A Multitude of Possibilities

In 1980, Richard Baxter famously detailed how international law manifested itself in “her infinite variety.”Footnote ² Baxter was referring, in part, to the bewilderingly wide variety of ways that states and other actors expressed international legal normativity, ranging from traditional instruments like treaties and joint communiqués to resolutions of international organizations and innovative forms of soft law. Baxter usefully explained that different forms of international law contained different levels of precision, enjoyed varying degrees of acceptance, and reflected various shades of normative commitment.

Surveying the field of legal ethics in international adjudication nearly four decades later, it is tempting to conclude that ethical rules likewise exist in a nearly infinite variety.Footnote ³ Variation is found in the scope and precision of rules; their normative weight; their genesis, specifically whether they are promulgated by states, judges, or other actors; and whether they are consolidated in a separate code or, rather, scattered across various texts not specifically devoted to ethics issues.

International arbitration, in particular, is marked by fragmented and decentralized clusters of “institution specific” rules, as each arbitral institution determines whether and how to address ethics issues. The International Centre for Settlement of Investment Disputes, UNCITRAL, and the International Chamber of Commerce, for example, have each adopted their own set of rules, although each addresses only a narrow set of ethical issues. Adding to the complexity, a variety of soft law instruments exist, including the International Bar Association Guidelines on Conflicts of Interest in International Arbitration (IBA Guidelines) and the Hague Principles on Ethical Standards for Counsel Appearing Before International Courts and Tribunals.Footnote ⁴

As calls for ethics reform gain momentum, attention should focus on lessons learned from past drafting attempts and on strategies for consolidating this multiplicity of forms and contents into one viable code—how to move, in other words, from a plurality to one.

In this essay, we do not attempt to resolve all the complex and longstanding normative debates over the codification of ethics norms in international litigation.Footnote ⁵ Instead, we identify and explore three threshold questions that we believe should shape discussions over a common code of ethics for disputes involving a state as a party. First, in terms of form, we explore the legal nature of the proposed norms, and specifically whether or not they should be binding.Footnote ⁶ Second, we examine whether one general code is preferable to numerous institution-specific codes; we explore particularly how an ethics code's substantive content has implications for the code's scope of application. Third, we delve into the issue of process and thus how a code could become part of the applicable procedural law.

We highlight these threshold issues because absent reasonably clear agreement on them, it is difficult to engage meaningfully in any discussion of whether it is descriptively possible or normatively desirable to develop ethical standards applicable to international adjudication. Issues of form, content, scope, and process in these debates are intrinsically related, and changes in one of these variables has unavoidable implications for the others.

Soft or Hard Law?

The initial question we consider is whether a binding code is preferable to nonbinding guidelines. There are strong arguments in favor of regulating ethics through nonmandatory guidelines. Chiefly, guidelines can be more flexible—an important consideration given the wide variety of international adjudication fora. Moreover, nonbinding ethical rules could attract a larger and more diverse set of parties.

However, at least three arguments tip the balance towards a mandatory code. First, several previous attempts by well-respected institutions to develop nonbinding guidelines have had only limited success. Although the IBA Guidelines have been invoked more frequently than other soft law instruments, they cover only a small fraction of the ethics universe, and even they are applied relatively rarely. Absent the authority and legitimacy that attaches to state-negotiated solutions, parties have proven reluctant to incorporate ethics codes into the resolution of their disputes. Second, ethical issues are too important to be left to voluntary standards. Recent cases that focused on ethics questions, and the widespread public interest in certain ethical issues, strengthen the argument that a comprehensive and binding code is needed. Third, and most importantly, tribunals may, and do, disregard soft law when they consider ethical issues.

The recent Chagos arbitration is instructive in this regard.Footnote ⁷ In 2010, Mauritius filed a claim alleging that the United Kingdom had violated the 1982 UN Convention on the Law of the Sea (UNCLOS) by establishing a Marine Protected Area in the Chagos Archipelago.Footnote ⁸ A tribunal was established under Annex VII UNCLOS and administered by the Permanent Court of Arbitration (PCA).

The United Kingdom appointed Judge Sir Christopher Greenwood as arbitrator, triggering a challenge by Mauritius, based on Sir Christopher's “long, close, and continuing relationship with the Government of the United Kingdom,” including prior work as counsel for the British government. Mauritius argued that in applying the obligation of independence and impartiality, the tribunal should refer to the standards reflected in international arbitration rules and statutes, including those found in soft-law instruments such as the IBA Guidelines.Footnote ⁹

After careful review, the tribunal determined that the law and practice of tribunals “not seized of inter-State disputes” was “not directly relevant.” It further determined that Mauritius had not demonstrated that the rules adopted by non-governmental institutions such as the IBA have been expressly adopted by States, that they form part of a general practice accepted as law, or fall within any other of the sources of international law enumerated in Article 38(1) of the Statute of the ICJ.Footnote ¹⁰ The tribunal therefore concluded that the IBA rules were neither directly applicable nor “relevant for the purposes of its analysis in the present proceedings.”Footnote ¹¹

In sum, the tribunal declined to consider soft law instruments, even for analytical purposes. The tribunal's conclusion shows that, absent a clear prior agreement by the parties, a tribunal can decide not to rely on soft law even when requested to do so by one of the parties. To guarantee consistency and foreseeability, ethics rules should be adopted via mandatory code that can be known and readily identifiable by the parties and the tribunal alike before any disputes ensue.

General or Specific Code?

A second threshold question concerns the scope of the applicability of an ethics code, and specifically whether a series of specialized codes, each intended for a particular forum, is preferable to a general code that can be used by a multiplicity of fora addressing disputes involving states. On balance, we favor an approach that would broadly apply to all proceedings involving a state.

First, proceedings where at least one party is a state share a number of similarities, and these proceedings differ in a number of important respects from those not involving states. Most obviously, these disputes all involve a sovereign, a unique entity under international law that is regulated by various rules and legal regimes not applicable to other actors. As such, cases that involve a state often share similarities in terms of applicable rules of procedure, applicable provisions addressing immunities, the tribunal's power, and the collection of evidence, among other topics. Moreover, disputes that involve states inevitably possess a public character, which increasingly captures the interest of broader publics, and as such these proceedings also often raise recurring issues concerning transparency, third party access, cost and funding, and related issues.

The reasoning of the Chagos challenge decision is illustrative in this regard. In looking for an appropriate standard, the tribunal referred to ICJ practice and other interstate arbitrations, including a number of PCA-administered arbitrations, such as the Eritrea-Ethiopia Boundary Commission, the OSPAR case, and the MOX Plant case. Each of these disputes used rules based on the UNCITRAL Arbitration Rules, but with certain modifications “to reflect the public international law character of disputes between States, and diplomatic practice appropriate to such disputes.”Footnote ¹² The tribunal looked to these actions because the standards that they employed “can be considered to form part of the practice of inter-State arbitral tribunals.”Footnote ¹³

A second, and even more important, reason to regulate ethics generally in disputes involving a state as a party is systemic: One of the most potent critiques of international litigation is that judges and arbitrators are often appointed repeatedly and that they and counsel may engage in double-hatting. Repeat appointments of the same person by the same party, the same counsel, or even in the same kind of proceedings may result in unacceptable conflicts of interest or, at a minimum, the appearance of a lack of impartiality. Likewise, double-hatting, the practice by which individuals act both as international arbitrators and as counsel in separate proceedings, is roundly criticized but remains unregulated.Footnote ¹⁴

In the current institution-rich environment, only a general code of conduct that applies to all proceedings involving a state can properly address systemic concerns. A general code could bring at least two specific advantages. First, it could require full disclosure of all cases in which an arbitrator is involved regardless of the forum, thus providing comprehensive information to all parties of any possible conflicts and addressing both repeat appointments and double-hatting. Second, a general code could regulate comprehensively the number of cases in which an arbitrator may simultaneously participate. For example, a general code could include provisions that limit the number of cases on which a potential arbitrator can concurrently sit, or the number of appointments that could come from the same party or counsel. Alternatively, a general code could also specify a period of time—for example, two or three years—during which an arbitrator could not work as counsel or a counsel could not accept appointments.Footnote ¹⁵ For current purposes, the details of any rules eventually adopted are less important than the fact of a system-wide approach. If the goal is to tackle repeat appointments and double-hatting, only a generally applicable framework can provide the necessary and effective regulation for international litigation that involves states.

The Choice of Process

A final threshold issue to consider is what process should be adopted to ensure the implementation and success of the code.

At first blush, it is reasonable to think that incorporating a code into a treaty would give it the strongest support and chance to succeed. Treaties are strictly regulated sources of international law, whose applicability and enforcement processes are clear.

In this context, the multifaceted approach to adoption used for the Rules on Transparency in Treaty-based Investor-State Arbitration (the Rules) is instructive.Footnote ¹⁶ The Rules were developed by UNCITRAL and aim at increasing transparency in the investment arbitration process. Interestingly, they can be incorporated into arbitral proceedings in a variety of ways. As a matter of treaty law, they apply to investor-state arbitration initiated under the UNCITRAL Arbitration Rules pursuant to a treaty concluded on or after April 1, 2014.Footnote ¹⁷ States have negotiated about fifty such new treaties, but only sixteen have entered into force.Footnote ¹⁸ Also, parties to the UN Convention on Transparency in Treaty-based Investor-State Arbitration (Mauritius Convention) agree to apply the Rules to disputes arising under treaties concluded before April 1, 2014. However, the Mauritius Convention has only entered into force for five countries.Footnote ¹⁹ The implementation of the Rules through the treaty option seems to be of limited impact so far. This is unsurprising, as substantial time often passes between the conclusion of a treaty and the start of treaty-based disputes.

In conjunction with the treaty-based approach, arbitral institutions could adopt the Code in their rules so that it could apply to all future proceedings administered by that institution.Footnote ²⁰ By way of example, in 2013 the UNCITRAL Arbitration Rules were amended to incorporate the new transparency rules. Adoption of a common code of ethics by arbitral institutions would promote a more expeditious and uniform application of the Code.