A Reply to Sir Frank Berman

I am very grateful to Sir Frank Berman, Anthea Roberts, and Federico Ortino for their wide-ranging observations about my book. As much as I would like to give an exhaustive reply to each of them, I could not do justice to all of their comments without exceeding the natural limitations of time and space imposed by the forum. I will therefore focus on a limited number of issues, that either seem to me to pose the most serious challenges to the argument that I make in my book, or to which I have given most thought. I will respond to Sir Frank Berman’s arguments in this post, and will address Anthea Roberts’ and Federico Ortino’s arguments in the next post.

Sir Frank Berman raises a number of questions about the style and substance of my argument. In particular, he seems doubtful about the practical validity of the distinction between treaty and customary rules in the area, and suggests that a clearer distinction between interpretation and application, the latter concept opening a wider scope for case-by-case variation, may dispose of my concern about accommodating investment arbitration in the structure of sources and interpretation of international law. I will address these two issues in turn. Briefly, I believe that the treaty/custom distinction is of considerable practical importance. While I entirely agree with everything that Sir Frank Berman says about the distinction between interpretation and application – and I would be so bold as to say that I do not ignore it in my book – it does not affect my argument that the great reliance on earlier arbitral pronouncements suggests a process of interpretation and application taking place regarding the same rule of (general) international law. (Sir Frank also makes certain observations about the effect that different structures of dispute settlement have on development of law; I will address that point in my response to Federico Ortino).

On the Treaty/Custom Distinction

Is it important whether a ‘fair and equitable treatment’ clause is read as a technical term of reference to custom (or a term that does not refer to custom as such but that does not preclude it from being taken into account in the interpretative process), or rather a term that neither requires nor permits reliance on general international law? In technical terms, an answer to such a question would depend on whether the rules of customary law are identifiably different from those that may be determined by interpretation of the treaty language. It is relatively easy to give an affirmative answer regarding rules on administration of justice, where (as I argue in Chapter 8) the basic principles are well-established: in particular, a wrongful act of denial of justice is complete only when the whole judicial system is exhausted, denial of justice focuses on procedure, and denial of justice by substance of the judgment is exceptional. It is not at all obvious, to say the least, that an interpretation not involving general international law would identify different requirements of exhaustion of remedies as a matter of primary obligation because conduct by a particular organ is concerned, or draw the distinction between procedural and substantive unfairness (the fact that most Tribunals do draw these distinctions strongly suggests that they are implicitly relying on customary law).

It is more complicated to give an affirmative answer regarding mistreatment of property outside the judicial context. The standard of outrage and impropriety from the 1926 Neer award is usually cited when pre-World War Two law is discussed, by both those invoking it and those dismissing its authority. In Chapter 2, I suggest that both sides are wrong: Neer is certainly not an effectively irrebutable statement of law, but it is also not an accidental irrelevance. It is an important stage in development of the international standard by means of analogy, deriving criteria of procedural outrage from the better-established rules of denial of justice, and then applying it more generally. Once Neer is accepted as what it is – a default argument, perfectly open to clarification or replacement by specific rules – the solution lies not in defending its shorthand or finding a new shorthand, but in elaboration of more specific rules on particular issues (as the Neer Commission itself proceeded to do in a subsequent case regarding the taking of life). Within the four corners of this argument, has the default focus on arbitrariness been replaced by a new rule requiring respect for legitimate expectations of investors? In Chapter 9, I suggest that an affirmative answer, at least at this point, is not obviously called for, although it equally does not mean that an inquiry into lawfulness of abrupt changes that is articulated in terms of arbitrariness would necessarily lead to results significantly different from a consideration of legitimate expectations. To conclude, there is a significant practical difference between including and excluding custom: for administration of justice, significantly different answers; for other aspects of mistreatment, at least significantly different questions.

On the Interpretation/Application Distinction

Sir Frank Berman also suggests that I might have benefited from a clearer distinction between interpretation and application, since the latter might open wider scope to case-by-case variations. I entirely agree with the distinction and its implications (as recently elaborated by Gourgourinis). To loosely adopt the formulation of Judge Ehrlich ([1927] PCIJ Series A no 9 35, 39), a Tribunal interpreting a fair and equitable treatment clause will determine that its meaning requires a State to provide due process in the treatment of an investment; a Tribunal applying the clause will determine the consequences that it attaches to an e-mail sent by a secretary of the Minister of Energy to an investor’s representative at 4.57 am, notifying the latter of a 9.00 am meeting, where a decision on an extension of the investor’s licence to drill for oil will take place, and asking to provide comments to a 300-page memo beforehand. The further a Tribunal proceeds in its reasoning (does fair and equitable treatment include due process? Does due process include a duty of consultation in matters affecting investment? Is such a duty complied with by an e-mail of the suggested kind? etc.), the more deeply it will be involved in the application of the treaty, and that is precisely what is should be doing. A few years ago, Vaughan Lowe pointed out the similarity between fair and equitable treatment and the process by which the law of the sea identified the relevance and weight of factors affecting the provisional equidistance line ((2006) 100 ASIL Proceedings 73, 73). I find the analogy very apt. Far from focusing on abstract interpretation, one of the main theses of my book is that law in the area should develop more detailed rules through application to particular cases, rather than a priori extrapolations of jurisprudential or related character (Robert Jennings’ quotation from the concluding paragraph of my introductory post speaks precisely to this point).

The point that Chapter 5 makes is that interpretation and application are still interpretation and application of a particular rule. That is not a call for a dogmatic or hierarchical approach to treaty interpretation (it seems to me that the argument made in the book is entirely in line with how Sir Frank Berman described treaty interpretation and application a few years ago, (2006) 26 Statute L Rev 1, 9-12). I merely suggest that VCLT and analogous customary law set limits on the list of admissible authorities (albeit somewhat open ones, since Article 32 is non-exhaustive), and that unrelated pari materia treaty rules and their arbitral interpretation and application do not obviously fall within that limit, even if very widely and flexibly drawn. I am perfectly happy not to criticise the Tribunal imagined in the previous paragraph for finding that the early-morning e-mail is in breach of due process as part of fair and equitable treatment, the next Tribunal for relying on this ruling to find that an announcement posted on the relevant Ministry’s website a month before a meeting is in compliance with the obligation, and the third Tribunal for considering both awards and finding that a notice by a trained carrier pigeon a week before a meeting may be a tad eccentric, but not unlawful. In methodological terms, that is how Tribunals generally interpret and apply fair and equitable treatment, and that is a perfectly plausible way of going about it. However, for me to withhold the critical sting, all three Tribunals have to be interpreting and applying (and developing) the same rule, and for Tribunals interpreting different treaties that can only mean a rule of customary international law. That is everything that Chapters 5 says. (Chapter 6 provides the other side of the normative coin, demonstrating in positive terms how the reference to custom takes place. Chapters 5 and 6 only set up the legal framework within which the development of the international standard and fair and equitable treatment takes place, and it is Part Three that spells out the legal implications for the content of the rule.)