Printer Friendly

Codifying custom.

Codifying decentralized forms of law, such as the common law and customary international law, has been a cornerstone of the positivist turn in legal theory since at least the nineteenth century. Commentators laud codification's purported virtues, including systematizing, centralizing, and clarifying the law. These attributes are thought to increase the general welfare of those subject to legal rules and therefore to justify and explain codification. The literature, however, overlooks codification's distributive consequences. In so doing, it misses a common motive for codification: to define legal rules in a way that advantages individual codifying institutions, regardless of how it affects the general welfare.

This Article fills the gap in the literature by examining three rationales for why states codify customary international law: (1) a desire to clarify the substantive content of customary law in order to promote cooperation (the Clarification Thesis); (2) a desire to enhance compliance through mechanisms such as monitoring, enforcement, and dispute-resolution provisions (the Compliance Thesis); and (3) a desire to define the content of customary rules for a state's individual benefit (the Capture Thesis). While codification's proponents conceive of the enterprise in terms of the Clarification and Compliance Theses, I argue that states frequently use codification to capture customary international legal rules to benefit themselves at the expense of the general welfare. As states with divergent views on how to interpret a customary rule pursue conflicting codification efforts, they entrench schisms in the law along regional or ideological lines, thereby delegitimizing customary rules and increasing fragmentation. Thus, far from being an unqualified boon to benevolent legal ordering, codification can replicate, magnify, or alter the power dynamics present in forming bare customary law. Indeed, the fragmentation of customary law that can result from codification actually prevents a unified understanding of customary law from emerging--the exact opposite of codification's ostensible purpose. This Article uses the Capture Thesis to explain important developments in customary international law, including the outlawing of the slave trade in the nineteenth century, the rise of bilateral investment treaties, and the inability to reach an agreement on a multilateral investment treaty.
     A. Clarification
     B. Compliance
     A. The Intuition
     B. Codification as Commitment
        1. Commitment
        2. Piracy and the Outlawing of the Slave Trade
     C. Codification and Procedural Rules
        1. Procedure and the Codification of Customary
        2. Procedure in the Amendment of Codified
           Customary Obligations
     D. Treaty Membership
     E. Conclusion
     A. Suboptimal Rules
     B. Fragmentation
        1. How Codification Leads to Fragmentation
        2. International Investment Law as Fragmented
           a. The Codification of the Calvo Doctrine
           b. International Investment Agreements and the Failure
              to Reach a Multilateral Investment Agreement
           C. Conclusion


In 2004, the World Trade Organization (WTO) abandoned its most recent effort to reach a multilateral agreement governing investment. (1) Such efforts have been part of an on-again, off-again push by developed countries to codify and harmonize international investment law, with the goal of reducing barriers to international capital flows and facilitating regional integration in places like North America and Europe. Developed states have pushed for a multilateral agreement since the end of World War II and indeed produced draft conventions in 1948, 1967, and 1998. (2) But none of these draft conventions has gained traction. With efforts at the WTO unsuccessful, trade in capital remains the key pillar of international economic law still largely outside the purview of WTO disciplines. International investment law thus remains governed by customary international law and the "spaghetti bowl" of bilateral, regional, and sectoral agreements that partially codify investment law.

This pattern of failure is puzzling. States long ago agreed to a robust set of multilateral rules governing trade in goods and services. They reached these agreements despite a relative lack of preexisting multilateral law on trade liberalization and protection, at least when compared with investment law. Prior to the 1947 General Agreement on Tariffs and Trade (GATT), (3) for example, there was little multilateral law in the area of trade barriers; prior to the 1994 General Agreement on Trade in Services, (4) there was little independent law on trade in services.

Prospective codifiers of investment law, by contrast, have a wealth of sources on which to draw for their multilateral codification efforts. For at least a century, customary international investment law has provided a multilateral legal standard governing the protection of property owned by aliens. (5) Moreover, the explosion of bilateral investment treaties (BITs) and similar regional and sectoral investment agreements--the structure and content of which are similar across the roughly 2600 agreements in force--have reinforced preexisting customary law and created a type of multilateral regime that reaches most states in the world. (6) Yet as the many efforts to negotiate a viable multilateral investment agreement at both the WTO and the Organisation for Economic Cooperation and Development (OECD) attest, states do not view this patchwork of treaties with limited membership and customary law as an adequate substitute for a truly multilateral convention.

What, then, explains the failure to reach a multilateral agreement on investment? I argue that this failure is part of a larger phenomenon in which codification--by which I mean the process of reducing law to a written instrument that elaborates established customary doctrines--can actually interfere with the development of truly global governance. Specifically, I contend that states often use codification to capture customary international legal rules to benefit themselves at the expense of the general welfare of the global community.

This suggestion may seem radical: the codification of customary international law--along with the rise of international organizations, one of the pillars of the twentieth century movement toward legalization in international affairs--has been a part of the legalization project since the late nineteenth century. (7) Spurred by the rise of legal positivism and a desire to avoid the pan-European and global conflicts that marred the nineteenth and early twentieth centuries, states convened major codification conferences in 1899, 1907, and 1930. (8) After World War II, the U.N. Charter explicitly tasked the General Assembly with the development and codification of international law as a means of promoting international cooperation, (9) a job the General Assembly has largely delegated to the International Law Commission (ILC). Moreover, codification continues to occupy a central place in contemporary international legal practice. Since 1990, states have codified or have tried to codify customary rules of investment law in negotiations at the WTO and OECD, (10) international criminal law in the Rome Statute of the International Criminal Court, (11) and customary norms of environmental protection in a variety of regional and global agreements. (12) The wars in Iraq and Afghanistan have highlighted customary law on the use of force and its exceptions, as codified in the U.N. Charter. Furthermore, the treatment of detainees in the war on terror has brought to the fore the customary international law on the treatment of prisoners, as codified in the Geneva Conventions, (13) and the ban on torture, as codified in the Convention Against Torture (CAT). (14)

The conventional wisdom is that states codify customary law to clarify the law and promote compliance. Claims that codification clarifies the law, in particular, are legion. (15) According to these arguments, codification allows states to specify more precisely what customary international law requires, thereby facilitating deeper cooperation and avoiding costly disputes over vague legal rules. (16) Codification also introduces the possibility of attaching compliance-promoting mechanisms--such as protocols granting international tribunals jurisdiction over disputes--to customary rules. (17) Such mechanisms increase sanctions for violating the law and are thereby thought to boost compliance rates.

These rationales, which I refer to as the Clarification and Compliance Theses, only partially explain the allure of codification. Both the Clarification and Compliance Theses focus primarily on efficiency. They suggest that states will take actions that increase overall global welfare. For example, all else equal, agreeing to submit disputes to a tribunal should boost compliance by exposing violations. But rational states generally are interested first and foremost in increasing their own welfare. Increasing overall welfare is merely a means to that end, since it increases the size of the pie to be divided. For states to do what is in the interest of overall welfare thus requires certain assumptions about interstate bargaining--such as low transaction costs--which will often not hold in the real world.

I therefore introduce what I call the Capture Thesis of codification. (18) I argue that codification is often driven by distributional concerns, rather than efficiency concerns. States seek to codify customary rules that benefit their interests as a way to define and capture those rules. Codifying customary rules is an attractive strategy because it allows states to influence the content of rules that bind all states without having to negotiate a universal treaty, with all the costs and compromises such negotiations entail.

States capture customary international law through codification in two ways. First, they use codification as a commitment device in situations in which the interpretation of a customary rule is unsettled. Codifying their understanding of the customary rule among themselves binds like-minded states to that understanding by raising the costs of adopting alternative interpretations. This credible commitment in turn pressures noncodifying states that wish to cooperate with codifying states to adopt the codified understanding of the customary rule, even when the noncodifying states would prefer an alternative interpretation. Second, codification introduces explicit power-based bargaining dynamics that are absent from, or at least muted in, the formation of bare (i.e., uncodified) customary rules. In particular, codification exchanges a largely unstructured customary lawmaking environment for a bargaining-based forum in which (1) influence can be limited through exclusion and agenda setting, and (2) onerous amendment processes can create veto power over changes in the law.

The Capture Thesis makes clear that codification can replicate, magnify, or alter power dynamics present in the formation of bare customary international law. Thus, notwithstanding the benefits flowing from clarification and the creation of compliance mechanisms, codification may result in legal rules that reduce global welfare. This conclusion is important for the literature on whether decentralized forms of law, such as the common law or customary international law, evolve toward efficiency. I argue that the use of power in the codification of custom can actually undercut, rather than promote, custom's efficiency. Even where rules are not suboptimal from a welfare perspective, allowing powerful states to capture the gains from cooperation may delegitimize customary rules. Finally, the Capture Thesis counter-intuitively suggests that codification can sometimes increase fragmentation: if states with divergent views on how to interpret a customary rule pursue conflicting codification efforts, they can entrench schisms in the law along regional or ideological lines. (19)

The remainder of this Article proceeds as follows. Part I introduces the concept of codification and its history and explains how I use that concept here. Part II develops the Clarification and Compliance Theses, discussing why, as a descriptive matter, they do not fully explain the codification of customary international law. Part III sets forth the Capture Thesis, including the limits on codification's use as a device to advance one's self-interest, and illustrates the theory with examples such as the British campaign to outlaw the slave trade. Part IV examines the unintended consequences of codification in light of the Capture Thesis, arguing that codification can actually lead to suboptimal customary rules and exacerbate the fragmentation of international law by entrenching differences along regional or ideological lines. I illustrate these ill effects by analyzing how codification has prevented states from negotiating a multilateral treaty on investment. I then conclude.


Prior to the twentieth century, most international law was customary law. Customary international law, as the commonly cited definition goes, "results from a general and consistent state practice" done out of "a sense of legal obligation." (20) This definition, although easily stated, turns out to be terribly difficult to apply. (21) For example, there is little agreement as to how widespread the practice must be or how consistently states must follow the practice. (22) More fundamentally, scholars and commentators do not agree on what kinds of practice are relevant. (23) Do votes in the U.N. General Assembly count, or are more concrete actions required? What about domestic laws? As to the second requirement, how is one to tell whether a state took some action out of a sense of legal obligation, rather than expedience, habit, or some other reason? Moreover, the definition is circular, insofar as it appears to require states to act out of a sense of legal obligation before such an obligation exists. (24) The uncertainty injected into customary law by these practical difficulties raises serious questions about the utility of customary law in regulating interstate relations.

While scholars have spilled much ink attacking and defending customary international law as such, states have largely responded to custom's difficulties by turning customary law into treaty law; that is, they have responded with codification. By codification, I mean the formulation and reduction to a written instrument of rules of law that elaborate established doctrines and precedents, which, even if nonbinding, have legal consequences. (25) Codification solves some of the practical problems with customary law by clearly delineating the steps that mark the creation of a legal obligation. For example, a codifying treaty might state that it enters into force upon a certain number of ratifications, an objective metric for the creation of legal obligations otherwise lacking in the unstructured environment in which bare customary law forms. (26)

Indeed, codification has been the elder and underappreciated sibling of the twentieth century movement to establish international organizations and legalize international relations. Codification and international organizations both seek to use substantive legal and procedural rules to structure and shape what would otherwise be unrestrained political interactions among states. International organizations have received the lion's share of scholarly attention because, among other reasons, they tend to be durable, provide a venue to deal with multiple related issues over a period of time and in a way easily amenable to study, and raise concerns about state sovereignty that are not as directly implicated in treaties that merely create rules of conduct.

But codification is no less important. The codification movement preceded and in many ways laid the groundwork for the development of international organizations. Jeremy Bentham is credited with initiating the codification movement in the late eighteenth century. (27) At the international level, the movement gained steam during the late nineteenth century as a way to manage states' increasing interrelationships. Some of the earliest efforts to codify international law occurred in 1873, when both the International Law Association (ILA) and the Institut de Droit International were founded. (28)

The earliest intergovernmental bodies aimed at codification were the Hague Conference on Private International Law--founded in 1893 and still the leading body for the codification of private international law--and the Pan-American Union (now the Organization of American States). (29) In public international law, the Hague Peace Conferences of 1899 and 1907, which predated the creation of the League of Nations, presented the first multilateral opportunities to consider codification. (30) After its creation, the League of Nations renewed these codification efforts, adopting a series of resolutions identifying a need for the progressive codification of international law. (31) In the wake of World War I, Judge Manley Hudson of the Permanent Court of International Justice noted that the "revival of interest in the codification of international law" was driven by the perceived need to "formulat[e] and re-establish[] and clarify[] international law." (32) This pressure culminated in the 1930 League of Nations Conference on the Progressive Codification of International Law, which, much like the League that sponsored it, has not been counted a success. (33)

Prior to World War II, then, neither codification efforts nor international organizations such as the League of Nations were terribly successful in legalizing international relations. Subsequent codification efforts under the auspices of the United Nations have met with greater success. In modern legal practice, the ILC is by far the most important organ of codification, despite the existence of many regional institutions devoted to codification, some of which are older than the ILC. (34)

In 1947, the U.N. General Assembly created the ILC pursuant to its mandate under article 13 of the U.N. Charter to "encourag[e] the progressive development of international law and its codification." (35) The General Assembly charged the ILC with "the promotion of the progressive development of international law and its codification." (36) The ILC stands at the front end of a state-centric process of lawmaking, preparing draft articles for states to consider adopting as treaties. (37) The ILC has produced codifications of many of the foundational areas of international law, including the law of treaties, diplomatic and consular law, and the law of the sea. (38) Today, codification continues to be a critical lawmaking tool at states' disposal, as evidenced by the debates on whether the United States should ratify the U.N. Convention on the Law of the Sea and by the 2010 negotiations over codifying the crime of aggression in the Rome Statute. (39)

Despite codification's continued significance in international relations, codification as a lawmaking technique has largely been relegated to an afterthought in contemporary scholarship. This is puzzling, because practicing international lawyers pay an enormous amount of attention to the dynamic relationship between treaties and custom. The United States' Model Bilateral Investment Treaty, for example, goes out of its way to include an annex confirming that the treaty's substantive rules reflect the parties' view of customary law. (40) Similarly, delegates spent much time at the 2010 International Criminal Court (ICC) Review Conference negotiating interpretive understandings governing the relationship between the adopted definition of the crime of aggression and the background customary law. (41)

Nevertheless, the scholarly commentary's casual references to treaties codifying existing customary norms seem to suggest that codification is simply the practice of reducing an uncontested legal rule to written form. The real work is done, one might infer, in the customary process that produced the rule being codified. Alternatively, the difficult work of international lawmaking might have been done in treaty negotiations that eventually developed into new rules of customary international law. In these situations, customary law is almost conceived of as a passive recipient of settled norms worked out through treaty negotiations against a tabula rasa. In either conception, treaties and custom interact gingerly, each disturbing the other only after appropriate deliberation and development.

The sharp dichotomy of this approach is reflected in the way codification is conceived of in the context of adjudication, as compared to its conception in the actual practice of lawmaking. In dispute resolution, when determining whether a treaty codifies customary international law, the question is generally reduced to whether the precise treaty rule in question was a customary rule when the treaty came into force. (42) In the context of lawmaking, however, this static conception of codification has generally been rejected in favor of a more dynamic understanding of the relationship between abstract customary rules that are agreed to be law and the more precise formulations of those rules that may be contested. For example, although the ILC Statute makes a distinction between the codification of international law and its progressive development, (43) in practice the distinction has had little effect. (44) As Hersch Lauterpacht, who served both on the ILC and as a judge on the International Court of Justice, put it,
   [O]nce we approach at close quarters practically any branch of
   international law, we are driven, amidst some feeling of
   incredulity, to the conclusion that although there is as a rule a
   consensus of opinion on broad principle ... there is no semblance
   of agreement in relation to specific rules and problems. (45)

Thus, in practice, codification has been an exercise in identifying areas of custom and attempting to fill in the gaps. It is in this dynamic sense that I use the term "codification." Seen in this light, codification plays a role often thought to be reserved for adjudication in common law systems. In the domestic context, for example, the creation of a standard is understood as a delegation to courts to fill in the content of a law ex post. (46) In the international context, though, delegations to courts are relatively rare, and international courts lack the procedural rules, such as compulsory jurisdiction and stare decisis, that give domestic judicial decisions expansive effect. (47) Given the weakness of most international tribunals, negotiations between states are the international legal system's most important method of developing the content of vague or ambiguous customary rules in most areas of international law. (48)

This elaborating effect also differentiates codification from treaty negotiations that do not occur against the background of existing customary rules. States have at their disposal a variety of political and economic tools to induce other states to accept legal obligations. What distinguishes codification is that it works against a backdrop of existing state beliefs. Influencing state beliefs about marginal changes in a rule--about a rule's specifics--is easier than persuading states that an entirely new rule exists or should be adopted. All states believe, for example, that customary international law requires that diplomats and diplomatic missions receive certain types of privileges and immunities. What they disagree about is not the general rule, but the specifics: for instance, does the inviolability of diplomats prevent airport security from searching them? (49) Arguments about the answer make reference to an existing body of law on the subject of diplomatic immunity. And because the arguments rest on established law, they increase the costs for failing to adhere to the opposing side's arguments. These costs--the costs of being judged in noncompliance with the generally applicable customary rule--are not present when states use political pressure to advocate adoption of a pure treaty rule. Codifications, as I use that term, are signals about what codifying states believe customary law requires and are therefore signals about the legal (as opposed to political or economic) costs of failing to adhere to the codified rule.


The problem of codification is, of course, not a new one. The extant literature contains several rationales as to why states might codify customary international law. I discuss what I take to be the two dominant rationales here: what I term the Clarification Thesis and the Compliance Thesis. Although the Clarification and Compliance Theses explain much about the drive to codify customary international law, they are undertheorized. In this Part, I develop these two theses and illustrate why they fail to fully explain states' interest in codification.

A. Clarification

Customary international law has a reputation for vagueness and ambiguity. (50) This reputation is in some respects well deserved. Purported customary rules, such as the precautionary principle, are so vague that they offer little guidance in application. (51) The drive to clarify customary rules to promote cooperation between states (and increase their overall welfare) has long been offered as a rationale for codification. (52) The "Clarification Thesis" is the notion that codification is driven by a desire to clarify what customary rules require in order to promote cooperation and increase global welfare. (53) I develop an account of when clarity in legal obligations increases overall welfare. Clarification is not an unalloyed good, however, and I illustrate how the need for clarity does not fully explain the codification of customary international law.

Customary international law usually fails to define precisely the legal fights or obligations it creates. The rights and obligations are ambiguous in ways that create disputes or lead to coordination failures. These disputes, in turn, are costly to states, creating incentives to clarify what the law requires. By way of example, consider the diplomatic pouch or "bag," the device that foreign missions use to transport official items free of search by the host government. According to the leading commentator on diplomatic law, no "limits on size or weight [of the bag can] be deduced from international practice." (54) As a result, states are not infrequently drawn into disputes about whether a particular shipment is too large to qualify for protection as a diplomatic bag. In 1985, to cite one public example, the Federal Republic of Germany refused to recognize a Soviet truck with a total load of 9000 kilograms as a single diplomatic pouch. (55) Even today, some countries (for example, China) impose limits on the size and weight of the diplomatic pouch, which arguably violate customary international law. (56) These limitations, in turn, create costly conflicts, as states are unable to coordinate their behavior and must expend resources negotiating for rights they believe they already possess and on which they base their global operating procedures. (57)

Although clarity is often beneficial, it is not always so. The literature on rules versus standards allows us to think more carefully about the precise conditions under which clarity (i.e., a rule) is likely to increase overall welfare. (58) To assess whether a rule or standard is more appropriate, we must weigh the ex ante net costs of elaborating the rule against the net costs of having to apply, comply with, and subsequently clarify an imprecise standard. (59) Clarification should increase overall welfare when the net present costs of clarification are less than the net present costs of deferring clarification. However, as I explain below, even where the net present costs of clarification are less than the net present costs of deferring clarification, the clarification rationale still may not descriptively explain codification. The reason is that familiar bargaining problems, such as transaction costs and holdouts, may prevent states from converting a welfare-improving change in the law into one that makes all parties at least as well off as they would be under the status quo.

The costs of ex ante clarification in the international context flow from (1) transaction costs of negotiations, (2) investing in complying with a clearer rule, and (3) violations that would not have occurred but for the clearer rule. The chief benefit of ex ante clarification is increased cooperation. First, consider the associated transaction costs. Discount rates--the idea that receiving something of the same value in the future is worth less than receiving it today--mean that, viewed in isolation, the deferral of the transaction costs of clarification is cost effective if those costs do not increase. By contrast, where transaction costs are expected to rise, such as where more parties will have to participate in a negotiation in the future, ex ante clarification will be cost effective. Thus, transaction costs do not necessarily favor clarification, particularly when states expect transaction costs to be stable or decrease over time.

Second, clearer rules promote compliance by delineating what counts as compliant conduct. (60) In doing so, clearer rules may avert some subsequent noncompliant actions, an obvious benefit of clarification (and a cost of deferring clarification). They may also save the cost of conflicts that arise in disputing the meaning of unclear rules. However, clarifying the law ex ante also has its costs: it will frequently force states to move away from their status quo behavior. For example, clarifying the precautionary principle to allow states to ban the sale of genetically modified organisms (GMOs) would place costs on those nations that regularly use GMOs in agriculture. (61) The costs of complying with a clearer rule are unlikely to be borne evenly by all states. Rather, the allocation of the costs and benefits of compliance with a clearer rule is likely to depend on the specific rule chosen.

Finally, clearer rules may make unlawful certain actions that would otherwise be undeterrable. Making undeterrable actions unlawful results in a net loss to the parties, both because there is a penalty for violating the law that would not otherwise be incurred and because there is no offsetting gain to nonviolating states. (62) Ex ante clarification can thus impose net costs on the parties where the costs associated with making undeterrable actions unlawful outweighs the benefits from greater cooperation under the clarified standard. Moreover, like the costs of compliance, the costs of undeterrable violations are likely to be unevenly distributed among the parties.

Taken together, these criteria suggest that clarification will be desirable from an overall welfare standpoint when (1) the transaction costs of clarification are expected to increase over time, (2) clearer rules are likely to influence the behavior of states, (3) the costs of complying with clearer rules are low, and (4) the benefits of increased cooperation are high. In some instances, the joint net costs of ex ante clarification will not justify any substantial state investment in clarifying the law.

Moreover, even where clarification can be justified in cost-benefit terms, clarification frequently may not occur. The foregoing analysis has focused on overall welfare. Put in terms of efficiency, I have asked whether a clarifying change in the law is Kaldor-Hicks efficient, meaning that it increases net welfare even if some states are made individually worse off. (63) International law, though, generally requires that states consent to a rule before they are bound. (64) The result of the consent requirement in international law is that changes in the law that are merely Kaldor-Hicks improvements will not survive the codification process. (65) States will not agree to changes in customary rules that make them worse off, even if states as a whole are made better off by the change. States will agree to a change in the law only if it benefits them individually; that is, in the consent-driven process at the core of international law, states will agree only to Pareto-improving changes in the law, meaning changes in the law that make no state worse off and at least one state better off. This fact is particularly important given that the costs of clarification are unlikely to fall evenly on states, meaning that in many instances changing the law will make some state worse off.

As a matter of bargaining theory, of course, any Kaldor-Hicks efficient improvement in the law can be transformed into a Pareto-improving change through transfers to states that are otherwise net losers. (66) But in the real world, transaction costs are often quite high. Since transaction costs reduce the surplus available for transfers to net losers, such costs may prohibit the conversion of Kaldor-Hicks improvements into Pareto improvements. Moreover, once the possibility of transfers has been introduced, states may hold out for a greater share of the benefits from cooperation, thus further delaying or derailing agreement.

Thus, the Clarification Thesis--understood to be the descriptive claim that states codify customary international law because clarity is to their joint benefit--is limited by the realities of the way states behave when making legal rules. Rational states are concerned with overall welfare only to the extent that increasing overall welfare enlarges the pie in a way that will allow them to increase their individual share of the benefits from cooperation. The Kaldor-Hicks efficiency of international law is not a primary concern of rational states, only a derivative one. The Clarification Thesis thus explains the drive to codify customary international law when clarification increases overall welfare and when transaction costs are sufficiently low to allow the conversion of welfare-enhancing gains into Pareto-improving gains. But the Clarification Thesis by itself cannot explain codification in areas in which there are sharp distributional consequences to clarifying the law and when familiar bargaining problems prevent states from creating a legal regime that redistributes the benefits of cooperation.

Consider, as an example, the Arab Charter on Human Rights. (67) The most recent version of the Charter was adopted by the League of Arab States in 2004 and came into force in 2008. (68) The U.N. High Commissioner for Human Rights greeted the Arab Charter with guarded enthusiasm. While welcoming the ratification required to "bring the Arab Charter on Human Rights into force," the High Commissioner noted that the Charter conflicted with universal human rights norms in its treatment of the legality of the death penalty for children and by equating Zionism with racism. (69) In particular, the Charter states that the "[s]entence of death shall not be inflicted on persons under 18 years of age, unless otherwise stipulated in the laws in force at the time of the commission of the crime." (70) The Charter also "[r]eject[s] all forms of racism and Zionism, which constitute a violation of human rights." (71)

The codification of these particular human rights reflects a disagreement over the correct way to interpret established customary rules. To be sure, the Arab Charter does clarify the Arab League's interpretation of racism, which the Restatement (Third) of Foreign Relations clearly identifies as a violation of customary international law. (72) But this clarification is hardly likely to improve cooperation among states in stamping out racial discrimination or to reduce conflict in the way that the Clarification Thesis imagines. The question of whether Zionism is a form of racism banned by custom has been hotly contested over the years. In 1975, the U.N. General Assembly passed Resolution 3379, which "determine [d] that Zionism is a form of racism and racial discrimination." (73) That resolution was repealed in 1991 by Resolution 46/86. (74) By codifying their preferred interpretation of the rule, Arab states attempted to influence the content of the customary prohibition on racism for their own benefit. Specifically, Arab states acted defensively to codify their minority interpretation of a customary rule. As I argue below, used in this way, codification acts as a commitment device, binding a group of states to a particular interpretation of customary rules, rather than as an effort to clarify universally applicable legal rules.

B. Compliance

Having explained the logic and limits of the Clarification Thesis, I now explain why concerns about promoting compliance also fail to explain codification. Customary international law is for the most part comprised of rules of conduct. That is, customary international legal rules create legal rights and obligations, such as the right to exercise sovereignty over the territorial sea and the obligation to refrain from committing genocide. Treaties, however, can include a variety of mechanisms, such as monitoring mechanisms or jurisdictional protocols, that can further promote compliance. As governments have frequently recognized, codification can therefore improve compliance with customary rules by appending to these rules compliance-promoting mechanisms that generally do not exist in customary international law. (75) I refer to this rationale for codification as the Compliance Thesis.

The mere existence of rules of conduct can change the way states behave when, for example, decentralized penalties, such as reputational sanctions, exist for violating the rules of conduct. But while bare rules of conduct can affect state behavior, their mere existence may be insufficient to promote an ideal level of compliance. Bare rules do not address international law's overarching problem: the problem of enforcement. International law is mostly a self-help system, in the sense that there is no centralized enforcement mechanism. States may therefore wish to create certain mechanisms that are likely to raise the costs of noncompliance with the substantive rules of conduct.

Codification allows states to solve the enforcement problem in a variety of ways. Most importantly, codification gives states the chance to provide advance consent to dispute resolution provisions. For example, the Optional Protocol to the Vienna Convention on Consular Relations grants jurisdiction over disputes arising under that Convention to the International Court of Justice. (76) Similarly, the U.N. Convention on the Law of the Sea establishes the International Tribunal for the Law of the Sea to resolve disputes. (77) Bilateral investment treaties and the investment chapters of regional free trade agreements are noted less for the manner in which they codify customary international law and more for the fact that their arbitration provisions create a private cause of action against host states that did not exist under customary international law. And in defining the crimes over which the ICC has jurisdiction, the Rome Statute codifies much of the substantive international criminal law developed in the twentieth century. (78) The codification exercise, however, was less about agreeing on the substantive rules of conduct and considerably more about creating an international mechanism for the enforcement of customary rules.

Codification also allows states to create ancillary monitoring and enforcement obligations that are specific to the customary rule codified. Treaties can create self-reporting obligations for states that allow other states and monitoring bodies to judge whether a state is honoring its customary obligations. For example, the International Convention on Civil and Political Rights, which is thought to codify certain customary human rights obligations, (79) contains self-reporting requirements. (80) The CAT, which codifies the customary prohibition against torture, also requires states to report on measures taken to comply with the Convention. (81) The CAT requires states to criminalize acts of torture in their domestic law and to investigate and either extradite or prosecute those who have committed such acts. (82) These treaties thus create a variety of enforcement obligations that do not necessarily accompany the bare customary obligations that the treaties codify.

Finally, codification can have an effect on custom's status as domestic law. For example, in the United States, some scholars have suggested that in the wake of Erie Railroad Co. v. Tompkins (83) customary international law should be treated as state, rather than federal, common law. (84) To the extent this view is adopted by courts, codification restores custom to its status as federal law by operation of the Supremacy Clause. (85) This, in turn, may allow federal courts to act as enforcers of international legal obligations.

Although the compliance benefits of codification are significant--and may, in individual cases, drive the decision to codify--they do not fully explain the phenomenon of codification. Many of the most important codifying treaties fail to employ any of the features identified above that can make codification attractive. The Vienna Conventions on Diplomatic and Consular Relations, for example, do not create mandatory dispute resolution procedures or enforcement or monitoring obligations. (86) And while treaties may clarify custom's status in domestic law, that benefit is specific to countries that make a distinction between custom and treaties in terms of domestic effect. In Germany, for example, customary international law trumps inconsistent statutes. (87) Moreover, even where dispute resolution, enforcement, and reporting obligations exist, institutional design and international politics may render those devices ineffective. Consider the attempts to define the crime of aggression in the Rome Statute. After years of negotiation, the 2010 ICC Review Conference held in Kampala succeeded in adopting a definition, (88) but this success was complicated by the inclusion of restrictive 'jurisdictional paths" through which the ICC could actually obtain jurisdiction over allegations of criminal aggression. (89)

Individual states' resistance to compliance-inducing mechanisms is hardly a surprise. First, just as with the Clarification Thesis, there is a threshold question of whether the joint costs of compliance-enhancing mechanisms outweigh the benefits. Because sanctions are negative sum in the international system, there will be instances in which the joint expected costs of compliance-enhancing mechanisms outweigh the joint expected benefits. (90) Second, the same distributive issues must be addressed even in situations in which compliance mechanisms are justified by reference to joint welfare.

Compliance mechanisms raise the costs of codification most for those states that are likely to be noncompliant. Those states need to be compensated for agreeing to higher penalties for noncompliance. But just as with the Clarification Thesis, transaction costs and holdouts may make the necessary transfers or side payments impossible. The United States, for example, refuses to join the Rome Statute. In such situations, where states are unable to address the distributional issues that arise from creating greater sanctions for noncompliance (or clarification), neither the Compliance nor Clarification Theses can adequately explain codification. Instead, we must look to distributional considerations as the primary motivator of codification. Part III of this Article turns to this analysis.


The Compliance and Clarification Theses are, in economic terms, efficiency theses. Clarifying the substantive content of legal rules or bolstering substantive rules with compliance-promoting mechanisms can increase the aggregate welfare of states. Assuming transaction costs are low, states should be able to create transfers sufficient to make every state at least as well off under the codified regime as it would be under the uncodified regime.

In the real world, however, transaction costs are frequently quite high, and rational states are not interested in increasing aggregate welfare as an end in itself. They are interested in maximizing their own individual welfare, and increasing aggregate welfare is only a means to that end. We should thus expect states to pursue codification as a means to attaining customary rules that serve their own interests, regardless of whether aggregate welfare is increased. I refer to the thesis that codification is driven by distributive concerns--rather than efficiency concerns--as the Capture Thesis. In this Part, I develop the Capture Thesis and explain how states can use codification as a mechanism to lock in customary rules favorable to their interests. In Part W, I examine some of the ramifications of this strategy, including the codification of globally suboptimal rules and the fact that, perversely, codification can actually increase the fragmentation of international law.

A. The Intuition

I begin this Section by sketching out the intuition behind the Capture Thesis. I develop the thesis more fully in the sections that follow.

To start, it is useful to say a few words about how customary international law is formed in the absence of codification. As discussed above, customary international law has a rule of recognition: the existence of a rule of custom is evidenced by a consistent state practice done out of a sense of legal obligation. (91) Custom does not, however, have any procedures for establishing the requisite practice or sense of legal obligation. There is no legislature to vote and no deposit of a ratification instrument to signal consent to be bound. Customary international law is, in short, formed in an environment that is not structured by the procedural rules and explicit commitments that mark the formation of most legal rules, including treaty rules.

In such an environment, the formation and content of customary rules are uncertain. As one of the leading theories of the formation of custom has it, bare custom is essentially formed through a process of legal claims backed by actions. (92) These legal claims may be isolated events. That is, state A makes a claim against state B, and state C makes an unrelated claim against state D. (93) As the resolution of these claims coalesces around a rule, we say that a customary rule has formed. The more frequently disputes are resolved in accordance with a putative customary rule, the more likely we are to say that such a rule exists. Custom can also change through a similar process of accretion as legal claims are resolved contrary to a status quo customary rule. (94)

Powerful states play a disproportionate role in the formation of bare custom. (95) The influence of the powerful may be due, in part, to the fact that powerful states are more likely to prevail in individual disputes--due to the application of political or economic pressure--and are therefore more likely to obtain favorable precedents leading to the conclusion that a customary rule exists. Powerful states are also generally more effective at publicizing their legal positions and, indeed, formulating claims as legal claims in a way likely to impact custom. (96)

Bare custom, then, is the product of uncoordinated or loosely coordinated action that results in states coming to believe, based on each other's actions, that a legal rule exists. The powerful may be able to exploit such an environment to develop beneficial legal rules.

As I explain below, codification changes the dynamic of customary law formation by allowing states to coordinate and commit to interpretations of customary rules, without the need to do so in the context of particular disputes. The basic idea behind the Capture Thesis is that states use treaties, and the explicit bargaining processes through which treaties are made, to shape customary rules in a way that works to their individual benefit. More specifically, states will seek to codify customary rules when they think codification will allow them either to move the customary rule in a direction favorable to their interests or to lock in their preferred rules to guard against future changes in the customary rule.

Both powerful states and weak states may use codification. States should use codification when they think the resulting customary rule will be more favorable to their interests than the rule resulting from the formation of bare custom. Put differently, states seek to maximize the extent to which rules operate in their favor. Thus, states that have obtained a desirable rule of bare custom may still use codification to improve the rule or to entrench it defensively against future changes; weak states may use codification to attempt to gain greater influence over customary rules or to resist the influence of the powerful in the formation of bare custom.

Codification facilitates the development of favorable customary rules in two ways. First, by framing a treaty rule as an interpretation of an existing customary rule, codifying states can change the calculus of states inclined to object to the rule's adoption. (97) Instead of simply disagreeing with the proposing state, objecting states must object to the views and practices of all parties to the codifying treaty. Moreover, the act of codifying is a commitment device. The treaty itself is a commitment by the codifying states to each other, to interpret the background customary rule in a particular way. To disrupt the customary rule as interpreted in the codifying treaty, objecting or noncodifying states must persuade codifying states collectively to abandon their interpretation of the codified rule. In effect, codification creates a collective bargaining framework. No longer can objecting states bargain individually with states over the correct interpretation of the codified rules. Instead, objecting states must bargain against the collective weight of all codifying states.

Second, states explicitly bargain over treaty rules, whereas bargaining over bare customary rules is, at best, implicit. (98) States can therefore deploy procedural rules and bargaining tactics in treaty negotiations that cannot be deployed in the same fashion in the formation of bare customary rules. This is not to say that customary rules are not also bargained for. The process of claiming that a particular state practice represents a customary rule, as opposed to merely a behavioral regularity, is itself a tacit bargaining process. (99) But it is a process that is relatively unstructured. There are no firm, objectively verifiable rules as to precisely when a putative rule of customary international law is ripe for acceptance by states or has passed into the law. Not so with treaties. International conferences can deploy a variety of voting rules--ranging from simple majority voting to consensus--that govern when the conference adopts a proposed rule. As will be seen below, states can use codification as a tool to introduce explicit bargaining over treaty rules when they expect the resulting rules will serve them better than rules emerging through the ordinary customary process.

Seen in this light, the codification of international custom is consistent with accounts of law formation in which repeat players "play for the rules." (100) Lawmaking processes are a battleground on which international actors--principally, but not exclusively, states--compete on the basis of their political and policy interests to define and interpret rules that will advantage them in their interactions with other actors. Theorizing the role of power politics in making law is important, particularly in international law, because it allows us to differentiate between law and politics. The two are related, but they are not coextensive. Previous work in international law and international relations has examined the role of power in the formation of international institutions, (101) treaty design, (102) and the choice of a treaty versus a soft law instrument. (103) An interest-based understanding of codification--as the attempt by the states to define international law in a self-serving manner--adds to our understanding of this phenomenon. Indeed, codification is of particular interest because it lies at the intersection of different sources of international law. Codification exploits the overlap between these different sources to leverage lawmaking ability in ways that put even greater pressure on the already stressed consent-based paradigm of international law.

It is precisely this overlap that distinguishes codification from other manners of forum shopping. In ordinary forum shopping, states seek to resolve an issue in the international forum that is most likely to yield the desired outcome. Powerful states thus often seek to move particularly important issues out of international bodies with broad membership and equal voting into bodies with narrower memberships or weighted voting. (104) Weak states, by contrast, seek out institutions that favor strength in numbers. (105) In both instances, rulemaking states try to create pressures for states excluded or marginalized in the rulemaking process nevertheless to adhere to the resulting rules. Driven by the need to cooperate, marginalized states accept such rules. In effect, forum shopping in international lawmaking has a coercive effect when marginalized states would rather cooperate under a less desirable rule than risk a failure to cooperate. For example, developing states joined the WTO after the United States and the EU withdrew from the GATT in 1947 because doing so was necessary to maintain most-favored nation trade obligations. (106)

Codification is a kind of forum shopping, insofar as states seek to shift international lawmaking from the relatively unstructured world of custom to the more highly structured world of treaties. But codification is also coercive in a way that most forum shopping is not. Should they resist cooperation, marginalized states risk not only the sanctions of foregone cooperation; they also risk legal sanction for violating the customary rule. Consider, for example, customary rules governing compensation for expropriation. In 1938, in response to Mexico's nationalization of oil and agrarian assets, U.S. Secretary of State Cordell Hull articulated the Hull Rule, (107) which requires prompt, adequate, and effective compensation in the event of expropriation. (108) The United States, as well as many developed countries, has long considered it to be the customary rule on compensation for expropriation. (109)

In the 1970s, developing countries sought to weaken the Hull Rule (along with the broader international investment regime) through a series of U.N. General Assembly resolutions. (110) In response, developed countries, led by the United States, began employing BITs to shore up the existing regime. (111) Arguably, and in my view, this wave of bilateral codifications has solidified the Hull Rule as a rule of customary inter national law. (112) Moreover, as I explain in greater detail later, this process of codification has worked in the way the Capture Thesis suggests. (113) The Hull Rule's codification in BITs weakened the position of states that signed the BITs and yet claimed that the customary rule was not the Hull Rule. And as the number of outliers dwindled, the pressure on the holdouts grew. (114) Despite this codification strategy's success, the recent economic downturn has caused some states to consider reviving the Calvo doctrine (115) and thereby "renationalizing" investment disputes. Ecuador, Bolivia, and Venezuala, for example, have denounced a number of their BITs. (116) Argentina has been particularly active in this area. For example, the Argentine legislature has considered a series of bills that would require the government to denounce its international investment treaties and shift investment disputes to Argentine courts. (117) Should Argentina breach its treaty commitments and shift international investment disputes to its national courts, and should those courts fail to apply international investment standards, Argentina would face major risks. These risks include reduced foreign direct investment, reputational sanctions, and retaliation from capital-exporting countries for violating the customary rules of international investment law.

In the remainder of this Part, I describe the Capture Thesis more fully and illustrate my argument with examples. At its core, codification involves two or more parties applying a particular interpretation of a more general rule between themselves. Rational parties implementing their preferred interpretation of a rule would also prefer that their form become the generally applicable version of the rule. In diplomatic law, for example, two states that agree on the size limits of the diplomatic pouch might prefer that the customary international law governing the pouch also contain the same limits.

While it is clear that states have an incentive to encourage the general adoption of their preferred rules, the mechanism by which they do so is less obvious. Below, I explain how states can use codification as a device to commit to a single interpretation of customary rules. Codification allows groups of states that individually cannot enforce their wills to create incentives for noncodifying states to adhere to the first group's interpretations of rules, giving codifying states collectively more influence than they would otherwise have individually.

B. Codification as Commitment

1. Commitment

Earlier academic work in international law and international relations has analyzed how an individual state's capacity to make a credible threat to withhold participation from an international agreement or organization impacts negotiations. (118) When an individual state can impose dramatic costs on other states for failing to cooperate on its preferred terms, it can disproportionately influence the content of cooperative rules. Much of the time, however, no individual state can exert that kind of influence. That is, individual states lack the ability to cause other states to change their views of the law simply by threatening to withhold cooperation or to sanction other states for violating their view of customary law. Codifying custom presents a solution to this difficulty, allowing groups of states to use treaties to commit themselves to a particular interpretation of a customary rule. By providing a commitment device, codification allows groups of states to make collectively credible threats to withhold cooperation from those states that violate the codified rule, or, at a minimum, sanction those that do not adopt their interpretation of a customary rule. Using codification this way is particularly valuable in situations in which threats by individual states are either incredible or unlikely to influence other states. In short, the collectively credible threats and sanctions that emerge from codified custom create incentives for noncodifying states to adhere to the codified interpretation of the customary rules.

Costly commitment by codifying states to a particular interpretation of a rule is important because, absent such commitment, individual codifying states might be pressured into or inclined to strike bilateral arrangements that deviate from the codified interpretation. In other words, codification is a bit like a Prisoners' Dilemma. Like-minded states collectively do better if they band together and adopt the same interpretation of customary rules against both codifying and noncodifying states. Such a commitment maximizes their ability to influence the behavior of noncodifying states by increasing their costs of not adhering to the codified rule. But states may also individually benefit in some instances from defecting from the group and applying interpretations of customary rules at odds with the codified interpretation. Codification can solve this problem by making it costly for states to apply alternative versions of customary rules, even with codifying states.

These increased costs result from several factors. First, the very act of agreeing to an alternative interpretation of a customary rule prejudices a treaty's effort to codify that rule. (119) Since the codified customary rule is valuable for codifying states in resolving future disputes, weakening it by providing a countervailing precedent causes harm, not only to the individual codifying state, but also to all codifying states. (120) This weakening leads to a second cost, namely that a codifying state that agrees to an alternative, weaker version of a customary rule may face sanctions (most likely reputational) from other treaty members. Codifying treaties thus can be understood to involve two separate promises by parties. The first promise is to comply with the obligations set forth in the treaty. The second promise--implicit in the claim that a treaty rule codifies custom--is to interpret the customary rule in accordance with the treaty. Since treaty members are bound by the codified version of the customary rule with respect to one another and have claimed that the rule is universal by virtue of being customary, this second promise is necessarily a commitment to apply the codified version to nontreaty members. Violation of this second promise--cooperating with a noncodifying state under an alternative understanding of a customary rule--can therefore subject codifying states to sanctions similar to those they might face for violating their treaty obligations. (121)

This promise not to discriminate between codifying and noncodifying states in the application of a codified rule is sufficiently important that it sometimes explicitly appears in the text of agreements. Article 311 of the U.N. Convention on the Law of the Sea, for example, expressly precludes states parties from being "party to any agreement in derogation" of the putatively customary principle of the common heritage of mankind, as defined in the Convention. (122) Likewise, the OECD's failed Multilateral Agreement on Investment was in part an effort to generate a legal obligation to coordinate OECD countries in sanctioning unlawful expropriations. (123) To that end, the United States and the EU agreed in 1998 to establish a registry of claims for illegally expropriated property and to apply disciplines to property listed on the registry. (124) Finally, as discussed at greater length below, members of the Andean Community codified the Calvo doctrine and, in doing so, specifically prohibited member states from entering into agreements with foreign governments that accorded foreign investment treatment superior to national treatment. (125)

For these reasons, members of a codifying treaty face higher costs for cooperating on terms other than those in the codifying treaty than they would have in the treaty's absence. These costs act as a commitment device, tying treaty members to the codified version of the customary rule. In effect, treaty members agree to sanction one another for cooperating with noncodifying states on terms that conflict with those contained in the treaty. This mutual sanctioning regime reduces the attraction of alternative rules. Treaties thus provide signatories with a credible threat to sanction other treaty members for violating the codified customary rules and, at the extreme, to withhold cooperation from nontreaty members if they do not adhere to the codified rule.

2. Piracy and the Outlawing of the Slave Trade

The process of outlawing the slave trade in the nineteenth century is an excellent example of how codifying treaties can be used as a commitment device. The key state in the movement to outlaw the slave trade was Great Britain, the world's premier maritime power. Throughout most of the eighteenth century, the British government had tolerated the slave trade, but the late eighteenth and early nineteenth centuries saw two changes in British attitudes toward the slave trade. The first was the rise of an abolitionist movement based on humanitarian concerns. (126) The second was a change in Britain's economic incentives. American independence and the revival of slavery in Napoleonic France meant that two of Britain's rivals were profiting from a trade that no longer benefitted Britain economically and that sections of British society found morally abhorrent. (127) Britain thus embarked on a decades-long legal campaign to ban the slave trade and to develop ship-boarding rights, called "visitation" fights, to enforce the ban. (128)

Britain resorted to the partial codification of custom as a technique to develop a legal regime that permitted it to enforce a ban on the slave trade. The codification was clearly driven by an effort to define customary law in a way that was advantageous to British interests, rather than simply to clarify the applicable customary rules for the sake of promoting cooperation. Specifically, the consensus by the early nineteenth century was that the law of nations outlawed piracy. (129) Despite this general agreement, piracy was not a clearly defined concept; the British sought to use this ambiguity to define, through treaties, the slave trade as an act of piracy. (130) Britain first codified such an interpretation of the slave trade in an 1826 treaty with Brazil. (131) The Quintuple Treaty of 1841--to which Britain, France, Prussia, Russia, and Austria were parties--declared in article 1 that engaging in the slave trade was an act of piracy. (132) Belgium later acceded to the treaty. (133) The import of this codification was that, if the slave trade were accepted as an act of piracy, slave ships would be "stateless" in the same way pirate ships are, and thus would be susceptible to boarding by any nation's warships--a rare exception to exclusive flag-state jurisdiction over ships on the high seas.

Fearing the expansion of British naval power, the United States actively resisted the codification. American efforts included an anonymous campaign in Paris to undermine French support for the Quintuple Treaty. (134) These efforts bore fruit, as France never ratified the treaty. (135) Ultimately, however, the United States conceded visitation rights in the 1862 Treaty of Washington, (136) which lead to the customary rule's emergence shortly thereafter. A century later, the 1958 Geneva Convention on the High Seas included, as a self-described codification of custom, boarding rights in the event that a ship is suspected of engaging in the slave trade. (137) Indeed, the only situations in which the Convention permits boarding a ship flying the flag of another country (other than a situation in which the boarded ship is in fact of the same nationality as the warship) is when the boarded ship is engaged in acts of piracy or the slave trade. (138)

The development of this right underscores the influence codification of custom can have on the development of bare customs that are binding on all states. Britain did not simply create boarding rights through treaty rules. Rather, Britain employed a codification strategy in which it obtained treaty commitments from like-minded states to both grant boarding rights and interpret a bare rule of customary law permitting boarding of pirate ships as applying to ships engaged in the slave trade. This codification had clear distributional consequences, given how disadvantageous it was to states still engaged in the slave trade. Despite these repercussions, the British codification efforts played a role in pressuring holdouts, particularly the United States, to eventually agree that the slave trade should join piracy as an exception to exclusive flag-state jurisdiction. The codification helped isolate the United States, so that when the pressures of the Civil War came to bear, the United States ultimately accepted the British-sponsored customary rule. (139)

C. Codification and Procedural Rules

Codification has an additional benefit for states trying to capture customary rules: it allows states to use procedural bargaining tactics, such as agenda setting and voting rules, that are not available in the ordinary processes of making customary law. These procedural devices can be used either to ensure that a favorable interpretation of a customary rule is codified, or, once a codifiying treaty exists, to create some measure of veto power over changes in the law. These tactics are, of course, also available in noncodifying treaty negotiations. But because codification is a kind of forum shopping, these tactics are particularly noteworthy. States attempt to pick the lawmaking venue in which they expect to obtain the most favorable rules. Thus, a discussion of the procedural tactics available in the formation of treaties, but unavailable in the formation of bare custom, is worthwhile. I discuss these two features of codification in turn. (140)

1. Procedure and the Codification of Customary Obligations

Customary international law lacks procedural rules. It does, of course, have a rule of recognition: state practice undertaken out of a sense of legal obligation evidences a putative customary obligation. (141) But this differs from most rules of recognition, which tend to focus on agreed procedures as a way to manifest adequate consent to a legal obligation. (142) For example, there are multiple accepted procedures on how states may manifest consent to a treaty, with one of the procedures typically chosen for inclusion in the "Final Provisions" section of a treaty. But, where customary international law is concerned, we know neither how individual states should convincingly demonstrate that their practices are done out of a sense of legal obligation, nor precisely how many states must act with such a sense before a customary obligation is created. Instead, states make claims about legal rights, often in the context of specific disputes, while other states observe how those confrontations are resolved and react accordingly in their own future disputes. (143)

The codification of custom allows states to escape this trap. First, states can make clear that a legal obligation exists, at least as between the parties to the treaty. Second, the introduction of procedural rules ensures that the precise rules codified are favorable to those who control the procedures of the treaty negotiations. In other words, codification allows states to take advantage of explicit bargaining tactics that are not available in the relatively unstructured world of customary law. These tactics include agenda setting, voting rules for adopting proposed treaty terms, and exclusion.

Agenda setting can be used to determine the precise form the codifying articles will take when they are put before a diplomatic conference for a vote. Frequently, this is done through the preparation of draft articles by organizations such as the ILC. As a formal matter, such draft articles do not prevent states from reopening a debate or conversation on any particular point if it seems worthy of negotiation. In practice, however, draft texts that have been negotiated in a smaller group often have a certain inertia that resists substantial amendment. For example, the definition of the crime of aggression emerging from the Kampala Review Conference of the Statute of the International Criminal Court in 2010 was first transmitted to the Assembly of States Parties from a working group in February 2009. The working group's substantive definition, which enjoyed considerable support, was adopted at the Review Conference without amendment. (144)

Voting rules can be wielded to similar effect. They can be used either to override a dissident minority when adopting draft articles or, in the case of supermajority voting, to block proposed rules that the minority disfavors. Voting rules, it is important to note, are different from the rules governing the entry into force of treaties. Generally, proposed treaty rules cannot be applied to a state as treaty rules unless the state individually accepts or ratifies the rules. (145) The voting rules I refer to here are rules for adoption by a diplomatic conference. Such adoption is a condition precedent to the draft articles being adopted as such. (146) Moreover, the actions of the diplomatic conference may underscore or undermine any claims that the draft articles are representative of customary law. Thus, voting rules are critical in advancing or defending against a codification agenda.

The Third U.N. Convention on the Law of the Sea (UNCLOS III) illustrates the importance of voting rules. In convening UNCLOS III, there was considerable dispute as to what the voting rules should be. (147) Eventually, the General Assembly selected consensus as the means of decision, with an admonition that voting was to be a last resort, used only after all efforts at generating consensus had been exhausted. (148) Developing nations that wanted to ensure they could hold out for favorable terms on seabed mining supported consensus. (149) The idea was that consensus gave states hold-up power, thus conferring more leverage on otherwise weak states for the issues important to them. (150)

Finally, exclusion is another tactic that states can use to bolster their ability to influence customary rules. Exclusion can take several different forms. For one, states may simply exclude a state from a conference during which they plan to negotiate a purportedly codifying treaty. This kind of exclusion is at the heart of the Capture Thesis of codification. States try to use treaties as devices to create a bloc sufficiently large to compel outsiders to adhere to customary rules that they had little say in creating. Exclusion can take the form of negotiating a regional treaty that purports to codify customary rules, such as the Montevideo Convention on the Rights and Duties of States, (151) or inter-American agreements on the law of the sea, such as the 1952 Santiago Declaration on the Maritime Zone, (152) and the 1970 Montevideo Declaration on the Law of the Sea, (153) in which groups of Latin American states asserted exclusive jurisdiction two hundred nautical miles out to sea. (154)

States can also exclude by concluding a series of bilateral agreements that codify a favorable interpretation of customary international law. By conducting a series of one-on-one negotiations, states with greater bargaining power may deny weaker states the opportunity to leverage either strength in numbers or the ability to link issues. (155) Powerful states may thus be able to obtain more advantageous codification of legal rules. The rise of BITs can be understood as a manifestation of this kind of tactic. (156) Developed countries claimed that customary international law requires strong protections for foreign investment, and in particular "prompt, adequate and effective" compensation in the event of expropriation. These claims were weakened by Resolution 3201's New International Economic Order (NIEO). (157) Thus, developed countries turned to bilateral negotiations as a form of exclusive rulemaking. (158)

This second kind of exclusion better reflects how multilateral negotiations actually happen. Large multilateral conferences rarely involve all states present actually negotiating over the text of the agreement. The transaction costs of such a negotiation would be so large that reaching an agreement would likely be unworkable. (159) Instead, primary negotiations are generally delegated to working groups, committees, or individual states. (160) Exclusion from a key working group or committee can therefore be tantamount to a near total exclusion from the negotiations. Moreover, in negotiations that cover a large number of issues, resource limitations may effectively exclude some states from negotiating across the full range of concerns. (161) A state excluded from negotiations does, of course, have the opportunity to vote on the final product and may in some instances have influence through friendly nations on the committee. Moreover, every state retains the ability not to ratify the treaty. But exclusion from negotiations reduces a state's influence over the content of the proposed agreement.

Although examples of the use of committees are legion, UNCLOS III again illustrates the dynamic. At UNCLOS III, there were three main committees, each a committee of the whole, meaning that all represented parties were present. (162) However, negotiations with all states present proved unwieldy. Committee I, the most contentious committee, was pared down into a working group with ten developed nations, ten developing nations, and China. (163) Committee II, which dealt with traditional law-of-the-sea issues, also had private working groups, including one convened to address the exclusive economic zone. (164) While creating a working group has an efficiency rationale because it speeds up negotiations, it also privileges the input of states actually in the working group. Moreover, at least one participant in the UNCLOS III negotiations indicated that the negotiations were so sprawling, and the issues so numerous, that only a few of the most well-staffed delegations were able to track and effectively participate in the entire set of negotiations. (165)

2. Procedure in the Amendment of Codified Customary Obligations

Another advantage (or possible disadvantage) of the treaty mechanism is that treaties, unless they contain clauses to the contrary, can only be amended through the treaty-making process. (166) But precisely because customary international law lacks clear procedural rules governing its creation, its evolution is much more fluid. Treaties, by contrast, generally create veto points for amending treaty rules. Unless a treaty specifies otherwise, a state will not be bound by an amendment without its affirmative consent. (167) It is therefore potentially easier to disrupt a pure rule of custom than to amend a treaty rule. Codification thus has the potential to slow down the development of custom, (168) and, more importantly, to endow states that are parties to a codifying treaty with a veto on a rule's repeal.

In large part, the increased difficulty in changing codified customary rules stems from the fact that a greater number of states must, as a matter of procedure, weigh in on the change in the law. Where bare customary rules are concerned, the number of states participating in the formation or change of a customary rule may, by necessity, be limited to those few states particularly interested in the rule. By contrast, if the rule has been codified in a large multilateral treaty, under the Vienna Convention on the Law of Treaties each party will have to decide whether to accept the new rule. (169) In that context, rejections of an amendment to a customary rule could defeat the effort to change the codified customary rule. (170) Thus, treaty-amendment procedures can force greater participation in the customary lawmaking process, potentially to the detriment of custom's ability to evolve.

To be sure, this veto power is not absolute. Codified customary rules can continue to evolve outside of the codifying treaty. (171) Even treaties that codify customary international law sometimes include clauses meant to preserve custom's ability to evolve outside of the treaty process. For example, article 10 of the Rome Statute states, "Nothing in this Part shall be interpreted as limiting or prejudicing in any way existing or developing rules of international law for purposes other than this Statute." (172) Such statements are meant to preserve space for customary rules to evolve outside of a treaty regime that codifies then-existing custom, although it is unclear how successful these statements are in preventing an understanding of the customary rule from coalescing around the treaty rule. To take but one example, the Nuremberg Tribunal applied the 1907 Hague Convention on the Laws and Customs of War on Land to defendants as customary international law despite the fact that the Convention contains a si omnes clause that states, "The provisions contained in the Regulations referred to in Article 1, as well as in the present Convention, do not apply except between Contracting Powers, and then only if all the belligerents are parties to the Convention." (173) The Judgment of the Tribunal on this point rested explicitly on the fact that the Hague Convention sought to codify international humanitarian law by "revis[ing] the general laws and customs of war." (174)

Another way codifications can avoid stunting the development of custom is through the interpretation of a codifying provision as referencing existing customary law at the time the provision is applied, rather than the body of customary law from the time the treaty was adopted. Arbitral tribunals have taken this view when deciding investment disputes under both BITs and the investment chapters of regional free trade agreements. In applying the international minimum standard of treatment--the level of treatment that foreign investors are entitled to as a matter of international law--or the fair and equitable treatment obligation as part of that standard, tribunals under the relevant treaties have repeatedly held that the standard is evolving and should be applied as it exists at the time of the dispute. (175) In Pope & Talbot v. Canada, for example, Canada argued that Neer v. Mexico, a 1920s arbitration, dictated the international minimum standard of treatment. Neer held that the international minimum standard was breached by treatment that "amount[ed] 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." (176) The Pope tribunal found, however, that the customary standard had evolved since the 1920s and that NAFTA article 1105 required application of the evolving, rather than the 1920s, standard. (177)

D. Treaty Membership

Thus far I have shown how the codification of custom allows groups of states to band together and credibly sanction noncodifying states for failing to adhere to a codified version of customary rules. What, though, are the constraints on this practice?

The chief constraint on using codification to control the background customary rule is the requirement of a minimum level of participation in the codifying treaty. (178) Ultimately, to successfully induce noncodifying states to adhere to the codified rules, the treaty must have enough members to make nonadherence unattractive. Large memberships in codifying treaties necessarily increase the cost of nonadherence for noncodifying states in two ways. First, as membership grows, nonadhering states will be brought into conflict with an increasing number of codifying states because codifying states may choose to withhold reciprocal benefits from nonadhering states. At a minimum, the transaction costs of cooperation between these states will rise, as the codified rule will provide no default rule for cooperation. More codifying states mean more of these costly relationships for nonadhering states, thereby making adherence more attractive.

Second, as codifying states unify their interpretation of the underlying customary rule, the potential for reputafional sanctions for nonadhering states rises. All else equal, more states holding the same view of what custom requires will increase the reputational sanctions for those who do not comply with that view of custom. Thus, nonadhering states are faced with the transaction costs presented by trying to cooperate with codifying states, as well as additional reputational sanctions if they fail to adhere.

Although a larger number of codifying states should increase the number of noncodifying states that adhere to the treaty, increasing the number of codifying states is not costless. There are at least two significant drawbacks to expanding the number of codifying states: increased transaction costs and dilution of the rules agreed to in the treaty.

First, increasing the number of parties to a negotiation is commonly thought to increase the transaction costs involved in the negotiation. (179) As more parties become involved, the act of negotiating becomes more costly because more time is necessary for each state to set out its positions, make its proposals, and define its views on other states' proposals. (180) Moreover, a larger number of parties can make it more difficult to control the agenda, increasing the likelihood of cycling among alternatives and the resulting delays. These increased transaction costs affect the probability of reaching an agreement. Costlier negotiations mean a greater likelihood that bargaining will break down or that states will be deterred from opening negotiations in the first place. (181) Transaction costs also reduce the cooperative surplus created in the event an agreement is reached. (182)

Second, scholars have recognized that there is a tradeoff between the breadth of an agreement's membership and the depth of its substantive terms. (183) More specifically, when the standard of conduct required by a proposed rule does not vary among states, an increase in the number of states participating in the negotiations may dilute the standard of conduct upon which those states can agree. (184) The reason for this tradeoff is rather straightforward. Under any given rule of decision (e.g., majority rule, supermajority, unanimity), there is a marginal vote that is necessary to bring the proposed rule into force. The proposed standard of conduct thus must be set at a level that attracts that marginal vote. As more states become party to the vote, more votes become necessary to implement a proposed rule. And as state preferences become more heterogeneous, the proposed rule must be diluted to attract additional votes up until the point at which the rule attracts the marginal vote necessary to bring the rule into force. (185)

As an example, a state must generally consent to a treaty before it can be bound by it. (186) The decisional rule to bind all necessary parties is therefore unanimity. (187) Unanimity becomes a considerably more imposing requirement in negotiations as the number of states increases because the diversity of objections to stringent standards is likely to expand as well. That is, larger groups of necessary parties are likely to have a greater diversity of preferences, making distributional conflicts more severe. (188) To accommodate objecting states, the proposed standard has to be diluted or made vague so that a variety of actions can be considered compliant.

The negotiation of the definition of the crime of aggression in the Rome Statute illustrates the costs of expanded membership. While there is general consensus that "aggression" is contrary to customary international law, there is strong disagreement about what specifically constitutes the crime of aggression. (189) The major fault line in the negotiations was how much of a gap there should be between an "act" of aggression and a "crime" of aggression. (190) One group of states wanted no separation; every act of aggression could be prosecuted as a crime of aggression. (191) A second group of states advocated a higher standard, seeking to limit the crime of aggression to "flagrant" breaches of the U.N. Charter, wars of aggression, and "unlawful" uses of force or acts of aggression geared toward occupying or annexang territory. (192)

Instead, a standard of "manifest" violation of the U.N. Charter was adopted. The term was adopted precisely because it is indeterminate: it allowed states to resolve a serious distributional conflict with a vague standard that avoided clarifying the law in a way that benefitted one side or the other. This need to compromise on ambiguous language resulted from the fact that the voting rules on adopting amendments required a near unanimous vote at the Kampala Conference. Specifically, the rules on amendments required that for an amendment to be adopted, two-thirds of the full membership of the Rome Statute vote in favor. (193) At the time, the Rome Statute had 111 members, requiring seventy-four votes to adopt any amendments. (194) Initially, the Credentials Committee indicated that only seventy-two states had presented credentials entitling them to vote, although during the conference some maneuvering increased that number to eighty-five. (195) Thus, while unanimity was not strictly required, only a small number of defections could occur before the adoption of amendments would not be possible. In short, expanded membership, coupled with the voting rules at the Review Conference, introduced greater distributional conflict and made adopting a clearer definition of aggression impossible.

E. Conclusion

This Part has described how codification can be used to capture customary rules and the limits on that practice. Specifically, codification allows a group of like-minded states to commit to a single interpretation of a disputed customary rule. Codifying the treaty raises the costs of applying alternative versions of codified rules to noncodifying parties. This, in turn, raises the cost to noncodifying states of refusing to adhere to codified rules they may have played little role in forming. By refusing to adhere, noncodifying states risk being unable to cooperate with codifying states and being sanctioned for violating the customary obligation as codified. Moreover, codification can allow states to deploy explicit bargaining tactics that are unavailable in the unstructured environment in which bare rules of custom are formed. This practice is constrained, on the one hand, by the need to include a minimum number of states in order to create sufficient incentives to pressure noncodifying states to adhere to the codified rule. On the other hand, states wish to include no more states than necessary in the codification process to avoid the dilution of the codified rule that can result from broader participation in negotiations.


In this Part, I discuss the most important ramifications of the Capture Thesis. First, I argue that codified customary rules will often be Pareto inferior to uncodified custom because codification can be used to capture the content of customary legal rules. This result is highly counterintuitive because states must consent to be bound by a treaty, suggesting that a codifying treaty should be Pareto improving. Second, and more importantly, the fact that codification can be used to capture customary rules increases the fragmentation of international law. Codification increases fragmentation by creating an incentive for states to codify their own preferred interpretation of customary rules. In other words, groups of states with different interpretations each have an incentive to codify their own interpretations. Multiple codification efforts increase fragmentation by entrenching disagreements about what custom requires along regional or ideological lines. Codification can thus have precisely the opposite of its intended effect. Rather than unifying customary law, codification can actually create deeper interpretative divisions.

A. Suboptimal Rules

One of the central questions about customary international law, as with other forms of decentralized law, is whether it evolves toward efficiency. (196) Professor Eugene Kontorovich, for example, has argued that customary international law is unlikely to evolve toward efficiency because states do not share the attributes of a community likely to produce efficient norms. (197) In the international context, codification could in theory offer a way to ensure that customary law evolves toward efficiency. As Professor Kontorovich says, "Treaties are presumably welfare maximizing under the demanding Pareto criterion: they have the unanimous consent of the parties to the treaty, and thus presumably make all parties better off." (198) Put differently, norms that evolve through contractual mechanisms, such as treaties, might have a presumption of Pareto superiority that should not be accorded to customary norms not explicitly bargained for.

Contrary to the presumption that consent ensures that codifying treaties are Pareto-improving relative to bare custom, I argue that codified custom will often be Pareto inferior to bare custom. The difficulty with attaching a presumption of Pareto superiority to codifying treaties is that doing so minimizes the role of power in bargaining. Specifically, while voluntary agreements are presumptively Pareto improving, that need not mean that they are Pareto improving relative to the status quo, bare custom. Instead, powerful states or groups of states will often have the ability to define another state's options by threatening to withhold cooperation or by sanctioning states for failing to comply with the more powerful states' preferred interpretation of a customary rule. States with a credible threat to withhold cooperation or penalize nonconformance have the ability to take the status quo off the table. In such situations, a treaty need not be Pareto superior to the status quo; it only needs to be Pareto superior to the alternative, which may be no cooperation at all, cooperation under an even less desirable set of rules, or cooperation with a sanction for violating the powerful states' understanding of custom. (199) It follows from this logic that powerful states or groups of states have the ability to codify interpretations of customary rules that are Pareto inferior to a status quo interpretation or to the views of weak states.

The following example illustrates how a codified customary rule can be Pareto inferior to preexisting bare customary rules: the exemption from taxation enjoyed by the "premises" of a diplomatic mission, codified in article 23 of the Vienna Convention on Diplomatic Relations (VCDR). (200) The exemption from taxation of the mission's premises was not a customary rule in the nineteenth century, but by the 1930s, a number of foreign ministries and courts appeared to accept the exemption as a customary rule. (201) In 1969, after the VCDR and the Vienna Convention on Consular Relations (VCCR) had entered into force--but before the United States was a party to either treaty--the State Department argued during litigation that customary international law required exempting consular mission premises in New York from taxation. (202) The New York Court of Appeals adopted the State Department's position, suggesting that the exemption had hardened into a customary rule no later than its codification in those two treaties. (203) Thus, in codifying the rule on tax exemptions for mission premises, codifying states also changed the rule to require exemption where the status quo rule had not.

States reciprocally apply the exemption on taxation of mission premises, such that the exemption both benefits and burdens each state. The precise benefits and burdens, however, depend on the amount of property a government owns abroad that qualifies as "premises of the mission" within the meaning of the codified customary rule. Exemptions favor states with large diplomatic corps. Under a regime with no exemption, those states would be net payers of taxes. States with small diplomatic corps would be net recipients of tax receipts. That is, they would have to pay taxes on their relatively few, small missions abroad but could tax the relatively large number of missions they host. Codifying a rule requiring tax exemption for mission premises therefore worked an almost entirely distributional change in customary law. Tax receipts were reallocated from governments with small diplomatic corps to those with large diplomatic corps. Because some states lost tax receipts, this change by definition is not Pareto superior to the status quo rule that permitted but did not require exemption.

Nor is there any doubt that states themselves think of the codified customary rule in precisely these terms. The public record from a recent dispute over the taxation of permanent missions to the United Nations makes clear that states do indeed view the codified customary rule as principally distributional. City of New York v. Permanent Mission of India to the United Nations presented the question of whether residences for mission staff located on the same property as clearly tax-exempt office space were themselves tax-exempt "premises of the mission." (204) New York City initially filed suit against a number of foreign missions, including India's, seeking a declaratory judgment to validate millions of dollars in tax liens against the foreign missions. (205) The missions, for their part, claimed that the exemption for staff residences is required by the customary rule codified in article 23 of the VCDR. (206) Specifically, the VCDR defines the premises of the mission as "the buildings and the land ancillary thereto, irrespective of ownership, used for the purposes of the mission including the residence of the head of the mission." (207) Consistent with state practice that generally accorded such residences tax exemption, the missions maintained that mission staff residences were indeed "used for purposes of the mission/consulate." (208)

The case went to the Supreme Court and, on remand, the State Department intervened to "resolv[e] a dispute" that it called "a major irritant in the United States' bilateral relations [that] threatens to cost the United States hundreds of millions of dollars in reciprocal taxation." (209) As the State Department explained in its determination granting tax-exempt status to residences of staff of missions to the United Nations and the Organization of American States, "As the largest foreign-government property owner overseas, the United States benefits financially much more than other countries from an international practice exempting staff residences from real property taxes, and it stands to lose the most if the practice is undermined." (210) Simply put, the State Department publicly justified the tax exemption for mission premises as a distributional issue. The exemption benefits the United States implicitly at the expense of other states that lose the benefit of tax receipts from countries--like the United States and, perhaps, India--that have large diplomatic corps. In other words, the United States accepted that the tax-exempt status of diplomatic property crystallized by the VCDR and the VCCR is not a Pareto superior improvement over the rule permitting but not requiring exemption. (211) At best, the rule simply redistributed the gains from cooperation to powerful states and possibly reduced overall welfare in the process.

The fact that power and its application in procedural settings that magnify the influence of certain actors can interfere with the evolution of customary law toward efficiency is a familiar result. Repeat players, for example, can strategically litigate in ways that cause the law to evolve to reflect their interests, regardless of whether such self-interest-driven law is efficient. (212) Understanding how international law is subject to the same self-interested pressures is important in evaluating the normative desirability of particular rules and rule-making processes. In particular, the finding that the consent paradigm does not actually protect states from changes in the law that make them individually worse off further undercuts the normative support for retaining a strong consensual requirement in international law.

B. Fragmentation

Codification creates an incentive for noncodifying states to adhere to codified customary rules. It therefore becomes a tool to make international law without the consent, or at least the direct participation, of other states. Despite the drawbacks of codification--in particular the lack of participation by noncodifying states and the possibility of suboptimal rules becoming universal--there is at least one potential benefit to using codification to, effectively, legislate: international law will be harmonized, with a single standard prevailing over multiple standards. Codification, then, can aid cooperation by solving a coordination problem. In essence, this is the Clarification Thesis.

Unfortunately, the very fact that codification can entrench a particular interpretation of customary law, but that international law does not locate the authority to codify in any single institution, undercuts codification's value as a coordination device. Instead, the fact that codification can change noncodifying states' behavior creates incentives for groups of states to engage in diverging codification efforts in an attempt to entrench each group's own preferred version of a customary rule. These multiple codifications--a species of what has been termed "fragmentation" (213)--dramatically increase the cost of coordination going forward, which is even worse than simply leading to the breakdown of coordination on a single legal standard. Moreover, competing codification efforts will often be linked to broader disagreements about both the content and the appropriate role of international law. Multiple codifications therefore will tend to exacerbate regional and ideological tensions in international law and international relations.

Fragmentation--"the emergence of specialized and (relatively) autonomous rules or rule-complexes, legal institutions and spheres of legal practice" (214)--is a concept that has occasioned much debate in the international legal community. (215) On the one hand are those who have lauded fragmentation as an indication of the increasing relevance of international law. (216) In particular, fragmentation is thought to encourage pluralistic legal dialogue. Professor William Burke-White has argued that "an increasingly loud interjudicial dialogue.... provides actors at all levels with means to communicate, share information, and possibly resolve potential conflicts before they even occur." (217) Additionally, Professor Kal Raustiala has argued that transnational networks of government officials can support the effectiveness of, and compliance with, treaties through similar cooperative dynamics, as well as the "regulatory export" of "rules and practices from major powers to weaker states." (218)

On the other hand are those worried that fragmentation "may jeopardize the unity of international law and, as a consequence, its role in inter-State relations." (219) Fragmentation creates the possibility, and in many cases the reality, of conflicting judicial decisions based on similar or the same law. (220) Because international law lacks a hierarchical court system, there is no mechanism to resolve these conflicts. Moreover, multiple overlapping international institutions create the possibility of inconsistent treaty-based obligations and forum shopping. (221) These overlapping obligations can reduce international law's effectiveness by rendering international legal obligations unclear or making it difficult to comply with all of the obligations governing a single activity. The availability of forum shopping, and indeed fragmentation itself, as litigation strategies can also raise the transaction costs for states that want to change international legal rules. (222) Of particular note, Eyal Benvenisti and George Downs argue that fragmentation prevents logrolling--exchanging votes across unrelated issue areas--and coalition building by weaker states. (223) Weak states cannot logroll because fragmentation means that rules in different issue areas are being worked out in isolation. (224) In this respect, forum shopping inhibits learning. Powerful states remain in one forum until weak states begin to gain control of the forum; the powerful then decamp to another forum to conduct negotiations. (225)

I make a third argument. Fragmentation can result from a race to capture customary rules through codification. The fragmentation of codified custom undermines customary law as a whole by removing any pretense of a unified customary law binding on all nations. Divisions over how to interpret and apply customary rules become enshrined in treaties that have little hope of becoming universal. (226) Moreover, the fact that the divisions have now been codified significantly increases the cost of future coordination. Instead of bringing other states along one at a time, states that have codified a competing version of custom face higher costs if they abandon their treaty in favor of another codifying agreement. Below, I describe how codification can lead to fragmentation and then explain the difficulties of fragmented but codified customary law.

1. How Codification Leads to Fragmentation

Codification can lead to fragmentation in at least two ways. First, fragmentation can result from a straightforward failure to coordinate on the same interpretation of customary law, even when coordination is in each group's interest. If different groups of states simultaneously attempt to codify their own preferred versions of a customary rule, then they may end up codifying different versions of the rule. This inefficient outcome occurs when states prefer cooperation, but distributional tension exists over what standard states that would like to cooperate should adopt. (227) Each group of states codifies its preferred version, even when both groups would prefer a single interpretation of the customary rule to two different ones.

In the real world, this outcome is unlikely. Codification does not happen instantaneously. It takes time, and information about ongoing codification efforts among states is relatively good. Thus, a first-mover can establish a focal point with an initial codification and thereby reduce unintended fragmentation.

Early codification efforts following decolonization likely benefitted from this type of first-mover advantage. Newly independent states did not pursue separate codification efforts because they either did not realize the effects of codification or did not develop a firm view on what customary rules favored their interests. (228) Once the VCDR or the VCLT codified interpretations of those two laws, states preferring coordination had little choice but to adhere to those conventions. (229) A separate codification no longer had the chance to attract support, making it less desirable than simply coordinating on existing rules.

The second way in which codification can lead to fragmentation--and the more prevalent in my view--is when two groups of states in sharp distributional conflict over the content of legal rules each group tries to use codification as a means to capture the customary rules. The intuition is that the codification of custom can have a polarizing effect, similar to that of political parties on domestic politics, thereby stymieing the will of the median voter (or state, in this case). (230) This situation differs from a mere failure to coordinate because the two groups do not necessarily prefer cooperation with each other. Rather, codification is part of a strategy that each group of states employs to lock in its preferred rules and attract states that might otherwise resist. However, if two groups of states employ this strategy simultaneously, they can prevent a single uniform understanding of a rule from emerging.

To illustrate how this dynamic can lead to fragmentation, suppose two groups of states that are each considering codifying an area of customary law. Each group (or individual state within a group) would like the codified rule to maximize its own utility. A single codifying state's utility is a product of two factors: how close the codified rule is to its preferred rule, and how many states (and also which states) are expected to adhere to the codified rule. These factors reflect the fact that, as discussed above, codifying states face a tradeoff between admitting more states to the codification effort and codifying their preferred rule. Having more states cooperate on the codification effort not only translates into more pull for the codified rule among noncodifying states, but it also likely leads to a weakening or generalizing of the codified rule. (231) An individual state would only willingly move away from its preferred rule if the reduction in utility was offset by the utility gained from additional adherents.

This tradeoff has significant implications for codification efforts. Groups of states with sharply divergent views of the ideal version of a customary rule are unlikely to moderate their ideal to attract states on the opposite extreme. When states' ideal points are too far apart, attracting states with diametrically opposed views involves giving up too much value in terms of the rules of cooperation, even factoring in benefits from additional adherents. Codifying states with extreme views are thus unlikely to be interested in expending significant effort to attract other states with which they have a sharp distributional conflict. Rather, the tradeoff between the rule and the number of expected adherents is likely to cause codifying states to focus on moderate states.

This dynamic might at first blush seem good for cooperation. The median voter theory suggests that where extreme groups compete for the support of moderates, proposed rules should collapse to the proposal the median voter favors. (232) Codification should therefore have a unifying effect. Unfortunately, as experience with domestic politics shows, intermediate institutions can interfere with this result. (233) An intermediate institution, as I use the term here, refers to an institution that makes a rule or policy choice that binds only its participants but acts as a proposal to the general population. Political parties and chambers of Congress are intermediate institutions in this sense. In nominating a candidate, for example, a party selects its representative in the general election, but that selection is merely a proposal to the general population. Intermediate institutions tend to force proposals to the median of participants in the institution. Where those participants are not representative of the general population, intermediate institutions can have a polarizing effect. Put differently, where the median member of an intermediate institution differs from the median member of the general population, the intermediate institution can prevent policy choices from collapsing on the general population's median. Thus, the median member of Congress may be rendered impotent because he is far from the median member of his respective chamber, and it is the median member of the chamber who controls that chamber's policy position.

A codifying treaty is an intermediate institution in the sense I use the term here. The treaty establishes a rule binding on its parties, but the codified interpretation of the customary rule is really a proposal to noncodifying states to adopt a rule as custom. The tradeoff between moving away from a preferred rule and attracting adherents will determine how close to the center codifying states are willing to move. Codifying states with outlying preferences will resist changing the codified rule if attracting moderates is too costly. This resistance creates space for an alternative codification effort in the other extreme. Two (or more) different codification efforts ultimately raise the cost of reconciling states to a single, unified understanding of customary law.

In this way, codification can entrench differences. Two groups of states flame their preferred rule in an effort to attract moderate states. The very purpose of codification is, as set forth above, to negotiate a rule without including those who would dilute the rule. But resistance to dilution and inclusion results in an interpretation of a customary rule that cannot gain universal acceptance. Moreover, once there are two or more different codifications of the same rule, achieving a universal interpretation of the rule becomes even harder than it was prior to codification. Codifying states can no longer adopt a new interpretation of the customary rule without cost. Instead, they incur either the political and legal costs of exiting or violating the codifying treaty or the costs of persuading other treaty parties to amend the rule. Either way, these costs inhibit the creation of a unified understanding of custom.

2. International Investment Law as Fragmented Codification

The evolution of international investment law illustrates how fragmented codifications can impede the development of a truly universal understanding of customary rules. Below, I discuss two aspects of the codification of international investment law. First, I address the codification of the Calvo doctrine in the 1970s, particularly by Latin American states, and the role it played in inhibiting the development of modern investment law. Second, I analyze how the codification of investment law in BITs and regional international investment agreements (IIAs) has prevented a multilateral investment framework from emerging. Investment law has evolved from primarily customary law prior to the 1960s to an area dominated by BITs. (234) I argue that the fragmented codification of international investment law is a key, and underappreciated, reason that states have been unable to agree on a truly multilateral set of investment rules.

The failure to reach a multilateral investment agreement, despite repeated efforts, is puzzling for a number of reasons. Unlike international trade in goods and services, which was only lightly regulated on a multilateral basis prior to the various GATT/WTO agreements, international investment law has an extensive background of multilateral customary law on which to draw in a multilateral codification effort. Moreover, states appear willing and able to reach agreements codifying international investment law in BITs. The convergence among BITs has led some commentators to observe that BITs themselves might properly be termed a multilateral regime, even if they are not multilateral in form. (235) Why, then, have states failed to reach a substantially similar multilateral agreement?

Prior work on BITs has suggested that, in bilateral negotiations, developing states may compete with each other by establishing legal rules designed to attract capital. (236) Developing states resist strong investment rules in multilateral settings because their strength in numbers allows them to hold out for more favorable terms. (237) By contrast, bilateral settings isolate developing states and render them unable to engage in logrolling, issue linkages, or any of the other tactics that contribute to their success in multilateral environments. (238)

While some empirical evidence suggests that this competitive dynamic may indeed have driven the diffusion of BITs, (239) this story cannot fully explain the failure to coalesce around a single multilateral investment agreement. Developed states have tried several times to reach a multilateral agreement on investment among themselves at the OECD that would be open for accession--but not renegotiation--by developing countries. (240) Accession to such an exclusively negotiated multilateral agreement would have sent the same signal about the future treatment of foreign direct investment as signing a series of BITs, and developing states would still have been unable to use their strength in numbers to obtain favorable terms. Despite this, a multilateral agreement on investment has never even been opened for signature. (241) I argue that conflicting codification efforts, such as the codification of the Calvo doctrine and the variations in substantive protections afforded by BITs, enhance our understanding of the obstacles to concluding a multilateral agreement on trade in capital.

a. The Codification of the Calvo Doctrine

International investment law is primarily a descendant of international rules governing the treatment of aliens. (242) In its origins, it required that states treat aliens as well as they treat their own nationals--the so-called national-treatment obligation. (243) Over time, developed states claimed that an international minimum standard of treatment complemented the national-treatment obligation. (244) That standard established a floor for treatment of aliens, even if such poor treatment was consistent with a state's behavior toward its own nationals. (245) Many arbitral tribunals have held that the exact content of the international minimum standard evolves over time. At one time, the standard might have aligned with the tribunal's decision in Neer v. Mexico--that only outrage, bad faith, or willful neglect toward an alien violated the international minimum standards. (246) However, in recent years, tribunals--although not without exception (247)--have expressly rejected that standard in favor of standards based on more modern concepts of fair and equitable treatment and full protection and security. (248) This modern view includes rules requiring that expropriation be for a public purpose, nondiscriminatory, in accordance with due process, and accompanied by the payment of prompt, adequate, and effective compensation (the so-called Hull Rule). (249)

The developing world has contested the existence of the international minimum standard for decades. (250) As noted above in Section III.A, the Hull Rule itself comes not from a judicial decision but from a diplomatic note sent by the United States to Mexico in the midst of a dispute over the Mexican expropriation of American-owned oil fields and agrarian lands. (251) Mexican resistance to the Hull Rule was part of a larger opposition by both Latin American and Eastern Bloc countries to the idea that customary international law required more than national treatment. (252) The Calvo doctrine embodied this resistance to developed countries' claims about custom. This doctrine held that foreign investors were entitled only to national treatment, both in terms of their substantive legal rights and in terms of their ability to seek remedies at international tribunals. (253)

This resistance to the developed world's assertions about customary investment law manifested itself in several different types of codification efforts. First, developing states used their numerical superiority in the U.N. General Assembly to pass a series of resolutions aimed at undermining the international minimum standard. Two resolutions are the most significant in this regard. Resolution 3201, the NIEO, asserted a country's "right to nationalization or transfer of ownership to its nationals, this right being an expression of the full permanent sovereignty of the State." (254) The resolution made no reference to a responsibility to pay compensation in the event of nationalization. (255) This push to disrupt existing customary law received a further shot in the arm with the passage of Resolution 3281, the so-called Charter of Economic Rights and Duties of States. (256) The Charter adopted the Calvo doctrine, stating the right "[t]o nationalize, expropriate or transfer ownership of foreign property" was subject only to a duty to pay "appropriate compensation ..., taking into account [a State's] relevant laws and regulations and all circumstances that the State considers pertinent." (257)

To be sure, U.N. General Assembly resolutions are not legally binding. But these resolutions, representing as they do the views of a significant majority of states, can reasonably be thought of as soft law. That is, while not themselves legally binding, they have legal consequences because states use them to interpret, or correlate with, legal obligations. (258) In this case, of course, the legal obligations at issue are the customary rules of international investment law. And while there is some dispute as to whether the General Assembly resolutions did indeed disrupt the international minimum standard, (259) that debate itself is evidence that the soft codification of the Calvo doctrine increased uncertainty as to what customary law required.

A more important codification, and one that has largely been overlooked in the literature on the rise of BITs, was the decision by several Latin American countries to bind themselves to the Calvo doctrine through the Andean Pact. Specifically, Decision 24 of the Andean Commission, the so-called Andean Foreign Investment Code, forbade member states from "grant[ing] to foreign investors any treatment more favorable than that granted to national investors." (260) Moreover, the Code specifically required that member states not enter into any "instrument relating to investments ... that removes possible conflicts or controversies from the national jurisdiction." (261) Thus, Andean states codified both aspects of the Calvo doctrine--the substantive limitation of customary international law to national treatment and the refusal to allow international tribunals to take jurisdiction over disputes.

The Andean Commission's codification of the Calvo doctrine appears to have deterred the growth of BITs among Andean Pact countries. Indeed, members of the Andean Pact entered into only a single BIT from the time Decision 24 came into force until the partial repeal of its Calvo provisions in 1987. (262) Morever, Andean Pact states signed only one additional BIT before the total repeal of the Calvo provisions of the Andean Code in Decision 291 of 1991. (263) This is despite the fact that, according to U.N. Conference on Trade and Development, 385 BITs existed by the end of 1989. (264) The five countries in the Andean Community thus accounted for only one of those 385 during the period in which Decision 24 forbade acceptance and codification of the international minimum standard. By contrast, other Latin American states not bound by the Andean Code began entering into BITs considerably earlier. Panama, for example, had four BITs in force before Decision 291, (265) Uruguay had two, (266) and Paraguay had one. (267) Thus, while the explosion of BITs in the 1990s had many economic and political causes, it seems reasonable to infer that at least part of the hesitation members of the Andean Pact felt toward BITs stemmed from the codification of the Calvo doctrine in Decision 24.

b. International Investment Agreements and the Failure to Reach a Multilateral Investment Agreement

Perhaps not coincidentally, the codification of the Calvo doctrine, particularly in Latin America, and the pressure that such codification put on customary international law as understood by developed states led these states to embark on their own codification efforts to shore up the existence and content of the international minimum standard. As is well known, developed states' codification efforts took the form of BITs, regional agreements such as NAFTA, and sectoral agreements such as the Energy Charter Treaty. (268) Many commentators now believe that this system of IIAs has created a type of multilateral investment law. (269) Yet this codification effort displays the attributes of fragmented codification. While IIAs exhibit broad structural convergence, they tend to differ in the details, thereby enshrining variations in approaches to customary law. For example, in its Model BIT, the United States and its counterparties confirm that the fair and equitable treatment (FET) obligation is part of the international minimum standard. (270) Many other BITs, however, imply no connection between the FET standard and the international minimum standard, leading many commentators to argue that the FET obligation is purely a treaty obligation. (271)

Codifications of the concept of "indirect expropriation" have similarly varied. In 2004, the United States and Canada added annexes to their Model BITs to clarify the standard used in determining whether an indirect expropriation has occurred. (272) The annex to the United States' Model BIT asserts that "[e]xcept in rare circumstances, nondiscriminatory regulatory actions by a Party that are designed and applied to protect legitimate public welfare objectives, such as public health, safety, and the environment, do not constitute indirect expropriations." (273) This formulation attempts to strike a balance between customary international law concepts of police powers and indirect expropriation. This balance protects state police powers while still allowing for the possibility that nondiscriminatory regulatory action might amount to an indirect expropriation. India, Singapore, and China have included similar codifications in various IIAs. (274) Other states, however, have sought to omit the qualifier "except in rare circumstances" and codify a balance between these two customary concepts that is even more protective of state police powers. For example, the 2007 COMESA Common Investment Agreement among Southern and Eastern African countries provides,
   Consistent with the right of states to regulate and the customary
   international law principles on police powers, bona fide regulatory
   measures taken by a Member State that are designed and applied to
   protect or enhance legitimate public welfare objectives, such as
   public health, safety and the environment, shall not constitute an
   indirect expropriation under this Article. (275)

The 2009 ASEAN Comprehensive Investment Agreement is also absolutist in protecting state police powers. (276)

Reconciling the differences between the various standards codified in IIAs has become an obstacle to the development of a truly multilateral investment agreement. Having committed themselves to more favorable rules, both developed and developing states have been unwilling to make the kinds of concessions necessary to reach a multilateral agreement. Each has preferred its own codified understanding of custom to the version that would emerge from a multilateral renegotiation of investment law.

For example, since the failure of the Havana Charter of the International Trade Organization after World War II, efforts at a multilateral investment agreement have shifted between the GATT/WTO and the OECD. (277) The most serious effort to codify international investment law at the WTO began when investment was included as one of the "Singapore Issues" in 1996. (278) The WTO established the Working Group on Trade and Investment, but ultimately the topic was so contentious that it was removed from the WTO's agenda following the 2003 Cancun Ministerial Conference. (279) In large part, the WTO's efforts fell apart because certain developed and developing countries failed to support the WTO rules. (280) In particular, the United States felt that the results of the WTO negotiations would undermine rules it had successfully codified through its BIT program. (281) Developing countries such as India, China, Brazil, and Malaysia also opposed a WTO agreement on investment. At least one commentator has explicitly linked this resistance to a fear of a multilateral codification of developed states' interpretation of rules on expropriation and compensation." (283) Efforts to codify at the WTO ultimately failed because the negotiations required concessions that neither developed nor developing states were willing to make in light of their bilateral codification practices. In other words, preexisting codifications inhibited multilateral agreement.

The OECD's most recent effort to codify investment law, the 1998 Multilateral Agreement on Investment (MAI), failed for similar reasons. The 1998 MAI faced resistance from a number of quarters. Developing states, such as India, continued to oppose multilateral rules; their objections were based in part on substantive grounds and in part due to the exclusive forum in which the rules were being negotiated. On the other side, business interests in developed states were opposed to the MAI because protections in BITs struck a more advantageous balance between substance and adherence than the MAI. (285) Negotiators were also unable to agree on the relationship between the MAI and existing IIAs. (286) The need for clarification arose not only because of the existence of so many prior agreements covering the same subject matter, but also because both the MAI and existing IIAs either purported to or could reasonably be thought to codify custom. In the event of a conflict between treaties, rules on later-in-time treaties prevailing over earlier treaties could settle the dispute. But since the treaty rules in question were related to customary rules, it became even more important to identify the relationship of the MAI to the customary law as codified in existing IIAs. Indeed both the commentary to the draft MAI (287) and a Chairman's Note (288) identify just this issue. The commentary to the draft MAI points out that "[r]eference to international law is critical in [the articles on investment protection]," noting that the issue needed to be discussed throughout the MAI. (289) Despite this concern, however, the draft MAI failed to include a clause specifying precisely how it related to existing IIAs. (290)

Similar conflict issues have arisen as the European Union has moved toward integrating its investment policies. Specifically, during the accession or candidacy of eight Central and Eastern European states to the EU, the European Commission determined that a number of those countries' BITs were inconsistent with EU law. (291) These states' accession to the EU was held up while they renegotiated their BITs with the United States. (292) Despite this renegotiation, commentators remain concerned that the potential for conflict is considerably greater. (293) European states have BITs with many countries other than the United States, (294) and there are also intra-EU BITs that may conflict with European law and inhibit the ability of the EU to develop its competencies in areas related to investment. (295)

Investment law thus highlights the costs that codification can impose on attempts to create unified law. In many decentralized systems of law, rhetoric about the benefits of legal pluralism and competition among rulemakers appeals to the notion of a market for legal rules. But legal institutions are often path dependent and competitive in a way that can deprive decentralized legal systems of harmonization benefits at the appropriate time. Despite the existence of the ILC, no state or institution has a monopoly on codification. Codification can therefore increase the costs of competition over legal rules by reducing the future expected benefits from harmonization.

C. Conclusion

The virtue of codification as a strategy to capture legal rules is that the costs that codifying states incur in deviating from the rule when they deal with noncodifying states increases the pressure on noncodifying states to adhere to the codified customary rules. Unfortunately, these same costs can cut the other way. Costly commitments to particular interpretations of customary rules can make unifying the law more difficult if multiple codified interpretations of customary rules arise. And this is a likely eventuality. States with outlying preferences will not moderate the codified rules to attract states with very different preferences, creating the possibility of competing codifications. Thus, codification can, perversely, lead to customary law that is even weaker than uncodified custom. Finally, codifying custom can lead to suboptimal rules when powerful states are able to codify changes in the law that are Pareto inferior to the status quo.


The codification of customary international law has been one of the central projects of the international legal community since the beginning of the twentieth century. The drive for codification is part of a larger story about global governance increasing the welfare of states. The twentieth century, and in particular the post-World War II era, saw the rise of the multilateral treaty as the primary legal instrument in an international order that increasingly funneled international disputes into organizational channels. Codification seemed perfect for this enterprise, allowing states to translate large fields of amorphous customary law into written instruments that were both purportedly clearer and included compliance-inducing mechanisms to increase adherence to international law. Moreover, codification, employing the rules of state consent, appeared to address a central concern with customary international law--that states can be bound without their consent, and thereby made worse off.

But lawmaking is rarely without distributional consequences. Legal systems that erect high barriers to the creation of new legal obligations face pressure to devise processes that reduce the cost of making new rules. The negotiated elaboration of vague, universally applicable customary rules is one way in which states seek to reduce the transaction costs of making new law. Codification does this by limiting participation in the formulation of customary rules and, at the same time, creating incentives for excluded states to adhere to the negotiated terms. The consequences of this strategy of codification are profound. Codified customary rules, counterintuitively, may not evolve toward efficiency, thus casting further doubt on the thesis that decentralized systems of law offer a way out of the gridlock and messy politics that plague legislatures and legislative-like bodies. Moreover, codification can actually undercut the long-term benefits of competition among legal rules by making it more difficult to harmonize standards.

These consequences should lead us to question the desirability of codification as an across-the-board solution to the indeterminacy that can plague international legal obligations. In some instances, bare customary law may be superior in delivering on the promises of a universal and decentralized legal system. Perhaps more importantly, codification's distributional consequences highlight the importance of reconsidering the status of sacred cows. If the consent system does not protect against inefficient changes in and the fragmentation of customary law, then perhaps the consent requirement should be jettisoned in favor of a rule that more appropriately balances the competing objectives of international lawmakers.


(2) See STEPHAN W. SCHILL, THE MULTILATERALIZATION OF INTERNATIONAL INVEST MENT LAW 31-60 (2009) (discussing post-World War II efforts to negotiate a multilateral investment regime).

(3) General Agreement on Tariffs and Trade, Oct. 30, 1947, 61 Stat. A-11, 55 U.N.T.S. 184.

(4) General Agreement on Trade in Services, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1B, 1869 U.N.T.S. 183.

(5) See, e.g., Neer v. Mexico, 4 R.I.A.A. 60, 61-62 (U.S.-Mex. Gen. CI. Comm'n 1926) (holding that a foreign government's treatment of an alien "should amount to an outrage" in order to hold the government liable).

(6) See SCHILL, supra note 2, at 40-41 (chronicling the rise of, and subsequent changes to, bilateral and regional investment agreements); Andreas F. Lowenfeld, Investment Agreements and International Law, 42 COLUM.J. TRANSNAT'L L. 123, 125-28 (2003) (discussing developing countries' increased use of BITs in recent years to attract capital from multinational companies).

(7) For the jurisprudential origins of the codification movement, see "LEGISLATOR OF THE WORLD": WRITINGS ON CODIFICATION, LAW, AND EDUCATION (Philip Schofield & Jonathan Harris eds., 1998), which collects the letters and writings of Jeremy Bentham, the originator of codification. See also G.W.E HEGEL, ELEMENTS OF THE PHILOSOPHY OF RIGHT [section] 211, at 241 (Allen W. Wood ed., H.B. Nisbet trans., 1991) (stressing the need to codify custom as law in order to prevent the confusion and indeterminacy of common law systems). Bentham is credited with coining the term "codification" and arguing that it would promote clarity in the law. See Judith Resnick, BringBack Bentham: "Open Courts," "Terror Trials," and Public Sphere(s), 5 LAW & ETHICS HUM. RTS. 2, 19-20 (2011); see also infra note 27.


(9) U.N. Charter art. 13, para. 1.

(10) The OECD's failed attempt to pass the Multilateral Investment Agreement is a prominent example. Org. for Econ. Cooperation & Dev. [OECD], The Multilateral Agreement on Investment: Draft Consolidated Text, DAFFE/MAI(98)7/REV1 (Apr. 22, 1998) [hereinafter Draft MA1], available at ng987rle.pdf. The Draft MAI will be discussed below as an example of how codification may present barriers to interntational agreement. See infra notes 284-91 and accompanying text.

(11) See Rome Statute of the International Criminal Court art. 10, July 17, 1998, 2187 U.N.T.S. 90 [hereinafter Rome Statute] (disclaiming any interpretation of the treaty that would limit existing or developing rules of international law).

(12) See, e.g., United Nations Convention on the Law of the Sea pmbl., Dec. 10, 1982, 1833 U.N.T.S. 397 [hereinafter UNCLOS] (citing codification of the law of the sea as a main purpose of the convention); United Nations Convention on the Law of Non-Navigational Uses of International Watercourses pmbl., opened for signature May 21, 1997, 36 I.L.M. 700 (citing codification of this area of international law as a main purpose of the convention).

(13) Geneva Convention Relative to the Treatment of Prisoners of War, Aug. 12, 1949, 6 U.S.T. 3316, 75 U.N.T.S. 135.

(14) Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, Dec. 10, 1984, S. TREATYDOC. NO. 100-20 (1988), 1465 U.N.T.S. 85 [hereinafter CAT].

(15) See, e.g., R.R. Baxter, Multilateral Treaties as Evidence of Customary International Law, 41 BRIT. Y.B. INT'L L. 275, 293 (1965-1966) (arguing that treaties codifying law may influence, shape, and alter the law in signatory countries); Richard Falk, Reparations, International Law, and Global Justice: A New Frontier (noting that the purpose of international law is to "codif[y] behavioral trends in state practice and shift[] political attitudes on the part of governments with the intention of stabilizing and clarifying expectations about the future"), in THE HANDBOOK OF REPARATIONS 478, 480 (Pablo de Greiff ed., 2006); Bing Bing Jia, The Relations Between Treaties and Custom, 9 CHINESE J. INT'L L. 81, 108 (2010) ("[A] piece of authoritative work of codification ... [is] intended to clarify and settle applicable rules of international law...."); H. Lauterpacht, Codification and Development of International Law, 49 AM.J. INT'L L. 16, 19 (1955) ("It is probably a fact that the absence of agreed rules partaking of a reasonable degree of certainty is a serious challenge to the legal nature of what goes by the name of international law."); Iain Scobbie, The Invocation of Responsibility for the Breach of 'Obligations Under Peremptory Norms of General International Law,' 13 EUR. J. INT'L L. 1201, 1202 (2002) (equating the codification of custom with its clarification).

(16) See infra Section II.A.

(17) See infra Section II.B.

(18) See infra Part III.

(19) In so arguing, this Article contributes to the literature on strategic international lawmaking by states, and specifically on the relationship between different forms of legal commitments, such as treaties, soft law, and customary international law. See, e.g., Kenneth W. Abbott & Duncan Snidal, Hard and Soft Law in International Governance, 54 INT'L ORG. 421, 434-50 (2000) (detailing the advantages of soft law, such as protected sovereignty, increased certainty, and more frequent compromise); Andrew T. Guzman & Timothy L. Meyer, International Soft Law, 2 J. LEGAL ANALYSIS 171, 176-79 (2010) (putting forth four explanations for why states use soft law); Barbara Koremenos, Loosening the Ties that Bind: A Learning Model of Agreement Flexibility, 55 INT'L ORG. 289, 290-304 (2001) (developing a theoretical model for international agreements with respect to how states choose their international legal commitments); Kal Raustiala, Form and Substance in International Agreements, 99 AM. J. INT'L L. 581, 583-85 (2005) (discussing legality, substance, and structure as three dimensions of the institutional design of international agreements); Edward T. Swaine, Rational Custom, 52 DUKE L.J. 559, 564 (2002) (attempting to reconcile custom and rational choice theory).

(20) RESTATEMENT (THIRD) OF FOREIGN RELATIONS LAW OF THE UNITED STATES [section] 102(2) (1987); see also Statute of the International Court of Justice art. 38(1)(b),June 26, 1945, 59 Stat. 1055 (authorizing the International Court of Justice to apply "international custom, as evidence of a general practice accepted as law" in resolving international disputes).

(21) The literature critiquing the traditional formulation of custom is voluminous. For the most well-known critique, see ANTHONY A. D'AMATO, THE CONCEPT OF CUSTOM IN INTERNATIONAL LAW (1971). D'Amato famously argued that the traditional test for custom is circular: "But if custom creates law, how can a component of custom require that the creative acts be in accordance with some prior right or obligation in international law?" Id. at 53.

(22) See id. at 58 (decrying the lack of standard criteria to evaluate the amount of time required to define international law as customary).

(23) See Andrew T. Guzman, Saving Customary International Law, 27 MICH. J. INT'L L. 115, 124-28 (2005) (reviewing the controversies over what counts as state practice for purposes of determining customary international law); see also D'AMATO, supra note 21, at 88 (arguing that only physical acts count as state practice, whereas claims made by states do not).

(24) See D'AMATO, supra note 21, at 53, 66 (summarizing attempts to rationalize this apparent circularity problem).

(25) Although I focus here primarily on binding legal instruments, this definition allows for the possibility, frequently realized in modern international legal practice, that legal consequences can flow from instruments that are not themselves legally binding. Put differently, I consider "codifications" to be any nonbinding instruments that at least some states believe embody existing customary rules. These instruments, sometimes termed "soft law," include nonbinding declarations issued by states or certain draft articles issued by the ILC that are asserted to reflect customary law. See, e.g., Draft Articles on Responsibility of States for Internationally Wrongful Acts, G.A. Res. 56/83, Annex, U.N. Doc. A/RES/56/83 (Dec. 12, 2001) (codifying to some extent the customary law of state responsibility). Such instruments have legal consequences to the extent that they shape or reflect states' understanding of what custom requires. I would also include in this definition interpretative notes--such as those issued by authoritative bodies like the NAFTA Free Trade Commission--that are designed to give more precise content to legal obligations. Where these obligations are codified customary obligations, interpretative notes can serve the same function as a codifying treaty. I exclude from my definition nonbinding documents that do not reflect any state's understanding of custom, such as model laws or draft conventions that do not receive any state endorsement.

(26) See, e.g., Rome Statute, supra note 11, art. 126 (stating that the statute shall come into force upon ratification by sixty countries).

(27) Report of Sub-Committee Upon the History and Status of Codification, 4 AM. SOC'Y INT'L L. PROC. 208, 214-18 (1910); see also Arthur Watts, Codification and Progressive Development of International Law (providing a comprehensive overview of the history of international codification), available at subscriber_article?script=yes&id=/ epil/entries/law-9780199231690-e1380&recno= 1&author=Watts%20QC%20%20Arthur, in MAX PLANCK ENCYCLOPEDIA OF PUBLIC INTERNATIONAL LAW (Rudiger Wolfrum ed., 2012) (forthcoming).

(28) For an account of the early history of international codification efforts, see BOYLE & CHINKIN, supra note 8, at 163-64.

(29) Id.

(30) Id. at 165-66.

(31) See, e.g., Resolution of Sept. 25, 1931, 12 League of Nations O.J., Spec. Supp. 92, at 9 (1931); Resolution of Sept. 27, 1927, 8 League of Nations O.J., Spec. Supp. 53, at 9-10 (1927); Resolution of Sept. 22, 1924, 5 League of Nations O.J., Spec. Supp. 21, at 10 (1924).

(32) Manley O. Hudson, The Permanent Court of International Justice, 35 HARV. L. REV. 245, 256 (1922).

(33) See Guillaume Sacriste & Antoine Vauchez, The Force of International Law: Lawyers' Diplomacy on the International Scene in the 1920s, 32 LAW & SOC. INQUIRY 83, 101 (2007) (discussing the failure of this conference in the broader context of the interwar international law community).

(34) Examples include the Inter-American Juridical Committee, the Asian-African Legal Consultative Committee, and the European Committee on Legal Cooperation.

See, e.g., B. Graefrath, The International Law Commission Tomorrow, 85 AM.J. INT'L L. 595, 608 (1991) (discussing other codifying institutions).

(35) U.N. Charter art. 13, para. 1. This mandate is carried out in a variety of other ways, including through ad hoc committee reports to the Sixth Committee (Legal) of the General Assembly and by specialized organs of the United Nations, such as the United Nations Commission on International Trade Law (UNCITRAL) or the United Nations Environmental Programme (UNEP). For a discussion of these committees and organs, see BOYLE & CHINKIN, supra note 8, at 167-68.

(36) Statute of the International Law Commission, G.A. Res. 174 (II), art. 1, U.N. Doc. A/RES/174(II) (Nov. 21, 1947) [hereinafter ILC Statute].

(37) Given this fact, it is perhaps not surprising that the composition of the ILC has been a subject of contestation and revision. At its inception, the ILC had only fifteen members, eight of whom were from Europe or the United States, with an additional four from Latin America, three from Asia, and none from Africa. JEFFREY S. MORTON, THE INTERNATIONAL LAW COMMISSION OF THE UNITED NATIONS 8-9 (2000). To accommodate the growth in the number of U.N. member states, the ILC's membership has been expanded three times: in 1957 to twenty-one, in 1962 to twenty-five, and in 1981 to thirty-four. Id. Each expansion diluted the share of American and European seats, and there is evidence that this expansion increased partisan dynamics at the ILC. Using a dataset compiled from public records of ILC proceedings during the development of the Draft Code of Crimes Against the Peace and Security of Mankind, Professor Morton argues that the ILC members show high levels of geographic and ideological cohesion. Id. at 83-92. This cohesion, he asserts, is inconsistent with an apolitical view of the ILC membership advocated by many scholars and practitioners. Id.

(38) See, e.g., Vienna Convention on the Law of Treaties, May 23, 1969, 1155 U.N.T.S. 331 [hereinafter VCLT]; Vienna Convention on Diplomatic Relations art. 47.2 (b), Apr. 18, 1961, 23 U.S.T. 3227, 500 U.N.T.S. 95 [hereinafter VCDR]; Convention on the High Seas, Apr. 29, 1958, 13 U.S.T. 2312, 450 U.N.T.S. 11.

(39) Nonstate actors also play an important role in codification in at least two ways. First, they can influence states' negotiation of a codifying treaty, much in the way NGOs participated in the negotiation of the Rome Statute. See William R. Pace & Mark Thieroff, Participation of Non-Governmental Organizations (describing how many "governments, the Secretary General, other United Nations officials and media experts

have commented on the decisive role of NGOs at the Rome Conference"), in THE INTERNATIONAL CRIMINAL COURT: THE MAKING OF THE ROME STATUTE: ISSUES, NEGOTIATIONS, RESULTS 391, 392-93 (Roy S. Lee ed., 1999). Second, nonstate actors can "codify" the law themselves, although they generally lack the ability to make their "codifications" binding on states directly. See Jordan J. Paust, Nonstate Actor Participation in International Law and the Pretense of Exclusion, 51 VA.J. INT'L L. 977, 997 (2011) (discussing Professor Francis Lieber's codification of the customary laws of war in the Lieber Code, one of the most well-known private codification efforts). Nevertheless, states can adopt these nonbinding codifications as evidence of their views of custom. I defer detailed examination of the role nonstate actors play in codification to future work, focusing here instead on state-centric codification efforts.

(40) See, e.g., U.S. DEP'T OF STATE, 2004 U.S. MODEL BILATERAL INVESTMENT TREATY, annexes A, B, art. 5(1) (2004) [hereinafter 2004 U.S. MODEL BIT], available at http:// ("Each Party shall accord to covered investments treatment in accordance with customary international law, including fair and equitable treatment and full protection and security.").

(41) See Int'l Criminal Court [ICC], Assembly of States Parties of the International Criminal Court, The Crime of Aggression, ICC Doc. RC/Res. 6, annex III (June 11, 2010) (noting that the definition of the crime of aggression, consistent with article 10 of the Rome Statute, should not limit or prejudice existing or developing rules of international law).

(42) The International Court of Justice (ICJ) set forth ways in which treaties can be related to customary rules in the North Sea Continental Shelf Cases. There, the ICJ decided that treaty rules (1) may declare customary rules, (2) crystallize emergent customary rules, or (3) over time become customary rules. North Sea Continental Shelf (F.R.G./Den.; F.R.G./Neth.), 1969 I.CJ. 3, 39, 4142 (Feb. 20). Treaty rules in category (3) are not properly thought of as codifications, because they are not initially claimed to be custom.

(43) ILC Statute, supra note 36, art. 15 (offering separate definitions for the progressive development of international law and the codification of international law).

(44) See Lauterpacht, supra note 15, at 17 ("[T]here is very little to codify if by that term is meant no more than giving, [in the language of ILC statute, article 15], precision and systematic order to rules of international law....").

(45) Id.

(46) See, e.g., Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 DUKE L.J. 557, 591 (1992) (noting that certain "standard[s] make[] ... ex post, case-by-case determinations").

(47) See, e.g., Guzman & Meyer, supra note 19, at 201-06 (describing the ICL as making "nonbinding rulings or standards"); see also Andrew T. Guzman & Jennifer Landsidle, The Myth of International Delegation, 96 CALIF. L. REV. 1693, 1694 (2008) ("When one turns to examine instances of international delegation, what becomes immediately apparent, at least to the authors, is how little of it there actually is."). In the domestic context, the adjudication of questions likely to reappear before the courts is a strategic enterprise because of the path dependence of adjudicated law. Repeat litigants try to ensure that early decisions set forth favorable rules that they can employ in future litigation. See, e.g., Catherine Albiston, The Rule of Law and the Litigation Process: The Paradox of Losing by Winning, 33 LAW & SOC. REV. 869, 877-86 (1999) (describing strategic decision-making in the U.S. litigation process); Stephen J. Choi & G. Mitu Gulati, Contract as Statute, 104 MICH. L. REV. 1129, 1145-58 (2006) (discussing the problems present in litigating interpretations of boilerplate contract terms).

(48) See Lauterpacht, supra note 15, at 22 ("[T]he task of codifying international law ... must be primarily one of bringing about an agreed body of rules...."); see also Joel P. Trachtman, Regulatory Jurisdiction and the WTO, 10 J. INT'L ECON. L. 631, 633 (2007) (arguing that some rules are better worked out through adjudication, while others are better worked out through negotiation). The proliferation of international tribunals has, to some extent, changed this practice. Some courts, such as the European Court of Justice or the European Court of Human Rights, play a major role in interpreting supranational, if not international, obligations. Nevertheless, compared with domestic courts, the level of delegation to international tribunals, where states are constrained by the tribunal's decisions going forward, remains quite low.

(49) See India: Airport Pat-Down Draws Protest, N.Y. TIMES, Dec. 10, 2010, at A8 (reporting that the Indian foreign minister considered the pat-down of an ambassador to be "unacceptable").

(50) See, e.g., George Norman & Joel P. Trachtman, The Customary International Law Game, 99 AM.J. INT'L L. 541, 548 n.34 (2005) (considering the suggestion that customary international law may be less detailed than treaty norms).

(51) See Jonathan Remy Nash, Essay, Standing and the Precautionary Principle, 108 COLUM. L. REV. 494, 500 (2008) ("Some argue that the precautionary principle is hampered by a lack of clarity, or, at least, agreement as to the principle's meaning." (footnote omitted)).

(52) For example, Lauterpacht has argued,
   [T]he absence of agreed rules partaking of a reasonable degree of
   certainty is a serious challenge to the legal nature of what goes
   by the name of international law. That circumstance alone supplies
   cogent proof of the justification, nay, of the urgency of the task
   of codification of international law.

Lauterpacht, supra note 15, at 19; see a/s0John O. McGinnis, The Appropriate Hierarchy of Global Multilateralism and Customary International Law: The Example of the WTO, 44 VA. J. INT'L L. 229, 239 (2003) ("The first advantage of multilateral agreements over customary international law is that they provide a more precise definition of the agreed upon rule for the simple reason that the provisions to which states have agreed are written down in the text of the agreement.").

(53) See, e.g., Kaplow, supra note 46, at 608-11 (discussing how codification enhances predictability and agreement between two branches of government).


(55) Id,

(56) See Zhu Lijiang, Chinese Practice in Public International Law: 2008, 8 CHINESE J. INT'L L. 493, 537 (2009) (stating that Chinese law provides that "[d]iplomatic pouches sent or received by a diplomatic mission shall ... comply with the relevant provisions of the Chinese government on the weight and dimensions of diplomatic pouches").

(57) Moreover, as Lauterpacht observed, conflicts that arise from uncertainty about the content of customary international law are potentially even costlier than similar disputes in the domestic system because states engage in reprisals rather than turning to courts to resolve disputes. See Lauterpacht, supra note 15, at 20 ("Within the state obscurity and uncertainty of the law are a drawback, but it is a drawback which is provisional inasmuch as the uncertainty can be removed with regard to a particular controversy by the decision of a court endowed with compulsory jurisdiction. This is not the position among states.").

(58) For an analysis of the advantages and disadvantages of rules and standards, see Kaplow, supra note 46, at 585.

(59) See, e.g., id. at 579 (advocating an economic analysis of costs and benefits at each stage of rule creation and enforcement).

(60) See id. at 608 & n.138 (explaining that rules provide clearer notice to actors, allowing for increased compliance).

(61) For an excellent analysis of the dispute between the United States and the European Union over genetically modified organisms, see generally MARK A. POLLACK & GREGORY C. SHAFFER, WHEN COOPERATION FALLS: THE INTERNATIONAL LAW AND POLITICS OF GENETICALLY MODIFIED FOODS (2009).

(62) Cf. Andrew T. Guzman, The Design of International Agreements, 16 EUR. J. INT'L L. 579, 582 (2005) (arguing that international sanctions create a loss to one party without an offsetting gain, yielding a net loss).

(63) Technically, a change in the law is Kaldor-Hicks superior to the status quo if, assuming zero transaction costs, it is possible to imagine transfers such that the change in the law is Pareto superior to the status quo. The transfers need not actually occur, however, and thus some states can be left worse off by a Kaldor-Hicks improvement in the law. Andrew T. Guzman, The Consent Problem in International Law 6 (June 14, 2011) (unpublished manuscript), available at

(64) See id. at 1 n.2 (citing a number of authorities for the proposition that international law is predicated on consent); see also Laurence R. Helfer, Nonconsensual International Lawmaking, 2008 U. ILL. L. REV. 71, 72-73 (discussing the advantages and disadvantages of requiring state consent to international agreements).

(65) See Guzman, supra note 63, at 6 ("In less technical language, requiring consent frustrates many potential arrangements that would improve the lot of the states as a whole.").

(66) Guzman & Meyer, supranote 19, at 198.

(67) League of Arab States, Arab Charter on Human Rights 2004, May 22, 2004 [hereinafter Arab Charter], translated in 12 INT'L HUM. RTS. REP. 893, 895 (2005). The negotiation of the crime of aggression in the Rome Statute also illustrates how codification may not result in clarification. Codifying states may deliberately obscure a rule as a compromise between two different interpretations. See infra Section III.D.

(68) Mervat Rishmawi, The Arab Charter on Human Rights and the League of Arab States: An Update, 10 HUM. RTS. L. REV. 169, 169 (2010).

(69) Press Release, U.N. High Comm'r for Human Rights, Statement by UN High Commissioner for Human Rights on the Entry into Force of the Arab Charter on Human Rights (Jan. 30, 2008), available at nsf/0/6C211162E43235FAC12573E00056E19D?opendocument.

(70) Arab Charter, supra note 67, art. 7 (emphasis added).

(71) Id. pmbl.


(73) G.A. Res. 3379 (XXX), [paragraph] 6, U.N. Doc. A/RES/3379 (Nov. 10, 1975).

(74) G.A. Res. 46/86, U.N. Doc. A/RES/46/86 (Dec. 16, 1991).

(75) For example, the U.S. delegate to the conference negotiating the Vienna Convention on Consular Relations proposed a jurisdictional protocol, arguing that "the codification of international law and the formulation of measures to ensure compliance with its provisions should go hand in hand." 1 United Nations Conference on Consular Relations, Mar. 4-Apr. 22, 1963, Vienna Conference on Consular Relations, at 249, U.N. Doc. A/CONF.25/16 (Mar. 26, 1963); see also U.N. SCOR, 49th Sess., 3453d mtg. at 7, U.N. Doc. S/PV.3453 (Nov. 8, 1994) (transcribing the Czech delegate's remark that the creation of the International Criminal Tribunal for Rwanda as a "breakthrough" in the codification of international law).

(76) Optional Protocol to the Vienna Convention on Consular Relations Concerning the Compulsory Settlement of Disputes, Apr. 24, 1963, 21 U.S.T. 77, 596 U.N.T.S. 261 [hereinafter VCCR Optional Protocol on Disputes].

(77) UNCLOS, supra note 12, art. 188; Statute for the International Tribunal for the Law of the Sea, Dec. 10, 1982, 1833 U.N.T.S. 561.

(78) See Leena Grover, A Call to Arms: Fundamental Dilemmas Confronting the Interpretation of Crimes in the Rome Statute of the International Criminal Court, 21 EUR.J. INT'L L. 543, 569 (2010) (describing the Rome Statute's effect as "the crystallization of certain [international] norms").

(79) See, e.g., Jordan J. Paust, Antiterrorism Military Commissions: Courting Illegality, 23 MICH.J. INT'L L. 1, 12 (2001) (arguing that the International Covenant on Civil and Political Rights sets forth a minimum standard of due process and human rights guaranteed to all people); Kweku Vanderpuye, Traditions in Conflict: The Internationalization of Confrontation, 43 CORNELL INT'L L.J. 513, 539 (2010) (noting that article 14 of the Covenant sets forth universally recognized human rights obligations).

(80) See International Covenant on Civil and Political Rights art. 40, opened for signature Dec. 16, 1966, 999 U.N.T.S. 171 (entered into force Mar. 23, 1976) (providing details of required reports that all states must submit upon adopted measures).

(81) CAT, supra note 14, art. 19 (outlining who submits these mandatory reports and what they must contain). 82 Id. arts. 5-9.

(83) 304 U.S. 64 (1938).

(84) See Curtis A. Bradley & Jack L. Goldsmith, Customary International Law as Federal Common Law: A Critique of the Modern Position, 110 HARV. L. REV. 815, 870 (1997) (asserting that customary international law should be treated as state law, absent federal common law); Curtis A. Bradley et al., Sosa, Customary International Law, and the Continuing Relevance of Erie, 120 HARV. L. REV. 869, 891-92 (2007) (providing a more updated analysis after Sosa v. Alvarez-Machain, 542 U.S. 692 (2004), and other case law dealing with the treatment of international law as nonfederal common law). But see Harold Hongju Koh, Commentary, Is International Law Really State Law?, 111 HARV. L. REV. 1824, 1827 (1998) (arguing that customary international law is federal law).

(85) See U.S. CONST. art. VI, cl. 2 ("[A]ll Treaties made, or which shall be made, under the Authority of the United States, shall be the supreme Law of the Land ....").

(86) Each convention contains an optional protocol on disputes. See, e.g., VCCR Optional Protocol on Disputes, supra note 76.

(87) See Karl M. Meessen, Antitrust Jurisdiction Under Customary International Law, 78 AM.J. INT'L L. 783, 790 (1984) (using conflicting antitrust law to demonstrate Germany's reliance upon customary international law).

(88) ICC, supra note 41, at 18.

(89) See id. at 19 (addressing the way in which jurisdiction over the crime of aggression may and may not be exercised); see also Diane Marie Amann, The Crime of Aggression, the United States, and the Value of Peace 8-11 (Oct. 15, 2010) (unpublished manuscript) (on file with author) (analyzing potential jurisdictional paths that could arise if states adopted certain amendments to international criminal law).

(90) See Guzman, supra note 62, at 582 (outlining how a state may do itself more harm than good through adopting compliance-enhancing mechanisms).

(91) See supra text accompanying note 20.

(92) See D'AMATO, supra note 21, at 92-98 (discussing the process of custom formation and modification through the use of several "hypothetical conflict-of-custom situations").

(93) Id. at 92-93.

(94) See id. at 97-98 ("The number of disconfirmatory acts required to replace [an] original rule is a function partly of the number of acts that established the original rule ..., the remoteness in time of the establishing acts, the legal authoritativeness of the participating states, and other possible factors....").

(95) Id. at 96.

(96) See id. at 96-97 (citing Great Britain as an example of a nation that "speaks with a greater authority in international law than its military position might warrant").

(97) See infra subsection III.B.1.

(98) See infra Section III.C.

(99) Cf. D'AMATO, supra note 21, at 90-98 (describing a process of implicit bargaining the results in the formation of a customary rule).

(100) See Marc Galanter, Why the "Haves" Come Out Ahead: Speculations on the Limits of Legal Change, 9 LAW & SOC'Y REV. 95, 97-104 (1974) (listing nine distinct advantages that "repeat players" have over "one-shotters" in the formation of rules through both rulemaking and litigation proceedings).

(101) See, e.g., LLOYD GRUBER, RULING THE WORLD 5-10 (2000) (putting forth a theory of the interplay between power politics and international institutionalization where "[t]he losers acquiesce because they know that the winners are in a position to proceed without them").

(102) See, e.g., Guzman, supra note 62, at 591-94 (discussing various motivations for inclusion or exclusion of different treaty design elements).

(103) See, e.g., Guzman & Meyer, supra note 19, at 197-201 (explaining how states choose soft, nonbinding law over hard law, including a desire for soft law when states can act to make "unilateral amendments" and re-coordinate expectations about what constitutes compliant behavior); Timothy Meyer, Soft Law as Delegation, 32 FORDHAM INT'L L.J. 888, 917-21 (2009) ("In the decision between hard and soft law ... power can be important in determining the form of a legal agreement."); Gregory C. Shaffer & Mark A. Pollack, Hard vs. Soft Law: Alternatives, Complements, and Antagonists in International Governance, 94 MINN. L. REV. 706, 728-30 (2010) ("[T]he interaction of hard and soft law will be shaped primarily by the preferences of powerful states such as the United States and the EU....").

(104) See Eyal Benvenisti & George W. Downs, The Empire's New Clothes: Political Economy and the Fragmentation of International Law, 60 STAN. L. REV. 595, 610-12 (2007) (discussing the common practice of "serial bilateralism," by which powerful states create "narrow negotiation venues as a means of limiting the ability of weaker states to form countervailing coalitions"); Kal Rausfiala, The Architecture of International Cooperation: Transgovernmental Networks and the Future of International Law, 43 VA. J. INT'L L. 1, 86-87 (2002) ("Rather than agree on a global standard, competing standard-setting states may simply opt for a 'miniliateral' solution....").

(105) See, e.g., Laurence R. Helfer, Regime Shifting: The TRIPs Agreement and the New Dynamics of International Intellectual Property Lawmaking, 29 YALEJ. INT'L L. 1, 6, 13-18, 53-63 (2004) (discussing developing states' efforts to move international intellectual property lawmaking away from the WTO and into more hospitable fora).

(106) Richard H. Steinberg, In the Shadow of Law or Power? Consensus-Based Bargaining and Outcomes in the GATT/WTO, 56 INT'L ORG. 339, 359-360 (2002).

(107) Jason Webb Yackee, Pacta Sunt Servanda and State Promises to Foreign Investors Before Bilateral Investment Treaties: Myth and Reality, 32 FORDHAM INT'L L.J. 1550, 1560 (2009).

(108) Nicholas DiMascio & Joost Pauwelyn, Nondiscrimination in Trade and Investment Treaties: Worlds Apart or Two Sides of the Same Coin?, 102 AM.J. INT'L L. 48, 52 (2008) (internal quotation marks omitted).

(109) See, e.g., SCHILL, supra note 2, at 83 (discussing modern treaty practice with respect to the Hull Rule); DiMascio & Panwelyn, supra note 108, at 52 (noting that developed countries tried to ensure the survival of the customary "prompt, adequate, and effective compensation" rule by including it in BITs); Andrew T. Guzman, Why LDCs Sign Treaties that Hurt Them: Explaining the Popularity of Bilateral Investment Treaties, 38 VA. J. INT'L L. 639, 645-46 (1998) (observing that there was international consensus surrounding

rounding the Hull Rule for the first half of the twentieth century); Yackee, supra note 107, at 1560 (explaining that since the 1930s, the United States "has insisted that the customary rule is and always has been" the Hull Rule).

(110) See Yackee, supra note 107, at 1550-64 (describing the resolutions passed as seeking to establish the concept of "permanent sovereignty" and, ultimately, a "New International Economic Order").

(111) DiMascio & Pauwelyn, supra note 108, at 52.

(112) See, e.g., Jose E. Alvarez, A BIT on Custom, 42 N.Y.U.J. INT'L L. & POL. 17, 20 (2009) ("[T]hose who see investment treaties as ... unconnected to ... customary international law ... are wrong.").

(113) See infra Part IV.

(114) Cf. Zachary Elldns et al., Competing for Capital: The Diffusion of Bilateral Investment Treaties, 1960-2000, 2008 U. ILL. L. REV. 265, 280 (arguing that as more and more capital-importing states adopted bilateral investments, competitive pressures drove late adopters to sign BITs as well).

(115) The Calvo doctrine asserts that foreigners are not entitled to special privileges, such as the right to take investment disputes to international tribunals. For more information on the Calvo doctrine, see Wenhua Shan, Is Calvo Dead?, 55 AM.J. COME L. 123, 124-30 (2007).

(116) Jeswald W. Salacuse, The Emerging Global Regime for Investment, 51 HARV. INT'L L.J. 427, 469-70 (2010).

(117) See Christopher M. Ryan, Meeting Expectations: Assessing the Long-Term Legitimacy and Stability of International Investment Law, 29 U. PA.J. INT'L L. 725, 747-48 (2008) (describing

three bills proposed by the Argentine legislature in 2004 and 2005 aimed at "limit[ing] Argentina's participation in the international investment law system").

(118) See, e.g., GRUBER, supra note 101, at 37 (arguing that for such a coercive negotiation strategy "to be effective, 'compellent' threats must be credible"); Timothy Meyer, Power, Exit Costs, and Renegotiation in International Law, 51 HARV. INT'L L.J. 379, 400 (2010) (arguing that both the quality of an individual state's outside options, other than an existing agreement, and its ability to make "a credible threat" to the existing scheme shape the course of such "renegotiation").

(119) This statement assumes that the codified rule and the alternative interpretation conflict. However, some codifying treaties provide that states may reach bilateral agreements containing protections in excess of the codified customary standards. For example, the Vienna Convention on Diplomatic Relations provides that states accord other states treatment that differs from the Convention standards where required "by custom or agreement." VCDR, supra note 38, art. 47.2(b); see also VCCR Optional Protocol on Disputes, supra note 76, art. 73 ("Nothing in the present Convention shall preclude States from concluding international agreements confirming or supplementing or extending or amplifying the provisions thereof."). The United States has availed itself of these provisions to conclude agreements granting greater immunities to its personnel stationed in the Soviet Union/Russia and China. DENZA, supra note 54, at 407.

(120) Cf. D'AMATO, supra note 21, at 175-77 (arguing that "[t]he force of even a single precedent is magnified" in the modern world and that international law that does not remain constant through time may lose its significance).

(121) Similar arguments apply to nonbinding codifications such as General Assembly Resolutions. See infra subsection IV.B.2. The principle difference between nonbinding and binding codifications is that the sanction for deviating from nonbinding codifications will not be as strong because the level of commitment demonstrated to the codified interpretation will not be as strong. See generally Guzman & Meyer, supra note 19 (arguing that soft law obligations create less of an expectation of compliance and therefore generate lower sanctions in the event of a violation). In principle, nonbinding codifications created by nonstate actors could also work in a similar fashion, although the inability to bind states to legal obligations may change the dynamic with nonstate actors somewhat. I defer extended consideration of the role of nonstate actors to subsequent work.

(122) UNCLOS, supra note 12, art. 311.6. The customary status of the principle of common heritage is controversial. Compare Jonathan I. Charney, The Antarctic System and Customary International Law (arguing that the principle of common heritage is not a "rule of general international law applicable to all areas outside of national jurisdiction" because there is no wide consensus as to its general applicability), in INTERNATIONAL LAW FORANTARCTICA 51, 75 (Francesco Francioni & Tullio Scovazzi eds., 2d ed. 1996), with Rudiger Wolfrum, Common Heritage of Mankind (arguing that "[t]he common heritage principle is part of ... customary international law ... providing general but not specific legal obligations with respect to the utilization of areas beyond national ... jurisdiction"), in 1 MAX PLANCK ENCYCLOPEDIA OF PUBLIC INTERNATIONAL LAW 692, 694 (Rudolf Bernhardt ed., 1992).

Similar to UNCLOS article 311.6, article 47 of the VCDR includes an obligation not to discriminate in the application of the Convention, subject to exceptions. VCDR, supra note 38, art. 47. By its terms, this obligation applies only to parties to the Convention. In practice, however, the implemented legislation often makes no distinction in the extension of privileges between parties to the VCDR and nonparties, instead distinguishing on the basis of reciprocal treatment. See DENZA, supra note 54, at 405 (noting that legislation in the United Kingdom, Canada, the Netherlands, and Belgium allows these countries to withdraw certain diplomatic privileges and immunities from diplomatic missions of countries that had not granted reciprocal privileges and immunities to British, Canadian, Dutch, or Belgian missions).

(123) See Sol Picciotto, Linkages in International Investment Regulation: The Antinomies of the Draft Multilateral Agreement on Investment, 19 U. PA. J. INT'L ECON. L. 731, 753-54 (1998) (asserting that OECD countries could have used multilateral agreements "on an investment protection standard ... as the basis for common action against breaches of" this standard).

(124) Understanding with Respect to Disciplines for the Strengthening of Investment Protection, U.S.-E.U., May 18, 1998, available at summit/Summit9805/invest.htm.

(125) See infra Part IV.

(126) See WILHELM G. GREWE, THE EPOCHS OF INTERNATIONAL LAW 555-56 (Michael Byers ed. & trans., 2000) (discussing the role humanitarian concerns played in the rise of British disapproval of the slave trade, but noting that these concerns were conveniently tied to economic changes).

(127) Id. at 556-57.

(128) See generally Michael Byers, Note, Policing the High Seas: The Proliferation Security Initiative, 98 AM.J. INT'L L. 526, 534-36 (2004).

(129) See, e.g., United States v. Smith, 18 U.S. (5 Wheat.) 153, 161-62 (1820) (holding that piracy is against the law of nations based on the universal view of the relevant authorities).

(130) See GREWE, supra note 126, at 562.

(131) Id.

(132) Treaty for the Suppression of the African Slave Trade art. 1, Dec. 20, 1841, 92 Consol. T.S. 437, 441.

(133) Treaty for the Accession of Belgium to the Treaty of 20 December 1841 for the Suppression of the Slave Trade, Feb. 28, 1848, 102 Consol. T.S. 95, 96-97.

(134) See GREWE, supra note 126, at 565 (noting that the United States "successfully in cited a campaign by opposition parties in France against that country's accession to the Quintuple Treaty," which included the publication of an anonymous pamphlet in Paris that "strongly condemned the humanitarian Polish or British claims for dominion of the sea").

(135) Id. at 566.

(136) Lyons-Seward Treaty art. 1, Apr. 7, 1862, 12 Stat. 1225.

(137) Convention on the High Seas, supra note 38, art. 22.1 (b).

(138) Id.

(139) See supra text accompanying note 136.

(140) A related argument is that treaties are superior to customary international law because treaties are made through democratically accountable procedures. See John O. McGinnis, The Comparative Disadvantage of Customary International Law, 30 HARV. J.L. & PUB. POL'Y 7, 9-11 (2006) (asserting that a problem of customary international law is its neglect of democratic decisionmaking, and listing five different democratic deficits); John O. McGinnis & Ilya Somin, Democracy and International Human Rights Law, 84 NOTRE DAME L. REV. 1739, 1771-73 (2009) (arguing that "[d]omestic political actors cannot create norms by failing to object," unlike in customary international law). I do not address these arguments here.

(141) Statute of the International Court of Justice, supra note 20, art. 38 (1) (b).

(142) See Harlan Grant Cohen, Finding International Law: Rethinking the Doctrine of Sources, 93 IOWA L. REV. 65, 78 (2007) (remarking that in the prevailing positivist view, international law is based primarily on the consent of sovereign states instead of custom).

(143) See D'AMATO, supra note 21, at 90-98 (setting forth this description and providing hypothetical cases).

(144) See Beth Van Schaack, Negotiating at the Interface of Power and Law: The Crime of Aggression, 49 COLUM.J. TRANSNAT'L L. 505, 520-21 (2011).

(145) There are exceptions to this rule, such as when a treaty permits a majority or supermajority to actually enact amendments to a treaty. See Helfer, supra note 64, at 84-86 (discussing treaties, most notably the Montreal Protocol to the Vienna Convention on the Protection of the Ozone Layer, that permit majorities or supermajorities to bind dissenting states).

146 States could, of course, enter into a treaty that consisted of terms rejected by a diplomatic conference.

(147) See MOHAMED EL BARADEI & CHLOE GAVIN, CROWDED AGENDAS, CROWDED ROOMS: INSTITUTIONAL ARRANGEMENTS AT UNCLOS III, at 3-7 (1981) (explaining the terms of "the Gentlemen's Agreement" of 1973 that established the decisionmaking framework of UNCLOS III).

(148) Id.

(149) Id. at 9.

(150) Id.

(151) See Convention on the Rights and Duties of States, Dec. 26, 1933, 49 Stat. 3097, 165 L.N.T.S. 19 (codifying the customary rules of statehood); see also D.J. HARRIS, CASES AND MATERIALS ON INTERNATIONAL LAW 102 [paragraph] 2 (5th ed. 1998) (observing that the Montevideo Convention is widely accepted as codifying the customary requirements of statehood).

(152) (Declaration on the Maritime Zone, Aug. 18, 1952, 1006 U.N.T.S. 325.

(153) Declaration of Montevideo on the Law of the Sea, May 8, 1970, 9 I.L.M. 1081.

(154) See INTERNATIONAL LAW IN THE WESTERN HEMISPHERE 58-64 (Nigel S. Rodley & C. Neale Ronning eds., 1974) (detailing the history of seabed agreements between Latin American countries).

(155) See Benvenisti & Downs, supra note 104, at 611 (noting the increased frequency of "serial bilateralism" by large states in areas of "vital interest").

(156) See supra text accompanying notes 110-17.

(157) Declaration on the Establishment of a New International Economic Order, G.A. Res. 3201, U.N. GAOR, 6th Sess., Supp. No. 1, U.N. Doc. A/9959 (1974).

(158) For a discussion of the fall of the Hull rule's "prompt, adequate and effective" standard and the subsequent rise of the BIT, see generally Guzman, supra note 109, at 646-56. Guzman's claims that the Hull Rule was custom prior to the NIEO, but was then demolished by it, have both been criticized as incorrect. That is, some believe there was no customary law to be demolished, and others believe the customary law in question both existed and survived the NIEO's challenge. Compare Jeswald W. Salacuse & Nicholas P. Sullivan, Do BITs Really Work?: An Evaluation of Bilateral Investment Treaties and Their Grand Bargain, 46 HARV. INT'L LJ. 67, 68 (2005) (characterizing post-World War II customary investment law as "an ephemeral structure consisting largely of scattered treaty provisions, a few questionable customs, and contested general principles of law"), and Yackee, supra note 107, at 1565 & n.66 (discussing and citing scholars who "argue quite plausibly that customary international law has never fully reflected the Hull formulation"), with Alvarez, supra note 112, at 39 ("[T]he traditional customary rules of state responsibility, including the international minimum standard, were not displaced by some [developing countries'] efforts to establish the NIEO."). For my purposes, it is enough that the developed world claims that NIEO undermined customary investment rules, prompting them to engage in codification. Given the indeterminacy of legal rules, the strength of claims during negotiations about what legal rules require is more important than the binary distinctions that tribunals make about whether a law exists and covers the challenged conduct.

(159) See Gabriella Blum, Bilateralism, Multilateralism, and the Architecture of International Law, 49 HARV. INT'L L.J. 323, 351 (2008) ("The introduction of additional parties to treaty negotiations is hardly ever cost-free. It potentially increases barriers to efficient agreements and exacerbates problems of information asymmetry, strategic barriers, psychological barriers, and institutional constraints.").

(160) See id. at 351-52 (reporting that the "real negotiation" at Bretton Woods was a bilateral U.S.-U.K. negotiation). For this reason, the use of membership in a treaty negotiation as a proxy for transaction costs is imperfect. See, e.g., Daniel J. Gervais, Reinventing Lisbon: The Case for a Protocol to the Lisbon Agreement (Geographical Indications), 11 CHI.J. INT'L L. 67, 120 (2010) ("A massive addition of new members to the Lisbon system would naturally entail administrative obligations...."). While transaction costs certainly do increase with new members, the marginal increase in transaction costs probably falls as the overall number of members rises.

(161) See Peter K. Yu, Sinie Trade Agreements, 44 U.C. DAVIS L. REV. 953, 977 (2011) (noting that countries with limited resources "may not have the ability to dedicate efforts to normmaking in a multitude of competing fora").

(162) EL BARADEI & GAVIN, supra note 147, at 14-15.

(168) Id. at 19.

(164) See Erik Franckx, The 200-Mile Limit: Between Creeping Jurisdiction and Creeping Common Heritage?: Some Law of the Sea Considerations from Professor Louis Sohn's Former LL.M. Student, 39 GEO. WASH. INT'L L. REV. 467, 470 n.6 (2007) (noting the efforts of two prominent ambassadors to form a "private negotiating group ... competent to deal ... with the [exclusive economic zone]").

(165) See EL BARADEI & GAVIN, supra note 147, at 11 (explaining the pragmatic difficulties of" It] he package deal and consensus" voting in UNCLOS III negotiations).

(166) See Joel R. Paul, The Rule of Law Is Not for Everyone, 24 BERKELEY J. INT'L L. 1046, 1052 (2006) (book review) ("[T]he Vienna Convention on the Law of Treaties provides that the parties to a treaty are bound to it unless they act in conformity with the treaty to amend its express terms.").

167 See VCLT, supra note 38, art. 40 ("The amending agreement does not bind any State already a party to the treaty which does not become a party to the amending agreement.").

(168) See Grover, supra note 78, at 570 (arguing that codification may have the effect of "freezing the development of customary international law by 'photographing' the law at a moment in time").

(169) VCLT, supra note 38, art. 40.

(170) See Grover, supra note 78, at 570 (arguing that the static, coded nature of rules may stifle "the natural growth of international criminal law").

(171) See Arthur M. Weisburd, Customary International Law: The Problem of Treaties, 21 VAND. J. TRANSNAT'L L. 1, 11 (1988) (noting that the importance of state practice to international legal obligations means that treaties cannot freeze customary law simply by codifying it).

(172) Rome Statute, supra note 11, art. 10. The adoption of the definition of the crime of aggression was similarly qualified by interpretive understandings, which the United States designed with the goal of "undermin[ing] any tendency to refer to these definitions as evidence of the progressive development of customary international law." Beth Van Schaack, "The Grass That Gets Trampled When Elephants Fight"." Will the Codification of the Crime of Aggression Protect Women?, 15 UCLAJ. INT'L L. & FOREIGN AFF. (forthcoming 2012) (manuscript at 32 n.179). This effort was itself undermined by the fact that the United States proposed to include an explicit reference to "customary international law" in the "understandings" and was rebuffed. See Beth Van Schaack, Understanding Aggression I, INTLAWGRRLS (Jun. 24, 2010, 6:00 AM), 2010/06/understanding-aggression.html (reviewing and discussing the debate over the interpretive "understandings" that occurred in advance of the ICC Kampala Conference on May 31, 2010).

(173) Hague Convention Respecting the Laws and Customs of War on Land art. 2, Oct. 18, 1907, 36 Stat. 2277 [hereinafter Hague Convention] (consuming the applicability of Article I's mandate to issue orders to respect the regulations and customary law).

(174) International Military Tribunal (Nuremberg), Judgment and Sentences, Oct. 1, 1946 (quoting Hague Convention, supra note 173, pmbl.) (internal quotation marks omitted), reprinted in 41 AM.J. INT'L L. 172, 248-49 (1947).

(175) See, e.g., BG Grp. Plc. v. Argentina, Final Award, [paragraph] 302 (UNCITRAL Arb. Trib. 2007), (finding that the minimum standard is not fixed in time); Int'l Thunderbird Gaming Corp. v. Mexico, Award, [paragraph] 194 (NAFTA UNCITRAL Arb. Trib. 2006), Award.pdf (condoning treating the minimum standard as evolving law); KENNETH J. VANDEVELDE, BILATERAL INVESTMENT TREATIES: HISTORY, POLICY, AND INTERPRETATION 194, 228-29 (2010) (describing arbitration decisions that found an evolving standard of treatment). But see Glamis Gold, Ltd. v. United States, Award, [paragraph] 21 (NAFFA UNCITRAL Arb. Trib. 2009), (agreeing that "fair and equitable treatment" remains subject to the standard articulated in Neer v. Mexico, 4 R.I.A.A. 60 (U.S.-Mex. Gen C1. Comm'n 1926), but suggesting that the standard has evolved over time).

(176) Neer, 4 R.I.A.A. at 61-62.

(177) Pope & Talbot Inc. v. Canada, Award on the Merits of Phase 2, [paragraph] 118 (NAFTA Arb. Trib. 2001), Canada/PopeFinalMeritsAward-pdf (rejecting Canada's argument for the standard of egregious conduct Neer would require); see also Tecnicas Medicambientales Tecmed S.A.v. United Mexican States, IC-SID Case No. ARB(AF)/00/2, Award, [paragraph] 116 (May 29, 2003), 43 I.L.M. 133 (noting that international law is not "frozen in time"); Mondev Int'l Ltd. v. United States of Am., ICSID Case No. ARB(AF)/99/2, Award, [paragraph] 125 (Oct. 11, 2002), 6 ICSID Rep. 192 (2004) (finding that customary law refers to law as "it stands no earlier than the time at which NAFTA came into force").

(178) See Blum, supra note 159, at 359-61 (discussing the minimum effective coalition necessary to sustain cooperation); see also Helfer, supra note 64, at 98-100 (arguing that problem structure affects the minimum number of states whose participation is required to sustain cooperation).

(179) See, e.g., Barbara Koremenos, Contracting Around International Uncertainty, 99 AM. POE. SCI. REV. 549, 551 (2005) (noting that involving more states in negotiations may make the negotiations last longer than those between fewer states).

(180) Id.

(181) Id. at 551-52.

(182) Id.

(183) See, e.g., Michael J. Gilligan, Is There a Broader-Deeper Trade-Off in International Multilateral Agreements?, 58 INT'L ORG. 459, 461 (2004); Helfer, supra note 64, at 92.

(184) Gilligan, supra note 183, at 461-62.

(185) See id. at 462-63 (creating a model to predict required levels of dilution).

(186) See VCLT, supra note 38, arts. 34-37 (explaining when treaties may be binding on third states).

(187) There are a variety of ways for states to soften the unanimity rule, such as empowering diplomatic conferences to adopt draft articles with less than unanimous support or allowing a treaty to come into force upon some minimum number of ratifications. See, e.g., Guzman, supra note 63, 14-34 (discussing several modes of creating nonunanimous international agreements). Although these devices allow proposed treaty obligations to advance, they do not alter the fundamental unanimity rule imposed by the consent doctrine--to be bound, a state must consent. See id. at 14-15. Treaties frequently recognize the significance of individual states by making either particular states' ratification or ratification by states engaged in a minimum percentage of the regulated activity a condition precedent to the treaty's entry into force. See, e.g., Protocol to the International Convention for the Prevention of Pollution from Ships art. 5, para. 1, Feb. 17, 1978, 1340 U.N.T.S. 61 (providing that the Convention "shall enter into force twelve months after the date on which not less than fifteen States, the combined merchant fleets of which constitute not less than fifty per cent of the gross tonnage of the world's merchant shipping, have become Parties to it").

(188) Of course, a large group of like-minded states will be able to agree on "deep" standards of conduct more easily than a small group in which there is a diversity of interests. Most treaties, however, begin with a state or group of states pushing for a resolution to a problem. As the number of states necessary to ratify that resolution expands, it becomes increasingly likely that the additional states will diverge in preferences from the original group. For this reason, it makes sense to think about the breadth-versus-depth tradeoff as related to the number of parties to a treaty.


(190) See Van Schaack, supra note 144, at 522-23 (illustrating how the definitions of acts and crimes of aggression were "open to endless interpretation" as a result of a compromise in language struck by the two principal factions).

(191) See ICC, Assembly of States Parties, 5th Sess., Special Working Grp. on the Crime of Aggression, Informal Inter-Sessional Meeting of the Special Working Group on the Crime of Aggression, [paragraph] 18, ICC-ASP/5/SWGCA/INF.1 (Nov. 23-Dec. 6, 2006), available at http://

(192) See id. [paragraph][paragraph] 18-20.

(193) Van Schaack, supra note 144, at 518.

(194) Id.

(195) See id. ("Additional states submitted 'information concerning the appointment of representatives,' bringing the number of potential voting states up to 85...." (quoting Valerie Oosterveld, Final Day in Kampala, INTLAWGRRLS (June 11, 2010), http://

(196) See Todd J. Zywicki & Edward Peter Stringham, Common Law and Economic Efficiency (George Mason Law & Econ. Research Paper Series, Paper No. 10-43, 2010), available at (reviewing the literature on whether the common law evolves toward efficiency).

(197) See Eugene Kontorovich, Inefficient Customs in International Law, 48 WM. & MARY L. REV. 859, 889-94 (2006) (offering a list of circumstances under which a community is most likely to develop efficient norms--such as small group size and frequent group member interaction--and subsequently arguing that international states do no possess such attributes).

(198) Id. at 903. Despite this concern, Professor Kontorovich does not believe that codified customs are any more efficient than bare customs because treaties often create new rules that are merely called custom. See id. at 904. Treaties have no expectation of enforcement, and thus there are no additional costs to codifying inefficient customary rules. Treaties are rarely global in the way that customary law is, and even codifying treaties usually contain new rules, such that a deal might be struck to codify inefficient rules of custom in exchange for Pareto-improving new rules. Id. at 903-04.

(199) See Meyer, supra note 118, at 393-94 (discussing different treaty provisions that "affect a state's legal ability to exit an agreement," including conditional rights to exit and imposition of "costly bargaining procedures on would-be exiters").

(200) See VCDR, supra note 119, art. 23 ("The sending State and the head of the mission shall be exempt from all national, regional or municipal dues and taxes in respect of the premises of the mission....").

(201) See DENZA, supra note 54, at 151-52 (discussing the Canadian and American acceptance of the exemption as a customary rule).

(202) See Republic of Argentina v. City of New York, 250 N.E.2d 698, 700 (N.Y. 1969) ("The Department of State is of the opinion that under recognized principles of international law and comity the several states of the United States, as well as their political subdivisions, should not assess taxes against foreign government-owned property used for public noncommercial purposes." (quoting Letter from Richard D. Kearney, Acting Legal Adviser, Dep't of State, to the Comptroller of the City of N.Y. (Sept. 2, 1965)) (internal quotations omitted)).

(203) See id. at 701 (describing a provision in the VCCR similar to article 23 of the VCDR as "codifying or declaring existing law" and "intended to reflect the rules to which nations generally conform").

(204) 618 F.3d 172, 174-75, 197 (2d Cir. 2010).

(205) Id. at 175-76. A related issue on the jurisdictional immunities of foreign states was recently decided by the International Court of Justice. See Jurisdictional Immunities of the State (Ger. v. It.), [paragraph] 107 (Feb. 3, 2012), available at docket/files/143/16883.pdf (upholding Germany's immunity under international law from judgments against it rendered by Italian courts).

(206) See Permanent Mission of India, 618 F.3d at 177; see also Republic of Argentina, 250 N.E.2d at 700 n.2 ("In view of the many consulates and other government offices which it maintains abroad, the United States unquestionably has a real interest in having the court find that such property is immune from taxation under international law.").

(207) VCDR, supra note 38, art. 1(i).

(208) See Permanent Mission of India, 618 F.3d at 175; see also Designation and Determination under the Foreign Missions Act, 74 Fed. Reg. 31,788, 31,788 (July 2, 2009) (discussing the state practice of granting residences exemption from taxation and designating it as a benefit under the Foreign Missions Act).

(209) Permanent Mission of India, 618 F.3d at 178.

(210) 74 Fed. Reg. at 31,788.

(211) The dispute in Permanent Mission of India involved the tax-exempt status of a permanent mission to the United Nations, 618 F.3d at 175, a slightly different question than the tax-exempt status of diplomatic missions. Nevertheless, the broader point about the effect of tax exemptions remains the same.

(212) See Albiston, supra note 47, at 870 (noting that repeat players can control the outcome of rules due to "unequal resources" and "greater strategic knowledge"); Choi & Gulati, supra note 47, at 1152 (noting that in the case of boilerplate contract litigation, "the resources of the specific parties are unlikely to match the resources of the group of all contracting parties," and that the resources and abilities of the particular repeat litigant weigh heavily in the court's determination of which position to adopt).

(213) See Rep. of the Int'l Law Comm'n, 58th Sess., May 1-June 9, July 3-Aug. 11, 2006, [paragraph] 242-45, U.N. Doc A/61/10; GAOR, 61st Sess., Supp. No. 10 (2006) (describing the phenomenon of fragmentation in international law and the challenge it poses).

(214) Id. ([paragraph]) 143.

(215) See, e.g., Harlan Grant Cohen, binding International Law, Part II: Our Fragmenting Legal Community, 44 N.Y.U. J. INT'L L. & POL. (forthcoming 2012) (manuscript at 7), available at =1843566 (seeking the "root causes of fragmentation" in an effort "to diagnose the disease"); Andreas Fischer-Lescano & Gunther Tuebner, Regime-Collisions: The Vain Search for Legal Unity in the Fragmentation of Global Law, 25 MICH. J. INT'L L. 999, 1001-02 (2004) (discussing "[t]he issue of how to combat" fragmentation, as well as "all the problems of contradictions between individual decisions, rule collisions, doctrinal inconsistency and conflict between different legal principles ... increasingly concerning case law, expert committees, ICJ Presidents and academic controversies") (footnotes omitted).

(216) See, e.g., William W. Burke-White, International Legal Pluralism, 25 MICH. J. INT'L L. 963, 967 (2004) ("An alternate perspective on the increasing number of fora for international legal adjudication is that international law is today more relevant than it has ever been in the past."); see also ANNE-MARIE SLAUGHTER, A NEW WORLD ORDER 10 (2004) ("[W]e need global rules without centralized power but with government actors who can be held to account through a variety of political mechanisms.... [A] world order based on government networks, working alongside and even in place of more traditional international institutions, holds great potential."); Raustiala, supra note 104, at 3-4 (arguing that contemporary international cooperation "is not inter-national at all" but is in fact occurring at the state level and that the laws made at that level between cooperating domestic agencies "have growing international salience" in a "globalizing world").

(217) Burke-White, supra note 216, at 971.

(218) Raustiala, supra note 104, at 7.

(219) Martti Koskenniemi & Paivi Leino, Fragmentation of International Law? Postmodern Anxieties, 15 LEIDEN J. INT'L L. 553, 555 (2002) (quoting H.E. Judge Gilbert Guillaume, President, Int'l Court of Justice, Speech to the Gen. Assembly of the United Nations (Oct. 30, 2001) (transcript available at pr=82&pt=3&p1=1&p2=3&p3=l)) (internal quotation marks omitted); see also Benvenisti & Downs, supra note 104, at 625 (discussing how fragmentation can be used a strategy by powerful states to effectively disenfranchise weak states).

(220) See Caroline Henkels, Overcoming Jurisdictional Isolationism at the WTO-FTA Nexus: A Potential Approach for the WTO, 19 EUR. J. INT'L L. 571, 574 (2008) (explaining that jurisdictional overlap and the rise in adjudicative fora with compulsory jurisdiction increases the risk of conflicting decisions and fragmentation).

(221) See Benvenisti & Downs, supra note 104, at 630 ("Any serious effort on the part of international tribunals to define the nature of these ... obligations ... will inevitably bring [them] into conflict with the fragmentation strategies of powerful states...."); Kal Raustiala & David G. Victor, The Regime Complex for Plant Genetic Resources, 58 INT'L ORG. 277, 299 (2004) (arguing that actors, when given the choice of any of a multitude of fora for developing international rules, will select a forum that is best suited to their interests).

(222) See Benvenisti & Downs, supra note 104, at 599 (outlining four separate "fragmentation strategies" and noting the corresponding increase in transaction costs for weaker states).

(223) Id. at 610.

(224) See id. ("This proliferation of narrow agreements with few, if any, linkages makes logrolling among weaker states and cooperation virtually impossible.").

(225) See id. at 614 (maintaining that powerful states commonly switch to a competing venue when weaker states gain control); see also Cohen, supra note 215 (manuscript at 36-39) (arguing that fragmentation is not only a result of doctrinal inconsistencies, however strategic, but that it also stems from fundamental disagreements about legitimate means of international lawmaking).

(226) Cf. Cohen, supra note 215 (manuscript at 18-19) (arguing that interpretative disputes often mask deeper disputes over the legitimacy of international lawmaking procedures).

(227) This structure resembles the anachronistically labeled game known as the "Battle of the Sexes," in which players prefer to coordinate their activities but can fail to do so because of distributional tensions over which outcome to coordinate to achieve.

(228) See J. Patrick Kelly, The Twilight of Customary International Law, 40 VA. J. INT'L L. 449, 510 & n.254, 511 (2000) (detailing the "curious era" in which new states adopted the minimum customary standards of the once-controlling Western nations).

(229) See Curtis A. Bradley & Mitu Gulati, Withdrawing from International Custom, 120 YALE L.J. 202, 230-31 (2010) (arguing that Western countries developed the theory that one could not withdraw from customary international law during decolonization to ensure that customary international law developed by Western states would continue to bind new states).

(230) The median voter model is a theoretical model of voting in which the median member of a group determines the group's policy choices. See generally Duncan Black, On the Rationale of Group Decision-making, 56 J. POE. ECON. 23 (1948) (articulating and outlining the original rationale behind the theorem).

(231) See supra Section III.D.

(232) Jesse Richman, Parties, Pivots, and Policy: The Status Quo Test, 105 AM. POL. SCI. REV. 151, 152 (2011).

(233) See id. at 152-53 (describing several theories under which institutional veto players and political parties can alter the median voter result).

(234) See Jonathan B. Potts, Note, Stabilizing the Role of Umbrella Clauses in Bilateral In vestment Treaties: Intent, Reliance, and Internationalization, 51 VA. J. INT'L L. 1005, 1006 (2011) (noting that the United States is a party to 40 BITs and that approximately 2600 exist worldwide).

(235) See SCHILL, supra note 2, at 24 (asserting that BITs may serve the same goals as multilateral investment agreements).

(236) See Elkins et al., supra note 114, at 277 (noting that countries which historically have been viewed as having unreliable investment environments now compete to establish BITs favorable to foreign investment).

(237) See, e.g., Guzman, supra note 109, at 679 (discussing the different incentives developing countries face when considering signing a BIT or a multilateral agreement).

(238) See Benvenisti & Downs, supra note 104, at 610 (suggesting that developed countries often use narrow agreements to divide weaker states and hamper their ability to negotiate).

(239) See generally Elkins et al., supra note 114.

(240) See SCHILL, supra note 2, at 53 (noting that moving negotiations to the OECD was part of a strategy to exclude developing states from negotiations).

(241) See id. at 31-60 (describing various failed multilateral efforts to negotiate an agreement on investment).

(242) See id. at 25-28 (explaining the history of modern informational rules about aliens and their relationship to international investment law).

(243) Id.

(244) Id.

(245) Id.

(246) 4 R.I.A.A. 60, 61 (U.S.-Mex. Gen. C1. Comm'n 1926)

(247) See, e.g., Glamis Gold, Ltd. v. United States, Award, ([paragraph]) 21 (NAFTA UNCITRAL Arb. Trib. 2009), (holding that the standard for fair and equitable treatment had not changed since Neer).

(248) See VANDEVELDE, supra note 175, at 226-32 (discussing how the international minimum standard has evolved over the course of a series of NAFTA arbitration cases).

(249) Id.

(250) See SCHILL, supra note 2, at 27 (noting that a movement ot challenge the existence of an international minimum standard started after World War I).

(251), See id. at 26-27 (noting that the U.S. Secretary of State at the time, Cordell Hull, articulated an early version of the international minimal standard when he denied Mexico's right to treat American citizens on par with Mexican nationals in a series of diplomatic notes).

(252) See id. at 27 (asserting that the Latin American countries' movement "gained ground due to the successful communist revolution in Russia in 1917").

(253) Shan, supranote 115, at 126.

(254) G.A. Res. 3201, supra note 157.

(255) Id.

(256) G.A. Res. 3281 (XXIX), U.N. Doc. A/RES/3281 (Dec. 12, 1974).

(257) Id. art. 2.2(c).

(258) See Meyer, supra note 103, at 906 (defining soft law as "those agreements that are not themselves legally binding but are created with the expectation that they will be given the force of law through either domestic law or binding international agreements").

(259) Compare Alvarez, supra note 112, at 39-40 (arguing that the NIEO failed to disrupt existing customary rules of international law), with Guzman, supra note 109, at 651 (arguing that the NIEO demonstrated that certain customary rules of international law, such as the Hull Rule, no longer applied).

(260) Andean Commission, Common Regime of Treatment of Foreign Capital and of Trademarks, Patents, Licenses, and Royalties, Decision 24 art. 50, Nov. 30, 1976, 16 LL.M. 138, 153.

(261) Id.

(262) See U.N. Conference on Trade & Dev. (UNCTAD), Bilateral Investment Treaties 1959-1999, at 25-123, U.N. Doc. UNCTAD/ITE/IIA/2 (2000), available at http:// (listing all BITs that countries entered into between 1959 and 1999, including those to which Andean Pact nations were parties).

(263) The members of the Andean Pact during this period were Bolivia, Colombia, Ecuador, Peru, and Venezuela. The two exceptions were Ecuador's BIT with Uruguay and Bolivia's BIT with Germany, which both came into force in November 1990. Id. at 33, 48. Bolivia signed several other BITs during the period between Decision 220 and Decision 291 (i.e., between the partial and total repeal of the codification of the Calvo doctrine), but none of the other BITs came into force until after Decision 291. See id. at 33.

(264) Id. at iii.

(265) Id. at 89.

(266) Id. at 118-19.

(267) It. at 90.

(268) See SCHILL, supra note 2, at 42-43 ("Overall, the provisions of many of the regional and sectoral investment treaties closely resemble the standard content of BITs ....").

(269) See id. at 24 (arguing that BITs are simply "functional substitute[s]" for a multilateral scheme); Lowenfeld, supra note 6, at 128 (arguing that the existence of a large number of similar BITs is effectively "international legislation").

(270) See 2004 U.S. MODEL BIT, supra note 40, at 7 ("Each party shall accord to covered investments treatment in accordance with customary international law, including fair and equitable treatment and full protection and security.").

(271) See, e.g., Marek Wierzbowski & Aleksander Gubrynowicz, Conflict of Norms Stemming from Intra-EU B1Ts and EU Legal Obligations: Some Remarks on Possible Solutions (arguing that it would be difficult to get states to agree that FET is part of the international minimum standard), in INTERNATIONAL INVESTMENT LAW FOR THE 21ST CENTURY 544, 549 (Christina Binder et al. eds., 2009).

(272) See 2004 U.S. MODEL BIT, supra note 40, at 38; 2004 CANADA MODEL FOREIGN INVESTMENT PROMOTION AND PROTECTION AGREEMENT, annex B, art. 13 (1), available at http:/ / trade-agreements-accords-commerciaux/assets/pdfs/ 2004-fipa-model-en.pdf.

(273) 2004 U.S. MODEL BIT, supra note 40, at 38.

(274) See Suzanne A. Spears, The Quest for Policy, Space in a New Generation of International Investment Agreements, 13 J. INT'L ECON. L. 1037, 1051 (2010) (noting that a 2005 free trade agreement between India and Singapore and a 2006 BIT between China and India incorporated interpretative statements similar to those adopted in 2004 in the United States and Canada's model BITs).

(275) Common Mkt. for E. and S. Africa (COMESA), Investment Agreement for the COMESA Common Investment Area art. 20(8), May 23, 2007 available at wednesday/ExerciseMaterials/invagr eecomesa.pdf.

(276) See Ass'n of Se. Asian Nations (ASEAN), ASEAN Comprehensive Investment Agreement annex 2, Feb. 26, 2009, available at FINAL-SIGNED-ACIA.pdf ("Nondiscriminatory measures of a Member State that are designed and applied to protect legitimate public welfare objectives, such as public health, safety and the environment, do not constitute an expropriation...."); see also Spears, supra note 274, at 1052 (noting that the 2009 ASEAN Comprehensive Investment Agreement takes the same absolutist approach as the 2007 COMESA Common Investment Area Agreement).

(277) The effort to include investment in the Havana Charter was largely driven by the United States, which ultimately declined to ratify the Charter because the protections contained therein were too favorable to capital-importing states. See SCHILL, supra note 2, at 33.

(278) See id. at 58-60 (recounting the rise and fall of international investment law as a main topic in the WTO).

(279) See Pierre Sauve, Multilateral Rules on Investment: Is Forward Movement Possible?, 9 J. INT'L ECON. L. 325, 330-31 (2006) (acknowledging the impasse that developed between countries that supported investment negotiations in the WTO and those that opposed them).

(280) See Blum, supra note 159, at 341-42 (arguing that defeating a multilateral agreement on investment at the WTO may have been a "Pyrrhic victory" for poorer countries left to negotiate BITs with more powerful states).

(281) See Sauve, supra note 279, at 330 (indicating that the business community in the United States was concerned that a low-standard WTO agreement might lower the high standards of investment protection in U.S. BITs).

(282) Id. at 331. Brazil, for its part, has signed 15 BITs but has not brought a single one into force. See ICS1D Database of Bilateral Investment Treaties, ICSID, http://icsid. (follow "Publications" hyperlink; then follow "Search Listings of Bilateral Investment Treaties" hyperlink; then follow "Brazil (15)" hyperlink) (last visited Jan. 15, 2012).

(283) George Chifor, Caveat Emptor: Developing International Disciplines for Deterring Third Party Investment in Unlawfully Expropriated Property, 33 LAW & POL'Y INT'L BUS. 179, 271 n.326 (2002).

(284) See SCHILL, supra note 2, at 55 (indicating that India's opposition to the MAI was based, at least in part, on the fact that non-OECD members couldn't participate in the negotiations).

(285) See Stefan D. Amarasinha & Juliane Kokott, Multilateral Investment Rules Revisited (arguing that the failure of the MAI was partially attributable to a "lack of support from industry who found that ... existing BITs provided better protection"), in THE OXFORD HANDBOOK OF INTERNATIONAL INVESTMENT LAW 119, 127 (Peter Muchlinski et al. eds., 2008).

(286) See id. (asserting that one of the reasons for the MAI's failure was the "lack of clarity as regards the relationship between the MAI, BITs, and the GATS and TRIMs Agreements").

(287) OECD, Negotiationg Grp. on the Multilateral Agreement on Inv., The Multilateral Agreement on Investment: Commentary to the Consolidated Text, DAFFE/MAI(98)8/REV1 (Apr. 22, 1998), available at

(288) OECD, The MAI and Bilateral, Regional and Sectoral Agreements, Note by the Chairman, DAFFE/MAI(96)26 (Aug. 30, 1996), available at pdf/ng/ng9626e.pdf.

(289) OECD, supra note 287, at 29.

(290) See Draft MAI, supra note 10, [section] X (using language that clarifies the relationship only between the draft MAI and the Articles of Agreement of the International Monetary Fund).

(291) See Anca Radu, Foreign Investors in the EU--Which 'Best Treatment'? Interactions Between Bilateral Investment Treaties and EU Law, 14 EUR. L.J. 237, 241 (2008) ("The Commission ... found that a number of provisions in the BITs concluded by eight [Central and Eastern European] countries with the USA and Canada were inconsistent with the EU law and had to be renegotiated.").

(292) See id. (noting that the renegotiation of a BIT already in force is not an easy task and that the Commission had to assist the eight countries with the renegotiation process).

(293) See id. ("The eventual incompatibilities between Member States' obligations under the BITs and EU law may lead to disputes....").

(294) Id.

(295) See, e.g., Wierzbowski & Gubrynowicz, supra note 271, at 549 (acknowledging that the lack of a clear stance by the EU on the FET clause means there is a possibility that investors could use the clause in intra-EU BITs to challenge domestic measures that are necessary to meet obligations under EU law).

TIMOTHY MEYER, Assistant Professor of Law, University of Georgia School of Law. For helpful comments at various stages of this project, I am indebted to Curtis Bradley, Elizabeth Burch, Dan Coenen, Harlan Cohen, Robert Cooter, Jeff Dunoff, Jean Galbraith, Andrew Guzman, Matthew Hall, Duncan Hollis, Patti Heald, Laurence Helfer, Hillel Levin, Willow Meyer, Mark Pollack, Usha Rodrigues, Peter Rut]edge, Pammela Saunders, Jessica Stanton, Logan Sawyer, Christian Turner, David Zaring, and participants at faculty workshops at the Sandra Day O'Connor College of Law at Arizona State University and Temple University Beasley School of Law as well as a New Scholars panel at the annual meeting of the Southeastern Association of Law Schools, and the Junior International Law Scholars Association conference at Yale Law School. Thanks to David Klein and Mitchell Metcalf for research assistance.
COPYRIGHT 2012 University of Pennsylvania, Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Meyer, Timothy
Publication:University of Pennsylvania Law Review
Date:Mar 1, 2012
Previous Article:Triaging appointed-counsel funding and pro se access to justice.
Next Article:Getting more by asking less: justifying and reforming tax law's offer-in-compromise procedure.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters