Political scientists have finally begun to take international law seriously. Todd Landman's new book is an admirable example of this new trend. Situated at the intersection of international relations and comparative politics, it is a sustained study of the role of democratization, development, international socialization and international law on governments' human rights practices since the mid-1980s. It is a global quantitative study that comes to some very different conclusions from the existing literature in this genre. In contrast to the pathbreaking work of Cona Hathaway (“Do Human Rights Treaties Make a Difference?” Yale Law Journal 111 [8, 2002]: 101–99) and Linda Camp Keith (“The United Nations International Covenant on Civil and Political Rights: Does It Make a Difference in Human Rights Behavior?” Journal of Peace Research 36 [no. 1, 1999]: 95–118), the key finding in Landman's study is that international treaty law has contributed to improvements in human rights behavior. How does he come to this unorthodox conclusion, and what does this mean for our understanding of international law and democratic development?

The answer to the first question is slowly, and by way of empirical rather than theoretical argumentation. One of the strengths of the book is that readers are afforded a good long descriptive look at the data on rights practices and treaty ratification from many different angles. Ratification rates for each of the six core human rights treaties are graphed over time and by region. Their variance is examined. Unweighted (a three-point scale representing no action, signature, ratification) and weighted (multiplied by the extent of reservations added) measures are compared graphically and in tables. A similar approach is taken with the key dependent variable, for which Landman has five different measures (the political terror scale based on Amnesty International reports and U.S. State Department reports; Freedom House's measure of political rights and civil rights; and Hathaway's torture scale). Bivariate relationships are exhaustively explored. For those whose tastes tend toward visual displays of descriptive data, Chapters 4–6 are a feast. But the main findings are simple: Democracies ratify these agreements more readily than do autocracies, and third- and fourth-wave democracies ratify more readily with fewer reservations than do the established democracies. Unfortunately (though predictably), the third- and fourth-waves' actual rights practices tend to be far worse.

Landman's is one of the few accounts of treaty commitments that takes reservation making into account. Reservations—entered when governments ratify a treaty—are a way to “customize” an obligation to reflect the preferences of the ratifying actor. Many people think, as the author apparently does, that reservations reflect a lower level of commitment to a treaty and an attenuated intention to comply. Fully half of the scores of models in the book compare ratification with a ratification discounted by the extent of reservations. In this reviewer's view, this is an overly formalistic account of governments' intent. Reservations have practically nothing to do with intent to comply; in fact, the relationship might well be the opposite of that offered by Landman. For example, 15 African countries have signed all six core human rights treaties, but have never entered reservations or made declarations. They include Benin, Burkina Faso, Burundi, Cameroon, Cape Verde, Chad, Côte d'Ivoire, Gabon, Malawi, Namibia, Nigeria, Senegal, Seychelles, Sierra Leone, and Togo. In Central and East Asia the list of nonreserving treaty members continues with a variety of states that are not obvious nominees for the honor of most intensely committed to international human rights law: Armenia, Azerbaijan, Kyrgyz Republic, Tajikistan, Turkmenistan, Uzbekistan, Cambodia, and the Philippines. Similarly, in Europe, the list features Albania, Estonia, Georgia, Latvia, Lithuania, Moldova, Portugal, Yugoslavia, and Macedonia. In Latin America, Bolivia, Colombia, Costa Rica, Honduras, Peru, St. Vincent, and Uruguay round out the list of nonreserving six-treaty members. In fact, two of the strongest predictors of reservation making are a high commitment to the rule of law and GDP per capita. Reservations have much to do with legal capacity and the likelihood that the treaty will be enforced in domestic law.

It takes a long time to get to the controlled models that are standard in the social sciences. But they are important contributions to our understanding. Landman's general strategy is to use time-series cross-sectional analyses, examining first the ratification decision and then the rights indicators as dependent variables, followed by a two-stage estimation that endogenizes treaty ratification with “instrumental” variables to explain rights practices. In addition to treaty commitments, five types of explanatory variables are tested: democracy, wealth, interdependence, conflict, population, and three regional dummy variables. Tables 6.2 and 6.3 (pp. 138–41) contain the key results: The more committed a country is to a key rights treaty, the stronger the improvement in rights practices.

Since the contribution of the book is this particular empirical result, it is useful to take a close look at the model that produces it. Landman uses an “instrumental” variable to endogenize treaty commitments, but he runs into the same problems most people do who take this approach: He does not really have a valid instrument for ratification. Ratification is modeled as a function of democracy, wealth, and membership in intergovernmental organizations and the presence (not well defined) of nongovernmental organizations (p. 136). While these are correlated with ratification, unfortunately none of these can remotely be thought of as affecting human rights practices only through their relationship with ratification; quite the contrary, according to Chapter 4. I am not aware of anyone who has yet published a credible IV model of treaty ratification and compliance, and so Landman's effort should be considered progress even if he has not solved the problem of endogeneity.

Curiously for a comparativist, there is very little comparative politics in the author's account of treaty ratification and human rights behavior. There is but one domestic variable in his many specifications of rights norms and practices: democracy. He uses the polity scale as well as a categorization of regimes (“old,” “second,” “third,” and “fourth” wave) to stand in for domestic institutions. There are no varieties of autocracies. The institutions of governance among democracies are assumed not to be important, despite the fact that ratification procedures differ in important ways, federal structures may complicate ratification, and changes in governmental partisanship might have an important bearing on rights practices. Nevertheless, were these factors to be added to his quantitative analysis, they probably would not modify his central finding about democracy, which many scholars have found indeed to be robust. Whether the more controversial finding that treaties account for human rights behavior would stand up to a better specified model of domestic politics is less certain.

This underspecification of the political and/or sociological processes associated with the embrace of international law and good rights performance is related to my central concern with the book as a whole. Landman takes no clear position on why certain states make the counterintuitive move to commit themselves and ultimately to change their behaviors based on legal commitments. Do treaties function as screens, such that only the best rights practitioners are willing to sign? Such a view might be informed by theories of credible commitments with steep ex ante costs, but the author does not draw on this literature specifically. Are treaties effective constraints ex post? If so, how do commitments on paper become politically relevant to government actors? Or are treaties a manifestation of “world culture” that government officials imbibe and mimic as a result of their socialization through intergovernmental organizations? Landman has done an excellent job of undercutting a constant refrain of political realists by showing that the international human rights regime is relevant to real politics. Now we need to know why.