Final Thoughs on Discussion of How International Law Works

by Andrew Guzman

As this it my final post in connection with this discussion of my book, How International Law Works, I want to thank Opinio Juris for giving me this opportunity, and the commentators for so thoughtfully sharing their opinions. Much of the discussion has been about the methodology used in the book, and as I have had my say on that subject in my several prior posts I will not dwell on it now.

Let me instead mention a couple of things that I hope the book has achieved or will achieve as more people get the chance to read it. I hope the book provides a foundation for further rational choice analysis of international law. I, for example, have recently been working on international tribunals, a subject that is not covered in detail in the book. More could be said from a rational choice perspective on just about any international law topic.

I also hope the book contributes to the discourse between international law and international relations. These fields have grown closer in the last twenty years, and I think both have benefited. This book seeks to address both sides of this narrowing divide, and hopefully is useful to both.

I very much want this book to also speak to more traditional international law scholars. There is no natural tension between conventional views of international law and rational choice. It is true that rational choice, because of its emphasis on reasoning from assumptions can sometimes seem abstract and disconnected from reality, but that is exactly why it is important to have constant reminders of the need to address real question and real puzzles. A rational choice approach, it seems to me, strengthens the study of international law in part because it provides theoretical underpinnings with which international law and legal scholars can respond to critics, analyze hard questions, and debate the role and workings of the field.

The world needs international law now, perhaps more than ever before. Problems from climate change to nuclear proliferation to hunger to disease to terrorism will only be solved with the help of international legal structures and institutions. Whatever else international law scholars do, we need to be thinking about when international law can help with these problems and what structures or approaches are most effective. We, need, in short, to think very hard about how international law works.

Why Isn’t There More Soft Law? A Response to Raustiala

by Andrew Guzman

In How International Law Works I grapple with the question of how states make the trade-off among the various features of agreements, including hard and soft law. I am not sure I agree that Kal’s empirical puzzle actually exists, but let’s assume it does and see why that might be so. A very similar question is discussed in the book – why are dispute resolution procedures almost never used in soft law agreements? The argument in the book (pp. 157-161) is very close to what follows.

One possible explanation for why we do not see many large, high-profile multilateral agreements can be seen by imagining a move from a soft-law agreement with few clear substantive obligations (e.g., the Rio Declaration) to an agreement that involves a greater level of commitment. One can imagine doing this in two ways (in fact, there are more than two, but let’s keep it simple). First, the states could stick with a soft law form, but provide more specific obligations. States would not be legally bound, but there would be a clear statement of what is expected of them. A failure to live up to the agreement would have some costs, though presumably less than the costs of violating the same agreement if it were a treaty. Second, states could leave the language unchanged but turn the agreement into a treaty. This would leave obligations somewhat vague and therefore give states some flexibility in how they behave while still claiming compliance.

Kal’s empirical claim raises the question of why we don’t see more of the first category (soft law with clear terms) and less of the second (treaties with vague terms), at least in some high-profile contexts. Why was the Vienna Convention for the Protection of the Ozone Layer a vague treaty rather than a soft law instrument? A partial answer comes from thinking about what sort of state preferences would yield that result. We expect this outcome if we assume that states prefer to increase their level of commitment by increasing the formality of the promise rather than its precision. This suggests that the choice of hard law offers a better cost-benefit trade-off to states than does the choice of more exact language. This, of course, begs the question of why they have that preference, and I do not have a satisfying response. The good news is that there is a lesson to be learned here. Perhaps Kal’s empirical puzzle is not such a puzzle at all, but instead tells us something we did not previously know about state preferences and the trade-offs states face.

Raustiala Reply

by Kal Raustiala

Andrew’s reply makes a good point. I share his skepticism about my proffered explanation for the choice of hard law over soft law. The problem is that we have an empirical puzzle that demands explanation, and we lack a good one. If there is a better explanation than the one I put forward, I have not yet seen it.

The puzzle is that under Andrew’s theory we ought to see lots of soft law agreements. To some degree we do, and there is a selection bias problem to boot: soft law agreements are not always listed in the usual treaty databases, and thus it is hard to know how many there are. However, it seems hard to argue that there are more multilateral soft law agreements than hard law agreements. And it is clearly the case that major multilateral negotiations more frequently result in a hard law agreement, and when they do not they are usually viewed as failures.

The landmines issue is a good example. Spurred on by NGOs, landmines regulation advocates convinced many governments to negotiate a major agreement. That agreement resulted not in a soft law accord but a treaty. We observe this pattern again and again. Some big UN meetings do yield soft law accords (e.g., the Rio Declaration of 1992) but again, advocates in the area generally deride these as weak and vacuous. So I think the puzzle remains: why so few soft law accords? As I noted in my first post, we see many soft law accords in technocratic areas that are not politically salient, such as banking. That is consistent at least with the theory that domestic/civil society preferences might be playing some role in the choice between the two. But it is certainly plausible that other factors are at play, or that I am entirely wrong about this.

Let me briefly comment on something else Andrew said. He wrote that international law is perhaps just as oxymoronic as soft law, in that the former lacks coercive enforcement of the sort we see in the domestic context. Yet it is interesting to note that in constitutional law we don’t see very much coercive enforcement either. When President Truman lost in Youngstown, what forced him to comply with the ruling? It certainly wasn’t that the military threatened to lock Truman up if he failed to follow the Supreme Court.

We could multiply the instances easily; anytime a federal court orders a government agency to do X or not do Y, why does the executive branch follow the court and obey the law? Again, it cannot be coercive power that explains it, since the executive wields the coercive power. The point here is that the difference between international law and domestic law is not always very great, and in fact may be illusory in some instances—instances that are at the core of our domestic legal system. It is worth bearing this in mind as we think about how international law works.

Treaties and Soft Law Stories, A Reply to Raustiala

by Andrew Guzman

Kal’s comment raises some interesting questions about the nature of international law and what it means to be a treaty rather than soft law. He juxtaposes hard law and soft law, as I did in my opening comments. Let me suggest another comparison – hard law in the international context and domestic law. The contrast here is stark and obvious – domestic law usually comes with the potential of coercive enforcement while international law rarely does. So based on that comparison, you might say that not only is the term “soft law” oxymoronic, as Kal says, but so is the term “international law.” International law does not “bind” a state in anything like the way domestic law binds domestic actors. The point here is that whatever soft law is, and whatever international law is, worrying about whether they should be called “law” makes little sense unless we know what “law” is. How International Law Works is, in significant part, an effort to help us understand what international law is.

Kal argues that there may be more to the distinction between hard and soft law than my book suggests. He says that governments pay a lot of attention to whether or not an agreement is legally binding under international law. Clearly states do worry about the legal status of agreements, but they also worry about a bunch of other things as well. Soft law agreements (Helsinki Final Act, Basel Accord, Universal Declaration of Human Rights, cooperation agreements in competition law and securities, etc) are common and often appear to generate high levels of compliance. So I think I am a little less persuaded than Kal that there is a bright line dividing treaties from soft law.

That is not to say that I think there is no difference between the two. I would quibble with Kal’s statement that I believe it is “not very significant” whether an agreement is hard or soft law. Differences of degree are nevertheless differences, and I think it does increase the level of commitment when states enter into treaties.

There is a literature in both law and political science about why states choose treaties rather than soft law, and some of it focuses on the domestic implications of treaties. I largely agree with the suggestion in this literature (and in Kal’s post) that these domestic effects matter. To prove it to you let me quote from my book: “Perhaps the most important influences on the decision [to choose hard law rather than soft law] are the domestic law implications of selecting a treaty. This is certainly the most conventional account of the choice between hard and soft law, and is surely an important influence on the decision.” (p. 145). So I agree with the importance of these effects. This is an area in which work by legal scholars, informed by the existing international relations literature, could make a big contribution.

I find Kal’s second hypothesis – that some domestic constituencies prefer hard law over soft law because they think international law is quite a bit like domestic law and so want to see “real” treaties – more problematic. The argument would have to go something like the following. Human rights groups like Amnesty International lobby to get human rights treaties drafted, and to get states to join. They have a strong preference for treaties because they, in effect, confuse international law and domestic law, and so think that a treaty is much more effective, relative to soft law, than it actually is. It is implausible to think that Amnesty itself is fooled in this way, so we have to back up a step and look to those that support Amnesty. So it must be those that contribute time or money to Amnesty who are fooled in this way, and the folks that run Amnesty International pander to the preferences of that population, even when doing so does not serve the goal of advancing human rights. So the story ultimately relies on an assumption that the public is both the engine that drives outcomes (a treaty rather than soft law) and is fooled. But of course we can explain any actual or imagine outcome through this line of argument. If, for example, we observed that states seem to enter into soft law agreements more often than theory predicts, the explanation would be that the public just wants to see that the agreement contains desirable substantive terms and that the difference between a treaty and soft law is lost on them. Indeed, something very close to this argument is sometimes advanced to explain human rights treaties – the states that sign realize that the treaties will have little impact, but they sign anyway because domestic constituencies want to see that their governments are working to improve human rights and the treaties fool those domestic groups.

Arguments that rely on an unsophisticated but powerful public start to look like arguments that rely on unsophisticated policy makers, since the latter become a pass-through for the goals of the former. If it is assumed that the public (or other important domestic constituencies) think of international law as similar to domestic law, we are very close to an assumption that the public has a preference for compliance with international law generally. If this is acted on by decision-makers, we have recreated the traditional assumption that states have a propensity to comply with international law. This is a long-standing view among international law scholars and practitioners, and it retains significant support. For reasons explained in the book and familiar in debates about international law, I think it is a good idea to think about international law without making this assumption. For the same reasons I am reluctant to make assumptions about democratic constituencies that recreate the propensity to comply.

Raustiala on How International Law Works

by Kal Raustiala

The conversation thus far has been interesting, but I want to bring up some new issues that we have not yet discussed. I have long been interested in how international agreements work, in part because since the end of the Second World War (at least) international agreements have been the most significant source of international law, and in part because this is one of the more natural areas of overlap between international legal theory and international relations theory. For a long time both fields studied the same topic, but used different nomenclature: “regimes,” “treaties,” “institutions,” and so forth. Regime was the label preferred in IR circles, but as the late Abe Chayes often said, at the core of almost every regime is a treaty. Indeed, he claimed that political scientists were basically just afraid to use “the L word.” That fear is generally gone now, and one of the results is that we have an increasing supply of papers on the topic of how agreements actually work and why they look the way they do.

Andrew’s book delves into these topics on a number of fronts. I want to focus briefly an issue he raises in his opening post: the distinction between hard law and soft law. At one level is this is a narrow question about why we see different types of international agreements and what difference these types make.

At another level, however, this topic is emblematic of a larger issue, or claim, at the heart of Andrew’s approach. For the most part his analysis of international law proceeds as if “law” does not matter very much. What does matter is commitment, and he explores the different ways states make commitments and, most importantly, the ways they make those commitments credible to other actors. Law per se is not all that important to this discussion. Consequently, whether an agreement is soft law or hard law is, in his view, not very significant. As he says in his opening post, “there is no stark difference between the two.”

I find the term soft law largely oxymoronic. But to make things easier, I will use it here. Let’s first consider whether there is really is little difference between hard and soft law in terms of effects. The standard lawyers view is that a hard law agreement is more meaningful (more “binding”) than a soft law agreement. But since international law lacks the enforcement mechanisms of domestic law, what binding means in this context is hard to say.

For Andrew the answer is more commitment. He suggests that hard law is really no different than writing down a treaty on special red paper that everyone agrees means that a commitment is really, really serious. But at the same time he points out that the stringency of commitment can be varied by manipulating a whole host of treaty design features and by staking reputation in other, extra-treaty ways. As a result, whether an agreement is hard or soft doesn’t really matter all that much; it is the overall level of commitment that matters. It is definitely worth thinking about whether this claim is correct—that “law” in the international system is really just a way of saying, “I really mean it this time.”

To some degree I think he is right. In other work I have made similar claims to his about the design of treaties (see e.g. Form and Substance in International Agreements, AJIL 2005). But one problem with Andrew’s approach is that there is substantial evidence that governments pay a lot of attention to whether an agreement is legally binding or not, and they are very careful about labeling, or at least signaling, which is which. And since they possess other levers—other treaty design features—that enable them to calibrate the level of commitment, it isn’t clear why the hard law/soft law distinction matters so much to states, at least in Andrew’s worldview.

In short, the empirical behavior of governments suggests that there is some difference between the two that matters more than his theory suggests it should. The question is, from a rational choice perspective, what is the difference? First, a legally-binding commitment often leads to significant domestic effects, such as the need to get legislative approval. Governments have an incentive to choose hard or soft law in part because of these domestic legal effects. These effects are largely left out of his analysis, but they are clearly important.

Second, it may be that domestic actors, civil society, and other “constituencies for cooperation” think that international law is quite a bit like domestic law, and so naturally want to see “real” treaties rather than soft law substitutes. Governments rationally supply these treaties. There is some evidence for this claim; soft law agreements seem most common in technocratic areas in which domestic political groups don’t really have a strong presence. But this claim requires some further assumptions about these constituencies being unsophisticated about international law, and these claims maybe aren’t all that warranted.

There is more to be said about this, of course, but it worth thinking more about what role law per se actually plays in international cooperation and especially in international agreements. This question is a big one.

The Puzzle of Human Rights Treaties

by Andrew Guzman

I confess that I am a little surprised by the focus on methodological issues that has emerged in the discussion of my book. My intention was to adopt non-controversial rational choice assumptions widely used across both law and the social sciences. This approach to modeling behavior is dominant, for example, in both economics and political science. Equally relevant, there is an enormous literature debating the merits and demerits of the approach and the debates on the subject are, by now, well-worn and familiar. I fear that we are simply rediscovering what others, (international relations scholars in particular) have known for a long time. I do not mean by this that questioning the methods I or anyone else uses is out of bounds. But if I have indeed carried out a rational choice analysis of international law without doing violence to that methodology, I hope we can get beyond the question of whether rational choice is useful (it seems everyone agrees that it is) and whether it is the only thing that is useful (nobody thinks so).

Alex’ most recent post addresses the question of how reputation interacts with treaty ratification. As I said in my earlier post, the issue of ratification is an interesting and important one, but not one I focus on in my book. The most natural way to think of the decision to ratify is to imagine a trade in which each country makes a promise in exchange for the promise of its treaty-partner. With human rights treaties, however, this is sometimes not a persuasive explanation. Jeff Dunoff reminded us below that at least some human rights treaties feature a form of compensation in exchange for joining, but I doubt we can explain all treaties in this way. I suggested another explanation in an earlier post – perhaps for many states the cost of joining is very small, meaning the even modest benefits are enough to cause states to join. There was also a discussion of how reputation might influence the ratification of treaties. A country whose reputation understates its willingness to comply with international law (or perhaps with human rights treaties) might, under certain conditions, enjoy a reputational boost by signing a treaty and then complying with it.

All of these explanations, and more, are available to explain participation in human rights treaties. I remain, however, of the view that we lack a good explanation for these treaties generally. It may be possible to explaining them one at a time, but developing “just so” reasons for ratification does not amount to a theory. So where does that leave us? I think it leaves us with a puzzle about why states join human rights treaties. Alex asks if this fact signals a weakness in rational choice theory. That is hard to say. It may be that the solution to the puzzle simply needs to be identified, and that it will be found within a rational choice model. A persuasive critique of a rational choice approach on this basis requires, at a minimum, an explanation of how this puzzle is explained through some alternative methodology.

Reputational Limits Continued

by Alex Geisinger

I thank everybody for their comments regarding the reputational posts. I think this conversation has been excellent up to now and hope this post will add some substance as we approach, what seems to me, to be some agreement regarding the reputational element of rational choice. Before I begin, let me also be clear about my terminology. I agree with Andrew regarding the notion of internalization. To me, internalization is analogous to a preference change within a state (to the extent we assume a state can have a preference). As Andrew points out in his response to my earlier post, I believe there may be many different mechanisms at play in this process. However, given that I take my charge here to be to focus on the explanatory value of Andrew’s rational choice account of international law, I do not fully consider these alternative models.

So, to the extent we are testing the rational choice framework, I move forward with my thoughts in hopes to approach some understanding of the theory’s benefits and limits in the area of reputation. As I understand Andrew’s response, we are in general agreement that some force which we loosely call “reputation” exists and that it probably does a lot of “heavy lifting” when it comes to garnering human rights treaty ratification. I’m not sure there’s agreement on this but it seems fair to at least assume that if reputation affects willingness to ratify in human rights treaties, it also works similarly with regards to other types of treaties and perhaps in other areas of international law. I think, as well, we agree that a very parsimonious rational choice theory cannot fully account for the reputational force or that, at least, Andrew has not provided a full account of the reputational force as it applies to ratification in his book.

If we agree that the reputational force cannot be fully captured in a parsimonious theory, then the question I raise is how this limitation affects the explanatory power of the the theory as a whole? That is, if the reputational force influences state decision-making but we do not have a sense of exactly how this force works, doesn’t this limit the explanatory power of the rest of the rational choice theory? This is particularly the case if we assume the reputational force to be a powerful and pervasive one and is, of course, less the case if we assume the reputational force to be weak and of limited influence. My own sense is that the ability of the reputational force to do the “heavy lifting” in human rights treaty ratification suggests it can be a relatively powerful force. My own work in the area of norm development and origin also suggests to me that reputational forces are pervasive.

There is, of course, no doubt that Andrew’s elegant description of how rational self-interested states may create and comply with international law has great value. Among other things, he has provided us with a springboard for an important discussion of how rational choice theories explain international law. In his response to my earlier post, Andrew suggested that he believes reputational influences can, ultimately, be explained fully by rational choice theories. As my earlier post suggests, I believe that other processes may account for the reputational influence as well and I look forward to a continued dialogue regarding these issues. Before entering such a dialogue, however, I will ask for everyone’s thoughts on the issue I raise here regarding the limits of reputation in rational choice. I am, of course, also interested in knowing whether I have properly identified areas of agreement and disagreement and any other issues with my analysis.

Rational Choice and Broader Conceptions of International Law, A Response to Jeff Dunoff

by Andrew Guzman

Thank to Jeff for this thought provoking comment and his kind words about my book. I am not sure that my response addresses all his concerns, but I hope it at least speaks to some of them!

Let me first state that while I agree with much of what Jeff has stated in his post, I do not agree when he says that my conception of compliance presupposes a particular understanding of international law. My book’s central focus is an attempt to explain how a rule of international law can affect state behavior. I think that question would be relevant under any of the conceptions of international law Jeff mentions. Everybody agrees that the impact of treaties, soft law, CIL, and norms on behavior is a central question in international law. In that sense, I think the book sheds light on the field regardless of one’s view of the nature, purpose, or functions of international law.

I think Jeff’s post poses two quite different questions. The first is the familiar question of whether rational choice models are the one and only way to think about international law. The answer to this is clearly “no” as I say in the book and have said in earlier posts here at Opinio Juris. Liberal theories and constructivist theories, as Jeff mentions, have some claim to informing our understanding of international law. In some ways these approaches are also compatible with rational choice models because they can help to explain state preferences which serve as the input for those models. In other ways the different approaches are at odds with rational choice, of course, so to some extent we are forced to choose among methodologies.

The second question seems to be whether a rational choice approach can address international law issues that do not focus on compliance with legal obligations or that do not define compliance in the way I do. Jeff asks if rational choice accounts can be build upon conceptions of international law “that view it more as a process of authoritative decision making than as a bundle of rules, that focus more on the evolving trajectory of complex legal regimes than on the ratification of treaty norms.”

Rational choice, because it has an underlying theory and because it is not committed to inflexible doctrinal categories (e.g., hard law v. soft law), offers a way to think through problems and questions fairly systematically. So, for example, if one thinks of international law as a process by which decisions are made, a rational choice model insists that you make some assumptions about how that process works. Once you specify the decisions that interest you, and how they are made a rational choice approach gives you the tools to move from those assumptions to some conclusions. Other approaches might be able to do the same thing, of course. I find reasoning about international law problematic, regardless of the methodology, when it fails to identify its assumptions and does not tell the reader how it gets to its conclusions.

I do not know of any way to think about the “evolving trajectory of complex legal regimes” without a much stronger sense of what causes the regime to evolve, what are its component parts, and so on. The main problem with complexity, I like to say, is that it is so darn complicated. Rational choice models sometimes do a poor job with complexity but this is not because they feature rationality, but rather because they are models. The human brain does not handle complexity well, and so we use models, or simplifications, or maps, even when we do not do so consciously. Rational choice is a way of bringing order to our thinking because it offers some simple and plausible, if imperfect, assumptions about behavior. The resulting conclusions will be mere approximations of reality, but that is the inevitable result of any modeling process. Complaints about the fact that the models do not yield perfect representations of reality (to be clear, Jeff is not making such complaints in his comment) are misguided. We have no choice but to simplify. The question is how we will do it.

A Parsimonious Approach to (How) International Law (Works)

by Jeffrey Dunoff

How International Law Works [HILW] is a terrific book. For current purposes, I am less interested in the specifics of Andrew’s arguments than I am in his larger project – the explanation of international law in rational choice/game theoretic terms.

HILW employs standard rational choice assumptions, for example, that states “are rational, self-interested, and able to identify and pursue their interests.” The book’s goal is to explore, in light of these assumptions, “how international legal obligations might influence state behavior.” Much of HILW’s analytical and rhetorical strength lies in its relatively parsimonious model of state preferences, interests and behavior.

My interest is in whether there is a connection between the parsimony of these rational choice assumptions and the relatively parsimonious vision of international law that emerges in HILW? Consider, for example, HILW’s the discussion of treaties. HILW tells us that “states make commitments of their own in order to extract commitments from other states.” We might ask if rational choice approaches compel this understanding of the purpose of treaties — and whether this account of states’ motivations for entering into treaties is incomplete. For example, states sometimes enter into treaties to lock in domestic reforms. Many argue that the Salinas government joined NAFTA to prevent future Mexican governments from adopting policies hostile to foreign investment. Other times, states enter into treaties for reasons that have little to do with the treaty’s underlying subject matter. Consider the Convention on Long-Range Transboundary Air Pollution; most states who signed this treaty had little interest in acid rain but viewed the treaty as a vehicle to promote East-West detente. Finally, states may enter into treaties for expressive reasons or to embarrass other states – think of the Apartheid Convention – and may harbor few illusions that the treaty will actually change state behavior.

Similarly, HILW is centrally concerned with “compliance,” which the book seems to take as referring straightforwardly to behavior in conformity with treaty obligations. But as Kal Raustiala, Ben Kingsbury and others have persuasively argued, the concept of compliance is hardly straightforward. Rather, the significance of behavior in conformity with rules will vary significantly depending on how one understands the nature, purpose and functions of “international law.” Traditional positivist accounts are rule based. But there are competing process-based views, including the New Haven School’s vision of law as a process of authoritative and controlling decision making, and recent accounts that understand law in terms of a transnational legal process. Other approaches would include liberal accounts, which focus on the compliance as resulting from particular constellations of domestic political forces, and competing constructivist accounts. The point is that any particular conception of compliance presupposes a contested and controversial understanding of international law.

So I think the larger methodological issues HILW raises relate to the relationship between any particular understanding of international law and efforts to construct game-theoretic accounts of how international law works. Can rational choice shed light on which of these competing theories of international law to embrace? More importantly, does the parsimonious rational choice account of state behavior drive the book’s conception of international law — or does the parsimonious conception of international law drive the game-theoretic analysis that follows? In other words, might the very definitions of relevant issues, actors, and preferences in rational choice accounts depend on an unacknowledged (and controversial) conception of the preexisting international order? HILW’s focus on, say, treaties rather than evolving international regimes; on sovereign states to the virtual exclusion of other relevant international legal actors; and on compliance as conformity with rules all presuppose a particular—and controversial— vision of the international legal order.

In short, can rational choice accounts be built upon conceptions of international law that view it more as a process of authoritative decision making than as a bundle of rules, that focus more on the evolving trajectory of complex legal regimes than on the ratification of treaty norms, and that understands compliance not simply as a matter of rational calculation but also as a product of internalized identities and norms of appropriate behavior. If the answer to this inquiry is “no,” than we can understand HILW as an important illustration of both the promise – and the limits – of rational choice explanations of how international law works.

More on Human Rights and Rational Choice

by Andrew Guzman

I think the discussion of rational choice explanations for human rights treaties has become a little muddled due to a lack of clarity in terminology. In particular, two terms need to be clarified. First, a “rational choice” approach implies an approach in which states have preferences which are complete, transitive, and stable. In my book and in almost all rational choice discussions, there is an additional assumption that states are unitary actors. This last assumption is sometime relaxed in an ad hoc way to explain some particular feature of the world that is otherwise difficult to get at. Like many others, I do this from time to time in my book.

Roger’s last comment on human treaties looks to a domestic law explanation for American participation in the CEDR. It is appropriate to call this a rational choice explanation, and it may well be the best explanation we have. The problem for creating a theory of international law is that this domestic influence does not generalize easily. To be sure one can observe that domestic politics matters and that virtually anything states do is affected by them. This is close to what liberal theories attempt to do – they look to domestic interest groups and politics to explain events. Ideally, we would have a model of domestic politics that would explain the preferences of states at the international level. Unfortunately we do not have a strong enough theory of domestic politics to go much beyond the statement that they matter. We have only a very crude understanding of when one domestic view will trump over another.

The other term I want to mention is “internalization.” When Alex used this term I took him to be referring to preferences formation, akin to the argument Harold Koh has advanced for why international law is complied with, the work of Goodman and Jinks, and the views of constructivists in political science. If preferences can change they lose their transitive property, and so are no longer rational. I discuss constructivism briefly in the introduction to my book, and in the interests of space will not repeat myself here. It is enough to say that constructivism is a difference approach that needs to elaborated and evaluated on its own terms. We cannot do that in a serious way here.

When Roger uses the phrase “internalization of a norm” he is thinking of the impact of domestic politics. When I use it I mean that domestic preferences can be changed by international law (or international norms). I do not think we disagree very much beyond this semantic difference.

Finally, I should add that my preference, as evidence in the book (and elaborated in the first chapter), is to stick with the assumption of a unitary state. To be sure, this reduces the flexibility of the model, but that is a virtue, not a vice. If domestic concerns are allowed in, almost anything can be explained with a wave to interest group politics. To think seriously about domestic politics, then, requires a careful inquiry into specific events; a practice best suited to positive explanations of individual past events rather than general models of behavior.

Rational Choice Explanations for Human Rights Treaties

by Roger Alford

Let me just offer a quick additional reflection on the question of whether rational choice theory may help explain the conundrum of why states sign human rights treaties.

The easiest explanation is when the human rights commitment is bundled together with other provisions in a treaty, and the cost of making the human rights commitment is offset by other benefits derived from the treaty. The Helsinki Accords are the obvious example, with the USSR receiving significant benefits from provisions such as the territorial integrity of States, while committing itself to respect for human rights and fundamental freedoms. The rational choice explanation is obvious within the confines of the treaty itself.

Most human rights treaties are not of this nature. As for free-standing human rights treaties, a rational choice assumption would be that the cost of adhering to a human rights treaty must be offset by some greater good external to the treaty. Two possibilities come to mind, one international and the other domestic. In some cases, a country may wish to sign a human rights treaty to procure some greater international benefit. For example, the choice of some eastern European countries to sign the European Convention on Human Rights can be explained by their desire to secure admission to the European Union. And of course, foreign aid is often conditioned on adherence to human rights treaties. It is quite rational for developing countries to make human rights commitments for the sake procuring foreign aid.

In other cases the benefit may have nothing to do with international relations. If I understand Andrew correctly from his last post, he assumes that the internalization of a norm is a departure from rational choice assumptions. I’m not sure this is correct. I would think that a state could make the rational choice to suffer the cost of adhering to a human rights treaty in order to secure a domestic benefit. The 1965 Convention on the Elimination of All Forms of Racial Discrimination is a good example. CEDR was pushed through the General Assembly by a majority bloc of newly-independent developing countries from Africa and Asia. But there is a rational choice explanation for why the United States quickly came on board: the international cost of adherence was offset by a domestic benefit. The Johnson Administration was under intense domestic pressure from the civil rights movement to show progress on racial equality, and adherence to an international treaty was one such clear signal.

Rational choice explanations of international law cannot divorce international costs from domestic benefits. All that matters is that there is some rational explanation for why states make international commitments. The explanation may be of a horizontal nature between nations, or of a vertical nature within the state itself.

Formation of Human Rights Treaties, A Response to Alex Geisinger

by Andrew Guzman

Alex Geisinger’s thoughtful comments return to some of the themes I touched on in my earlier response to Roger Alford. Alex is more forceful, however, challenging whether the theory I advance is able to account for the role of reputation in treaty formation.

I think the place to begin here is some terminology. The use of the term reputation invites confusion, and there may be some of that confusion going on here. In the book (and here) I use the term to refer to a reputation for compliance with international commitments. That is, the project takes as its focus the ability of states to make credible promises. Just as private parties find contracting valuable, states find it valuable to exchange promises. So the main thrust of the book is directed at the question of when those promises are credible.

Roger and Alex both pull at a different notion of reputation – one for cooperation more generally, or perhaps for participation in international human rights treaties. It is clear that there is no legal obligation to join a human rights treaty, but there may be political pressure or some other incentive at work. Alex seems to be on board with the notion that states join such treaties in order to advance their own interests, stating that states seek “material benefits somewhere down the line.” I agree. One reason I find human rights treaties so puzzling is that it is devilishly hard to figure out just what it is signatories think they are getting. Alex suggests some possibilities – to obtain aid, to signal a willingness to cooperation, to signal that membership in a group is valuable, to signal a low discount rate. We can certainly identify instances in which membership comes with some clear benefit of aid or other support, but there are many examples where that is not true. The problem with signaling stories was already discussed in my response to Roger: the signal must be costly and if you subsequently violate the agreement, the net impact on whatever it is you are trying to signal will be negative.

Alex mentions one additional possibility – the internalization of a norm. This is a departure from rational choice assumptions, and for that reason is not discussed in detail in the book. In some instances some sort of norm internalization (or, in economist-speak, a changing of preferences) seems a plausible explanation for some countries. It is hard to imagine anyone signing the CEDAW 50 years ago, for example. The norm internalization story, however, does not help us to understand why states sign treaties that they are almost certain to violate. If the relevant norm were internalized, would it not manifest itself in domestic policy too? It seems to me that not only is it hard to understand human rights treaties in a rational choice model, it is hard to explain them using any methodology.

Turning back to Alex’ question – can a rational choice theory account for treaty formation. I am confident that the answer is “yes” in the sense that thinking of treaty formation in rational choice terms makes sense to me. Indeed, that is how Roger, Alex, and I have been discussing it here. A slightly different question is whether I have addressed treaty formation in the book. To that question I would answer “sort of.” As the last paragraph in my response to Roger indicates, the book does explore some of the relevant issues, but not all of them. There is much more to be said on this issue, and it strikes me as important for our understanding of international law and international relations. And if someone has a good explanation for why states sign human rights treaties, I invite them let me know and we can co-author the paper on it!