means for the determination of rules of law’. In the Land and Maritime Boundary case, the ICJ made clear that Nigeria was not bound by decisions the Court had reached in previous cases. In its view, nevertheless, the real question was whether ‘there is cause not to follow the reasoning and conclusions of earlier cases’.9

5.06  Shabtai Rosenne has commented that the ICJ refers to its earlier decisions ‘not so much in the quality of binding precedent [but] as having persuasive influence’.10 Also relevant is Rudolf Bernhardt’s summary of ICJ practice:

The ICJ, like every court, hesitates to overrule former pronouncements; quite to the contrary, it often refers to previous decisions and to reasons developed in such decisions, whether these reasons have been essential for that decision or are only obiter dicta. In many judgments of the ICJ, the Court quotes extensively its own pronouncements in former cases with different parties. Even advisory opinions, which are formally not binding for any State or even for the international organ having requested the opinion, are often quoted in later advisory opinions and judgments.11

(p. 118) 5.07  Similar to Article 59 of the ICJ Statute, an ICSID award by virtue of Article 53(1) of the ICSID Convention binds only the parties to the arbitration.12 This provision was characterized by the SGS v Philippines tribunal as one ‘which might be regarded as directed to the res judicata effect of awards rather than their impact as precedents in later cases’.13

5.08  The El Paso tribunal said this about the significance of prior awards:

ICSID arbitral tribunals are established ad hoc, from case to case, in the framework of the Washington Convention, and the present tribunal knows of no provision, either in that Convention or in the BIT, establishing an obligation of stare decisis. It is, nonetheless, a reasonable assumption that international arbitral tribunals, notably those established within the ICSID system, will generally take account of the precedents established by other arbitration organs, especially those set by other international tribunals. The present Tribunal will follow the same line, especially since both parties, in their written pleadings and oral arguments, have heavily relied on precedent.14

5.09  More recently, consistency in FIAT decision-making has been emphasized by the Saipem tribunal:

The Tribunal considers that it is not bound by previous decisions. At the same time, it is of the opinion that it must pay due consideration to earlier decisions of international tribunals. It believes that, subject to compelling contrary grounds, it has a duty to adopt solutions established in a series of consistent cases. It also believes that, subject to the specifics of a given treaty and of the circumstances of the actual case, it has a duty to seek to contribute to the harmonious development of investment law and thereby to meet the legitimate expectations of the community of States and investors towards certainty of the rule of law.15

(p. 119) 5.10  The adoption of ‘must’ and ‘duty’ in the previous passage indicate a strong preference by this tribunal to show deference to other tribunals and align their decisions accordingly. This decision highlights the difference between investment arbitration and commercial arbitration. In the latter, arbitrators by and large concentrate on making determinations on the facts in dispute, not on developing the law. One reason for this difference is that commercial arbitration awards are rarely made public and consequently are not examined or referred to by subsequent arbitral tribunals.16 In contrast to the approach taken in Saipem, a more circumspect view was taken in SGS v Philippines:

although different tribunals constituted under the ICSID system should in general seek to act consistently with each other, in the end it must be for each tribunal to exercise its competence in accordance with the applicable law, which will by definition be different for each BIT and each Respondent State. Moreover there is no doctrine of precedent in international law, if by precedent is meant a rule of the binding effect of a single decision. There is no hierarchy of international tribunals, and even if there were, there is no good reason for allowing the first tribunal in time to resolve issues for all later tribunals. It must be initially for the control mechanisms provided for under the BIT and the ICSID Convention, and in the longer term for the development of a common legal opinion or jurisprudence constante, to resolve the difficult legal questions discussed by the SGS v. Pakistan Tribunal and also in the present decision.17

5.11  These words set the stage for this tribunal’s now famous decision in which it declined to follow the SGS v Pakistan tribunal’s interpretation of an umbrella clause similar to that which was at issue in SGS v Philippines. This, however, did not mean to say that the SGS v Philippines tribunal disagreed with all aspects of the SGS v Pakistan award.18

5.12  The question whether prior decisions should be followed represents an important question in investment treaty arbitration because FIATs frequently interpret treaty provisions that are identical or similar to those already interpreted by other FIATs.19 Indeed, there have been times where the different claimants in parallel arbitrations (p. 120) have been seeking rights against the same host State, under the same BIT and in respect of the same provisions of that BIT.20 The absence of any binding precedential effect of FIAT awards thus leaves the door open for inconsistency in decision making among FIATs, including in the interpretation of treaties.21

5.13  Notwithstanding this potential for inconsistent decisions, the bulk of FIAT practice demonstrates that prior FIAT decisions have a significant degree of influence on subsequent FIAT interpretations of foreign investment treaties. For example, it is common practice for ICSID tribunals to refer to other ICSID tribunal awards to confirm or support their interpretation or application of a treaty, particularly in relation to the interpretation of the ICSID Convention.22

5.14  Other reasons FIATs have employed to justify following or giving weight to prior FIAT decisions (for interpreting treaties and other decisional purposes) include considerations as to whether the prior decision appeared to be:

  1. (a)  convincing or persuasive;23

  2. (b)  ‘well founded’;24

  3. (c)  ‘instructive’;25(p. 121)

  4. (d)  ‘in harmony with applicable international jurisprudence’;26

  5. (e)  involving ‘similar circumstances’ or a ‘high level of similarity’;27

  6. (f)  ‘correct’;28 and

  7. (g)  ‘appropriate’ to consider.29

5.15  FIATs also have had regard to (a) the development of an ‘accepted meaning’ of a term through authoritative definition in ICJ judgments30 and (b) an ‘overwhelming weight of [ICSID] authority’ that points toward a certain interpretation of the ICSID Convention.31 The justifications for reliance or finding support in prior decisions enhance the uniformity, predictability, and authority of the very public FIAT decision-making process.32

5.16  In many cases, FIATs consider the prior awards of other FIATs in large part because heavy reliance is placed on these decisions in the parties’ pleadings.33 This practice is evident in the following passage in the Feldman award:

in view of the fact that both of the parties in this proceeding have extensively cited and relied upon some of the earlier decisions [of NAFTA tribunals], the Tribunal believes it appropriate to discuss briefly relevant aspects of earlier decisions. …34

5.17  FIATs have also confirmed or supported their interpretation or application of certain treaty provisions by reference to relevant interpretations rendered by a wide range of other international courts and tribunals, in particular the PCIJ,35 ICJ,36 European Court of Justice,37 European Court of Human Rights,38 Iran–United (p. 122) States Claims Tribunal,39 and ad hoc arbitrations.40 There have also been several references to World Trade Organization jurisprudence.41 In rare instances, references to domestic court decisions have been considered appropriate.42 An exception has been the approach taken in National Grid, in which the tribunal gave limited value to an OPIC decision because it was a decision of an organ of a government rather than one arrived at by an independent tribunal after hearing the parties.43 The Continental Casualty award is notable for the considerable weight it gave to GATT and WTO case law in its interpretation of Article XI of the Argentina–United States BIT:

Since the text of Art. XI derives from the parallel model clause of the U.S. FCN treaties and these treaties in turn reflect the formulation of Art. XX of GATT 1947, the Tribunal finds it more appropriate to refer to the GATT and WTO case law which has extensively dealt with the concept and requirements of necessity in the context of economic measures derogating to the obligations contained in GATT, rather than to refer to the requirement of necessity under customary international law.44

5.18  However, care must be exercised by FIATs not to place too much emphasis on interpretation or practices relating to provisions belonging to other legal regimes. For (p. 123) example, the context of the GATT and WTO trade-related provisions could be significantly different to the context of investment protection provisions in BITs. Accordingly, the interpretation of investment treaty provisions that, for example, deal with national treatment by reference to apparently similar provisions in the GATT (Article 3), requires caution. As Wälde notes, both investment and trade law treaty regimes

may have some similarities (economic matters; international discipline on national regulation; counteracting protectionist attitudes). But they are also substantially different (GATT is inter-State only; BITs are investor-State; GATT only provides prospective remedies; BITs provide compensation for past damage).45

5.19  One of the most important results in the Fauchald empirical analysis was the finding that prior case law was used in the interpretative process in 92 of the 98 decisions subject to that study.46 Other examinations of FIAT awards also indicate an evolving form of precedent (though not in the common law sense) in investment arbitration.47 Thomas Wälde has gone so far as to contend that

tribunals which deviate from established jurisprudence, without an extensive effort at reasoning, distinction and providing a full hearing to the parties on their intention to deviate, might be considered to commit severe procedural and material rule breaches that could bring them into the visor of the—limited—procedures for judicial review and annulment.48

5.20  In the context of the ICSID system, an award would have to be on the extreme end of the ICSID award spectrum for it to be annulled solely on the basis that it has not considered prior awards. No annulment by an ad hoc Committee has yet been made on such a basis.

(p. 124) 5.21  A discernible practice associated with FIAT reference to prior awards is that it is rare for distinctions to be made between the ratio decidendi and obiter dicta.49 As a consequence, FIATs typically give the same weight to statements of prior FIATs whether or not they constitute ratio decidendi or obiter dicta. Commission has contended that ‘[a] continued failure to distinguish between the ratio decidendi and obiter dicta of prior awards and decisions could threaten the integrity of the tribunals and legitimacy of the investment treaty system itself’.50 It remains to be seen whether the system will be jeopardized in such a fashion.

5.22  Despite the numerous cases that demonstrate a good degree of respect is shown toward prior FIAT decisions, a number of FIATs have been reluctant to adopt the interpretation of other FIATs too readily. The most prevalent reasons why other FIAT decisions are not followed or are not given great weight is that the antecedent case involved different facts, background, negotiating histories, or circumstances.51 For example, in the Enron proceedings, Argentina urged the tribunal to be consistent with the interpretation of Article 25 of the ICSID Convention made in Vacuum Salt. Among its reasons for rejecting this submission, the tribunal stressed that Vacuum Salt ‘was an entirely different case not comparable in any way with this one’.52 In the AES arbitration, the tribunal agreed with the following submission by Argentina:

Repeating decisions taken in other cases, without making the factual and legal distinctions, may constitute an excess of power and may affect the integrity of the international system for the protection of investments.53

5.23  The risks associated with use of awards as precedents have led Thomas Wälde and Todd Weiler to comment:

one should never assume an automatic precedential effect on any individual tribunal award. Modern investment treaties have many commonalities and their text is mostly derived from previous treaties, but there are often substantial differences—particular [sic] in terms of reservations and exceptions, but also in structure and the ‘fine print’ of the relevant text. One needs therefore to find a balance between common and (p. 125) emerging trends in interpretation, without failing to give sufficient counterweight to the peculiarities of the text, context and purpose of a particular treaty such as the NAFTA, ECT or bilateral treaties.54

5.24  Moreover, antecedent awards may not provide a sufficiently conclusive answer on a point.55 Likewise, the evidence submitted in the prior case may raise doubts about the assistance that prior case may provide, as the Aguas del Tunari case demonstrates. Both parties there referred to various ICSID cases in their submissions as to the meaning of ‘control’ in Article 25(2)(b) of the ICSID Convention. The Aguas del Tunari tribunal was unsure as to what evidence had been filed in these prior cases and to what extent the respective tribunals considered this evidence. Consequently, it took the view that ‘[w]ithout access to the full records of these cases, the Tribunal does not believe it possible to assess their significance for the present arbitration’.56

5.25  Another issue that is raised in FIAT jurisprudence is whether prior judicial or arbitral decisions constitute a ‘source’ of international law. Sempra Energy is notable for observing that

interpretation is not the exclusive task of States. It is also the duty of tribunals called upon to settle a dispute, particularly when the question is to interpret the meaning of the terms used in a treaty. This is precisely the role of judicial decisions as a source of international law in Article 38(1) of the Statute of the International Court of Justice, to which the Respondent refers.57

5.26  This statement calls for clarification. Article 38(1)(d) provides that the ICJ is to apply ‘judicial decisions … as a subsidiary means for the determination of rules of law’. Alain Pellet’s insightful view on this provision is that:

jurisprudence and doctrine are not sources of law—or, for that matter, of rights and obligations for the contesting States; they are documentary ‘sources’ indicating where the Court can find evidence of the existence of the rules it is bound to apply by virtue of the three other sub-paragraphs [i.e., Article 38(1)(a), (b) and (c)] … strictly speaking, the Court does not ‘apply’ those ‘means’, which are only tools which it is invited to use in order to investigate the three sources listed above.58

5.27  In the light of Pellet’s (correct) reading of Article 38(1) of the ICJ Statute, the Methanex tribunal’s approach is to be preferred to that taken in Sempra Energy extract. (p. 126) The Methanex tribunal commented that while prior NAFTA awards were neither ‘sources of law’ nor ‘legally binding’, the claimant was ‘entitled to adopt their legal reasoning as part of its case’.59

5.28  In Canadian Cattlemen the tribunal took the view that

Article 38 [paragraph 1.d.] of the Statute of the International Court of Justice provides that judicial decisions are applicable for the interpretation of public international law as ‘subsidiary means’. Therefore, they must be understood to be also supplementary means of interpretation in the sense of Article 32 of the [Vienna Convention].60

5.29  This statement fails to take into account the differences between the two provisions. As mentioned previously, Article 38 of the ICJ Statute refers to judicial decisions as ‘subsidiary means for the determination of rules of law’. This provision enables judicial decisions (or arbitral awards) to be used to identify appropriate rules. That is distinct from using prior decisions under Article 32 of the Vienna Convention to interpret particular provisions of investment treaties, which are of themselves rules of international law as between the treaty parties.61 The logic of the Canadian Cattlemen award is therefore problematic. This is perhaps one reason why it is relatively rare for FIATs to discuss or refer to the ICJ Statute when they refer to past awards when interpreting treaties.

5.30  To bring this section to a close, the Fauchald empirical analysis finding should be recalled. It concluded that references to case law in interpretative arguments constituted by far the most frequently used mode of interpretation.62 The following observation of Wälde and Weiler is therefore apposite:

An investment jurisprudence is at present emerging rapidly—perhaps too easily in view of the sometimes significant legal differences between the various treaties underlying the cases. The result of these trends may well be a burgeoning ‘common law’ of international trade and investment protection, where different tribunals work separately, but somehow symbiotically, to articulate the meaning of treaty obligations which share similar, if not identical, language.63

(p. 127) 5.31  This development in international investment law has parallels with decision trends in international law generally. Brown, for example, has concluded in a well-received work that there is an emerging ‘common law of international adjudication’ involving a practice where the resolution of an issue in one international court or tribunal is often aided by considering how that issue has been dealt with by other international bodies.64 But this practice, Brown observes, is not based on a sense of obligation to follow other decisions but, inter alia, on similarities in (a) the procedure and remedies to be applied, (b) the applicable rules or law, (c) the lawyers that are involved, and (d) the functions to be carried out by those bodies.65 These reasons for following prior decisions apply equally to international investment law.

C.  Treaties, Instruments, or Materials

1.  Treaties

5.32  FIATs often find interpretative guidance from comparisons with provisions in other treaties that are similar or identical to the provision in dispute or which evidence the practice of States.66 The Fauchald empirical analysis, for example, found 28 out of 98 FIAT decisions in which BITs not the subject of the dispute were utilized in interpretative arguments and fourteen others that referred to other investment-related instruments.67 This tendency may be attributable, in large measure, to the hundreds of BITs covering essentially the same types of subject matter, particularly where, in Wälde’s animated terminology, the drafters have quarried text blocks from other treaties.68 As a consequence, common threads are bound to weave through many investment treaties. In this regard, McLachlan’s articulation of the process by which BITs build upon antecedent BITs provides some understanding as to why prior treaty provisions are important:

Each state brings to the negotiating table a lexicon which is derived from prior treaties (bilateral or multilateral) into which it has entered with other states. The resulting text in each case may be different. It is, after all, the product of specific negotiation. But it will inevitably share common elements with what has gone before.69

(p. 128) 5.33  Consequently, the examination of prior treaty provisions may be a fruitful exercise in the FIAT interpretative process. In framing this issue within the context of the Convention Rules, Gardiner includes the discussion of this topic under Article 31(3)(c) but states that use of other treaties in the interpretive process ‘is such an accepted and established practice that it is hard to find any situation in which justification in terms of the [Convention Rules] has been presented’.70

5.34  An appropriate starting point for a discussion on relevant FIAT practice is the exposition by the AAP tribunal concerning interpretation by reference to other treaties. That tribunal neatly summed up the position in its self-styled Rule F:

‘When there is need of interpretation of a treaty it is proper to consider stipulations of earlier or later treaties in relation to subjects similar to those treated in the treaty under consideration’. … Thus, establishing the practice followed through comparative law survey of all relevant precedents becomes an extremely useful tool to provide an authoritative interpretation.71

5.35  Many FIATs have followed this approach. References have frequently been made to other treaties, whether or not antecedent or subsequent to the treaty in dispute,72 (p. 129) particularly to support or confirm an interpretation.73 In Pope & Talbot, for example, the tribunal noted that

the language of Article 1105 [of NAFTA] grew out of the provisions of [antecedent] bilateral commercial treaties negotiated by the United States and other industrialized countries … there are very strong reasons for interpreting the language of Article 1105 consistently with the language in the BITs.74

5.36  However, sight must not be lost of the context of the terms subject to interpretation.75 For example, jurisprudence as to national treatment provisions in the GATT has not been adopted by FIATs in interpreting similar provisions in investment treaties (including NAFTA Chapter 11) because of the different contexts in which GATT provisions operate.76 In this regard the Methanex tribunal quoted with approval the following passage of the Law of the Sea Tribunal’s Mox Plant decision:

the application of international law rules on interpretation of treaties to identical or similar provisions of different treaties may not yield the same results, having regard to, inter alia, differences in the respective contexts, objects and purposes, subsequent practice of parties and travaux préparatoires. …77

(p. 130) 5.37  Certain FIATs have exercised a degree of caution or altogether avoided reliance on other treaties as an aid to interpretation.78 Some reasons given for taking this position are as follows:

  1. (a)  the terms of the other treaty were different;79

  2. (b)  no reason was indicated to justify a difference in language between two treaties;80

  3. (c)  the parties to the other treaty were different;81

  4. (d)  the tribunal was not informed by the parties of all the investment treaties signed by the contracting States to the BIT in dispute;82

  5. (e)  the other treaty related to a different discipline of international law;83

  6. (f)  the meaning of the treaty in dispute was clear;84 and

  7. (g)  the BIT practice of both parties to the BIT at issue was of ‘limited probative value’.85

5.38  Obviously, if the treaty provisions subject to comparison are differently worded to a significant extent, then FIATs have refused to utilize them, or interpretations thereof. The issue of differences in treaties is dealt within the proceeding text.

5.39  Comparisons between the treaty in dispute and model BITs have also become issues in FIAT interpretation practice. A question raised in Siemens was whether special significance should be attached to a dispute settlement clause in the Argentina–Germany BIT because it departed from Germany’s standard BIT dispute settlement clause. The tribunal took the view that ‘it is not in its power to second-guess [the Argentina–Germany BIT drafters’] intentions by attributing special meaning to (p. 131) phrases based on whether they were or were not part of a model draft’.86 As Professor Schreuer points out, model treaties are the starting point for negotiations. The extent to which the concluded treaty will echo a certain model treaty will depend on many factors, including the negotiating power of the parties, whether both countries have their own distinct model treaties they wish to promote, and the circumstances surrounding the negotiations.87

2.  Instruments or Materials

5.40  In addition to treaties, FIATs have referred to the following instruments or materials as part of the interpretative process:

  1. (a)  the work of the International Law Commission;88

  2. (b)  the work of the Institute of International Law;89

  3. (c)  the work of the International Law Association;90

  4. (d)  the Harvard Draft Convention on the International Responsibility of States for Injuries to Aliens;91

  5. (e)  the Statute of the ICJ;92

  6. (f)  United Nations General Assembly Resolutions;93

  7. (g)  United Nations Conference on Trade and Development reports;94

  8. (h)  Organization for Economic Cooperation and Development reports;95(p. 132)

  9. (i)  the World Bank Guidelines on the Treatment of Foreign Direct Investment;96

  10. (j)  the American Law Institute’s 1987 Restatement of the Law Third, the Foreign Relations of the United States;97

  11. (k)  the submissions of a respondent State in another case relating to the same treaty provision;98

  12. (l)  official statements and other materials of governments (or organs) of States party to the treaty in question;99 and

  13. (m)  the general practice of the signatories to the BIT at issue in respect of other investment treaties.100

5.41  In relation to the use of government materials produced during ratification or implementation of a treaty, the Mondev tribunal stated that

[w]hether or not explanations given by a signatory government to its own legislature in the course of ratification or implementation of a treaty can constitute part of the travaux préparatoires of the treaty for the purposes of its interpretation, they can certainly shed light on the purposes and approaches taken to the treaty, and thus can evidence opinio juris.101

5.42  The Mondev tribunal referred specifically to the US government transmittal statement for the 1995 US–Albania BIT to its legislature. That treaty contained language (p. 133) similar to the NAFTA provision at issue in Mondev, and was considered to have shed light on that NAFTA provision.102 However, prudence should be exercised considering that these government statements may only represent the view or intention of one State party to the treaty.

5.43  In connection with ICSID arbitrations, the Report of the Executive Directors103 merits separate discussion. Essentially, the Report is a type of explanatory note or set of guidelines for the ICSID Convention. Numerous references to the Report have been made by ICSID tribunals in their interpretation of the ICSID Convention.104 It does not appear to be considered as part of the preparatory work of the ICSID Convention.105 Reference to the Report is sometimes made even before any other Article 31 analysis takes place.106 Although no FIAT has stated as such, the Report’s proper place in the framework of the Convention Rules is likely to be as part of the context, in the sense that it is an instrument connected with the conclusion of the treaty as provided for in Article 31(2)(a). Characterized in this way, an interpreter’s immediate recourse to the Report would be consistent with the Article 31 process.

5.44  Concerning government investment promotion materials, the SPP tribunal noted in relation to the interpretation of a domestic statute providing Egypt’s consent to ICSID arbitration that ‘[i]nvestment promotion literature does not create rights; it informs potential investors of the rights they will enjoy by virtue of existing law if an investment is made’ and that ‘[w]hile there is no question of investment promotion literature altering the terms of a statute, in the present case the materials published by the General Authority merely confirm the conclusion already reached by the Tribunal’.107

5.45  Care also needs to be exercised when using extrinsic treaties and materials as a guide to interpretation. In relation to other treaties, their texts may fail to reveal important nuances of the wording as understood by the signatories to the treaty at issue. As Wälde and Weiler have observed, dangers lurk in the practice of comparing (p. 134) seemingly similar treaties, especially the risk of overlooking the particularities of text, context, and purpose.108 In this regard, the caution expressed in Aguas del Tunari is apposite:

The practice of a state as regards the conclusion of BITs other than the particular BIT involved in a dispute is not of direct value to the task of interpretation under Article 31 of the Vienna Convention. The fact that a pattern might exist in the content of the BITs entered into by a particular state does not mean that a specific BIT by that state should be understood as necessarily conforming to that pattern rather than constituting an exception to that pattern.109

5.46  In TSA Spectrum, the tribunal refused to interpret a treaty provision with reference to how it is understood under national law. It held that because the term ‘final decision’ in Article 10(3) of the 1994 Argentina–Netherlands BIT

appears in an international treaty and not in a purely national context, it is appropriate to give it an autonomous meaning and thus interpret it independently of any meaning the same term may have in the national laws of the two Contracting States.110

5.47  Another issue requiring discussion in this section is the relationship between the ICSID Arbitration Rules and the ICSID Convention. Those Arbitration Rules add detail to the procedural provisions already contained in the ICSID Convention.111 Although no FIAT has used the ICSID Arbitration Rules to determine an interpretation of the ICSID Convention,112 FIATs have used those Rules to confirm or support the interpretation of that Convention. A reason why the ICSID Arbitration Rules may be considered to be an agreement of the parties is that they are adopted by ICSID’s Administrative Council, which is composed of one representative of each signatory to the ICSID Convention.113 Recourse to the ICSID Arbitration Rules to interpret the ICSID Convention may be permitted under Vienna Convention Article 31(3). Under this provision, the Arbitration Rules may constitute a subsequent agreement between the parties regarding the application of the (p. 135) ICSID Convention or a subsequent practice in the application of the treaty establishing the agreement of the parties as to its interpretation.114 The ADF tribunal’s interpretation of the ICSID Arbitration (Additional Facility) Rules is of interest in this respect. In that case, the tribunal referred to the ICSID Convention and the ICSID Arbitration Rules and stated that ‘while these two instruments are not applicable to Additional Facility cases, like the instant case, they often do supply, in our opinion, relevant, and even close, analogues for terms used in the Additional Facility Rules’.115

5.48  A further issue relevant to this section is whether the ICSID Arbitration Rules may supplement the ICSID Convention when there is a gap in the latter. This was a very real problem for the ad hoc Committee in the Vivendi (Challenge) case, which was required to determine a challenge made by a party against the Committee’s president. The ad hoc Committee found that the ICSID Convention contained no provision governing the challenge of a member of an ad hoc Committee. In contrast, this issue was addressed in ICSID Arbitration Rules 9 and 53. Before the ad hoc Committee could utilize the Arbitration Rules to fill the gap in the ICSID Convention, it was considered necessary to determine whether the drafters of the ICSID Convention deliberately omitted such a provision from the Convention, in which case, the Committee doubted whether ICSID’s Administrative Council, the drafters of the Arbitration Rules, had competence to achieve by those Rules what the Convention itself intentionally did not achieve.116

5.49  As part of its decision, the ad hoc Committee examined the history of the ICSID Rules, which, in their opinion, showed that the intention of the Administrative Council was to apply all the Arbitration Rules, so far as was possible, to annulment proceedings. It added:

[i]n our view the only reason why the procedure laid down in Arbitration Rule 9 could not be applied to members of ad hoc Committees mutatis mutandis would be if to apply such a procedure was inconsistent with the Convention, having regard to its object and purpose. We see no reason to regard it as such.

… [T]he travaux préparatoires of the Convention do not suggest that there was any particular reason for excluding the application of [the Convention’s provisions concerning the replacement and disqualification of arbitrators in annulment proceedings]. It appears that no State party at the time of the adoption of Arbitration Rule 53 suggested any such reason. That Rule was adopted unanimously and was treated by the Members of the Administrative Council as uncontroversial. In the circumstances, the unanimous adoption of Arbitration Rule 53 can be seen, if not as an (p. 136) actual agreement by the States parties to the Convention as to its interpretation, at least as amounting to subsequent practice relevant to its interpretation.117

5.50  In a footnote to this passage, the Committee made express reference to Article 31(3) of the Vienna Convention. The approach of the Committee (effectively to supplement the ICSID Convention using the ICSID Arbitration Rules) may be considered to be in conformity with international law treaty interpretation principles.

5.51  Chapter 3 supra showed that reference to other treaties, in the sense that they constitute subsequent agreements between the parties to the treaty subject to interpretation, may also be made under Article 31(3).118

D.  Scholarly Opinion

5.52  Scholarly opinions or ‘teachings of the most highly qualified publicists of the various nations’, as noted in Article 38(1)(d) of the Statute of the ICJ, are not a ‘source’ of international law. They, like judicial decisions, constitute a ‘subsidiary means for the determination of rules of law’.119 The case law of the ICJ reveals that the Court is reticent in citing authors of scholarly literature by name.120 Nonetheless, it has been commented that ‘the scarcely avowed use of the “teachings of publicists” in the Court’s case law probably does not accurately reflect the influence these “teachings” still have’.121

5.53  The ICJ’s reticence in indicating specific opinions of writers may be ascribed to (a) the diplomatic sensitivities that surround much of the ICJ’s work, (b) the ‘small world’ of international law in which the Court is ‘well-advised not to distribute good or bad marks’, and (c) the ‘abstract discussions’ found in scholarly works.122

5.54  FIAT practice is markedly different from the practice of the ICJ. It is commonplace for tribunals to indicate that their views are ‘shared’ or ‘confirmed’ by specific scholarly literature on the subject under determination.123 Fauchald’s empirical analysis found that ‘legal doctrine’ (largely including scholarly literature) was the second (p. 137) most frequently utilized criterion by ICSID tribunals in their interpretative arguments.124

5.55  Part of the reason for the difference in ICJ and FIAT practice may well lie in the commercial nature of investor–State arbitration and also the autonomy and different composition of each tribunal. It must be said that the FIAT practice in this regard is preferable to that of the ICJ because the former embodies a more transparent process, gives recognition to deserving academic literature and generally enhances the legitimacy of the adjudicative process.

5.56  Another reason for the frequency of FIAT recourse to academic literature may also lie in the history of foreign investment arbitration. In its early days, relatively few FIAT awards existed. Given this barren environment of judicial practice, FIATs turned to the relatively verdant source of scholarly opinion for jurisprudential sustenance. The recent growth of case law generated by FIATs has no doubt changed this dynamic: the accumulation of publicly available case law has greatly increased the ability of arbitrators to use cases in support of their decisions. This, in turn, means there is less need to rely on academic literature. The awards reviewed for this book have evidenced that over time a distinct shift has taken place: FIATs have started to refer to past awards more than scholarly literature. Arguably, prior decisions have more practical value than scholarly opinion because they represent ‘real-life’ applications of law to the facts.125

5.57  An illustration of FIAT use of scholarly literature is found in Wena Hotels, in which the tribunal, after citing several relevant academic works, concluded that

[i]n the absence of any direct evidence of the intent of the Arab Republic of Egypt and the United Kingdom in negotiating Article 8(1), the Tribunal was strongly convinced by this common academic interpretation.126

5.58  The first edition of Schreuer’s The ICSID Convention: A Commentary is the pre-eminent work on the ICSID Convention. It is a model of scholarship, providing an extremely fastidious and systematic study of every provision in the ICSID Convention. Academics, practitioners, and arbitrators in the field of international investment law have found it of great practical value and a vast number of ICSID awards refer to it.127

(p. 138) 5.59  In addition to Schreuer’s commentary, there are a number of other notable persons whose works are frequently cited in FIAT awards. Broche’s work on initiating, drafting, and finalizing the ICSID Convention has led to a high degree of authority being accorded to his writings on the preparatory work of the Convention.128 In fact, in his capacity of General Counsel of the World Bank, Broches signed the ICSID Convention.129 He was the chair of (a) the regional consultative meetings of legal experts that deliberated on the preliminary draft of the Convention and (b) the Legal Committee on Settlement of Investment Disputes that produced a revised draft that formed the basis of the ICSID Convention’s final text.

5.60  Other distinguished commentators that find prominence in FIAT awards include Dolzer and Stevens;130 Amerasinghe;131 and Delaume.132

(p. 139) 5.61  The Mihaly tribunal, in conformity with the Statute of the ICJ, acknowledged that scholarly opinion was a subsidiary means of determining rules of law.133 The tribunal’s award also qualified the reference to teachings of the most highly qualified publicists in Article 38 of the ICJ Statute. It stated without further elaboration that ‘[o]pinions of experts on the theory and practice of multinational corporations’ were not to be equated with the ICJ Statute Article 38 reference to teachings of publicists.134 The statement suggests that expert opinions prepared especially for a specific dispute cannot be regarded as Article 38(1)(d) teachings. There may thus be a considerable variance in the deference paid to the publications of highly regarded international scholars and commentators as opposed to their expert opinions prepared (on instruction by a party) for a particular case.135

E.  Inter-Temporal Aspects

5.62  An examination of the drafting history of the Convention Rules reveals that Waldock’s Third Report (1964) and the ILC Draft Articles 1964 included a rule requiring an interpretation to be made in the light of ‘international law in force at the time of the conclusion of the treaty’.136 A contrary proposal was made at the Vienna Conference by Czechoslovakia’s delegation, which submitted that the interpretation should be made on the basis of the rules in force at the time of the application of the treaty.137 Neither of these approaches were finally adopted and no reference to time was included in the final text of the rule requiring that interpreters take into account relevant rules of international law (Article 31(3)(c)). Eduardo (p. 140) Jiménez de Aréchaga explains the omission of any reference to a relevant time was that it

is a question which must remain open and depends on whether the parties intended to incorporate in the treaty some legal concepts with a meaning that would remain unchanged, or intended to leave certain terms as elastic and open ended, subject to change and susceptible of receiving the meaning they might acquire in the subsequent development of the law.138

5.63  The ILC’s position reflects Judge Huber’s famous dictum in the Island of Palmas case in which he observed that ‘a juridical fact must be appreciated in the light of the law contemporary with it’ but also that the existence of rights ‘shall follow the conditions required by the evolution of the law’.139

5.64  In the Mondev case, the interpretation of NAFTA Article 1105 required the tribunal to decide whether the NAFTA States were correct in suggesting that this provision’s content (particularly its reference to fair and equitable treatment and full protection and security) was correctly expressed by the Mexico–United States General Claims Commission in the Neer case.140 The standard adopted in Neer was high:

the treatment of an alien … should amount to an outrage, to bad faith, to wilful neglect of duty, or to an insufficiency of governmental action so far short of international standards that every reasonable and impartial man would readily recognize its insufficiency.141

5.65  One of the grounds upon which the Mondev tribunal declined to apply the Neer standard was that it was determined in the 1920s and did not take into account the intervening developments in the law:

Neer and like arbitral awards were decided in the 1920s, when the status of the individual in international law, and the international protection of foreign investments, were far less developed than they have since come to be. In particular, both the substantive and procedural rights of the individual in international law have undergone considerable development. In the light of these developments it is unconvincing to confine the meaning of ‘fair and equitable treatment’ and ‘full protection and security’ of foreign investments to what those terms—had they been current at the time—might have meant in the 1920s when applied to the physical security of an alien. To the modern eye, what is unfair or inequitable need not equate with the outrageous or the egregious. In particular, a State may treat foreign investment unfairly and inequitably without necessarily acting in bad faith.142

(p. 141) 5.66  The tribunal then referred to the sizeable number of modern investment treaties that, in its view, almost uniformly provided for a standard of protection that was higher than the standard established in Neer. In the light of this recent development, the Mondev tribunal concluded that ‘[i]t would be surprising if this practice and the vast number of provisions it reflects were to be interpreted as meaning no more than the Neer tribunal (in a very different context) meant in 1927’.143

5.67  Additionally, the Mondev tribunal commented on the need to consider evolutionary processes in the development of an international law rule over a long period of time:

A reasonable evolutionary interpretation of Article 1105(1) is consistent both with the travaux, with normal principles of interpretation and with the fact that, as the Respondent accepted in argument, the terms ‘fair and equitable treatment’ and ‘full protection and security’ had their origin in bilateral treaties in the post-war period. In these circumstances the content of the minimum standard today cannot be limited to the content of customary international law as recognised in arbitral decisions in the 1920s.

… [T]here can be no doubt that, by interpreting Article 1105(1) to prescribe the customary international law minimum standard of treatment of aliens as the minimum standard of treatment to be afforded to investments of investors of another Party under NAFTA, the term ‘customary international law’ refers to customary international law as it stood no earlier than the time at which NAFTA came into force. It is not limited to the international law of the 19th century or even of the first half of the 20th century, although decisions from that period remain relevant. In holding that Article 1105(1) refers to customary international law, the FTC interpretations incorporate current international law, whose content is shaped by the conclusion of more than two thousand bilateral investment treaties and many treaties of friendship and commerce. …144

(p. 142) 5.68  Similarly, in connection with the interpretation of NAFTA Article 1105, the Pope & Talbot tribunal held that customary international law standards of treatment have ‘evolved since 1926, and, were the issue necessary to the Tribunal’s decision here, it would propose a formulation more in keeping with the present practice of states’.145 However, the tribunal did not apply its interpretation because it considered that even under the minimum standard of customary international law proposed by Canada (i.e. according to the standard pronounced in the Neer case), damages would be owing to the claimant.

5.69  The effect of well over 2,500 BITs on the development of customary international law cannot be underestimated. Nonetheless, it is still prudent to strike a balance between subsequent developments and the concerns of textualists, such as Judge Fitzmaurice, ever vigilant against importing terms into treaties and imposing obligations on parties that are not strictly evident on the face of the text.146

5.70  Evolutionary considerations are found also in the Tradex case. Although the issue there concerned the interpretation of a domestic Albanian law that provided foreign investors recourse to ICSID arbitration, the approach adopted is similar to the evolutionary treaty interpretation positions taken in Mondev and Pope & Talbot. The tribunal in Tradex said this:

The interpretation of Art. 8 and Art. 9 of the 1993 Albanian Law adopted by the tribunal appears confirmed by the developments in the Albanian investment laws of 1990, 1992 and 1993. Clearly, there has been a continuous evolution of such investment laws to assure a constantly better protection of the investments; this explains why an investment made in 1990 does not remain submitted to the rules in force at the moment in which it was made, but is subsequently submitted to the new rules. The succession in time of dispute settlement mechanisms is to be evaluated in such a context of progressive evolution.147

(p. 143) 5.71  As to the evolution of the standard of compensation to be paid by host States to investors, the CME tribunal commented that:

[t]he possibility of payment of compensation determined by the law of the host State or by the circumstances of the host State has disappeared from contemporary international law as it is expressed in investment treaties in such extraordinary numbers, and with such concordant provisions, as to have reshaped the body of customary international law itself.148

5.72  In contrast to the FIAT cases mentioned here, the LG&E tribunal considered a more rigid position was required when determining whether the essential security measures under the US–Argentina BIT were self-judging. The tribunal took the view that

[t]he provisions included in the international treaty are to be interpreted in conformity with the interpretation given and agreed upon by both parties at the time of its signature, unless both parties agreed to its modification.149

5.73  The tribunal concluded that only after the US–Argentina BIT was signed did the United States begin to consider its essential security measures treaty provisions as self-judging. On this basis, Argentina’s contention that the security measures provisions were self-judging was rejected.

F.  Principle of Effectiveness

5.74  The interpretative principle of effectiveness does not find express reference in the Convention Rules. It is widely accepted and deployed by international courts and tribunals and operates on the presumption that parties intended that all terms in their agreement had a purpose and that they did not intend any part of it to be ineffective.150 The principle found a place in Waldock’s Third ILC Report but was removed in subsequent ILC drafts.151 Ultimately (but not altogether convincingly) the ILC considered the principle to be embodied in Article 31(1), particularly (p. 144) within the parameters of the good faith and object and purpose criteria, which were considered as always striving to give meaning to the text.152

5.75  The principle of effectiveness has been frequently utilized by FIATs.153 The AAP award spoke of the principle in the following terms:

Nothing is better settled, as a canon of interpretation in all systems of law, than that a clause must be so interpreted as to give it a meaning rather than so as to deprive it of meaning. … This is simply an application of the more wider legal principle of ‘effectiveness’ which requires favouring the interpretation that gives to each treaty provision ‘effet utile’.154

5.76  The ad hoc Committee in Klöckner described the principle as a ‘customary principle of interpretation’.155 According to the Eureko tribunal, the principle of effectiveness was a ‘cardinal rule’ of treaty interpretation that required

that each and every operative clause of a treaty is to be interpreted as meaningful rather than meaningless. It is equally well established in the jurisprudence of international law, particularly that of the Permanent Court of International Justice and the International Court of Justice, that treaties, and hence their clauses, are to be interpreted so as to render them effective rather than ineffective.156

(p. 145) 5.77  In applying this principle, FIATs have required ‘full effect’ to be given to a provision157 or to the intentions of the parties.158 A variant of this practice has been the tendency of FIATs to avoid interpretations that deprive a treaty or its provisions of

  1. (a)  ‘any practical value’;159

  2. (b)  ‘any meaning’;160

  3. (c)  ‘any effect’;161

  4. (d)  ‘practical content’;162

  5. (e)  ‘practical applicability’;163 or

  6. (f)  ‘any semantic content or practical utility of its own’.164

5.78  Other FIATs have declined to make an interpretation that would render a provision

  1. (a)  invalid;165

  2. (b)  ‘useless’;166

  3. (c)  ‘pleonastic’;167

  4. (d)  ‘meaningless’;168 or

  5. (e)  ‘destroy the internal logic of [a provision] and render much of that provision superfluous’.169

5.79  On the whole, these pronouncements have been faithful to the principle as it operates in international law.

(p. 146) 5.80  FIATs have been conscious of the similarities between the object and purpose criterion and the principle of effectiveness170 and also the limitations of the principle as the Banro tribunal has commented:

The Tribunal is certainly aware of the general principle of interpretation whereby a text ought to be interpreted in the manner that gives it effect—ut magis valeat quam pereat. However, this principle of interpretation should not lead to confer, a posteriori, to a provision deprived of its object and purpose a result that goes against its clear and explicit terms.171

5.81  This warning reflects the caution expressed by the ICJ in the Interpretation of Peace Treaties that the principle of effectiveness will not be applied if it is contrary to the letter and spirit of the provisions subject to interpretation.172 Other concerns have been expressed about the principle from prominent international judges and lawyers. Fitzmaurice noted that the principle had teleological leanings. He was concerned ‘to keep it within bounds, to prevent it from leading to judicial legislation’.173 The view is taken in Oppenheim that

the absence of a full measure of effectiveness may be the direct result of the inability of the parties to reach agreement on fully effective provisions; in such a case the court cannot invoke the need for effectiveness in order to revise the treaty to make good the parties’ omission.174

5.82  More recently, the Renta 4 tribunal has indicated that effect need not be given to every word of a provision if that is not possible. In that case it held that ‘[t]he search (p. 147) to give meaning to the eight (or eleven) words that follow “relating to” in Article 10(1) [of the Spain–USSR BIT] simply cannot be allowed to deprive the remaining text of its essential positive meaning’.175 However, an interpretation that deprives certain words of their effect should not be made lightly.

5.83  Some commentators consider the principle of effectiveness as possessing two limbs. To them, the first limb—‘la règle de l’effet utile’—involves a presumption that individual provisions of the treaty must have been intended to have some use or effect and that an interpretation that reduces provisions to mere surplusage is suspect.176 The second—‘la règle de l’efficacité ’—is that the treaty as a whole (and, necessarily, each of its provisions) must be presumed to have a purpose, and an interpretation denying that end would attract doubt.177

5.84  On a number of occasions FIATs have explicitly employed ‘la règle de l’effet utile ’.178 No instance was found where an explicit application of ‘la règle de l’efficacité ’ was made by a FIAT. A reason for the lack of reference to this aspect of the effectiveness principle may be due to its potential overlap with the tribunal’s application of the Article 31(1) object and purpose criterion. Nonetheless, the Sempra Energy tribunal appeared (perhaps unconsciously) to be applying ‘la règle de l’efficacité ’ when it stated:

If the purpose of the Treaty and the terms of its provisions have the scope the parties negotiated and accepted, they could not now, as has been noted, be ignored by the Tribunal since that would devoid the Treaty of all useful effect.179

5.85  The FIAT awards cited in this section demonstrate the importance and widespread use of the principle of effectiveness in the international law of treaty interpretation. As useful as the principle is, care still must be exercised that it does not unduly override the text of the treaty.

(p. 148) G.  Legal Maxims

5.86  Legal maxims are not anachronistic phrases of a bygone age. They encapsulate reasoning or rules of logic that have survived centuries of use. Many of them continue to be relevant in a large number of today’s legal systems. An international tribunal’s use of maxims to interpret treaties could be justified on the grounds that they are relevant rules of international law under Article 31(3)(c) of the Vienna Convention or supplementary means of interpretation under Article 32.180 Wälde has expressed the view that classification of these rules within the confines of Article 32 may restrict their weight too much and the better approach may be to consider them as tools for identifying the ordinary meaning and context under Article 31(1) of the Vienna Convention.181

5.87  The ILC declined to include a reference to maxims in the Convention Rules as recourse to them was considered discretionary rather than obligatory. It observed that the interpretation of treaties by international tribunals at one time or another used almost every ‘maxim of which use is made in national systems of law in the interpretation of statutes’,182 and added:

Thus, it would be possible to find sufficient evidence of recourse to principles and maxims in international practice to justify their inclusion in a codification of the law of treaties, if the question were simply one of their relevance on the international plane. But the question raised by jurists is rather as to the non-obligatory character of many of these principles and maxims. They are, for the most part, principles of logic and good sense valuable only as guides to assist in appreciating the meaning which the parties may have intended to attach to the expressions that they employed in a document. Their suitability for use in any given case hinges on a variety of considerations which have first to be appreciated by the interpreter of the document. … Even when a possible occasion for their application may appear to exist, their application is not automatic but depends on the conviction of the interpreter that it is appropriate in the particular circumstances of the case.183

5.88  FIATs have provided several practical applications of legal maxims, particularly relating to the maxim generalia specialibus non derogant (i.e. what is general does not (p. 149) restrict (or detract from) what is special) or lex specialis.184 Take, for instance, SGS v Philippines. In that case, the BIT provision subject to interpretation prescribed that if consultations failed to resolve a dispute, the investor may submit the dispute either to the national courts of the host State or to ICSID or UNCITRAL arbitration. In contrast, Article 12 of the private agreement between SGS and the Philippines required, in part, that ‘[a]ll actions concerning disputes in connection with the obligations of either party to this Agreement shall be filed at the Regional Trial Courts of Makati or Manila’. The tribunal, in following the maxim generalia specialibus non derogant, held that it was not to be presumed that the general provision in the BIT ‘has the effect of overriding specific provisions of particular contracts, freely negotiated between the parties’.185 Similarly, in SPP (Jurisdiction No. 2), the tribunal used the maxim to determine that

[a] specific agreement between the parties to a dispute would naturally take precedence with respect to a bilateral treaty between the investor’s State and Egypt, while such a bilateral treaty would in turn prevail with respect to a multilateral treaty such as the Washington Convention.186

5.89  Importantly, the principle was deployed in the Fraport award to determine the relation of a BIT to Article 25 of the ICSID Convention. There, the tribunal held that the ‘language of the pertinent BIT … serves as a lex specialis with respect to Article 25 of the Washington Convention’.187 In his declaration in SGS v Philippines, Crivellaro contended that a qualification should be made to the generalia specialibus and lex posterior maxims:

I doubt whether the Roman adagios generalia specialibus non derogant and lex posterior deragat legi priori (paras. 141 and 142 of the Decision) may be extended to the comparison between a treaty and a contract. As the Decision admits in respect of the second maxim (but the same should also apply to the first maxim), for such maxims to apply it is required that the two instruments have the same legal character and, I would add, be made by or be directly binding on the same parties.188

(p. 150) 5.90  No authority is cited for this call for equivalence in legal character. No comment was elicited by the majority on this issue. This limitation imposed on the two maxims in the declaration has not been adopted in other FIAT decisions.

5.91  In SGS v Pakistan the tribunal held that the claimant’s interpretation of the umbrella clause at issue was exceedingly expansive. It took the view that the clause required a considerably higher degree of specificity before it could accept such an interpretation. The tribunal’s conclusion was that the interpretative approach to be adopted was summed up by the maxim in dubio mitius (i.e. the interpretation of treaties in deference to the sovereignty of states).189 The Eureko tribunal criticized this finding on the following grounds:

This Tribunal feels bound to add that reliance by the Tribunal in SGS v Pakistan on the maxim in dubio mitius so as effectively to presume that sovereign rights override the rights of a foreign investor could be seen as a reversion to a doctrine that has been displaced by contemporary customary international law, particularly as that law has been reshaped by the conclusion of more than 2000 essentially concordant bilateral investment treaties.190

5.92  The doctrine of interpreting treaties restrictively is discussed in Chapter 6, Section H infra.

5.93  The expressio unius est exclusio alterius rule (i.e. if something is expressed it must be taken to exclude something else) was applied by the Waste Management tribunal in the following way:

Where a treaty spells out in detail and with precision the requirements for maintaining a claim, there is no room for implying into the treaty additional requirements, whether based on alleged requirements of general international law in the field of diplomatic protection or otherwise.191

5.94  The flaws that may be associated with the type of logic shown here were exposed in Siemens, where the tribunal took the view that

Article 4 is the only article in the [1991 Germany–Argentina BIT] that deals with compensation in case of expropriation and of war or civil unrest. If the Treaty should be interpreted as alleged by Argentina—by excluding from its application every specific situation that has not been included—, we would be bound to reach the conclusion that, in cases of discrimination, arbitrary measures, or treatment short of the just and equitable standard, there would not be a right to compensation under the Treaty—an unlikely intended result by the Contracting Parties given the Treaty’s purpose. If a matter is dealt with in a provision of the Treaty and not specifically mentioned under (p. 151) other provisions, it does not necessarily follow that the other provisions should be considered to exclude the matter especially covered.192

5.95  This inherent problem in the expressio unius maxim calls for care to be exercised if it is employed. An injudicious application may lead to an unwarranted limitation on the application of a treaty.193

5.96  Other maxims applied or discussed by FIATs include ejusdem generis (i.e. of the same genus);194 ut res magis valeat quam pereat (i.e. let the thing stand rather than fall);195 and lex posterior derogat legi priori (i.e. a later law overrides an earlier one).196

5.97  The jurisprudence referred to previously shows that FIATs have helped provide a practical context for the maxims and enhance the understanding of how those maxims operate on the international law plane. However, none are expressly included in the Convention Rules; they must be considered as guides, not rules that must be followed slavishly.197(p. 152)


Sinclair, The Vienna Convention on the Law of Treaties (1984), at 153. See also Gardiner, Treaty Interpretation (2008), at 6 (commenting that the Convention Rules are not ‘an exclusive compilation of guidance on treaty interpretation, other skills and principles that are used to achieve a reasoned interpretation remaining admissible to the extent not in conflict with the [Convention Rules]’).

See, e.g. the numerous awards and decisions made available on websites such as <http://icsid.worldbank.org>; <www.investmentclaims.com>; and <http://ita.law.uvic.ca>.

See Weeramantry, supra note 2, at 112–3.

See United States Diplomatic and Consular Staff in Tehran, Judgment, ICJ Reports (1980), at 18, para. 33; and Interpretation of Agreement of 25 March 1951 between the WHO and Egypt, ICJ Reports (1980), at 87, para. 33. Judge Shahabuddeen observed in his Lauterpacht Lectures that the Court’s jurisprudence ‘has developed in the direction of a strong tendency to adhere closely to previous holdings’. Shahabuddeen, supra note 6, at 238.

ICJ Reports 275 (1998), at 292, para. 28.

10  Rosenne, The Law and Practice of the International Court, 1920–2005, Volume III: Procedure (2006), at 1553. A similar respect for prior decisions is held by the WTO Appellate Body. As an example, it stated in the Shrimp Turtle II case that ‘[a]dopted panel reports are an important part of the GATT acquis. … They create legitimate expectations among WTO Members and, therefore, should be taken into account where they are relevant to any dispute’. United States Import Prohibition of Certain Shrimp and Shrimp Products, WTO Appellate Body Report, 12 October 1998, WT/DS58/AB/R, at para. 108.

11  Bernhardt, supra note 7, at 1244.

12  In Methanex (Amicus Decision), at para. 51, the claimant argued that by interpreting Article 15(1) of the UNCITRAL Arbitration Rules to permit the acceptance of amicus submissions, the tribunal would be setting a precedent for other tribunals. The response of the tribunal was to state that it ‘can set no legal precedent, in general or at all. It has no power to determine for other arbitration tribunals how to interpret Article 15(1)’.

13  See SGS v Philippines, at para. 97. See also AES, at para. 23(d); and Enron (Ancillary Claim), at paras 25–6.

14  El Paso, at 39. See also Amco (Jurisdiction), at para. 14; Bayindir, at para. 76; and AES (Jurisdiction), at paras 30–2. As to NAFTA cases, see, e.g. Waste Management, dissenting opinion of Keith Highet, introductory (unnumbered) paragraph (‘[t]he precedential significance of this Award for future proceedings under the North American Free Trade Agreement (NAFTA) cannot be underestimated’); and Methanex (Partial Award), at para. 141. Correspondingly, a prior tribunal’s rejection of interpretations submitted by a party may also influence subsequent tribunals. See, e.g. Sempra Energy (Jurisdiction), at para. 145.

15  Saipem (Jurisdiction), at para. 67 (emphasis added). The identical words were repeated in the Saipem (Award), at para. 90. Very similar words were used in Duke v Ecuador, at para. 117. See also ADC, at para. 293 (‘cautious reliance on certain principles developed in a number of those cases, as persuasive authority, may advance the body of law, which in turn may serve predictability in the interest of both investors and host States’); and Wälde, ‘Investment Arbitration under the Energy Charter Treaty: An Overview of Key Issues’, 1 Transnational Dispute Management (2004), Issue 1 (‘The reasoning of almost all modern arbitral awards demonstrate the great care investment arbitral tribunals apply to ensure they are positioned in the mainstream of emerging jurisprudence. While there is no formal “stare decisis” rule, there is a de facto and very strong pressure on each tribunal to heed what other tribunals have done with identical or very similar legal language. This does not necessarily prevent contradictory awards …’); and Lowenfeld, ‘Investment Agreements and International Law’, 42 Columbia Journal of Transnational Law 123 (2003–04), at 128 (‘[b]ut since so many of the recent treaties are so much alike, it is proper in a BIT case for arbitrators who are called upon to construe terms such as “fair and equitable treatment,” “adequate compensation,” or “equal access” to draw on the awards in similar disputes under similar treaties’).

17  SGS v Philippines, at para. 97 (footnotes omitted).

18  See, e.g. SGS v Philippines, at paras 111 and 157.

19  See, e.g. Azurix (Jurisdiction), at para. 73 (the tribunal held it ‘concurs with decisions of tribunals that have interpreted the same provision in the same BIT or similar provisions in other BITs to which Respondent is a party’); LG&E (Jurisdiction), at para. 74; and Lanco, at para. 43. See also Kaufmann-Kohler, ‘Interpretation of Treaties: How do Arbitral Tribunals Interpret Dispute Settlement Provisions Embodied in Investment Treaties?’, in Mistelis and Lew, Pervasive Problems in International Arbitration (2006), at 258 (observing that ‘all investment treaties protect investments by granting investors rights which are materially identical or comparable’).

20  See, e.g. AES, Azurix, LG&E, Enron, CMS, and Lanco arbitrations, which all invoked the 1991 Argentina–US BIT. In this regard, see particularly AES, at para. 29.

21  For a detailed discussion on prominent FIAT cases that have not followed prior FIAT awards on the same or similar provisions, see Franck, ‘The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions’, 73 Fordham Law Review 1521 (2004). See also Gill, ‘Inconsistent Decisions: An Issue to be Addressed or a Fact of Life?’, in Ortino, Sheppard, and Warner (eds), Investment Treaty Law: Current Issues Volume I (2006), at 27 (inconsistent decisions in investment arbitration is no means unique, ‘domestic courts reach inconsistent decisions on a regular basis … the thousands of domestic, consumer and industry-specific arbitrations undoubtedly give rise to inconsistent decisions from time to time’).

22  See, e.g. Amco (Annulment), at paras 22 and 44; CMS (Jurisdiction), at paras 70–6; Vacuum Salt, at 337, para. 29, n. 9; Lanco, at para. 46; LG&E, at paras 77–8; Vivendi (Jurisdiction), at para. 94; Bayindir, at paras 96, 98, 102 and 128; Olguín (Jurisdiction), at para 26; Eureko, at para. 186; Gas Natural, at para. 36, 39, and 52; Azurix (Jurisdiction), at para. 73; AES, at paras 30 and 32; and Tokios (Jurisdiction), at para. 91. By way of interest, a glimpse of the extent to which some tribunals cite other ICSID cases is provided by a perusal of the endnotes in Enron (Ancillary Claim). See also Schreuer, supra note 3, at 617 (noting that the early years of the ICSID Convention, references to ICSID decisions was scant due to ‘their relatively small number and the difficulty in gaining access to them’). See also Weeramantry, supra note 2, at 115–9; and Wälde and Weiler, ‘Investment Arbitration under the Energy Charter Treaty in the Light of the new NAFTA Precedents: Towards a Global Code of Conduct for Economic Regulation’, in Kaufmann-Kohler and Stucki (eds), Investment Treaties and Arbitration, Swiss Arbitration Association Conference, 25 January 2002 (ASA Special Series No. 19) 159, at 166 (in the context of interpreting the ECT).

23  Mitchell (Enforcement Stay), at para. 23; Metalclad (Award), at para. 108; and Methanex (Final Award), at Part II, Chap. B, para. 6 (‘the Tribunal may remain open to persuasion based on legal reasoning developed in the GATT and WTO jurisprudence’).

24  Amco (Annulment), at para. 44.

25  LETCO (Award), 2 ICSID Reports 346, at 352.

26  Amco (Annulment), at para. 44.

27  AES (Jurisdiction), at para. 28 (‘if the basis of jurisdiction for these other tribunals and/or the underlying legal dispute in analysis present either a high level of similarity or, even more, an identity with those met in the present case, this Tribunal does not consider that it is barred, as a matter of principle, from considering the position taken or the opinion expressed by these other tribunals’); and Mitchell (Enforcement Stay), at para. 23.

28  Enron (Jurisdiction), at para. 40.

29  Bayindir, at 76 (finding that ‘it is not bound by earlier decisions, but will certainly carefully consider such decisions whenever appropriate’); and Plama (Jurisdiction), at para. 218.

30  Lucchetti (Award), at para. 48.

31  Tokios (Jurisdiction), at para. 42.

33  See, e.g. Renta 4, at para. 16; Plama (Jurisdiction), at para. 210; Mitchell (Enforcement Stay), at para. 23; El Paso, at para. 39; Enron (Ancillary Claim), para. 33; Bayindir, at paras 73–6; and Azurix (Jurisdiction), at para. 73.

34  Feldman (Award), at para. 107.

35  See, e.g. Klöckner (Annulment), paras 59–61; and Lucchetti (Award), at para. 48. See also Schreuer, supra note 3, at 616, n. 233.

36  See, e.g. Mondev, at para. 43; Klöckner (Annulment), at paras 59–61; Amco (Annulment), at paras 41 and 44; Lucchetti (Award), at para. 48; Siemens (Jurisdiction), at paras 94–103; Casado (Provisional Measures), at paras 19–20; and Plama (Jurisdiction), at paras 213–7. See also Schreuer, supra note 3, at 616, n. 233.

37  Loewen (Jurisdiction), at para. 45. See also Schreuer, supra note 3, at 616, n. 234.

38  See, e.g. Lauder, at para. 200. Contrast Mondev, at para. 69, n. 17. See also Schreuer, supra note 3, at 616, n. 235.

39  CME (Partial Award), at para. 608; Loewen (Jurisdiction), at para. 45; Tokios (Jurisdiction), at para. 92; Casado (Provisional Measures), at paras 22–3. See also Schreuer, supra note 3, at 616, n. 236.

40  Klöckner (Annulment), at paras 59–61. AAP, at para. 40 referred to a number of early international arbitral awards in support of its summation of treaty interpretation rules. See also Schreuer, supra note 3, at 616, n. 236.

41  WTO references tend to be made more often by NAFTA tribunals than other FIATs. In Methanex (Partial Award), at Part II, Chap. B, para. 6, the tribunal held that when it comes to interpreting certain provisions of the NAFTA Chapter 11:

the Tribunal may derive guidance from the way in which a similar phrase in the GATT has been interpreted in the past. Whilst such interpretations cannot be treated by this Tribunal as binding precedents, the Tribunal may remain open to persuasion based on legal reasoning developed in GATT and WTO jurisprudence, if relevant.

See also ADF (Award), at para. 147; Pope & Talbot (Merits, Phase 2), at paras 46–63; and SD Myers (Partial Award), at paras 244 and 291. As regards, reference to WTO case law by ICSID tribunals, see SGS v Pakistan, at para. 171, n. 178. In Occidental Exploration (Award), at paras 174–5, a non-NAFTA investment arbitration, WTO case law was distinguished and not followed.

42  See, e.g. SD Myers (Partial Award), at para. 249, n. 44 (referring to a line of decisions emanating from the Supreme Court of Canada in its interpretation of NAFTA Article 1102 and commenting that ‘[a]lthough domestic law is not controlling in Chapter 11 disputes, it is not inappropriate to consider how the domestic laws of the parties to the dispute address an issue’); and Dr Asante’s dissenting opinion in AAP (Dissenting Opinion), at 306 (referring to an English court decision (Adams v Naylor [1946] 2 All ER 241) and commenting that the case was ‘certainly not binding in this arbitration’ but it ‘may be instructive’). In contrast, see Enron (Ancillary Claim), at para. 39, in which the tribunal, in its interpretation of the Argentina–US BIT, was reluctant to rely on the view taken on direct and indirect share ownership in the US Supreme Court decision in Dole Food Co. v Patrickson, 123 S. Ct. 1655 (2003). In Duke v Ecuador, at para. 183, reference was made to the English Queen’s Bench judgment in Ecuador v Occidental Exploration & Production Co (No. 2) [2006] EWHC 345 (Comm).

43  National Grid (Award), at para. 144.

44  Continental Casualty (Award), at para. 192.

45  Wälde, supra note 16, at 772. This point is also discussed at paras 35 and 44 supra.

46  Fauchald, supra note 2, at 335.

47  See, e.g. Commission, supra note 2, at 158 (‘The role that precedent has come to play in investment treaty arbitration today resembles the common law doctrine of stare decisis absent certain of the associated values advanced in a common law system of precedent.’); Cheng, ‘Precedent and Control in Investment Treaty Arbitration’, 30 Fordham International Law Journal 1014 (2007), at 1044 (concluding on the basis of examined FIAT decision trends that a credible hypothesis may be made that ‘in spite the absence of rules of precedent in investment treaty arbitration, there is a strong—albeit imperfect and informal—norm of accounting for prior relevant awards and providing reasons for following or departing from those awards.’); and Schreuer and Weiniger, ‘A Doctrine of Precedent?’, in Muchlinski, Ortino, and Schreuer, The Oxford Handbook of International Investment Law (2008), at 1196 (‘a de facto practice of precedent certainly exists. However, it is not identical to that prevailing within domestic common law systems’).

48  Wälde, ‘The Specific Nature of Investment Arbitration’, in Kahn and Wälde (eds), New Aspects of International Investment Law (2007), at 105–6. See also AES (Jurisdiction), at para. 22 and Schreuer (2006), supra note 31, at 15 (‘an application for annulment that alleges an excess of powers or a failure to state reasons because the tribunal has simply relied on earlier decisions without making an independent decision or developing its own reasons is entirely possible’).

49  See particularly Amco (Annulment), at para. 44. See also Fauchald, supra note 2, at 335. But see Loewen (Jurisdiction), at para. 49 (reference to the ratio); and Renta 4, at paras 23 and 95 (reference to obiter).

50  Commission, supra note 2, at 156.

51  See, e.g. Aguas del Tunari, at para. 288; Enron (Ancillary Claim), at paras 35 and 43–45; Plama (Jurisdiction), at paras 210, 211, and 224; Sempra Energy (Jurisdiction), at para. 145 and 155; and SGS v Philippines, at para. 110. See also the CME (Final Award), at para. 432, where the tribunal stated that although that arbitration and the Lauder arbitration concerned the same investment and the respective BITs at issue in each case granted similar investment protections, the bilateral treaties were not identical. As a consequence, res judicata was held to be not applicable as between the CME and Lauder tribunals.

52  Enron (Ancillary Claim), at para. 45. See also Occidental Exploration (Award), at para. 57 (‘what ultimately matters is that every solution must respond to the specific circumstances of the dispute submitted and the nature of such dispute’); and Canadian Cattlemen, at para. 209.

53  Quoted in AES (Jurisdiction), at para. 22. See also Schreuer (2006), supra note 31, at 15.

54  Wälde and Weiler, supra note 22, at 166, n. 19.

55  See Plama (Jurisdiction), at para. 217.

56  Aguas del Tunari, at para. 288.

57  Sempra Energy (Jurisdiction), at para. 147 (emphasis added).

58  Pellet, ‘Article 38’, in Zimmermann et al. (eds), The Statute of the International Court of Justice: A Commentary (2006), at 784. However, one cannot underestimate the role of judicial or arbitral decisions in shaping international law. As Professor Shaw observes, many decisions of international arbitral awards ‘have been extremely significant in the development of international law’. Shaw, International Law (2003), at 104. See also Pellet, ibid., at 789–90 (citing instances of ‘the deep influence that the [ICJ] has exercised on the evolution of international law’).

59  Methanex (Partial Award), at para. 141. See also Enron (Jurisdiction), at para. 40 (‘decisions of ICSID or other arbitral tribunals are not a primary source of rules’).

60  Canadian Cattlemen, at para. 50.

62  Out of the 98 ICSID decisions analysed in that study, 90 used ICSID case law in their interpretative arguments, and another 30 decisions referred to other tribunal decisions that used the UNCITRAL Arbitration Rules. The next most frequently used type of interpretative criterion was ‘legal doctrine’, e.g. the writings of scholars and other commentaries, which were referred to in seventy-three decisions. See Fauchald, supra note 2, at 356.

64  Brown, A Common Law of International Adjudication (2007), at 226–34. But also see his discussion of limitations to courts’ abilities to develop this ‘common law’, ibid., at 234–7.

66  See also Article 31(3) of the Vienna Convention, which addresses the use of subsequent agreements or practice in treaty interpretation.

67  Fauchald, supra note 2, at 345.

68  Wälde, supra note 16, at 777.

70  Gardiner, supra note 1, at 282.

71  AAP, at para. 40, quoting the Elton case, Mexico–US General Claims Commission of 1929, Repertory, Vol. II, p. 35. As regards the reference to similar provisions in other treaties (or past interpretations of such provisions) by other international law courts and tribunals see Gardiner, supra note 1, at 345; and Phillimore, Commentaries upon International Law (1882), Vol. II, at 98 (referring to ‘[t]he rule of instituting a comparison between the Treaty in dispute and other treaties, whether prior, posterior, or contemporary, upon the same subject and between the same parties’). In the Oil Platforms case, ICJ Reports 161 (2003), at para. 20 the ICJ was required to interpret Article XX of the 1955 Treaty of Amity, Economic Relations and Consular Rights between the United States and Iran. In interpreting this provision, the Court saw no reason to depart from its interpretation in the Nicaragua case (ICJ Reports 14 (1986), at 116, para. 222 and 136, para. 271) of an identical provision in the 1956 Treaty of Friendship, Commerce and Navigation between the United States and Nicaragua. A factor that should be kept in mind when this decision is contrasted with FIAT awards is that the prior interpretation was made by the same court. This is highly unlikely to happen in the case of FIATs, which comprise a different panel of arbitrators for each case (save for rare exceptions such as Camuzzi and Sempra Energy or where the case is consolidated (see, e.g. Canfor (Consolidation Order)).

72  See, e.g. Aguas del Tunari, at paras 292–3 (‘[m]ost relevant to an assessment of state practice possibly bearing on the 1992 Bolivia-Netherlands BIT are those BITs which were negotiated contemporaneously in the early 1990s’); SPP (Jurisdiction No. 2), at paras 102–3; Fedax (Jurisdiction), at para. 34 (referring generally to ‘most contemporary bilateral treaties’ in interpreting the BIT at issue); Gruslin, at para. 21.4; SGS v Philippines, at para. 132(e); Fedax (Jurisdiction), at para. 27 (referring to the Convention Establishing the Multilateral Investment Guarantee Agency 1985 interpreting the ICSID Convention); National Grid (Jurisdiction), at para. 85; and Enron (Jurisdiction), at para. 47 (‘[t]here is no evidence in this case that the intention of the parties to the Argentina–United States Bilateral Treaty might be different from that expressed in other investment treaties’). See also Schreuer, supra note 3, at 289 (commenting that as long as they are reasonable, national legislation or treaty-based definitions should play a prominent role in the interpretation of Article 25(2)(b)).

In SPP (Jurisdiction No. 2), at paras 78–82, Egypt argued that to translate the term ‘tatimmu’ to mean ‘shall be’ was overstating the nature of the Arabic term. In the tribunal’s interpretation of that term—found in Arabic legislation that was deemed to be an acceptance of ICSID jurisdiction—it was noted that those two terms were used interchangeably in certain treaties concluded by Egypt in both the Arabic and English languages. See also its references to other treaties ibid., at para. 103.

73  See, e.g. Methanex (Partial Award), at para. 140 (interpretation of NAFTA supported by reference to the 1958 New York Convention); Bayindir, at para. 113 (other BITs used to confirm interpretation of BIT at issue); Maffezini (Jurisdiction), at para. 68 (the relevant BIT provisions ‘complement and are consistent with’ the ICSID Convention); Tokios (Jurisdiction), at para. 42 (‘the definition of corporate nationality in the Ukraine–Lithuania BIT, on its face and as applied to the present case, is consistent with the [ICSID] Convention and supports our analysis under it’) and para. 79 (‘[t]he Tribunal’s finding under the BIT is also consistent with the ICSID Convention’); Brownlie’s separate opinion in CME (Final Award), paras 17–8 (BIT interpretation supported by reference to 1975 Final Act of the Helsinki Conference on Security and Co-operation in Europe); and LESI, at Part II, para. 25(ii) (other BITs and Energy Charter Treaty used to confirm BIT at issue).

74  Pope & Talbot (Merits, Phase 2), at paras 110–1 and 115. See also Professor O’Connell, who remarked that where a treaty subject to interpretation ‘forms part of a system of treaties it is permissible to interpret it in the light of the other treaties’. O’Connell, International Law (1970), at 260.

75  See, e.g. Enron (Jurisdiction), at para. 46 (‘Each instrument must be interpreted autonomously in the light of its own context.’).

76  See, e.g. Methanex (Final Award), at Part II, Chap. B, para. 16; and Occidental Exploration (Award), at paras 174–6. But see Pope and Talbot (Merits, Phase 2), at paras 45–63, 68–9.

77  Methanex (Final Award), at Part II, Chap. B, para. 16, quoting MOX Plant case (Ireland v United Kingdom), Provisional Measures, Order of 3 December 2001, ITLOS Reports 95 (2001), at 106, para. 51, 41 ILM 405 (2002), 413.

78  Similarly, in cases before the ICJ where the terms of another treaty or instrument are not identical but similar, the Court appears to exercise more caution. In the Anglo–Iranian Oil case, ICJ Reports 93 (1952), at 105, the United Kingdom asserted that Iran’s unilateral declaration of its acceptance of the compulsory jurisdiction of the ICJ should be interpreted on the basis that the clause at issue was copied from a unilateral declaration adopted by Belgium. This Belgian document had also been adopted by numerous other States. Accordingly, the United Kingdom asserted that the clause in the Iranian declaration should be understood in the same sense as the Belgian formula. The Court did not accept the UK argument because there was an interpolation of certain words in the Iranian declaration that altered the Belgian formula to such an extent that it was impossible to seek the ‘real meaning’ of the Iranian declaration in the Belgian formula.

79  Aguas del Tunari, at paras 310–4; Salini v Jordan (Jurisdiction), at paras 116–8; and CSOB (Jurisdiction), at para. 57. See also Loewen (Award), at para. 235 (the only relevance of the ICSID Convention to that proceeding was that the parties had elected to function under its structure and that election could not be used to change or supplement the substance of the NAFTA Treaty).

80  Aguas del Tunari, at para. 299.

81  Sempra Energy (Jurisdiction), at para. 144. Contrast Plama (Jurisdiction), at para. 195 (‘treaties between one of the Contracting Parties and third States may be taken into account for the purpose of clarifying the meaning of a treaty’s text at the time it was entered into’) and para. 155.

82  Tza Yap Shum, at para. 109.

83  Methanex (Final Award), at Part IV, Chap. B, at paras 29–30.

84  Gruslin, at para. 21.6.

85  Aguas del Tunari, at paras 309–14. Contrast Maffezini (Jurisdiction), at paras 58–60.

86  Siemens (Jurisdiction), at para. 106. See also Aguas del Tunari, at para. 298 (recognizing ‘the need for care’ in assessing differences between a treaty and a model BIT) and paras 310–1; Enron (Jurisdiction), at para. 46; Pan American (Preliminary Objections), at para. 108; El Paso, at para. 80; CMS (Award), at para. 368; and Plama (Jurisdiction), at para. 204. See also Pope & Talbot (Merits, Phase 2), at paras 110–1.

87  Schreuer (2006), supra note 31, at 8–9.

88  See, e.g. Eureko, at paras 187–8 (referring to the ILC’s work on state responsibility); Loewen (Jurisdiction), at para. 47 (referring to the ILC’s work on state responsibility); and Methanex (Partial Award), at para. 99 (referring to the ILC work on the law of treaties). The pronouncements of the ILC may also be indicative of international law rules dealt with in Vienna Convention Article 31(3)(c).

89  American Manufacturing, Arbitrator Golsong’s separate opinion, at para. 13.

90  World Duty Free Company, at para. 139.

91  See, e.g. Tokios (Jurisdiction), at para. 2. This draft convention is published in Sohn and Baxter, ‘Responsibility of States for Injuries to the Economic Interests of Aliens’, 55 AJIL 545 (1961).

92  See, e.g. Casado (Provisional Measures), at para. 1 (noting that because Article 47 of the ICSID Convention was ‘directly inspired’ by Article 41 of the ICJ Statute, ‘particular importance’ could be accorded to the judgments of the ICJ and PCIJ); and Goetz (Award), at para. 54 (noting that because Article 45(1) of the ICSID Convention, and Article 42 of the ICSID Arbitration Rules were inspired by Article 53 of the ICJ Statute, it was ‘appropriate to refer to the principles enunciated by the International Court of Justice’).

93  SPP (Dissent - 1992), at 254–5.

94  Lauder, at para. 292; and Gruslin, at paras 21.3 and 21.6.

95  Saluka (Award), at para. 284, n. 18; Pope & Talbot (Merits, Phase 2), at para. 78, n. 73. In that same case, reference was made to the 1968 OECD Draft Convention on the Protection of Foreign Property but this was distinguished and not relied on by the tribunal. Pope & Talbot (Merits, Phase 2), at para. 112.

96  Fedax (Jurisdiction), at para. 35.

97  Feldman (Award), at paras 99, 105–7.

98  See, e.g. Bayindir, at para. 129. This type of situation may extend beyond the contours of treaty interpretation and may be seen equally as an instance of the preclusive effects of a prior statement (see, e.g. Phillips Petroleum Company Iran v Iran, 21 Iran–US CTR 79 (1989), at para. 207) or a rule of evidence that contradictory statements or conduct of an interested party should be construed against that party (see, e.g. Woodward-Clyde Consultants v Iran, 3 Iran–US CTR 239 (1983), at pp. 246–7; and Weeramantry, ‘Estoppel and the Preclusive Effects of Inconsistent Statements and Conduct: The Practice of the Iran–United States Claims Tribunal’, 27 Netherlands Yearbook of International Law 265 (1996)). See also, Sempra Energy (Jurisdiction), at paras 132 and 143–5; Methanex (Partial Award), at para. 145; and Enron (Jurisdiction), at para. 47.

99  See, e.g. Aguas del Tunari, at paras 271 and 294; Mondev, at paras 111–2; CMS (Award), at paras 362, 369; Generation Ukraine (Award), at paras 15.4–15.7; Feldman (Award), at para. 181; and Ethyl (Jurisdiction), at para. 84, n. 32. In Gruslin, materials submitted by Malaysia to support its interpretation included a memorandum from the Malaysian Ministry of Trade and Industry to the Malaysian Attorney-General’s Chambers and records of interview of Malaysian officials involved in the negotiations of the investment treaty at issue. The tribunal expressed a reluctance to allow extrinsic materials to colour the meaning of particular terms of the investment treaty. The tribunal observed that if the meaning of the phrase ‘is found to be clear, the Tribunal will not reduce its reach by reference to general considerations or assumptions derived from extrinsic sources of the sort relied upon by the Respondent in its materials and arguments’. Gruslin, at paras 17.1 and 21.4–21.6. Similarly see Berschader (Award), at para. 158.

100  See, e.g. Maffezini (Jurisdiction), at paras 58–60; Lanco, at para. 32; Aguas del Tunari, at paras 310–4; Sempra Energy (Jurisdiction), at paras 144–5; and Camuzzi (Jurisdiction), at para. 134. This type of practice may also relate to Article 31(3). The Aguas del Tunari award, at para. 293, considered ‘[m]ost relevant to an assessment of state practice possibly bearing on the 1992 Bolivia–Netherlands BIT are those BITs which were negotiated contemporaneously in the early 1990s’.

In SPP (Jurisdiction No. 2), at para. 110, the tribunal made reference to a non-disputing State’s (Senegal) treaty practice.

101  Mondev, at para. 111.

102  Mondev, at para. 112. See also CMS (Award), at para. 362.

103  Approved on 18 March 1965 by the directors at the same time they approved the final text of the Convention. See History of the ICSID Convention, Vol. II, Part 2, p. 1039. The Report was transmitted to governments at the same time the text of the Convention was submitted for signature and ratification, acceptance or approval. History of the ICSID Convention, Vol. II, Part 2, p.1086.

104  See, e.g. Kaiser Bauxite, at para. 17; Bayindir, at para. 125; LESI, at Part 2, para. 13; CSOB (Jurisdiction), at para. 63; Wena Hotels (Jurisdiction), at 85; Lanco, at paras 42–3; Tokios (Jurisdiction), at para. 99; and Azurix (Jurisdiction), at para. 58. See also Fauchald, supra note 2, at 330.

105  See, e.g. Banro, at para. 20, which suggests that the preparatory work is distinct from the Report of the Executive Directors. The tribunal there first referred to the preparatory work in making its point and then said this was ‘also confirmed’ by the Report.

106  See, e.g. Goetz (Award), at para. 83, in which the tribunal immediately referred to the Report’s definition of ‘legal dispute’ in Article 25 of the ICSID Convention, prior to any other Article 31 analysis. Similarly, see Fedax (Jurisdiction), at 189, para. 15 and Joy Mining, at para. 42.

107  See SPP (Jurisdiction No. 2), at paras 112 and 115.

108  Wälde and Weiler, supra note 22, at 166, n. 19.

109  Aguas del Tunari, at para. 291.

110  TSA Spectrum (Award), at para. 101.

111  See, e.g. Articles 36–63 of the ICSID Convention. In this regard Article 44 provides ‘[a]ny arbitration proceeding shall be conducted in accordance with the provisions of this Section and, except as the parties otherwise agree, in accordance with the Arbitration Rules in effect on the date on which the parties consented to arbitration’.

112  Indeed, interpreters must be mindful that the ICSID Arbitration Rules are subordinate to the ICSID Convention and in the event of a conflict between the two instruments, the Convention should prevail. See, e.g. Vivendi (Challenge), at para. 10 and Schreuer, supra note 3, at 677.

113  See, e.g. Amco (Annulment), at para. 37; Amco (Resubmitted Jurisdiction), 1 ICSID Reports 543, at 567; and MINE (Annulment), at para. 4.07. Fedax (Jurisdiction), at para. 28 referred to the ICSID Rules Governing the Additional Facility in interpreting the ICSID Convention. See also Aguas Provinciales, at para. 14 (referring to the UNCITRAL Arbitration Rules in interpreting the ICSID Convention); Vacuum Salt, at 337, para. 29, n. 9; and Guinea v Atlantic Triton Company Limited, 26 October 1984, Cour d’appel, Rennes (Second Chamber), 3 ICSID Reports 3, at 8 (this judgment was overturned by the Cour de Cassation).

114  See Articles 4(1) and 6(1)(c) of the ICSID Convention.

115  ADF (Award), at para. 144. See also Methanex (Partial Award), at paras 125–6 (referring to the ICJ Rules of Procedure in its interpretation of the UNCITRAL Arbitration Rules).

116  Vivendi (Challenge), at para. 5. See also Schreuer, supra note 3, at 1042; and Chapter 6, Section J infra.

117  Vivendi (Challenge), at paras 10–12 (footnotes omitted).

118  See supra, Chapter 3, Section D supra.

119  Article 38 of the ICJ Statute (emphasis added). For a discussion on judicial decisions as a subsidiary means for determining rules of law within the context of Article 38 of the ICJ Statute, see Section B supra.

120  Pellet, supra note 58, at 1558–9. This is not the case in respect of individual opinions of ICJ judges. See ibid., at 791–2.

123  See generally, Schreuer, supra note 3, at 617.

124  Fauchald, supra note 2, at 356.

125  See, e.g. Pellet, supra note 58, at 791.

126  Wena Hotels (Jurisdiction), at 82. See also Aguas del Tunari, at para. 283 (taking the view that among other things ‘scholarly commentary’ indicated that the drafters of the ICSID Convention intended a flexible definition of ‘control’ in Article 25); Amco (Annulment), at para. 22 and at para. 34; and Pope & Talbot (Merits), at para. 113.

127  See, e.g. Goetz (Award), at para. 67, n. 14; Tokios (Jurisdiction), at paras 22, 26, 42, 88, 94, 98, 106; Impregilo v Pakistan, at paras 108, n. 57 and 133; Joy Mining, para. 53, n. 18; Aguas del Tunari, at para. 281; Fedax (Jurisdiction), at para. 21; Tokios (Opinion), at para. 19; Wena Hotels (Jurisdiction), at 82–3; Bayindir, at 127; Methanex (Partial Award), at para. 107, n. 8; Maffezini (Jurisdiction), at para. 31, n. 6, para. 74, n. 54, and para. 94, n. 69; Mitchell (Enforcement Stay), at para. 41, n. 14; CSOB (Jurisdiction), at para. 38, n. 7; Duke v Peru, at para. 130; LG&E (Liability), at para. 85, n. 3; Vivendi (Annulment), at paras 62, 64, 66, and 86; Vivendi (Challenge), at para. 12; LESI (Award), at para. 8(i); Aucoven (Jurisdiction), at para. 51; Aucoven (Award), at para. 91; Banro, at para. 19; ADF (Award), at para. 144, n. 151; AES, at para. 40, n. 18; Camuzzi (Jurisdiction), at para. 20, n. 7; MTD (Stay of Execution), at para. 27, n. 1; Salini v Jordan (Award), at para. 102; Sempra Energy (Jurisdiction), at para. 30, n. 7; SGS v Pakistan, at para 48, n. 48; SGS v Philippines, at para. 29, n. 5; Siemens (Jurisdiction), at para. 169, n. 157; and Wena Hotels (Interpretation Decision), at para. 81, n. 60. Some of these references are to Professor Schreuer’s contributions to Volumes 11 to 15 of the ICSID Review-Foreign Investment Law Journal (1996–2000). Altogether, these ICSID Review volumes contained eight articles by Professor Schreuer, which collectively represent a significant part of Schreuer’s Commentary. See Schreuer, supra note 3, Preface, at xviii.

128  In Vacuum Salt, at para. 37, he was described as ‘the acknowledged authority’ on the ICSID Convention. See, also Fedax (Jurisdiction), at para. 21; Aucoven (Jurisdiction), at paras 96–8; CSOB (Jurisdiction), at para. 17; Aguas del Tunari, at para. 284; Tokios (Jurisdiction), at paras 25, 46, and 69; Maffezini (Jurisdiction), at para. 79; Wena Hotels (Jurisdiction), at 82 and 83; and Klöckner (Award), at 15. His most frequently cited work is Broches, ‘The Convention on the Settlement of Investment Disputes between States and Nationals of Other States’, 136 Recueil des Cours 331 (1972-II).

129  See Schreuer, supra note 3, at 1290. See also History of the ICSID Convention, Vol. 2, Part II, p. 1040.

130  Eureko, at para. 251 describes their work Bilateral Investment Treaties (1995) as ‘[t]he leading work on bilateral investment treaties’. However, given the growth of BITs since the mid-1990s, this work would not be fully reflective of the current generation of BITs. See also Wena Hotels (Jurisdiction), at 83–4; Olguín (Jurisdiction), at para 26; Genin, at paras 367 and 368; Lauder, paras 200 and 308; Pope & Talbot (Merits, Phase 2), at para. 111, n. 105; and Mondev, at para. 79, n. 19.

131  He is often cited in relation to his studies on ICSID’s jurisdiction: ‘The Jurisdiction of the International Centre for the Settlement of Investment Disputes’, 19 Indian Journal of International Law 166, 214 (1979); and ‘Jurisdiction Ratione Personae under the Convention on the Settlement of Investment Disputes between States and Nationals of Other States’ (1974–75) 46 BYBIL 227. See, e.g. Tokios (Jurisdiction), para 68 (referring to what it termed ‘Dr. Amerasinghe’s corollary rule of interpretation’) and 97; Wena Hotels (Jurisdiction), at 82; Aucoven (Jurisdiction), at para. 62; Fedax (Jurisdiction), at para. 22; and Vacuum Salt, at para. 29, n. 9, paras 37 and 42.

132  See, e.g. his publications ‘ICSID Arbitration: Practical Considerations’, 1 Journal of International Arbitration 101 (1984); and ‘ICSID Arbitration and the Courts’, 77 AJIL 784 (1983). See Wena Hotels (Jurisdiction), at 82; Salini v Morocco (Jurisdiction), at para. 27; Fedax (Jurisdiction), at paras 22–3; Tokios (Jurisdiction), at para. 42, n. 27; and Vacuum Salt, at para. 29, n. 9.

133  See, e.g. Mihaly, at para. 58. Note, however, that this case indicated that ‘experts on the theory and practice of multinational corporations’ fell outside the ambit of ‘teachings of the most highly qualified publicists’. Ibid.

134  Ibid.

135  See, e.g. the CME tribunal’s evaluation of the expert opinion provided by Schreuer, one of the leading scholars in the field of ICSID arbitration. CME (Final Award), at paras 401–13 (e.g. ‘Prof. Schreuer’s conclusions are not adequately supported by his citations, when closely read. Prof. Schreuer’s conclusions are carefully drafted with reference to assumptions which are not in accord with the facts of this case.’). As to the opinions of authoritative public international lawyers on points of substantive law prepared specifically for a case, see Loewen (Award), at para. 150.

136  Draft Article 70(1)(b) of Waldock III (emphasis added). See Appendix II infra. See also draft Article 56 of Waldock III and draft Article 69(1)(b) of the ILC Draft Articles 1964. Fitzmaurice was a staunch supporter of this approach. See, e.g. Fitzmaurice, ‘The Law and Procedure of the International Court of Justice 1951–54: Treaty Interpretation and Other Treaty Points’, 33 BYBIL 203 (1957), at 212 and 226 (‘to interpret such treaties according to modern concepts, would often amount to importing into them provisions they never really contained, and imposing on the parties obligations they never actually assumed’). An initial draft in Waldock III included a similar provision on the application of a treaty in accordance with international law.

137  Vienna Conference, First Session, Committee of the Whole, 33rd meeting, para. 54. An initial draft in Waldock III included a provision on the application of a treaty in accordance with international law in force at the time it was applied (numbered as Article 56(2)). See YILC (1964-II), at 8–9. However, this Article 56(2) was deleted in the Waldock III final draft provisions.

139  (US v Netherlands) 2 RIAA 829 (1949), at 845.

140  United States (LF and PE Neer) v Mexico, United States–Mexico General Claims Commission, 15 October 1926, 21 AJIL 555 (1927) (Supplement).

141  Mondev, at para. 114, quoting Neer, at 556.

142  Ibid., at para. 116.

143  Ibid., at para. 117. See also Azurix (Award), at para. 361; LG&E (Liability), at para. 125; and CMS (Award), at para. 284. The influence of subsequent developments has also been recognized in other international law cases. See, e.g. Shrimp Products case (WTO Appellate Body) (1998), at para. 130 (stating that ‘[f]rom the perspective embodied in the preamble of the WTO Agreement, we note that the generic term ‘natural resources’ in Article XX(g) is not ‘static’ in its content or reference but is rather ‘by definition, evolutionary.’’); dissenting opinion of Judge Alvarez in the Competence of the General Assembly regarding Admission to the United Nations, ICJ Reports 4 (1950), at 18 (‘a treaty or a text that has once been established acquires a life of its own. Consequently, in interpreting it we must have regard to the exigencies of contemporary life, rather than to the intentions of those who framed it’); and Judge Azevedo’s dissenting opinion in Competence of the General Assembly regarding Admission to the United Nations, ICJ Reports 4 (1950), at 23. See also Hudson, The Permanent Court of International Justice, 1920–1942 (1943), at 658 (‘[w]ith the lapse of time, intentions entertained by the draftsmen of an instrument may lose some of their importance, and a course of action by those who must live with and under the provisions of the instrument may assume a correspondingly greater significance’).

144  Mondev, at paras 123 and 125. See also Eureko, at para. 258. Wälde and Weiler have likewise suggested the following:

We suggest that the right approach is not to allow oneself to become too entangled in historical controversies about the minimum standard, but rather to follow general principles of treaty interpretation in applying the relevant provisions to the facts of a given case. This approach requires interpretation of treaty terms in their prevailing literal meaning, supported by the context and purpose of the treaty. Notions of ‘fair and equitable’, ‘constant protection’ or ‘avoidance of unreasonable impairment’ cannot be understood in the sense the minimum standard was discussed in the 19th century (relying on the prevailing standards of good governance), but must rather be read with today’s prevailing standard of good governance. This means that one must take into account the decades of international treaty practice, authoritative soft-law instruments and state practice which have evolved since at least the Second World War as indicative of such standards.

Wälde and Weiler, supra note 22, at 187. But see the approach to the Neer case in Glamis Gold.

145  Pope & Talbot (Damages), at para. 65. But see Glamis Gold, at para. 616.

146  See, e.g. Fitzmaurice, supra note 136, at 226.

147  Tradex (Jurisdiction), at 67. See also Eureko, at para. 258 (noting the doctrine that presumes sovereign rights override the rights of a foreign investor ‘has been displaced by contemporary customary international law, particularly as that law has been reshaped by the conclusion of more than 2000 essentially concordant bilateral investment treaties’).

148  CME (Final Award), at para. 498.

149  LG&E (Liability), at para. 213. See also Czech Republic v European Media Ventures SA [2007] EWHC 2851 (Comm), at para. 36 (holding that the ordinary meaning should be at the time the treaty was concluded).

150  See, e.g. Pollux, ‘The Interpretation of the Charter’, 23 BYBIL 54 (1946), at 68–9 (‘the aim of interpretation is to give effect to the instrument; to give it an interpretation which, on the whole, will render it most effective and useful’) and at 70 (‘the aim [of interpretation] must be to endow international conventions with the maximum possible effect’); Lauterpacht, The Development of International Law by the Permanent Court of International Justice (1934), at 69–70 and the revised 1958 version of that book, at 228; and Gaffney and Loftis, ‘The “Effective Ordinary Meaning” of BITs and the Jurisdiction of Treaty-Based Tribunals to Hear Contract Claims’, 8(1) Journal of World Investment & Trade 5 (2007).

151  See Appendix III infra.

152  YILC (1966-II), at 219, para. 6. Wälde, however, has commented that the relation between the principle of effectiveness and the good faith principle is tenuous. Wälde, supra note 16, at 738–40. See also Gardiner, supra note 1, at 150.

153  The PCIJ and the ICJ have also invoked the principle many times. See Territorial Dispute (Libya v Chad), ICJ Reports 6 (1994), at 23, para. 47 and 25, para. 51 (referring to the principle as ‘one of the fundamental principles of interpretation of treaties, consistently upheld by international jurisprudence’); Corfu Channel case, ICJ Reports 4 (1949), at 24; Acquisition of Polish Nationality, Advisory Opinion (No. 7), PCIJ, Series B (1923), at 17; Free Zones of Upper Savoy and the District of Gex, Order of 19 August 1929, PCIJ, Ser. A, No. 22, p. 13; the dissenting opinion of Judge de Visscher in the International Status of South-West Africa, Advisory Opinion, ICJ Reports 128 (1950), at 187; and Border and Transborder Armed Actions, Judgment, ICJ Reports 69 (1988), at 89, para. 46. The WTO Appellate Body placed the principle within the scope of the Convention Rules when it observed that ‘[a] fundamental tenet of treaty interpretation flowing from the general rule of interpretation set out in Article 31 [of the Vienna Convention] is the principle of effectiveness (ut res magis valeat quam pereat)’. Japan—Taxes on Alcoholic Beverages II, AB-1996–2, WTO Appellate Body, WT/DS8,10&11/AB/R (4 October 1996), at 12. See also Gardiner, supra note 1, at 159–61.

154  AAP, at para. 40.

155  Klöckner (Annulment), at para. 62.

156  Eureko, at para. 248. No citations were made to specific international law cases. It could have referred to AAP, at para. 40 (Rule (E)), which expounds a similar rule and cites authority in support. See also Noble Ventures (Award), at para. 50 (commenting after a reference to the Convention Rules that ‘the principle of effectiveness (effet utile) … too plays an important role in interpreting treaties’). FIATs have also considered the Roman maxim ut res magis valeat quam pereat as having equivalence with the principle of effectiveness. See Section G infra. In relation to the use of the principle by domestic courts in interpreting BITs, see Czech Republic v European Media Ventures SA [2007] EWHC 2851 (Comm), at para. 37.

157  See, e.g. MINE (Annulment), at paras 4.05–4.06. See also Canfor (Preliminary Question), at para. 324 (‘under well-known principles of international law, every provision of an international agreement must have meaning, because it is presumed that the State Parties that negotiated and concluded that agreement intended each of its provisions to have an effect’).

158  See, e.g. Amco (Jurisdiction), at para. 29.

159  Kaiser Bauxite, at para. 24.

160  Maffezini (Jurisdiction), at para. 36; Lucchetti (Award), at para. 59; and El Paso, at para. 110 (holding that treaty provisions ‘must be considered to carry some legal meaning’). See also Joy Mining, at para. 50; Holiday Inns, 1 ICSID Reports at 674; and Waste Management (Opinion), at 476, para. 47, and 478, para. 55. The SPP tribunal, in relation to a domestic statute, held that ‘[u]nder general principles of statutory interpretation, a legal text should be interpreted in such a way that a reason and a meaning can be attributed to every word in the text’. SPP (Jurisdiction No. 2), at para. 94. See also Romak, at para. 180.

161  Camuzzi (Jurisdiction), at para. 56.

162  Noble Ventures (Award), at para. 52.

163  Ibid.

164  Tecmed, at para. 156.

165  Tza Yap Shum, at para. 188.

166  American Manufacturing, separate opinion of Golsong, at 42, para. 20. El Paso, at para. 76, spoke of the need to avoid rendering ‘the whole Treaty completely useless’. See also Sempra Energy (Jurisdiction), at para. 94.

167  SPP (Jurisdiction No. 2), at para. 94.

168  Occidental Exploration (Award), at para. 68; Joy Mining, at para. 50; and SPP (Jurisdiction No. 2), at para. 94.

169  SPP (Jurisdiction No. 2), at para. 94. Note must be made that this was in relation to the interpretation of an Egyptian statute that, under certain conditions, consented to ICSID jurisdiction. See also SGS v Pakistan, at para. 168; Generation Ukraine (Award), at para. 14.3; and Eureko, at para. 258.

170  SGS v Philippines, at para. 116 (‘[t]he object and purpose of the BIT supports an effective interpretation of Article X(2)’). See also MINE (Annulment), at para. 4.05; Noble Ventures (Award), at para. 52; Mondev, at para. 91; and Schreuer (2006), supra note 31, at 4.

171  Banro (Award), at para. 6. See also Noble Ventures (Award), at para. 52. This passage was quoted with approval in the dissenting opinion of Alberro-Semerena in Aguas del Tunari, at para. 32.

172  See the Interpretation of Peace Treaties (Second Phase), ICJ Reports 221 (1950), at 229 and the dissenting opinion of Judge Read, ibid., at 238. See also Jennings and Watts (eds), Oppenheim’s International Law (1992), at 1281 (‘[e]ffectiveness is relative to the object and purpose of the treaty, a decision as to which will normally first have to be made’).

173  Fitzmaurice, supra note 136, at 19. See also Fitzmaurice, ‘Vae victis or woe to the negotiators: Your treaty or our “interpretation” of it?’, 65 AJIL 358 (1971), at 373; the joint dissenting opinion of Judges Fitzmaurice and Spender in the South West Africa cases, Preliminary Objections, ICJ Reports 319 (1962), at 468, 511–3; the Laguna del Desierto case, 113 ILR 1, at 44; and Thirlway, ‘The Law and Procedure of the International Court of Justice 1960–1989: Part Three’, 62 BYBIL 1 (1991), at 47–8.

174  Jennings and Watts, supra note 172, at 1281. See also Thirlway, who has observed that if ‘a particular set of circumstances had arisen or could be imagined in which a treaty provision, according to an otherwise acceptable interpretation, would be ineffective or lead to improbable results, this is not necessarily a ground for rejecting that interpretation. The principle of effectiveness should be employed as an aid to assessment of likely intentions, rather than as a rigid canon of interpretation whereby the text must be deemed to be effective in all circumstances.’ Thirlway, supra note 173, at 46 (footnote omitted). Thirlway qualifies this statement by adding that an ineffective interpretation should be excluded only where there is an alternative interpretation that produces an effective result and does not offend against other cannons of interpretation. Ibid., at 47–8.

175  Renta 4, at para. 32. Article 10(1) of the Spain–USSR BIT is comprised of over sixty words.

176  See, e.g. Berlia, ‘Contribution à l’interprétation des traités’, 114 Recueil des cours 283 (1965), at 306 et seq. Examples of this rule may be found in the merits phase of the Corfu Channel case, ICJ Reports 4 (1949), at 24 and the Anglo-Iranian Oil case, ICJ Reports 93 (1952), at 105. See also Thirlway, supra note 173, at 44.

177  See, e.g. Berlia, supra note 176, at 308 et seq. Examples of this rule may be found in the Interpretation of Peace Treaties (Second Phase) case, ICJ Reports 221 (1950), at 229; and in the Ambatielos, ICJ Reports (1952), at 45. This limb appears to have a close relation to the Article 31(1) object and purpose criterion.

178  AAP, at para. 40 (Rule (E)) and para. 52; CSOB (Jurisdiction), at para. 39; Noble Ventures (Award), at para. 50; Salini v Jordan (Jurisdiction), at para. 95; El Paso, at para. 110; and Thunderbird, dissenting opinion, at para. 91.

179  Sempra Energy (Jurisdiction), at para. 94.

180  Jennings and Watts, supra note 172, at 1275 et seq., refers to a number of legal maxims in its section on supplementary means of interpretation.

181  Wälde, supra note 16, at 733.

182  YILC (1966-II), at 218, para. 3.

183  YILC (1966-II), at 218, para. 4. Jennings and Watts, supra note 172, at 1269–70, comments that Roman law maxims applied in municipal law systems:

are expressive of common sense and of normal grammatical usage, they commend themselves also in the interpretation of treaties. However, while international law permits recourse to many principles and maxims, it does not always require recourse to them. The appropriateness of applying many of them depends on a variety of considerations which will determine whether, although they are accepted in international law as potentially relevant, they are also suitable for application in all the circumstances of a particular case. In such cases the principle is not so much a rule of international law as a discretionary aid. …

184  See generally, Wälde, supra note 16, at 742.

185  SGS v Philippines, at para. 141. In a similar vein, Schreuer, supra note 3, at 362, has commented that ‘[a] document containing a dispute settlement clause which is more specific in relation to the parties and to the dispute should be given precedence over a document of more general application’. See also AAP, at para. 54 (‘in the absence of a specific rule provided for in the Treaty itself as lex specialis, the general international law rules have to assume their role as lex generalis’) and para. 65 (holding that the lex generalis applies to all situations that are not covered by a specific rule). See also Canadian Cattlemen, at para. 166; AES (Jurisdiction), at para. 23(b); ADF (Award), at para. 147; and AAP (dissenting opinion), at 302–3.

186  SPP (Jurisdiction No. 2), at para. 83, citing Grotius, De Jure Belli ac Pacis, Bk II, Chap. XVI, and Mavrommatis Palestine Concessions, Jurisdiction, PCIJ, Series A, No. 2 (1924), pp. 31-2; and Saudi Arabia v Aramco 27 ILR 117 (1963). However, the BIT cannot override the ICSID Convention in a way that it expands the provisions contained in that Convention. See, e.g. Aguas del Tunari, at para. 278.

187  Fraport, at para. 305.

188  SGS v Philippines, Supplementary Declaration, at para. 9.

189  SGS v Pakistan, at para. 171.

190  Eureko, at para. 258. See also the criticism in Wälde, supra note 16, at 733–6.

191  Waste Management II (Award), at para. 85. See also National Grid (Jurisdiction), at para. 82 (‘specific mention of an item excludes others’).

192  Siemens (Jurisdiction), at para. 140. See also Austrian Airlines (Award), at para. 131 (the principle ‘cannot alone determine the outcome of the interpretation when a treaty contains other relevant elements’); Vacuum Salt, at 337, para. 29, n. 9; Tokios (Jurisdiction), para. 30; Tza Yap Shum, at para. 207; Methanex (Final Award), Part IV, Chap. C, at para. 14; and Wälde, supra note 16, at 740.

193  See also Schreuer (2006), supra note 31, at 7 (the maxim is of limited use and that ‘[w]hether the mention of one item or a list of items in a provision really excludes the relevance of other items depends very much on the particular circumstances and cannot be answered in a generalised way’).

194  Maffezini (Jurisdiction), at 405–6, at paras 46–50; and Plama (Jurisdiction), at paras 36 and 189.

195  This maxim has been used synonymously with the principle of effectiveness. See, e.g. Banro, at para. 6; El Paso, at para. 110; American Manufacturing, separate opinion of Arbitrator Golsong, at 42, para. 20; Waste Management, dissenting opinion, at para. 55; and Pan American (Preliminary Objections), at para. 132. See also Section F supra.

196  SGS v Philippines, at para. 143 and at para. 145 (‘[i]n principle a later agreement between the same parties could override an earlier one’).

197  See Aust, Modern Treaty Law and Practice (2nd edn, 2007), at 249.