Constructing a straw person? The risk of misrepresenting normative IR theory

Normative IR theory, the body of scholarship that brings together insights from political theory, moral philosophy, and IR to explore moral expectations, decisions, and dilemmas in world politics, is not a central player in Moral Limit and Possibility.Footnote ¹⁰ The volume is, rather, about the potential contributions to ethical theorizing of IR's constructivism as contemplated by a particular group of scholars who identify with this label. Nevertheless, normative IR theory is the subject of an intermittent and often revealing refrain. The lines of this refrain are something like the following: normative IR theory is necessarily and exclusively prescriptive (in the sense that its purpose is to offer ‘ought’ rather than ‘is’ statements); it defines ethical enquiry extremely narrowly as (in Reus-Smit's words) the ‘logical deduction of ethical principles’ or the ‘reasoned formulation of precepts governing right conduct’ (Reus-Smit Reference Reus-Smit2008, 65);Footnote ¹¹ relatedly, it eschews, or seriously neglects, empirical engagement;Footnote ¹² and, it relies on an impartialist moral starting point to address global questions, thereby precluding recognition of the social construction of agency and identity.Footnote ¹³ Criticism of work done within normative IR theory, particularly from other areas of IR, is important and to be welcomed. However, this refrain – whether repeated in whole or in part by individual contributors, and whether recited explicitly as the reason for a dearth of practical moral reasoning in IR, or hummed more quietly in the background of an account of why it is incumbent upon constructivists to take a ‘realistic’ approach to ethics – hits a few false notes. These, in turn, threaten to drown out potentially valuable harmonies between prominent positions in normative IR theory and the explicit aims set out in Moral Limit and Possibility.

A detailed account of how each line in this refrain is out of tune with a careful and comprehensive account of normative IR theory is attempted elsewhere, and, for reasons of space, cannot be recounted here.Footnote ¹⁴ To summarize, normative IR theory, like constructivism, encompasses a range of positions. Any simplistic opposition of this body of work to ‘empirical’ or ‘explanatory’ theory, or narrow equation of it with ‘prescriptive’ theory, provides an inaccurate, or at least only partial, view of much of the work being done in this sub-field. Moreover, few, if any, positions within normative IR theory remotely resemble the formalism upon which Reus-Smit places so much emphasis in justifying a constructivist correction.Footnote ¹⁵ These points are important. Yet, the clarification most likely to repay close attention in the context of the argument to follow is one that responds to the last line in the volume's apparent refrain; namely, one that counters the depiction of normative IR theory as inextricably linked to an ‘impartialist’ ethical perspective.

An impartialist moral starting point is simply one that abstracts from particular social, historical, cultural, and political contexts. Proponents of such a perspective maintain that this abstraction avoids prejudice, grants everyone equal moral standing, and is therefore fundamental to addressing transnational ethical questions. Critics protest that such inclusion is purchased at the high cost of denying the significance of the particular communities, relationships, and practices in which moral agents are embedded.Footnote ¹⁶ Price joins the chorus of critics. He laments the apparent disregard for the social construction of identity found in approaches to international ethics that invoke ideas like Kant's ‘categorical imperative’ and Rawls’ ‘original position’ (Price Reference Price2008a, 10–11, 32). Regardless of which side of this debate one comes down on (and acknowledging that some philosophers would respond to their critics by maintaining that impartialist positions can leave room for context), Price is exactly right that impartialist ethical perspectives collide with constructivism's compelling account of identity and agency as socially constituted. After all, such perspectives demand that we bracket precisely what constructivists argue to be defining.

One strain within normative IR theory is best described as impartialist – including the work of consequentialists and those who defend variations on Kantian ethics, such as the group of liberal cosmopolitans influenced by the work of Rawls. Yet, the vital point that is all but overlooked in Moral Limit and Possibility is that much work done within normative IR theory not only relies on a distinct set of influences, but takes great pains to challenge, reject, and identify alternatives to exactly those features of impartialist positions that seem to hide from view the particular ties, loyalties, contexts, and circumstances that shape and color moral agency and judgment in world politics. Promisingly, at one point in his critique of impartialist ethical approaches to world politics (which he argues ‘have tended to black-box sociological descriptions or explanations of identify formation’), Price notes that ‘[c]ommunitarian and identity-based political theory is conspicuously different on this score, of course’. Yet, he quickly adds that, ‘such projects tend to struggle with the difficulty of how to deal with transcommunity morality such as international norms’, and thereby dismisses them as unworthy of further pursuit if one is concerned specifically with international ethics (Price Reference Price2008a, 32–33). This is unfortunate. A prominent and diverse body of work in normative IR theory not only examines these perspectives, but also poses just those questions (incisively articulated by Price) of how to reconcile their anti-impartialist commitments with the need to address practical problems at the international, or global, level.Footnote ¹⁷

My objection to such aspects of normative IR theory being overlooked is not merely the pedantic preoccupation of someone defending her own turf (although perhaps this plays a part). Rather, ignoring this common ground leaves valuable contributions in the volume vulnerable to the charge that they are building a straw person out of an existing body of scholarship in order to make an argument for establishing a more robust alternative. Moreover, and even more importantly, this neglect risks forfeiting a significant step towards the sort of intra-disciplinary, cross-‘camp’ collaboration that might contribute to meeting the particular challenge to constructivism posed by Price.

Possibility and limits: the difficult leap from descriptive to evaluative and prescriptive accounts of moral norms

The challenge to constructivism in Moral Limit and Possibility, as it is variously set by Price and interpreted by his respondents, has three prongs. First, contributors are encouraged to articulate and explore the latent ethical assumptions that motivate and inform their work. Second, and more provocatively, the participants in Price's project are prompted to make a substantive contribution to ethical theorizing about world politics by considering how their constructivist positions help us to answer the pressing question of ‘what ought we to do?’. This prong is sharper; the challenge has the potential to cut deeper into alleged shortcomings of constructivist scholarship. Indeed, a sceptic might warn that this challenge carries some risk for those who would accept it as it prods constructivists to do what is arguably precluded by their own intellectual commitments: engage in prescriptive ethical arguments. Finally, and relatedly, contributors are asked to provide some basis for presuming that the moral change meticulously uncovered in their empirical research is good, right, just, or otherwise ethically desirable in a way that lends support to their equating it with moral progress. In other words, if one of the contributions claimed by constructivists has been to demonstrate the possibility of progressive moral change in world politics, then it is important to be clear about the evaluative criteria that make such assessments possible.

The first challenge is the least onerous. We are all influenced by unacknowledged and unexamined assumptions and value judgments – regardless of attempts to avoid having them bias our work, and however successful we are in such endeavors. Interrogating these can be revealing. The subsequent two challenges are, however, much more difficult to meet. Whereas the constructivist IR research represented in this volume performs the important task of describing and explaining moral norms in world politics, it appears to lack the tools to champion, challenge, or offer alternatives to them. In other words, this constructivism is able to say that norms matter in a way that is extremely valuable; however, it is less equipped to say why certain norms are more or less just or ethical than others, or why their emergence or transformation constitutes moral progress or regress.

Indeed, constructivism's evaluative limits become particularly apparent in Moral Limit and Possibility with respect to discussions of progress. The whole question of whether or not constructivists are able to evaluate moral progress is one with which Price opens the volume. His framing of the conundrum faced by constructivists – who empirically document change in moral norms and presume some change to be ethically desirable, but do not examine the basis on which such presumptions can be made – is excellent. However, his casual quip that ‘the challenge of having to offer a convincing defence of the ethical desirability of norms like the abolition of slavery or torture would not exactly keep too many constructivist scholars up at night’ effectively foreshadows the volume's general lack of a convincing, affirmative response (Price Reference Price2008a, 3). If anything is to trigger insomnia among IR scholars concerned with confronting ethical questions in world politics, the task of providing such a defence is a prime candidate. Not only is the violation of such norms deeply consequential, but we have no reason to assume that their ethical desirability is immune to scrutiny and scepticism or shielded from blatant challenges. This vulnerability is demonstrated by the serious debates among both academics and political advisors over the moral and legal acceptability of torture during the so-called ‘war on terror’. We can neither take the evaluative criteria underlying our claims to moral progress for granted, nor talk about moral progress if we abandon such measures altogether.

I am not suggesting that there should, or can, be one single constructivist position on what would constitute these evaluative criteria. (In fact, I argue below that there cannot be.) Yet, unless constructivists are willing to either deny the existence of such criteria and succumb to charges of relativism, or maintain that such criteria are unknowable to us and revert to agnosticism about issues of moral progress (and speak instead of moral change), they are left with two options in response to this aspect of Price's challenge. This first option is for constructivist scholars to simply invoke a set of moral standards with which to start and concede that defending these standards has no place in their work, thereby implying that the defence of the grounds on which their assessments of moral norms might be made is best left to someone else (and raising important questions about scholarly divisions of labor). The second option is for them, instead, to defend a specific evaluative criterion for judging the moral norms that they document, and take seriously the problem of how to ground such an ethical standard in a way that is compatible with constructivism's defining theoretical assumptions. The first option is implicitly chosen by the contributors to this volume; the second is not acknowledged, but can be teased out by further exploring points of contact between specific constructivist assumptions and work within normative IR theory.

Option 1: ‘Outsourcing’ ethics

Kathryn Sikkink's honest (and refreshingly direct) reply to Price's invitation to both define moral progress and be explicit about how standards for assessing it are derived provides more than one variation on the first option, even while being drawn reluctantly towards the second. Sikkink convincingly suggests that constructivists can contribute to understanding the consequences of practices and policies through empirical research and then link these insights to moral judgments (Sikkink Reference Sikkink2008, 86). Of course, this exercise requires some frame of reference for assessing better or worse consequences. Sikkink recognizes this: ‘[i]n order to weigh consequences, we must first have specified what principles we intend to use to evaluate which consequences are most valued and beneficial’. She proposes ‘to start with international human rights principles’ (Sikkink Reference Sikkink2008, 86).Footnote ¹⁸ Sikkink then goes on to define progress as ‘the improvement in the enjoyment of any of the human rights listed in international human rights law as compared to an earlier period if such improvement does not cause a commensurate regress in other places’ (Sikkink Reference Sikkink2008, 89). At first glance, this definition has an inherent conservative quality, as it seems simply to accept existing institutionalized standards and treat them as both starting points and evaluative criteria for making ethical judgments – a move that would make any critical engagement with these norms impossible. Sikkink identifies an important role for empirical research, but in the absence of some ethical framework or set of assumptions upon which to piggyback, which would establish prevailing human rights as a good, such empirical research can only document moral change rather than assess moral progress.

However, it becomes clear that Sikkink is not simply invoking existing moral norms as ethical standards for judgment. In addition to starting with human rights principles, she is also starting with an assumption of their value in order to link her evidence of change and accounts of consequence to moral judgments. Sikkink is content to leave the defence of the moral value of these principles to others, variously borrowing Rawls’ notion of an ‘overlapping political consensus’ (which she invokes via Jack Donnelly) and giving qualified praise to the work on rights and capabilities of Amartya Sen and Martha Nussbaum. To be clear, I do not think that there is anything wrong with starting with criteria for evaluating progress that rely on someone else's philosophical justifications (just as I think that it makes eminent sense for normative IR theorists engaged in defending such criteria to draw on the empirical work of others in making practical judgments about world politics). It is, however, notable, given Price's request to reveal the nature and source of these criteria, that Sikkink implicitly opts for a strict ‘division of labor’ response and effectively outsources this particular aspect of ethical reasoning to those committed to more philosophical approaches. Her constructivist analysis contributes the empirical research necessary to arrive at certain moral judgments about particular cases, but it does not provide the ethical framework requisite to forming these judgments.

Interestingly, Sikkink confesses to being ‘always tempted to make more foundational claims for the [human rights] norms’. ‘I believe’, she reveals, ‘that nothing does greater or more long-lasting harm to people than when other people intentionally and directly inflict bodily harm on them…’ (Sikkink Reference Sikkink2008, 89). Yet, she emphasizes that her argument ‘does not rest, nor depend, on such foundational claims’ – by which I understand her to mean that it is not within her remit to defend these sorts of foundational claims, and that her argument does not rely on these particular claims that she finds so intuitively appealing (Sikkink Reference Sikkink2008, 90). Whether we look to her commitment to an international overlapping consensus in support of the value of the existing list of human rights encoded in international law, or to her desire for ‘some basic deontological principles so that we don't have to reopen the ethical debate about consequences at every point’, it is abundantly clear that her assessment of moral progress does rely on foundational claims (Sikkink Reference Sikkink2008, 103). Yet, Sikkink seems convinced that defending her own evaluative criterion for either calibrating moral progress or prescribing particular norms is not a valid, or perhaps viable (she cites lack of philosophical training), endeavor in the context of her own research. Moreover, this sentiment is reflected by other contributors to the volume, who, although less forthcoming than Sikkink in providing direct responses to Price's questions about progress, also seem to accept a strict division of labor when it comes to establishing the basis upon which moral judgments in world politics are made.

There is good reason for this. Social constructivism, as a school of thought and an important theoretical approach to world politics in the discipline of IR, cannot yield a single, substantive set of ethical prescriptions or evaluative statements such as ‘we have a duty to respect and promote basic human rights’, ‘slavery (in all its forms) should be abolished’, and ‘torture is immoral’. Individual constructivist scholars might, of course, subscribe to ethical perspectives from which these commitments follow, and such commitments might then inform their work (by defining what they see as the most pressing research problems, or providing an intuitive sense of what constitutes moral progress). However, such prescriptions and evaluative statements do not somehow flow from constructivism's underlying assumptions. Many constructivists (such as those in Moral Limit and Possibility) happen to adhere to worldviews that champion the equal moral worth, and inalienable rights, of all human beings. However, constructivists could just as easily subscribe to ethical perspectives that are chauvinistic, parochial, and illiberal, and such scholars could then be moved to document the emergence and evolution of moral norms that best suit and reflect such sentiments. What is crucial is that a yardstick for evaluating particular moral norms as just or unjust, progressive or regressive, moral or immoral, is not somehow inherent in constructivism. Constructivists can (and, as this volume colorfully illustrates, do) appeal to substantive ethical positions as starting points in their empirical analyses. By combining these positions with detailed studies of moral norms, in contexts ranging from sanctions and humanitarian intervention to immigration and the self-determination of people, they make profoundly important contributions to ethical theorizing about world politics.Footnote ¹⁹ To use Sikkink's own formulation, ‘[j]udgement is a result of a combination of the premises and commitments that we begin with and the empirical research results about the consequences of action (Sikkink Reference Sikkink2008, 110)’.Footnote ²⁰ But from where does one adopt these premises and commitments?

With this query in mind, one might return to Reus-Smit's characterization of the nature of ethical reasoning in normative IR theory. As already suggested, defending principles against which moral judgments can be made is not by any means all that normative IR theorists do. We are just as concerned as our constructivist counterparts with problems of, inter alia, studying the consequences of action or inaction, identifying relevant moral agents, and examining their particular contexts and capacities – despite Reus-Smit's puzzling suggestion that such pursuits comprise a uniquely constructivist contribution to ethical reasoning when set against normative IR theory's single-minded aim to logically deduce principles of conduct (Reus-Smit Reference Reus-Smit2008, 67–70, 75). We simply approach these problems differently. Yet, somebody needs to defend evaluative criteria (however grounded) for moral judgment if there is any hope of the happy marriage between the ethical and the empirical in IR envisaged in this volume. Moreover, the commitment of many normative IR theorists to considering just this aspect of ethical reasoning in relation to practical, concrete problems in world politics make them obvious candidates for collaboration.

Important work within normative IR theory has set itself the task not only of identifying and describing moral norms in world politics, but also of identifying and defending criteria against which these principles and practices can be evaluated. Whether such criteria are defined by the ‘shared understandings’ of a particular community, an international ‘overlapping consensus’, or some standard external to and independent of the values and mores of any specific time and place, their defence is far from straightforward – and always contested. However, such an endeavor is both consistent with this sub-field's own understanding of its aims and scope and accords with accepted measures of scholarly rigor of those working within it. Normative IR theorists explicitly explore the ethical dimension of world politics, are often uninhibited in making philosophical arguments about how to define the right and the good, and (given methodological commitments that depart from those of mainstream IR) are less concerned to defend the empirical assumptions that form an important part of our work with reference to strict social scientific scholarly standards. (This is very different from Reus-Smit's assertion that we are either careless or willfully manipulative of empirical data, and highlights normative IR theorists’ own ‘division of labor’ assumptions when it comes to ethical and empirical enquiry.) The constructivists represented in Price's volume, by contrast, combine underlying ethical convictions and curiosity with detailed empirical analyses and a perceived imperative to prove their social science credentials, which seems to result in an accompanying tendency to downplay any ethical element of their work. Here, Sikkink's candid remark that when she began working on human rights, ‘the choice of topic alone was a sufficiently normative signal that I felt obliged to spend the rest of my time demonstrating that I was being rigorous in my theory and method’ is particularly poignant (Sikkink Reference Sikkink2008, 83).

There are reasons for inhibitions and points of emphasis in both ostensible camps – many of which are rooted in deeply consequential movements within the discipline, and discrete developments on different sides of the Atlantic. The constructivist contributors to this volume have firm roots in mainstream American IR, where the behavioralist revolution and the commitment to science (variously defined) that endures in its wake have exerted a powerful force on the study of world politics; normative IR theory has flourished among scholars with close ties to the United Kingdom, where philosophy, history, and law, traditionally at the heart of IR, have remained central influences.Footnote ²¹ It would be interesting for both normative IR theorists and these ‘mainstream’ constructivists to reflect on how the values, expectations, and norms within the broader academic contexts in which we are embedded have helped to shape our identities, interests, and behaviors as scholars. Indeed, an important question which Price's volume elicits (even if it does not raise as such) is whether engaging in evaluative and prescriptive analyses is actually precluded by the theoretical underpinnings of the constructivist project (in its different manifestations), or, conversely, simply constrained by self-imposed limitations brought on by disciplinary expectations and the internalized methodological biases of many of those working in this area. If it is the latter, then engaging in explicitly ethical enquiry requires scholars to question the way that mainstream American constructivism has come to define itself – which is exactly what I take Moral Limit and Possibility to be doing.

Option 2: Toward a constructivist ethics?

This line of questioning raises intriguing possibilities. It might be that ethically engaged constructivists have a choice that is not acknowledged in Moral Limit and Possibility, even though the challenges posed by the volume open up space for its serious contemplation. Namely, constructivists might eschew the sort of division of labor option described above and dispense with the notion that any criterion for moral judgment must be external to their own research. The logic here is that if constructivism is neutral when it comes to substantive moral positions (as I have maintained), but nevertheless aspires to the sort of ethical enquiry proposed by Price, then it can do one of two things. It can, as suggested above, start with one of any number of ethical frameworks, imported from elsewhere and presumably suited to the individual researcher's own intuitions. Alternatively (and this is the possibility that is, oddly, unexplored in Moral Limit and Possibility), it can make establishing a set of moral standards against which to judge prevailing norms, prescribe policies and practices, and assess moral progress an aim of its empirical research. This would seem an exciting and potentially groundbreaking reply to Price's challenge. However, even if establishing such evaluative criteria were accepted as a legitimate task for IR constructivists, there appears to be a further hurdle to clear – one that Price identifies early in the volume, but which receives scant attention in subsequent chapters.

Price and others observe at a number of points that the constructivists in this volume embrace a liberal cosmopolitan ethic (Price Reference Price2008a, 3; Reference Price2008b, 290).Footnote ²² However, as already highlighted, Price also alludes to a tension between this liberal cosmopolitan bent and fundamental constructivist assumptions (Price Reference Price2008, 1, 3, 32). Indeed, the way that liberal cosmopolitan positions tend to appeal to impartialist starting points does seem radically at odds with constructivism's own assumption that the identities of actors are defined by the institutionalized norms and values of their social contexts, making political agency (and arguably moral agency) radically situated and socially determined. This apparent obstacle to establishing an internally generated constructivist ethic deserves attention.

Of course, one might suggest that the most sensible way to deal with this proposed hurdle is to walk around it. In other words, perhaps this is not an issue that should concern the contributors to Moral Limit and Possibility (making their general lack of engagement with it unproblematic). Suggesting otherwise assumes that because constructivism as an explanatory approach to world politics appeals to the idea that identity and agency are socially constituted, then constructivism as an ethically evaluative approach must be consistent with this. But need we worry about this sort of consistency? One might object that we should abandon the goal of reconciling the underlying assumptions of constructivism in its guise as an explanatory approach with how constructivists’ chosen evaluative criteria are derived. However, this objection seems compelling only if the constructivist IR scholars in question remain content to adopt from elsewhere standards for ethical analyses, rather than employing the tools of their own theoretical perspective to defend or establish them. Endorsing just this role, one might maintain that constructivism makes a profound, and sufficient, contribution to ethical questions in IR by bringing to the table detailed empirical analyses that variously demonstrate the significance of moral norms in world politics and their scope for change. Such constructivist contributions not only have the potential to work together with (externally conceived) ethical frameworks in order to arrive at moral judgments, but they can also feed back into how these ethical frameworks are refined and elaborated (if those who take on the task of defending them heed constructivist insights). As the contributors to Moral Limit and Possibility illustrate throughout the volume, their work thereby supplements (and by the same logic, is supplemented by) more philosophical analyses – undertaken by normative IR theorists, for example. In sum, walking around this perceived hurdle is a viable option if one accepts the necessity of a strict division of labor when it comes to ethical enquiry into moral norms in world politics – and, specifically, concedes that constructivists must surrender the territory of defending the evaluative criteria required to say whether certain norms are better than others, when change constitutes moral progress, and, by extension, what we ought to do.

It would be easy (and maybe wise) to stop here. However, doing so would seem to accept an ethically attenuated account of constructivism – at least when set against the aspirations that introduce Moral Limit and Possibility. It would also be to give up on what is the most arduous, but also the most intellectually interesting and potentially valuable part of Price's challenge. What if one were to look at alternatives to impartialist cosmopolitan ethics as guides to how constructivists might defend evaluative criteria against which to justify claims to moral progress and offer prescriptions for action? It is possible that greater attention to areas of commonality with (a more nuanced) normative IR theory would reveal the potential for constructivism to assess moral norms, and even advocate their transformation, without either outsourcing the defense of requisite evaluative criteria or contradicting its defining commitment to social and historical context. It is at this point that my initial plea to take a more careful and comprehensive look at normative IR theory may prove to be more than simply inter-camp defensiveness.

As highlighted above, there is a sophisticated body of work within normative IR theory that defends alternatives to impartialist ethical positions – and adapts and refocuses them precisely to issues of transnational morality and norms. Significantly, constructivism's theoretical assumptions have a close affinity with the defining features of one category of these alternatives: so-called ‘communitarian’ positions.Footnote ²³ Important points of contact include the insight that agents are defined by the social contexts in which they are embedded, the portrayal of agents and structures as mutually constituted, the understanding of moral norms as social facts, and an emphasis on intersubjective meanings, linked (for communitarians and many, although perhaps not all, IR constructivists) to an interpretive methodology. The modest suggestion here is that those constructivists inclined to accept the bold challenge set down in Moral Limit and Possibility might do well to look to work in normative IR theory that has explored both the potential and pitfalls of adopting this anti-impartialist moral starting point to judge and prescribe moral norms in world politics.

The possibility of a communitarian-inspired constructivist ethics requires more attention and closer scrutiny than can be afforded here. Nevertheless, it is worth considering a few key characteristics of the constructivist ethics that communitarian thinking would direct us towards. First, such an approach would require one to investigate and interpret (rather than invoke) evaluative standards. The objects of interpretation would be the underlying values and ‘shared understandings’ of the community that one were researching.Footnote ²⁴ In other words, this would entail an empirical study of the broader systems of meaning and value within which particular moral norms are situated, negotiated, and debated. Second, the evaluative standards thereby uncovered could be used to judge prevailing policies and practices, identify and champion plausible alternatives, and make judgments about moral progress and regress.Footnote ²⁵ Significantly, such standards could not be independent of the social contexts within which they would be employed. (A radically situated, constructivist moral starting point could not support such independence; this interpretive approach requires adopting a perspective from within the community in question.) Yet neither need these standards be reduced to the only alternative to impartialist reasoning that Price cites as open to constructivists: ‘standards for evaluation…dependent upon existing moral norms’ (Price Reference Price2008, 3). (Such an alternative would leave constructivists wide open to charges that their evaluative criteria have no critical bite.) Instead, the constructivist ethics being contemplated here would represent a third option – one arguably less vulnerable to charges of both conservatism and relativism. According to a critical stream of communitarianism, one can be simultaneously committed to the social construction of values, agency, and identity, and still challenge the prevailing practices and espoused values within a particular community – by exposing their inconsistency and tension with underlying, even ‘latent’, values and social meanings.Footnote ²⁶ If constructivist scholars were to adopt this means of arriving at internal evaluative criteria, they could arguably lend empirical support and rigor to philosophical reflections on what Michael Walzer describes as the process of distinguishing ‘deep and inclusive accounts of our social life from shallow and partisan accounts’ (Walzer Reference Walzer1983b, 43). Third, and perhaps less appealing to some scholars, the evaluative criteria of a constructivist ethics thus conceived would remain contingent upon the results of such empirical research and vulnerable to change over time.

This final feature is strikingly similar to Price's concluding observation in Moral Limit and Possibility that ‘the ultimate ethical position developed here is contingent, and open to empirical challenge’ – but the reasoning is radically different. Price is referring back to the necessary contingency of Sikkink's adopted consequentialism, according to which ethical prescriptions are necessarily dependent on understandings of outcomes, which are themselves subject to empirical investigation. But, Sikkink's consequentialism is not an inherent feature of constructivism, nor is the evaluative criterion of consequentialism contingent upon empirical findings – only the resulting prescriptions are. The sort of constructivist ethics envisaged here, by contrast, would establish an evaluative criterion itself contingent upon a social reality revealed through empirical research – and subject to incremental change along with shifting underlying mores and values. Depending on the underlying ‘shared understandings’ of the community in question (or the most accurate ‘deep and inclusive account of our social life’) at any particular point in time, this evaluative criterion might mesh effortlessly with the strong liberal cosmopolitan commitments of those academics who happen to be conducting this empirical research. Indeed, the moral yardstick thereby uncovered might support prescriptions and evaluative statements such as ‘we have a duty to respect and promote basic human rights’, ‘slavery should be abolished’, and ‘torture is immoral’. But, then again, it might not. The point is simply that such an approach could rely on no independent, ahistorical, pre-social set of standards against which to make moral judgments – or gauge progress. The important caveat that constructivist scholars who would opt for this route should therefore bear in mind is that they might not arrive at the ethical standards that they want, or already subscribe to, through such a process of ‘deep’ interpretation.

There is no suggestion being made here that such an empirical excavation of evaluative criteria would be at all straightforward. Nor would it necessarily serve the laudable aims of the contributors to Price's volume given the moral predispositions so clearly set out in their respective chapters. (If it did, it could only do so contingently.) Such a project could be critical of the status quo, but critical in the sense of revealing inconsistencies and incoherence between a particular community's policies and espoused standards and its deeper – and even latent – value systems. Many would find the lack of an evaluative criterion external to the particular social context under investigation profoundly worrying. My point is simply that if we take seriously the more demanding version of Price's challenge to IR's empirically minded constructivists (that they defend the evaluative criteria necessary both to judge and to champion alternatives to the moral norms that they currently map), and if we take the underlying assumptions of IR's constructivism to a logical conclusion in the context of such an endeavor (thereby avoiding the tension Price valuably notes between constructivist insights and an impartialist ethical stance), then the critical communitarian perspectives revealed in debates within normative IR theory provide a guide to how a self-sufficient, empirically grounded constructivist ethics is conceivable. Of course, these debates also highlight what might be seen as pitfalls of adopting such an approach. (For example, reliance on such internal standards in multifaceted and rapidly changing societies, where one could never hope to determine their ‘definitive’ interpretation, would create challenging circumstances for moral evaluation – not to mention labor-intensive conditions for the researcher.) For some, such apparent pitfalls will reinforce the appeal of the first, ‘outsourcing ethics’ option implicitly chosen by the contributors to Price's volume and bring us back to important division-of-labor questions. Notably, whichever option appears more viable or appealing, consideration of both is prompted by the sophisticated, probing questions raised in Moral Limit and Possibility, and by a closer look at the relationship between IR's constructivism and normative IR theory offered in response to the volume's explicit call to ‘bridge-building’.