CRITICAL evaluations of international relations research increasingly question its applicability to alnalysis of the developing world. Pervasive great power bias and assertions of universal patterns are frequently cited as egregious problems with this body of scholarship.(1) International relations specialists provide the critics with plenty of ammunition. Kenneth Waltz writes that "a general theory of international politics is necessarily based on the great powers," while Jack Levy claims "the balance of power framework and its associated propositions were intended primarily as a theory of Great Power behavior." In a recent controversial book, John Mearsheimer focuses "mainly on great powers... because these states dominate and shape international politics." Vesna Danilovic begins her study of conflict and deterrence with a claim that "International relations cannot be well understood without paying attention to those states capable of making a difference," that is, the great powers.(2) In addition to their self-admitted preoccupation with the great powers, international relations researchers generally ignore the possibility that the relationships they hypothesize about might not apply across all regions of the world. It is extremely uncommon to find either theoretical arguments or research designs incorporating the possibility of regional variation. International relations researchers appear content to assume the patterns they study are global.(3) Great power bias and untested assumptions of the universality of patterns across the international system have resulted in a deep rift between standard international relations research and its critics. Standard researchers have constructed a body of theory about international relations, supported by impressive empirical evidence, at odds with what the critics observe in the developing world. For example, whereas much standard international relations scholarship anticipates that states will either balance against or bandwagon with the source of threats, students of Third World politics observe that developing states rarely ally with anyone, whether in balancing or bandwagoning coalitions.(4) Similarly, standard international relations scholarship suggests the maximization of state power is a central goal of political elites, but there are many examples of developing world political elites actually dismantling the edifices of state power.(5) I believe this gap exists between standard international relations research expectations and developing world reality because the developing world has been largely omitted from standard international relations research. There are a variety of ways in which this exclusion manifests itself. Critics suggest that the context within which developing states confront the world differs from that of developed states. If correct, then international relations theories must stop ignoring distinct contextual differences corresponding to different levels of development. Hypothesized relationships between variables may well be contingent upon the context pertaining to the state(s) in question.(6) But such theory elaboration is warranted only if the critics are correct that the contexts of developing states differ meaningfully from those of developed states. Thus a more fundamental question asks whether the critics are correct. In the next section of this article I offer evidence of important differences across regional groupings of states and thus across levels of development. An immediate statistical objection against such evidence is that cross-regional differences might be the result of cross-regional variation in data availability or data quality rather than of differing contexts. Complicating the situation greatly is the possibility that cross-regional variation in data availability and data quality could be caused by the different contexts within which developing states operate. Consequently, we cannot yet evaluate whether cross-regional variations are the result of reasonably straightforward data-related issues, of contextual differences, or of both. Thus far little has been said specifically about Africa. African international relations are an especially important part of the puzzle, because African international relations constitute the developing world activity most likely to be excluded from international relations research. The difficulty of collecting data., concerns about the quality of data that are collected, and the profound underdevelopment and fragility of African states are mutually reinforcing reasons for Africa's omission from international relations research. My goal is to offer specific steps to improve existing international relations research designs so that Africa and the developing world more generally no longer go missing. Toward that end I begin by describing some evidence suggesting empirical patterns are not uniform globally and that Africa is disproportionately likely to be omitted from standard data-based research. I discuss how Africa's variance from other regions might be due to data availability and/or differences in data quality. I then discuss the four books central to this review, with an eye toward uncovering what about African international relations might distinguish them from developed world international relations. Anticipating that discussion a bit, I suggest that a major distinction between African international relations and those elsewhere is that so many of Africa's states are states in name only-legal entities that have failed to consolidate political power within the territories over which they are the legally recognized authorities. Instead, political power is exercised by a variety of state and nonstate actors in Africa. Because standard international relations research theorizes about and collects data only for official states, much of Africa's international relations are left out. The wide variety of what we might conceive of as "effective political actors" within Africa could account for why Africa differs from great power inspired expectations about international behavior. It could also explain why data on African international relations are disproportionately likely to be missing and, when present, to be of inferior quality. My analysis and review suggest three mutually reinforcing changes in research design that I believe international relations researchers should implement. Throughout this article I focus on Africa, because it is the region most egregiously omitted from international relations research. But the problems associated with omitting Africa are also present when researchers similarly omit other developing areas from their research. Consequently, we likely must implement the same research design changes across the globe so as to include all developing regions.
