Nicholas Southwood

Claims about what justice “requires” and the “requirements” of justice are pervasive in political philosophy. However, there is a highly significant ambiguity in such claims that appears to have gone unnoticed. Such claims may pick out either one of two categorically distinct and noncoextensive kinds of requirement that we call 1) requirements-as-necessary-conditions for justice and 2) requirements-as-demands of justice. This is an especially compelling instance of an ambiguity that John Broome has famously observed in the context of claims about other requirements (notably the requirements of rationality and morality). But it appears to have been overlooked by political philosophers in the case of claims about the requirements of justice. The ambiguity is highly significant inasmuch as failing to notice it is liable to distort our normative thinking about politics and make us vulnerable to certain kinds of normatively consequential errors: both mistakenly drawing inferences about what justice demands of us from claims that certain states or societies are not just; and mistakenly drawing inferences about what states or societies are or would be just from claims that justice does not demand of states or societies that they do certain things. Paying greater attention to the distinction between these two different kinds of requirements and the ways in which they come apart is helpful, not merely in avoiding these distortions and errors, but also in resolving, or at least clarifying, a number of other notoriously murky meta-normative debates, especially various important debates about realism and idealism in political philosophy.

An important theme in recent experimental philosophy is that certain judgements (e.g. our judgements involving intentional action and causation) exhibit a kind of normative penetration whereby, in spite of a not-obviously-normative subject matter, they turn out to be sensitive to, and co-vary with, our normative attitudes in interesting and surprising ways. We present the results of several new experimental studies that suggest that our judgements about feasibility also appear to exhibit this kind of normative penetration in at least some cases; that the best explanation of it involves attributing to subjects a certain kind of extensional error that results from conflating the question at hand with another question that is more salient given their normative attitudes; and that these conclusions have significant, though not straightforward, implications for our understanding of both normative penetration and feasibility, respectively.

It is tempting to think (1) that we may sometimes have hopelessly utopian duties and yet (2) that “ought” implies “can.” How might we square these apparently conflicting claims? A simple solution is to interpret hopelessly utopian duties as duties to "pursue" the achievement of manifestly unattainable outcomes (as opposed to duties to "achieve" the outcomes), thereby promising to vindicate the possibility of such duties in a way that is compatible with “ought” implies “can.” The main challenge for this simple solution is to say what the relevant “duties to pursue” are supposed to involve. We survey several existing candidates (duties to try, duties to approximate, duties to do our part, and dynamic duties) and argue that none of them succeeds in delivering on the promise of the simple solution. We then propose a previously untheorized class of duties that we call "duties to devote ourselves" to achieving an outcome, and argue that such duties provide us with an interpretation of hopelessly utopian duties that is up to the task.

The familiar complaint that some ambitious proposal is infeasible naturally invites the following response: Once upon a time, the abolition of slavery and the enfranchisement of women seemed infeasible, yet these things were actually achieved. Presumably, then, many of those things that seem infeasible in our own time may well be achieved too and, thus, turn out to have been perfectly feasible after all. The Appeal to History, as we call it, is a bad argument. It is not true that if some desirable state of affairs was actually achieved, then it was feasible that it was achieved. “Actual” does not imply “feasible,” as we put it. Here is our objection. “Feasible” implies “not counterfactually fluky.” But “actual” does not imply “not counterfactually fluky.” So, “actual” does not imply “feasible.” While something like the Flukiness Objection is sometimes hinted at in the context of the related literature on abilities, it has not been developed in any detail, and both premises are inadequately motivated. We offer a novel articulation of the Flukiness Objection that is both more precise and better motivated. Our conclusions have important implications, not only for the admissible use of history in normative argument, but also by potentially circumscribing the normative claims that are applicable to us.

The open borders view is frequently dismissed for making infeasible demands. This is a potent strategy. Unlike normative arguments regarding open borders, which tend to be relatively intractable, the charge of infeasibility is supposed to operate as what we call a "normative argument-stopper." Nonetheless, we argue that the strategy fails. Bringing about open borders is perfectly feasible on the most plausible account of feasibility. We consider and reject what we take to be the only three credible ways to save the charge of infeasibility: by proposing an alternative account of feasibility; by proposing an alternative, more circumscribed interpretation of the subject-matter of feasibility claims; and by proposing a more expansive account of the addressees of the demand for open borders. The first fails to vindicate the claim that infeasibility is a normative argument-stopper. The second does not provide an argument against open borders at all. The third underestimates the power of at least some non-state actors. We conclude by drawing some lessons for the open borders view and the use of feasibility in politics more generally.

Are institutional principles of justice subject to a minimal realism constraint to the effect that, in order to be valid, they must not make wildly unrealistic demands? Most of us say “yes.” David Estlund says, “no.” However, while Estlund holds that 1) institutional principles of justice are not subject to a minimal realism constraint, he accepts that 2) institutional principles of justice are subject to an *attainability constraint* to the effect that, in order to be valid, they must not make demands we are unable to meet; and 3) what he calls “institutional proposals” are subject to a minimal realism constraint. I argue that these three theses do not represent a plausible combination, at least given Estlund’s account of the principle/proposal distinction. Given this account, Estlund is either wrong to reject a minimal realism constraint on institutional principles of justice, or wrong to accept an attainability constraint on institutional principles of justice and/or a minimal realism constraint on institutional proposals. Either way, this has significant implications for the plausibility of his overall case against the minimal realism constraint on institutional principles of justice.

Empirical support is offered for the claim that the original Knobe effect, whereby our intentional action ascriptions exhibit certain asymmetries in light of our moral attitudes, can be successfully cancelled. This is predicted by the view that the Knobe effect can be explained in purely pragmatic terms (Adams and Steadman 2004a, 2004b, 2007). However, previous cancelling studies (Adams and Steadman 2007; Nichols and Ulatowski 2007) have failed to identify evidence of cancellability. The key to the successful cancelling strategy presented here is to provide subjects with the opportunity to assent to statements that involve sufficiently strong forms of moral censure.

A difficult problem for contractualists is how to provide an interpretation of the contractual situation that is both subject to appropriately stringent constraints and yet also appropriately sensitive to certain features of us as we actually are. My suggestion is that we should embrace a model of contractualism that is structurally analogous to the “advice model” of the ideal observer theory famously proposed by Michael Smith (1994; 1995). An advice model of contractualism is appealing since it promises to deliver a straightforward solution to the so-called “conditional fallacy.” But it faces some formidable challenges. On the face of it, it seems to be straightforwardly conceptually incoherent. And it seems to deliver a solution to the conditional fallacy at the cost of being vulnerable to what I shall call “the concessional fallacy.” I shall consider how, if at all, these challenges are to be met. I shall then conclude by considering what this might mean for the so-called “ideal/non-ideal theory” issue.

How should contractualists seek to accommodate and respond to the existence of radical pluralism within contemporary liberal states? Ryan Muldoon has recently argued that a) the dominant Kantian liberal model of contractualism is hopelessly ill equipped to do so but that b) there is a particular kind of Hobbesian contractualism that can do much better. I raise some problems concerning the capacity of Muldoonian contractualism to respond appropriately to the problem of radical pluralism. I then propose a very different kind of solution that involves embracing an advice model of contractualism. Keywords: contractualism; contractarianism; pluralism; diversity; disagreement

It is commonly taken for granted that questions of feasibility are highly relevant to our normative thinking – and perhaps especially our normative thinking about politics. But what exactly does this preoccupation with feasibility amount to, and in what forms if any is it warranted? This article aims to provide a critical introduction to, and clearer characterization of, the feasibility issue. I begin by discussing the question of how feasibility is to be understood. I then turn to the question of feasibility’s role, suggesting that there are two quite different kinds of role questions that may be at play, though they are often run together: a question about feasibility’s normative significance; and a question about its proper use in informing our normative thinking. Finally, I consider how the feasibility issue differs from certain other related issues: the demandingness issue; the issue of whether “ought” implies “can;” and the “ideal versus non-ideal theory” issue.

Breaking a promise is generally taken to involve committing a certain kind of moral wrong, but what (if anything) explains this wrong? According to one influential theory that has been championed most recently by T.M. Scanlon, the wrong involved in breaking a promise is a matter of violating an obligation that one incurs to a promisee in virtue of giving her assurance that one will perform or refrain from performing certain acts. In this paper, we argue that the “Assurance View”, as we call it, is susceptible to two kinds of counterexamples. The first show that giving assurance is not sufficient for incurring the kind of obligation of fulfillment that one violates in breaking a promise. The second show that giving assurance is not necessary. Having shown that the Assurance View fails in these ways, we then very briefly sketch the outline of what we take to be a better view—a view that we claim is not only attractive in its own right and that avoids the earlier counterexamples, but that also affords us a deeper explanation of why the Assurance View seems initially plausible, yet nonetheless turns out to be ultimately inadequate.

The object of this paper is to explore the intersection of two issues. The first concerns the role that feasibility considerations play in constraining normative claims – claims, say, about what we (individually and collectively) ought to do and to be. The second concerns whether normative claims are to be understood as applying only to actions in their own right or also non-derivatively to attitudes. In particular, we argue that actions and attitudes may be subject to different feasibility constraints – and hence that how we conceive of the role of feasibility in an account of normativity will depend in part on how we conceive of the role of actions and attitudes in normative theorising.

Imperialism seems to be deeply antithetical to democracy. Yet, at least one form of imperialism – what I call “hands-off imperialism" – seems to be perfectly compatible with the kind of self-governance commonly thought to be the hallmark of democracy. The solution to this puzzle is to recognize that democracy involves more than self-governance. Rather, it involves what I call self-rule. Self-rule is an example of what Philip Pettit has called a modally demanding value. Modally demanding values are, roughly, values the instantiation of which depends not only on what actually happens, but on what would happen in certain non-actual circumstances. Self-rule is the modally demanding counterpart of self-governance, since it requires, not merely that the members of a state actually govern themselves, but that they would continue to do so across a range of non-actual situations. Moreover, the value of self-rule (and hence democracy) is not reducible to the value of self-governance. Understanding the modally demanding character of democracy allows us to appreciate what is democratically objectionable about occupation by a foreign power, even if there is no prospect of the foreign power intervening in the governance of the occupied state by its members.

Despite an impressive philosophical pedigree, contractualism (or contractarianism) has only been properly developed in two ways: by appeal to the idea of an instrumentally rational bargain or contract between self-interested individuals (Hobbesian contractualism) and by appeal to the idea of a substantively reasonable agreement among individuals who regard one another as free and equal persons warranting equal moral respect (Kantian contractualism). Both of these existing models of contractualism are susceptible to apparently devastating objections. In this article, I outline a third, `deliberative' model of contractualism, which is based on the idea of a deliberatively rational agreement and which, I argue, represents a significant improvement on the Hobbesian and Kantian models in certain important respects. Key Words: contractualism • contractarianism • deliberation • deliberative rationality • Scanlon • Gauthier.

How does it happen that our beliefs about what we ought to do cause us to intend to do what we believe we ought to do? This is what John Broome calls the "motivation question." Broome’s answer to the motivation question is that we can bring ourselves, by our own efforts, to intend to do what we believe we ought to do by exercising a special agential capacity: the capacity to engage in what he calls enkratic reasoning. My aim is to evaluate this answer. In doing so, I shall focus on three core aspects of Broome’s overall account: his account of ought, his account of enkratic rationality, and his account of enkratic reasoning in particular. In each case I suggest there are problems.

The conditional analysis of ability faces familiar counterexamples involving cases of volitional incapacity. An interesting response to the problem of volitional incapacity is to try to explain away the responses elicited by such counterexamples by distinguishing between what we are able to do and what we are able to bring ourselves to do. We argue that this error-theoretic response fails. Either it succeeds in solving the problem of volitional incapacity at the cost of making the conditional analysis vulnerable to obvious counterexamples to its necessity. Or, it avoids the counterexamples to its necessity but fails to solve the problem of volitional incapacity.

Many of us feel internally conflicted in the face of certain normative claims that make infeasible demands: say, normative claims that demand that agents do what, given deeply entrenched objectionable character traits, they cannot bring themselves to do. On the one hand, such claims may seem false on account of demanding the infeasible, and insisting otherwise may seem to amount to objectionable unworldliness – to chasing “pies in the sky.” On the other hand, such claims may seem true in spite of making infeasible demands, and insisting otherwise may seem to amount to treating the agents in question with undue lenience – to mistakenly letting them “off the hook.” What is going on? One possibility is that we are making a mistake. I explore the alternative hypothesis that our ambivalent reactions, far from involving any mistake, are entirely consistent and appropriate. Rather than some single privileged ought such that the idea that “ought” implies “feasible” is either true or false, there are simply different oughts that are supposed to be capable of operating in the service of, and playing distinct roles associated with, what I shall call different core normative practices. In particular, there is a) some salient core practice-serving ought for which it’s true that “ought” implies “feasible” and b) some other salient core practice-serving ought for which it’s false that “ought” implies “feasible.” I sketch a framework for understanding different core practice-serving oughts in general and then use this framework to consider which particular core practice-serving oughts might be capable of vindicating our ambivalence. I begin by considering and rejecting a prevalent and prima facie promising account according to which the relevant oughts are the prescriptive ought and the evaluative ought. I then propose a different account that holds that the oughts we need are instead the deliberative ought and the hypological ought.

The so-called "Human Nature Constraint" holds that if an agent is unable, due to features of human nature, to bring herself to act in a certain way, then this suffices to block or negate the claim that the agent is required to act in that way. David Estlund (2011) has recently mounted a forceful objection to the Human Nature Constraint. I argue that Estlund’s objection fails – but instructively, in a way that gives Estlund resources for a different way of resisting attempts to negate normative claims by deploying the Human Nature Constraint.

Given constructivism’s enduring popularity and appeal, it is perhaps something of a surprise that there remains considerable uncertainty among many philosophers about what constructivism is even supposed to be. My aim in this article is to make some progress on the question of how constructivism should be understood. I begin by saying something about what kind of theory constructivism is supposed to be. Next, I consider and reject both the standard proceduralist characterization of constructivism and also Sharon Street’s ingenious standpoint characterization. I then suggest an alternative characterization according to which what is central is the role played by certain standards of correct reasoning. I conclude by saying something about the implications of this account for evaluating the success of constructivism. I suggest that certain challenges that have been raised against constructivist theories are based on dubious understandings of constructivism, whereas other challenges only properly come into focus once a proper understanding is achieved.

I raise three objections to Pettit’s republican account of justice: 1) that it fails to account adequately for the role of certain values such as substantive fairness; 2) that it represents an uncomfortable hybrid of egalitarianism and sufficientarianism; and 3) that it fails Pettit’s own “eyeball test”. I then conclude in a more constructive vein, speculating about the kind of account of justice it is supposed to be and suggesting that, construed a certain way, it may have resources for answering the three objections.