If Nuno Monteiro and Keven Ruby (Reference Monteiro and Ruby2009) are right, International Relations (IR) scholars have for too long been permitted to use the philosophy of science as a lethal weapon in their internecine wars for disciplinary supremacy. Monteiro and Ruby think that this is a mistake, that the philosophy of science has been used foundationally, as if it could secure scientific status for some theories and deny it to others. They are certainly correct that the philosophy of science cannot play this imperial role, and philosophers have long ago abandoned the pretension that their discipline could somehow state the necessary and sufficient conditions for scientific knowledge and thus be legislative for the sciences. The current reigning position among philosophers is rather the opposite, as new forms of naturalism take center stage. As Quine (Reference Quine1969) put it in ‘Epistemology Naturalized’, epistemology and, by analogy, all other philosophical fields simply fall ‘into place as a chapter of psychology and hence of natural science’.Footnote 1 Even among the positivists, the idea of foundations was already suspect, with Neurath’s metaphor of science as a boat that is rebuilt at sea. That philosophy is not legislative for the sciences is hardly news, but not for the reasons that Monteiro and Ruby give. For philosophers, it is hard to understand their claim that all foundations require ‘a leap of faith’. Neither Cartesian certainty nor positivist protocol sentences involve ‘leaps of faith’ in any obvious way; the problem with them is that they do not yield anything as useful as naturalized epistemology.
In fact, naturalized epistemology is only rarely post-foundationalist in more or less the sense that Monteiro and Ruby suggest. The prevailing view in naturalized epistemology, however, is anti-foundationalist, as in Quine’s remark that ‘the only foundation for science is science itself’. The advantage of anti-foundationalism over post-foundationalism is that the anti-foundationalist sees no need for prudence about foundations, as if borrowing foundations might sometimes be fruitful independent of their worth in defining the standards of the discipline. Indeed, Monteiro and Ruby are of the opinion that the standards of scholarship should be defined by the discipline itself. ‘Philosophical foundations add nothing to the checks, balances, and safeguards the discipline has in place to ensure that the knowledge it produces is scientific. Consequently, scientific standards should not come from outside IR itself’ (Monteiro and Ruby, Reference Monteiro and Ruby2009: 37). Post-foundationalists in Monteiro and Ruby’s sense are concerned to take executive authority away from the external sovereign of the philosophy of science and hand it over to the internal sovereign, to the internal standards of the discipline. Exactly where any discipline begins or ends is a sociological and not a normative fact. In this way, post-foundational approaches thus seem to shift questions of science to questions of authority, as when Monteiro and Ruby confidently claim that procedures of the disciplinary community at any given time ‘should enable us to trust the results of that process, even if we cannot guarantee that, at the end of the day, the product of science is itself objective’. Very shaky and utterly ahistorical foundations, indeed.
Concerned with the pernicious effects of the philosophy of science on IR, Monteiro and Ruby are motivated to go post-foundational in order to make room for a variety of methods, theories, and purposes in the discipline, correcting the tendency of foundational approaches that seek to underwrite the ‘unitary character of science, obscuring the diversity of goals, practices and standards that characterizes scientific activity’ (2009: 38). If such diversity is the result of post-foundationalism, then it should only be applauded in the social sciences. Of course, this assumes that the standards and practices of the discipline will produce such diversity, so long as they are post-foundational. While consistent with post-foundationalism, the disciplinary community that decides the standards of relevance and does the checking and balancing may still seek uniformity in ‘standards internal to the discipline, taking into consideration the particularities of the objects of study’ (Monteiro and Ruby, Reference Monteiro and Ruby2009: 38). How does post-foundationalism make it anymore unlikely that these are shared within the discipline? The IR community will then have to relearn the lessons of mixed government. Where has the argument gone wrong?
Later, I argue that Monteiro and Ruby fall back on the traditional positivist distinction between ‘external’ and ‘internal’ questions in the philosophy of science. Here, I want only to show that they focus too narrowly on internal ‘scientific’ questions and on a political account of disciplinary authority. Monteiro and Ruby overlook not only the practical implications of anti-foundational arguments, but also many post-foundational alternatives important to the question of disciplinary relevance such as including practical criteria. In this regard, Monteiro and Ruby argue that ‘policy relevance is an odd requirement for IR scholarship, but a perfectly fine goal’ (2009: 38). In the context of a post-foundational philosophy of science, the oddity here is the puzzling positivist import of this statement, which makes science in general and IR in particular utterly distinct from other human endeavors. It is at the very least an important and unargued for assertion about the aims of science, as well as a commitment to a particular philosophy of science and a rather restricted idea of relevance. It seems to be more consistent with post-foundational methodological and theoretical diversity to hold that practical forms of verification are important in the social sciences. Understanding democratic practices and processes of democratization could very well have as it goal the improvement of democratic practices, where such improvements serve at the same time as the practical verification of the theoretical and explanatory claims of a social science. Monteiro and Ruby claim that such goals are external to the social scientific enterprise and thus not central to the criteria of its evaluation. On what basis? Post-foundationalism is barred from making such claims, as it cannot take on board such a foundational argument as a means to exclude an important alternative view of the practice of IR, that of pragmatism. This exclusion of practical aims from social science is implausible, not simply because they should know enough to be able to make policy recommendations, and also show these problems to be flaws in the very institutions that make policy. Many who study democracy in other social scientific disciplines, for example, do so in order to improve its practice and take the success in doing so as an epistemic achievement. It seems odd that post-foundationalists should relegate such approaches to the external goal of policy relevance.
One of the central ideas of both critical theory and pragmatist theories of knowledge is that epistemic claims are embedded in some practical context that in large part determines the relevant standards of justification and conditions of success.Footnote 2 A truth claim is thus to be judged in the light of its practical consequences. This ‘practical turn’ of epistemology is especially relevant for the social sciences, whose main practical contribution, according to pragmatism, is to supply methods for identifying and solving problems – admitting that there are many ways in which knowledge can work toward these ends.Footnote 3 From this perspective, there are many ways in which knowledge can be practical. Certainly, political science may inform the art of statecraft, and IR could consider itself to be a subset of the practical disciplines governing the relations of statesmen and citizens to each other, both of which aim at practical knowledge, as well as an understanding of the practical knowledge that actors in international setting employ. But, what could other disciplines contribute? The philosophy of social science, considered inconsequential by Monteiro and Ruby, can help us answer this question and also begin to discuss more fruitfully the interplay of practical, methodological, and theoretical questions than simple disciplinary relevance. I shall focus on the social sciences that study democracy, in which the practical claims show the fruitfulness of the explanations.
Improving democratic practice
To illustrate the difference between practical inquiry inspired by anti-foundationalist pragmatism and mere policy recommendations, we might consider the variety of ways in which the problems of democracy has been treated in the social sciences. First, it can be examined objectively, so that the social scientist seeks to explain the operation and impact of various features of democracy, such as the different ways in which political life can be organized, the effects of political parties on legislation, the voting behavior of various groups, and so on. One outcome of such inquiry may be various generalizations about democracy, concerning parliamentary systems, voting schemes, or the tendencies of a democracy to go to war or to prevent famines. Social scientists can also take a practical orientation and adopt goals that are not external to the goals of inquiry, in that they seek not just to explain or interpret what democracy is, but to change it.
Such a practical emphasis of the moral and political significance of realizing democratic ideals produces an equal and opposite reaction – skepticism about whether democracy is indeed such a means and location for realizing the goods at which social scientific reformers aim. Some of these skeptical accounts are themselves practical, even if in a negative sense; they aim at uncovering the obstacles to democracy under current circumstances, such as complexity, mass society, human irrationality, or globalization. They count as practical forms of inquiry because they help to develop or develop themselves alternatives to democratic practice that improve it by changing it. That such accounts are skeptical in a practical sense can be seen in the ways in which they lead to advocating that democracy can be replaced by alternative forms of rule, as Lippmann did in proposing technocracy as a solution to the invincible irrationality of mass rule. By identifying obstacles that democracy must overcome, such critics invite practical reinterpretations of democracy that address such problems, as Dewey did in his analysis of the emergence of new publics. Thus, in a field of inquiry aiming at improving democratic practices, there is an important dialectic between those who would improve democracy by changing it and skeptics who doubt whether such changes are possible.
The social scientific tools for improving democracy are varied: generalizations about features and outcomes of democratic practice, social facts in the pragmatic sense of obstacles and resources, responses to skepticism (or negative facts) about human reasoning, and the analysis of failures of democratic practice. However, no specific methods and theories can lay claim to having a special status in this regard. Rather, it is in the role of critic or reformer that they are put to practical use in the wider context of the purposes of a progressive or ameliorative politics. By whatever methods they employ, the social sciences are practical only if they develop a form of inquiry that shows how it is feasible to realize certain political ideals under contemporary circumstances of politics, as a form of ‘praxeology’ in Andrew Linklater’s (Reference Linklater2006) sense. At the same time, social facts can also outstrip particular institutional realizations of these ideals. Social science thematizes just these problematic features as both obstacles and opportunities for the further development of the democratic ideal, including globalization, war and its tendencies for domination at home and abroad, and the need to make decisions and promote good reasoning under conditions of complexity without the appropriate institutional structure. If it is merely skeptical, social science cannot solve such problems; and, at the same time, if it ignores social facts, it cannot meet the need to improve democratic practice.
Next, consider the usefulness of a similarly practical approach to the benefits of democracy. The difference such an approach could make can be seen by considering the two main social scientific generalizations about the beneficial effects of democracy, both of which concern primarily negative facts. The first is that there has (almost) never been a famine in a democracy; and the second is that democracies have (almost) never gone to war with each other. These facts show, ceteris paribus, that the relative absence of two great causes of human suffering – war and famine – can be tied to the operation of distinctive features of democracy. Without some fine-grained explanation of the mechanisms behind them, there is no reason to believe that these generalizations have always held or will always hold in the future, especially if the causes of famine and war are always changing and sometimes are brought about by democratic institutions themselves. Both generalizations have been hotly disputed, leading their defenders to introduce more and more ceteris paribus clauses to limit their scope. For example, Bruce Russett (Reference Russett1993) has argued that the generalizations have only held since the first half of the twentieth century, given the relative paucity of democratic states before them.Footnote 4 Yet, even with such ceteris paribus clauses, different mechanisms may do the explanatory work in the cases of famine and war.
More than in the case of the democratic peace hypothesis, these facts yield a robust explanation of why ‘there has never been a famine in a functioning multiparty democracy’, so that we may conclude that ‘famines are but one example of the protective reach of democracy’ (Sen, Reference Sen1999: 84). It would be tempting to associate this sort of security with the achievement of various instrumental freedoms or with one’s status as a subject or client of a state or similar institution with an effective and well-funded administration. But, even in the case of the protective function of the state, much more is required of democracy to create (or sustain in a crisis) the conditions of entitlement and accountability, as well as the reflexive capacity to change the normative framework. Once the explanation is put in the normative domain and not merely regarded as a lawlike generalization of dubious value, so is the practical understanding of remedies and solutions.
The practical effects of democracy are not directly tied to more effective administrative institutions or even to the consistent application of the rule of law, both of which democracy may achieve. As Sen (Reference Sen1986: 165–166) notes, there are limits to legality: ‘other relevant factors, for example market forces, can be seen as operating through a system of legal relations (ownership rights, contractual obligations, legal exchanges, etc.). In many cases, the law stands between food availability and food entitlement. Starvation deaths can reflect legality with a vengeance’. In this sense, the presence of famine must also be explained through the operation of social norms conjoined with the lack of effective social freedom of citizens with regard to norms’ content. The deplorable treatment of native populations in famines caused by colonial administrators has often been due to domination, manifested in their lack of substantive freedoms such as free expression or political participation. Thus, famine prevention can be gained through fairly simple democratic mechanisms of accountability such as competitive elections and a free press that distribute effective agency more widely than in their absence. The solution for these ills of democracy is not to discover new and more effective protective mechanisms or robust entitlements, as it is hard for some democracies to produce them. Rather, the solution is, as Sen puts it, ‘better democratic practice’ in which citizens are participants in a common deliberative practice and sufficiently protected and empowered to change the distribution of normative powers and take advantage of improved practices. It is certainly the case for women in many developing states and responsible for the unjust distribution of food within families.
To put it somewhat differently, the issue is not merely to construct a more protective democracy, but to create conditions under which an active citizenry is capable of initiating democratization, that is, using their power to extend the scope of democratic entitlements and to establish new possibilities of creative and empowered participation. Democracy is, according to this view, the project in which citizens (and not just the agents for whom the citizens are principals) exercise those normative and communicative powers that would make for better and more just democratic practice. This kind of enabling condition is essential to the explanation of the role of phenomena produced by democracy that serve as Sen’s explanations: citizens’ powers and entitlements.
The ‘democratic peace hypothesis’ is similar to Sen’s generalization about famines in that fairly minimal democratic conditions figure in the explanation of the absence of certain types of wars. The generalization is, however, more restricted in the case of war than famine. Democracies do go to war against non-democracies, although ‘almost never’ against other democracies. Many explanations have been offered for why this is the case, and many of these do not depend on any transformative effects of democratic institutions other than that they provide channels for influence and the expression of citizens’ rational interests and presume amity among democracies as the basis for trust. Seen in the light of the explanation of the absence of famines, democracy might reasonably be given a similar, more dynamic, and transformative role than is usually offered; by being embedded in democratic institutions, agents acquire the normative role of citizens, and the freedoms and powers that provide means by which to avoid the ills of war.
If this is the explanation of peace, it is important to make clear why war and the preparation for war often have the opposite effects. The institutional capability to wage war increases with the executive and administrative powers of the state, which often bypass democratic mechanisms of deliberation and accountability, and thus, work against democratization (where this is understood precisely as the widening and deepening of the institutional powers of citizens to initiate deliberation and participate effectively in it). At the same time, participating in national self-defense has often been accompanied by the emergence of new rights or their broader attribution to more of the population. Charles Tilly (Reference Tilly1990) has argued that warfare may have historically been an important mechanism for the introduction of social rights, as the state became more and more dependent on the willingness of citizens to accept the obligations of military service. As modern warfare became increasingly lethal and professionalized, however, the institutional powers of the state have outstripped this and other democratic mechanisms. The institutionally embedded normative powers of citizens are no longer sufficient to check the institutional powers of states to initiate wars, and these arrangements have left citizens vulnerable to expanding militarization that has weakened these same entitlements. A new dialectic between the capacities of citizens and the instrumental powers of states has not yet reached any equilibrium, so that there has now emerged a strong negative influence on democratic practices and human rights generally because of the use of state force for the sake of security. Liberal democracies have not only restricted some civil rights, but have become human rights violators, with the use of extralegal detention centers and torture in order to achieve security. As such, they might be said to have become less democratic, certainly in the active sense of creating enabling conditions for the exercise of normative powers.
These remarks indicate that the democratic peace generalization depends on a set of historically specific institutional and normative presuppositions having to do with states as the primary sources of organized political violence. When war is no longer the sole form of political violence, then the significance of the internal democracy of states as a means toward peace is greatly diminished. This is particularly true of the Kantian normative inference that democracies would somehow assure that the political federation of peaceful states is ever expanding. But, once the institutional mechanisms of war-making shift from representative bodies toward much less accountable administrative and executive functions and thus undermine the balance of institutional powers within a democracy, the expansive effect created by democratically organized institutions of domestic politics is less likely. This occurs when security requires limitations of freedoms and entitlements of one’s own citizens.
Beyond these internal effects, security brings to a halt the expansion of the zone of peace among liberal democracies. This means that the borders of the zone of peace will become a source of political conflict with those who are outside it. By this, I mean that various transnational publics are now increasingly aware of the ‘problematic fact’ of the zone of liberal peace and prosperity, and consider it as having inherent and systematic asymmetries. The increased potential for violence from those who are outside the zone of peace requires that democratic states adapt to these new threats to their security, often by restricting the liberties of their citizens and their own commitments to human rights, and thus leads to a tendency for democracies to restrict their own democracy and political inclusion within their own states. In this way, the conditions and institutions that promoted a democratic peace among states now act as part of a new negative feedback mechanism, affecting particularly the liberties and rights that have permitted an active citizenry to possess enormous influence over the use of violence. Instead of democracies making IR among states more peaceable, the new constellation of political violence is potentially making democratic states less democratic and less open to applying their internal standards of human rights and legal due process to those that they deem to be threats to security. Recent events show then that democratic peace depends on a positive feedback relation between the internal structure of states and the international political system.
If the practical import of these new feedback relationships undermines the prospect of expanding peace through a political union of existing democracies, peace and security are no longer reducible to the absence of war. Here, we need to modify some deep assumptions about the proper location for democracy and the exercise of the powers of citizenship, in order to determine what would help democratic states to avoid the problem of the weakening of internal democracy as a means to maintain security. One possibility is that some supranational institutions could exist that would make democratic states more rather than less democratic. In a word, peace requires not democracies, but democratization at positively interacting levels.
Both critical theory and pragmatism suggest that the social scientific study of democracy becomes one aspect of a practical theory or praxeology oriented to improving democratic practice. The central questions for a practically oriented social science of democracy are the following: What available forms of praxis are able to promote the transformations that could lead to new forms of democracy? What sort of practical knowledge is needed to make this possible and how might this knowledge be stabilised in institutionalized forms of democratic inquiry? What are the institutional means available to improve democratic deliberation, especially given the fact that certain contexts clearly promote worse rather than better reasoning and deliberation? One very robust finding is that deliberation within heterogeneous groups is less susceptible to framing effects than deliberation in homogeneous groups. Once again, such a finding has clear practical import for deliberative practice and would be a clear case of the sorts of investigation that IR theory as a practical social science with extradisciplinary aims ought to pursue. Monteiro and Ruby may claim to be pluralists in their philosophy of science; but, they do so only by employing an old positivist distinction between internal and external questions; they are clearly pluralists only about internal questions.
Improving IR as a social science: an external or internal question?
The forms of social science and types of social scientific reasoning discussed here are quite diverse. Together they provide a cumulative argument that what is at stake in such practical social science is not mere policy relevance as opposed to disciplinary relevance. Of course, such a study raises the questions concerning the status of the social scientist in proposing standards. It does so without stipulating what the proper standards or aims ought to be in advance. Of course, Monteiro and Ruby might respond that these considerations, fine as they are as goals in their own right, do not meet the criteria of disciplinary relevance, the key to legitimate knowledge in IR. They have, it seems gone too far, in attempting to achieve the laudable goal of ridding debates in IR of questions imported from the philosophy of science. When Monteiro and Ruby discuss science, they focus entirely on the philosophy of natural science, exactly as their opponents do. Their scant attention to the philosophy of social science has led them, in effect, to reproduce a version of Carnap’s (Reference Carnap1958) distinction between internal and external questions about linguistic frameworks, between questions about the framework itself, which ‘can only be judged as being more or less expedient, fruitful, or conducive to the aim for which it is intended, and those that assume a framework and ask question about entities within it’.
Given this distinction, Monteiro and Ruby are clearly satisfied with the first sort of inquiry as sufficient for disciplinary relevance, while the second sort of inquiry remains only an optional and ultimately external goal. As defined internally by disciplinary relevance, the goals of inquiry have no systematic place on such criteria; practical aims and goals are external questions. This is a huge assumption about the nature of social inquiry, which throws practical and critical social science out with the philosophy of science bath water. Quite apart from the philosophy of social science, the external questions of the aims of any discipline are very often in which many important disciplinary controversies are carried out. My point is to show that both external and internal questions as well as disciplinary and practical criteria are not as distinct as Monteiro and Ruby claim, but rather lie on a continuum. A post-foundational conception of IR ought to take on board the insight of pragmatically oriented social science that the one sort of question cannot be answered without the other. Because Ruby and Monterio make this sort of external and internal distinction typical of the philosophy of science they want to reject, they show why it better for diversity to be anti-foundational rather than post-foundational.
It could very well be that IR scholars adopt my recommendations, as it is ‘to be determined and debated by practitioners’. Exactly how they are to employ ‘the criterion of disciplinary relevance’ remains mysterious. Their own view is pragmatist, when they appeal to Wilfred Sellars’ view that ‘science is rational not because it has a foundation, but because it is a self-correcting enterprise which can put any claim in jeopardy, though not all at once’. In making this deeply pragmatist argument, Sellars is not defending anything like disciplinary relevance, which has often historically contradicted the very idea of ‘self-correcting science’. Despite the fact that Monteiro and Ruby quote Sellars approvingly, the disciplinary criteria they offer often appeal to forms of disciplinary authority that are too narrow to support the practical basis for self-correction.
