Self-Ownership, Libertarianism, and Impartiality Edward Feser [Paper presented at a conference on Impartiality and Partiality in Ethics at the University of Reading, England, December 1-2, 2006]
I. Impartiality is a central theme of contemporary liberal political philosophy, and certainly of the work of John Rawls, the most significant recent theorist of liberalism. The Kantian principle that persons are ‘ends in themselves’ and thus moral equals entitled to impartial concern underlies the argument of Rawls’s A Theory of Justice.[i] His later book Political Liberalism emphasizes another kind of impartiality important to liberals, namely the neutrality they think basic social institutions ought, as far as possible, to exhibit vis-à-vis the competing ‘reasonable comprehensive doctrines’ existing within a pluralistic society, and whose adherents these institutions are to govern.[ii] Rawls is also keen to argue, in that same book, that his conception of justice is impartial or neutral in a third sense too, insofar as it can be defended philosophically without having to appeal to controversial moral or metaphysical premises peculiar to some specific comprehensive philosophical doctrine. Rawls’s ‘political liberalism’ might be defined in terms of these three components: it is ‘liberal’ insofar as it seeks to ensure an equal and impartial concern for all persons as a matter of justice; and it is a ‘political’ conception rather than a ‘comprehensive’ one insofar as it attempts to fulfill this ambition in a way that is neutral both between the diverse moral and religious worldviews prevailing among citizens of the sort of society it is to govern, and between the various alternative philosophical doctrines (liberal and non-liberal) individual political theorists who could support it might be personally committed to. It rests instead on an ‘overlapping consensus’ between doctrines, while nevertheless sustaining a shared sense of justice between them, rather than a mere modus vivendi or truce between hostile factions. Libertarianism, which is commonly interpreted as a variant of liberalism,[iii] shares the liberal’s concern for treating persons as ends in themselves, though it understands this to entail a higher estimation of free markets and a lower estimation of government action than other modern liberals tend to exhibit. This is certainly true of Rawls’s critic Robert Nozick, who explicitly appeals to Kant’s principle in defense of laissez faire capitalism and the minimal state, and in criticism of Rawls’s difference principle.[iv] Libertarianism also appears to be concerned with impartiality; indeed, the view of libertarians seems to be that theirs is the most impartial of all political philosophies, in each of the three senses mentioned above: 1. With respect to our rights and duties as individuals: As usually interpreted, libertarianism is committed to a doctrine of universal full self-ownership, on which (as Nozick’s critic G. A. Cohen summarizes it) ‘each person enjoys, over herself and her powers, full and exclusive rights of control and use, and therefore owes no service or product to anyone else that she has not contracted to supply.’[v] In Nozick’s work, this principle is apparently understood as a concomitant of the Kantian principle of respect for persons; that is to say, to treat someone as an end in himself seems in Nozick’s view to entail treating him as a self-owner.[vi] And the principle of self-ownership seems, in turn, to entail that, at least as a matter of strict justice, we can have no positive obligations to other human beings – including not only the destitute, but even our own family members and friends – apart from those that we explicitly consent to take on. Justice requires only that we respect each individual’s (negative) right to be left alone. This is impartiality with a vengeance. In any event, it is in the view of libertarians a more impartial doctrine than Rawlsian liberalism is, given that Rawls’s difference principle entails (they would argue) discriminatory treatment of the rich and powerful as resources to be used for the benefit of the least well off members of society. 2. With respect to the various comprehensive doctrines prevailing within a pluralistic society: Libertarianism claims to provide, in Nozick’s words, a ‘framework for utopia,’ an ‘environment in which utopian experiments may be tried out [and] in which people are free to do their own thing.’[vii] Adherents of any comprehensive doctrine and the vision of life it embodies may within a libertarian polity live according to its dictates, even if those dictates are radically non-libertarian, so long as no one else is forced to do likewise. A hippie commune, Puritan commonwealth, Muslim umma,and Buddhist sangha can all co-exist within the boundaries of the minimal state, and can require of their members any degree of submission to their rules that they like, provided that membership in them is voluntary. Here too libertarians claim to be more impartial than other liberals insofar as their view does not require such private organizations to adhere to some egalitarian standard of admission. From the libertarian point of view, modern liberals’ promotion of anti-discrimination laws of various sorts imposes on all, in the name of neutrality, what is in fact nothing more than one particular comprehensive doctrine among others. 3. With respect to philosophical foundations: Libertarianism has been defended by philosophers committed to a wide variety of moral theories. Nozick, as we have seen, appeals to Kant’s principle of treating people as ends in themselves. Other libertarians have appealed to utilitarianism, contractarianism, Aristotelianism, natural law theory, Lockean natural rights theory, Ayn Rand’s ‘Objectivist’ philosophy, and Habermasian discourse ethics. Yet others eschew moral theory per se and emphasize economic arguments. Libertarian conclusions thus seem derivable from widely divergent premises. Nor is this surprising, libertarians might argue, given that their conception of justice is a very ‘thin’ one, requiring, again, only that everyone leaves everyone else alone. This is a polity (so the argument goes) that everyone, whatever the moral, religious, or philosophical premises he is committed to, can find reason to endorse. It is not surprising, then, that some libertarians have even tried to make a case for their view on Rawlsian grounds.[viii] For example, some have suggested that the unfettered free market best fulfills the egalitarian demands of Rawls’s difference principle, as a rising tide lifts all boats. More relevant to our purposes here, some libertarians have also suggested that it is their view that most fully realizes the neutrality Rawls sought to enshrine in his ideal of ‘political liberalism.’ ‘Political libertarianism,’ in their estimation, thus constitutes a superior vision of a just, liberal, and pluralistic society. Law professor Randy Barnett seems to be the most prominent libertarian to have defended this view, in a recent essay entitled ‘The Moral Foundations of Modern Libertarianism.’[ix] In what follows, I will argue that the purported impartiality of libertarianism is illusory, in each of the three senses of ‘impartiality’ described above. Barnett’s account will be my primary target, though what I will say is intended to apply to libertarian theories generally. For some libertarians, it may not matter whether their view really is as impartial or neutral as some have claimed; their commitment to libertarianism per se as a ‘comprehensive’ moral and political vision of the world may trump the desire to appeal to non-libertarians by suggesting that they need not give up anything essential to their own creeds by agreeing to accept a libertarian polity. Still, much of the rhetorical force of libertarian arguments lies in their suggestion that the libertarian worldview is compatible with any number of particular moral, religious, and philosophical systems. If many who would otherwise accept libertarianism were to conclude that they could not in fact do so consistent with their own deeper moral, religious, or philosophical commitments, the theory would lose much of its attraction.
II. Taking last things first, let us begin with the third sense in which some libertarians would regard their position as impartial or neutral, i.e. with respect to alternative moral-theoretic foundations. Barnett devotes almost half of his essay to defending this idea, which is reasonable given that, for reasons that will become evident, if this claim cannot be sustained, it is hard to see how libertarianism can possibly be neutral in either of the other senses in question. His focus is on the distinction between what he calls ‘moral rights’ and ‘consequentialist’ defenses of libertarianism, where by the former he has in mind natural rights approaches of the sort favored by Nozick, Rand, and Murray Rothbard, and by the latter, the economics-oriented arguments of writers like F. A. Hayek and Milton Friedman. As Barnett recounts, the history of libertarian political theory is partly a history of debate over whether these approaches are compatible, with libertarian rights theorists sometimes accusing consequentialist libertarians of being compromisers willing to trade liberty for economic well-being, and consequentialists accusing the rights theorists of utopian dogmatism. In Barnett’s estimation, this debate is misconceived. Rights and consequentialist theories are in his view best interpreted, not as rival claimants to ultimate moral truth, but rather as ‘fallible’ and complementary ‘modes of analysis,’ ‘useful heuristics,’ or ‘problem-solving devices,’ which by virtue of their ‘creative tension’ together afford us a more nuanced picture of the social world. In particular, rights theories make us attentive to the possible effects of our actions and policies on the private sphere of the individual, while consequentialist theories force us to consider the public sphere as well. The different theories also provide an ‘analytic check’ on one another: to the extent that they tend to converge on the same results, this should increase our confidence in those results; to the extent that they diverge, this could indicate a problem at the level of theory or application that calls for a reconsideration of some of our working assumptions. Moreover, their ultimate significance in Barnett’s view is as ways of refining our understanding of the deliverances of a third ‘problem-solving device,’ namely the gradual evolution of the legal norms enshrined in the common law. It is from these norms that moral rights and consequentialist theories ‘get their starting points’; and since legal decision-making is ‘casuistic,’ it takes account of more of the details of real-world circumstances than either philosophical or economic analysis are able to, and thus (Barnett seems to imply) has analytic priority over these other methods. Since the theories in question are ‘means, not ends,’ and specifically means of improving our ‘conventional legal principles,’ there need be no conflict between them. In this sense – and insofar as they tend to lead to similarly libertarian conclusions in any case – libertarianism can be said to be neutral between them. Barnett’s account seems to me to have a number of serious problems. The first is that the idea that natural rights and consequentialist theories are merely alternative ‘problem-solving devices’ that provide ‘analytic checks’ on one another is a non-starter. It certainly isn’t the way most actual rights theorists or consequentialists understand their own theories. And as usually understood the theories are indeed flatly incompatible. Suppose a rights theorist claims that I have an absolute right to X. Then in his view, no consideration of consequences could justify violating that right; indeed, any violation of that right would just be a bad consequence. For a consequentialist, on the other hand, there can be no such thing as an absolute right in the first place. So the theories are just at cross purposes. Since they differ fundamentally in their conceptions of the basic moral facts, they don’t agree on what we should be ‘checking’ for, and thus can hardly be construed as ‘devices’ or ‘checks’ or ‘modes of analysis’ aimed at solving the same problems. That Barnett doesn’t take seriously enough the self-understanding of actual natural rights and consequentialist theorists is also evident from his remarks about the limits he sees in these theories. The problem with rights theories, he says, is that they demand that justice be done ‘though the heavens fall,’ while the problem with consequentialist theories is that they are too ready to sacrifice individual liberty for the sake of the greater good. Now of course, these might well be good objections to each of these theories. But each objection is obviously only going to be valid from the point of view of the competing theory, or from the point of view of some third alternative theory. For a natural rights theorist might think that justice should sometimes be done though the heavens fall, while a consequentialist might say that liberty should sometimes be sacrificed for the greater good. From the natural rights or consequentialist perspective, these are not ‘problems’ at all. How, then, can a perspective which insists that these are problems for these theories be said to be ‘neutral’ between them? Surely what such a perspective really involves is just the belief that both theories are wrong, and ought to be replaced by some new theory, albeit one that incorporates certain elements of the ones superseded. At bottom, what Barnett really seems to be saying is not that ‘moral rights’ and ‘consequentialist’ libertarian theories are compatible, but rather that although they are incompatible as they stand, they can and ought to be reinterpreted – specifically, in an instrumentalist ‘problem-solving’ fashion – so that the incompatibilities disappear. But the ‘neutrality’ between moral theories that results is bogus, in two respects. First, it is not a neutrality between existing moral theories – which is surely what matters if libertarianism is to be impartial in some interesting sense – but only between theories that Barnett thinks should exist in the place of the ones that actually exist. Second, it appears surreptitiously to favor consequentialism over natural rights theories, since a piecemeal or ‘problem-solving’ approach to moral problems seems more in harmony with the consequentialist’s interest in maximizing some good or other (where the means of doing so are less important than that it is done) than with the interest of rights theories (as usually understood anyway) in respecting certain deontic constraints regardless of whether this conduces to problem-solving. A second problem with Barnett’s account is that the suggestion that the alternative theories in question converge on the same (libertarian) results, and thus confirm those results, is ambiguous. On one interpretation of this claim, it is true but trivial, while on another it is non-trivial but obviously false. To take the latter interpretation first, if Barnett means to say that consequentialism and natural rights theory per se tend to converge on libertarianism – or, for that matter, that Aristotelianism, natural law theory, etc. as such converge on it – then this is just clearly not true, since the vast majority of consequentialists, natural rights theorists, Aristotelians, natural law theorists, et al. are not libertarians. But if he means instead to say only that those consequentialists, natural rights theorists, Aristotelians, and so forth who happen also to be libertarians tend to converge in their arguments on libertarianism, then while true, the claim is (for obvious reasons) uninteresting. That some theorists (and a small minority at that) think that libertarianism can be defended by means of these various theories hardly suffices by to show that the theories themselves necessarily have a tendency to converge on the same conclusions (libertarian or otherwise) much less that libertarianism is neutral between these theories (most of whose proponents are probably hostile to libertarianism, and on grounds they would derive from the theories themselves). Finally, it is hard to see how Barnett’s advocacy of what he calls the ‘rule of law’ approach – taking the deliverances of the evolution of the common law to be the source of the norms that other approaches, including natural rights theory and consequentialism, must take as their starting points, and which it is their raisond’être to refine – can be anything other than the promotion of one ‘comprehensive’ philosophical doctrine over others. This is only more evident when we consider that Barnett is here endorsing and developing a set of ideas he derives in part from Hayek, ideas that some other libertarian theorists (whose views Barnett claims to be adjudicating in a neutral way) would object to. Of course, that doesn’t show that Barnett’s preferred Hayekian conception of the foundations of libertarian theory is mistaken. But it does show that it is implausible to present it as if it were a ‘neutral’ framework that favored no particular comprehensive philosophical doctrine (including no particular way of giving philosophical foundations to libertarianism) over any other. Barnett’s legal emphasis also smuggles in the common liberal conception of morality (or at least of justice, in Barnett’s case) as ultimately a matter of interpersonal conflict resolution – a conception that is by no means ‘neutral’ between alternative moral theories, and certainly not compatible with the Aristotelian and classical natural law approaches some libertarians claim to follow.[x] But this brings us to his treatment of the idea that libertarianism is neutral or impartial in the second sense we have identified, i.e. neutral between the various comprehensive doctrines prevailing within modern pluralistic societies. So let us now turn to that.
III. A further objection to Barnett’s account as summarized so far would be that his claim that the various alternative moral theories (consequentialist, rights-based, or what have you) are valuable insofar as they help us to improve existing legal conventions presupposes some standard by reference to which something counts as an improvement, and that this standard is unlikely to be a neutral one. This is an objection Barnett raises himself, and his answer is to reiterate the suggestion that the various comprehensive theories in question be interpreted, not as embodying competing ends (one of which gets to provide the criteria for improvement that the others must meet), but rather as ‘devices’ for solving a problem that is to be defined independently of any these particular doctrines. That problem is providing for the possibility of ‘social order,’ in the sense of conditions wherein individuals can pursue their own happiness in a way that is consistent with others doing the same. Indeed, this, Barnett asserts, is what justice is ultimately concerned with. Now the first thing to say about this is just to reemphasize two points made already. First, insofar as Barnett’s attempt to be neutral between competing moral-theoretic foundations for libertarianism requires that these theories be reinterpreted in a way their adherents might not approve of, the ‘neutrality’ thus achieved is specious. Second, Barnett’s insinuation that the end of providing for ‘social order’ in the sense he describes is definable apart from some particular comprehensive doctrine is dubious, especially given that he characterizes it as nothing less than the end of justice itself. For this characterization is distinctively liberal, and thus merely one conception of justice among others. Now in fairness to Barnett, it must be noted that he does acknowledge that his conception of justice is liberal, indeed libertarian, and thus controversial. So when he says that justice-as-social-order is an end that does not favor one comprehensive doctrine over another, what he presumably means, given the immediate context, is that it does not favor any particular libertarian comprehensive doctrine (rights-based, consequentialist, or whatever) over another. But again, this just raises the question of how Barnett’s conception of justice can provide a neutral framework for all the various non-libertarian moral, philosophical, and religious doctrines that prevail within a modern pluralistic society. In dealing with this question, Barnett acknowledges, as Rawls does, that no framework, even a merely ‘political’ rather than ‘comprehensive’ one, can be entirely neutral. Rawls claimed only that his own brand of liberalism is neutral between what he called ‘reasonable’ comprehensive doctrines, and Barnett also acknowledges that his position is bound to rule out any vision of the good that is incompatible with the possibility of ‘the pursuit of happiness of each person living in society with others.’[xi] Nazis and jihadists, then, are simply out of luck within a libertarian social order, but that’s a failure of neutrality that the rest of us can happily endorse. So far so good. The question, though, is whether Barnett’s proposed framework can nevertheless be neutral enough to encompass all of the reasonable people within modern pluralistic societies who deeply disagree with each other but who do not believe in settling their disputes by murdering the opposition. In particular, we need to ask whether it can accommodate: both those who believe that abortion is a basic human right and those who believe it is tantamount to murder; both those who believe that a right to view pornography is a corollary of the right of free speech and those who believe (on either feminist or traditional moral grounds) that pornography ought to be curtailed for the good of society; both those who believe that laws against discrimination in hiring, renting, admission to private associations, and the like are a requirement of basic justice, and those who believe such laws violate a natural right to freedom of association; both those who believe that the legalization of same-sex marriage is required by justice and those who regard it as an affront to the moral order that makes justice possible; both those who believe that religion ought to be kept out of public institutions and debate lest the rights of non-believers be threatened, and those who believe that public acknowledgement of at least a generic theism is a prerequisite of the very intelligibility of ‘rights’ talk; and so on and so forth. These examples represent some of the most contentious debates within modern pluralistic societies, the very sorts of debates that motivate Rawlsian liberals and their libertarian fellow travelers to seek a framework enshrining a sense of justice shared in common between alternative comprehensive doctrines. So if it turns out that Barnett’s framework isn’t neutral even between the sorts of views just described, it is hard to see how it can be neutral in any interesting sense at all. Hence, again, we need to ask: Is it neutral between them? To answer this question, let us first consider how Barnett fleshes out the libertarian conception of social order.[xii] First, he says, it acknowledges ‘the existence and value of individual persons.’ Second, it values ‘the ability of all persons to live and pursue happiness.’ Third, it recognizes that ‘only the protection of actions, rather than a guaranty of results, can potentially be afforded to everyone.’ Fourth, it emphasizes that the actions of any one person ‘may have both positive and negative effects on others.’ Fifth, it insists that a set of ‘ground rules’ are possible that would ‘provide all, or nearly all, persons living in society the opportunity to pursue happiness without depriving others of the same opportunity.’ Again, so far so good. There isn’t much on this list that many people would object to, though some egalitarians might raise questions about the third point. Though even there, Rawlsians, at least, would be happy with a purely procedural conception of justice, since they regard their egalitarian difference principle as a constraint upon action rather than a guarantee of any particular result. But there’s the rub. For libertarians, obviously, would not regard Rawls’s difference principle as something that ought to inform the framework within which adherents of competing comprehensive doctrines operate. They would regard it instead as the imposition of one particular, egalitarian, comprehensive doctrine on all. And yet Rawlsians would regard it as nothing of the kind. The trouble is that the idea that a neutral framework should entail only ‘the protection of actions, rather than a guaranty of results’ is itself susceptible of radically divergent interpretations. And so too are all of the other points Barnett lists. Take the first point, about ‘the existence and value of individual persons.’ The import of this principle obviously depends on what we count as a ‘person.’ Do slaves count as persons? Could a slaveholder in antebellum Virginia plausibly endorse Barnett’s ‘neutral’ libertarian framework and still keep his slaves, on the grounds that they are not (in his view) full persons and that anyone who claimed otherwise would be imposing some comprehensive moral or metaphysical doctrine on him? Presumably not, and Barnett would no doubt be untroubled by this limit on the neutrality of his system. Well and good; and the question is purely academic in any case, given that apologists for slavery are thin on the ground these days. But what about fetuses? If they count as persons, and if one remains unpersuaded by Judith Jarvis Thomson’s view that the question is irrelevant to the legitimacy of abortion, then quite obviously a practice that is now very common in Western pluralistic societies would have to be outlawed in any truly libertarian society. To fail to outlaw it on the grounds that not everyone thinks abortion is immoral would simply be to impose a particular comprehensive moral doctrine (i.e. a ‘pro-choice’ one) on unborn persons. On the other hand, if fetuses are in fact not persons at all, then it would seem that abortion must remain legal, and that it is those who want to restrict abortion who are trying to impose their comprehensive doctrine on others. Either way, it seems obvious that no polity, including Barnett’s purportedly neutral libertarian one, can possibly be ‘neutral’ between the ‘pro-life’ and ‘pro-choice’ sides on the abortion debate. Any polity will either regard abortion as tantamount to murder or not so regard it, and those who disagree with the view that prevails can quite plausibly complain that a certain contentious view about the nature of ‘persons’ is being imposed on all. When we add to the mix other contemporary controversies over matters of life and death – euthanasia, capital punishment, the treatment of animals, and so forth – it becomes even more obvious that to cite a recognition of ‘the existence and value of individual persons’ does nothing to give content to the suggestion that libertarianism provides an impartial framework between comprehensive doctrines. For what matters – what (some of) these different frameworks disagree about – is what counts as a person, and what counts as valuing a person, in the first place. Of course, Barnett might say that one side or the other in the debate over abortion (or capital punishment, euthanasia, or whatever) is just ‘unreasonable,’ in the Rawlsian sense, and that his view, like Rawls’s, is not intended to be neutral between ‘reasonable’ and ‘unreasonable’ comprehensive doctrines. But if he does say this, then the ‘neutrality’ of his system would be useless for the project of providing a framework within which the competing comprehensive doctrines that actually exist within contemporary pluralistic societies might abide by a shared sense of justice rather than an uneasy modus vivendi. For it would then appear that Barnett’s system is ‘neutral’ only between those comprehensive doctrines that are ‘reasonable’ in the sense of being consistent with the moral and metaphysical presuppositions that Barnett and those of like mind happen to consider reasonable. Those who do not share these presuppositions – surely millions of Barnett’s fellow citizens – would simply be written off as beyond the political pale, rather than (as we would expect from a view claiming to provide an interesting way of reconciling the hostile factions that characterize contemporary politics) accommodated within a more encompassing set of common principles. Here we see manifested a dilemma that I think afflicts Barnett’s entire project: either he keeps the statement of the principles guiding his favored polity extremely vague, so as to preserve a genuine neutrality between the main competing doctrines actually prevailing within existing pluralistic societies; or he makes them so determinate that they effectively marginalize a great many of these doctrines as ‘unreasonable.’ In either case, the ‘neutrality’ that results seems uninteresting and anticlimactic. A similar result follows from a consideration of Barnett’s condition that a liberal social order respects ‘the ability of all persons to live and pursue happiness’ – not only because what counts as a ‘person’ is just as problematic here as in the previous case, but also because what counts as ‘pursuing happiness’ is problematic. What is ‘happiness,’ after all? Is it a matter of ‘getting what you want’?[xiii] Is it a matter of maximizing pleasure and minimizing pain, as Bentham and Mill held (albeit in different ways)? On these conceptions, happiness seems to be definable in terms of the subjective states of the agent. On an Aristotelian or classical natural law conception, though, it is something objective, a matter of living in accordance with one’s nature, understood as a fixed essence entailing a natural end or purpose. Other conceptions are also possible. Now, on several subjectivist conceptions of happiness, it seems plausible that the sort of libertarian polity Barnett favors would be a just one. If we concede the libertarian argument that the unfettered free market maximizes the satisfaction of individual preferences, then it would seem that libertarianism does indeed promote the pursuit of happiness in the sense that it best allows people to ‘get what they want.’ Perhaps it even maximizes pleasure and minimizes pain, at least in Bentham’s crude sense of ‘pleasure’ and ‘pain.’ It isn’t so clear, however, that it maximizes pleasure and minimizes pain in Mill’s more refined senses of ‘pleasure’ and ‘pain.’ And it is very difficult to believe that it best promotes happiness in Aristotle’s sense of the term. For the market maximizes the satisfaction, not of all preferences, but rather of those backed by the most spending power. It is bound, then, to cater to the most vulgar tastes and passions – which are, by definition, the most common and thus the ones most people will pay to satisfy – rather than to more refined sensibilities. And since on an Aristotelian conception an individual’s moral character – his characteristic habits and sensibilities – is inevitably deeply influenced by the character types and sensibilities prevailing in the society around him, it follows that a commercial society is one in which the sort of refined moral character that most fully manifests the realization of human potentialities, and thus most fully guarantees human happiness, is bound to be very rare and difficult to achieve. But then, since on a classical Aristotelian or natural law view, a society cannot be just unless it makes the attainment of virtue a realistic possibility, a libertarian society would seem to be deeply unjust. It certainly isn’t neutral between an Aristotelian or classical natural law conception of justice and other conceptions. Moreover, since the conception of happiness operative in many religious traditions is far closer to the Aristotelian or natural law conception than it is to any subjectivist conception, a libertarian polity would also seem to be anything but neutral between these traditions either, at least where respecting ‘the ability of all persons to live and pursue happiness’ is concerned. If a society characterized by an unfettered free market is bound to be one in which all sorts of behaviors considered sinful from a religious point of view are prominent and celebrated, then from a religious point of view such a society cannot fail to be seriously unjust, since it will tend greatly to reduce the likelihood that individuals living within it will be able to lead virtuous lives and thus attain salvation and true happiness, whether this be conceived of in terms of the beatific vision, paradise, nirvana, moksha, or what have you. Barnett’s remaining principles – that a liberal society recognizes that individual actions ‘may have both positive and negative effects on others’ and aims to ‘provide all, or nearly all, persons living in society the opportunity to pursue happiness without depriving others of the same opportunity’ – are no less problematic than the ones already discussed, and for the same reasons. For what counts as a ‘negative effect’ on others, and what counts as ‘depriving others of the opportunity to attain happiness,’ depend on which comprehensive doctrine one is committed to. For the pro-choice advocate, one woman’s decision to have an abortion has no negative effect on anyone else, since others are free not to have abortions. But for the pro-life advocate, to have an abortion is to have the most negative effect imaginable on another person, since it involves a kind of murder. The pornographer claims to hurt no one since those who disapprove of his product are free not to purchase it. But many feminists, and people with traditional moral and religious views, would argue that when pornography becomes widespread and socially acceptable it creates what economists call a ‘negative externality,’ insofar as it radically alters the average person’s perception of women and of the meaning of sex, and effectively makes certain moral virtues extremely difficult or impossible to achieve on anything but the smallest of scales. And so forth. Now Barnett maintains that whatever failings traditional moralists and others might see in a libertarian society, such failings do not count as injustices and thus do not evidence a failure of neutrality on the part of a libertarian polity vis-à-vis the comprehensive doctrines in question. Thus he claims, for example, that while a libertarian society may not guarantee respect for the ‘imperfect rights’ we might be said to have as moral beings aiming to realize some vision of the good (i.e. rights that others cannot legitimately be forced to respect, even if they should respect them), it does guarantee the ‘perfect (i.e. enforceable) rights’ we have not to be harmed by others, and this is sufficient for justice.[xiv] Hence while it may in some sense violate my rights as a pursuer of moral perfection to have someone tempt me to view women as sex objects (for example), the rights in question are not perfect or enforceable rights, and thus the fact that their violation might be tolerated or even celebrated in a libertarian society does nothing to show that such a society is unjust. Furthermore, to try to promote the pursuit of moral virtue and other ‘spiritual goods’ via the coercive power of government would threaten to undermine the very ‘social order’ that is a prerequisite to their successful pursuit.[xv] It would lead to a ‘war of all against all’ insofar as those who ‘seek to impose by force their conception of the good’ on others are bound to be resisted by those with different conceptions.[xvi] The problem with this reply is that it begs all the important questions. The rigid distinction between the right and the good or between matters of justice and matters of virtue, and the related assumption that perfect rights and duties exactly correlate with the former and imperfect ones with the latter, are precisely what Aristotelians, classical natural law theorists, and many religious believers tend to reject. On their view, rights and justice are grounded in the good, and part of their very raison d’etre is to aid us in realizing moral virtue. For this reason, while most of what government does in securing justice and rights will involve merely keeping people from killing and stealing from each other and the like, government may, and sometimes should, step in to prevent at least large-scale threats to ‘private’ morality. As if in reply to this objection, Barnett complains that ‘many conservatives assume, usually implicitly, that force is justified whenever human conduct is found to be bad or immoral,’ and that ‘for comprehensive moralists of the Right or Left, using force to impose their morality on others might be their first choice among social arrangements.’[xvii] But surely no thinker of the Right or Left actually believes that force is justified whenever human conduct is bad or immoral, or is even advisable as a general policy. Even from a conservative point of view, literally sending the police into people’s bedrooms might be seen as counterproductive and indeed unjust. Even from a left-wing point of view, it might be unwise and unjust to try to force everyone to love people of other races. It is possible consistently to hold both that government should refrain from interfering in the small-scale, day-to-day private moral choices of individual citizens and also that it may and sometimes should interfere with business activity that may have a large-scale effect on the overall moral tenor of society – such as the marketing of narcotics or pornography, a refusal to cater to customers of a certain race, or what have you. The point is not to defend any particular piece of right-wing or left-wing morals legislation. The point is rather to note that certain conceptions of justice incorporate the idea that it is part of justice (even if not the whole of it) to guarantee the preconditions of the attainment of moral virtue, or at least of minimal public decency, so that any polity that rules this out a priori can hardly be said to be neutral between comprehensive moral doctrines or plausibly defensible from the point of view of an ‘overlapping consensus’ of such doctrines. Nor will it do for Barnett to pretend that only libertarianism can save us from the ‘war of all against all’ that must result if adherents of competing doctrines are allowed to ‘impose’ them on others – as if libertarianism alone existed above the fray of competing conceptions of justice, and as if libertarians cannot plausibly be seen by their critics as attempting to ‘impose’ their view of justice on non-libertarians. From the point of view of a traditional moralist, say, or a socialist, the libertarian is interested in nothing less than ‘imposing’ his idiosyncratic moral views on society as a whole – by refusing to allow government to defend the rights of the unborn, for example, or by denying to every citizen the equal share of the overall social product to which he is (in the view of some socialists) entitled as a matter of justice. Yes, the libertarian holds that these conceptions of justice are mistaken.[xviii] But then, that is precisely the point: Because he does take them to be mistaken, he cannot plausibly claim that his view is neutral in any interesting sense, or that it does not involve an imposition on others of a moral view at odds with their own. To be sure, Barnett’s ‘political libertarianism’ might differ from such conceptions of justice in aiming to be neutral and in claiming not to involve any such imposition – that is not in dispute. What is in dispute is whether the aim can actually be realized and whether the claim can be sustained, and Barnett gives us no reason to answer either question in the affirmative. Once again, Barnett might well reply by arguing that any comprehensive doctrine that has implications of the sort I’ve described is to that extent not only mistaken but ‘unreasonable’ (in the Rawlsian sense) and thus not a doctrine to which his analysis is meant to apply in the first place. But to do so would be to make manifest that his purportedly ‘neutral’ framework is really neutral only between comprehensive doctrines that have already incorporated into themselves the liberal premises definitive of the very sort of comprehensive liberalism that Barnett, like others who seek a ‘neutralist’ liberalism a la Rawls, claim not to be imposing on the various other comprehensive doctrines prevailing in modern pluralistic societies. Worse, it would be to make manifest that the resulting system is one that leaves out in the political cold enormous numbers of Barnett’s fellow citizens – indeed, probably a majority of them – who are committed to some variation or other of the moral and religious views in question. As I’ve argued, the ‘neutrality’ thus achieved would be bogus, and certainly useless for solving the problem of showing that a shared sense of justice, and not merely a modus vivendi, is possible between adherents of rival comprehensive doctrines under conditions of pluralism.
IV. We turn finally to the question of whether libertarianism is truly an ‘impartial’ view in the remaining sense of respecting all persons equally. Here too, I think, the impartiality that libertarianism can be said to exhibit is either trivial or specious, certainly in Barnett’s version and in any ‘political’ brand of libertarianism that aims to be neutral in the other two senses we’ve discussed. Part of the reason for this might be obvious from what has been said already. Libertarianism claims to treat all persons as self-owners. But as we have seen, what counts as a ‘person’ is itself a matter of contention. Do fetuses, antebellum slaves, and people in ‘persistent vegetative states’ count as ‘persons’ and thus as self-owners? Depending on the answer the libertarian gives, his view is bound to sound less than impartial from some points of view. Should he say, for example, that fetuses do not count as persons, then opponents of abortion will say that his view is not impartial between born and unborn persons. Should he say that fetuses are persons and should be protected by law accordingly, then people who seek abortions, or who want to become pregnant via in vitro fertilization (which typically entails the death of unneeded embryos), or who favor experimentation on embryos in the hope of finding new cures and the like, will say that his view is not impartial toward them, since it interferes with their liberty in the interests of upholding some controversial religious or metaphysical doctrine. Similarly, if the libertarian does not regard those in ‘persistent vegetative states’ as persons, euthanasia opponents will accuse him of discriminating against persons simply on the (in their view irrelevant) grounds that they are no longer capable of consciousness. While if he does regard them as persons, other critics will say that he is not truly impartial toward those whom he would burden with the care of human beings who (in their view) can no longer benefit from care and lack any entitlement to it. The libertarian could, of course, decide not to answer such questions at all, but if he does his view will lose much of its interest: If he refuses to tell us what counts as a person, his assurance that all persons must be treated as self-owners is of limited use. One problem with the claim that libertarianism upholds an impartial concern for all persons, then, is the same problem afflicting Barnett’s proposal that libertarianism be interpreted as a ‘political’ doctrine that is neutral between different possible philosophical justifications and competing comprehensive doctrines: It sounds plausible only if we keep the key concepts vague or indeterminate, and thus philosophically uninteresting and practically useless. Might not the claim be salvaged, though, if the idea of libertarianism as a ‘neutral’ or ‘political’ doctrine is abandoned and some rich comprehensive libertarian theory is put forward in its place? This might seem possible on some versions of libertarianism – for example, if Nozick’s very sketchy attempt to ground libertarianism in Kant’s principle of treating persons as ends in themselves could be fleshed out successfully, or if an old-fashioned Lockean natural rights theory could be revitalized.[xix] One problem, though, is that these deontological approaches to libertarianism do not seem to have many defenders today. Contractarian and consequentialist approaches appear to be favored by most contemporary libertarian theorists. This is certainly true of Barnett, who, as we’ve seen, interprets all libertarian theories as merely alternative heuristics or tools for refining conventional and evolving legal principles whose point is to coordinate the behavior of individuals as each pursues his own happiness. Barnett acknowledges that his approach is ‘consequentialist’ insofar as it provides only ‘hypothetical imperatives’ which, he thinks his analysis shows, we ought as a matter of empirical fact to follow if we happen to value the goal of allowing each individual to pursue ‘happiness, peace, and prosperity’ in harmony with others.[xx] Obviously, it is bound to be much harder to sustain the claim that libertarianism enshrines an impartial respect for all persons as such if the theory contains no deontic component.[xxi] For only if we are already impartially concerned with the happiness of all persons as such might we care to consider Barnett’s suggested method for securing it. By itself, Barnett’s consequentialist version of libertarianism gives us no reason to adopt such a concern, any more than a description of how to engineer a functioning bridge, however accurate, gives us a reason to construct one. Those who think, for example, that some groups have an inherent right to rule over others can simply thank Barnett for his advice but reply that they are not interested in it, given that they are not seeking the happiness of all but rather the aggrandizement of a special few. Barnett would rightly disapprove of this, but by his own admission he has given us no grounds for doing so. Things are even worse for impartiality where contractarian approaches to libertarianism are concerned.[xxii] The contractarian takes all moral obligations to derive from a tacit agreement to abide by certain norms in the interests of furthering the mutual advantage of the parties to the agreement. Since in general we all benefit from others’ leaving us alone in the pursuit of our own aims, it is to that extent to our advantage to agree to a general policy of leaving others alone, in the hope that they will extend to us the same courtesy. Notoriously, though, to the extent that some individuals cannot realistically benefit us at all, not even by leaving us alone (since they are too weak to threaten us anyway), it is hard for a contractarian to show that we have any grounds for treating them other than any way we feel like treating them. And those who are not parties to the agreement – either because they never entered into it in the first place or because they have broken it – can indeed be treated in principle any way we like. Contractarians have attempted to come up with various ways of avoiding the very nasty implications this view of morality seems to have (most of them ad hoc and unconvincing in the view of the theory’s critics), but the bottom line is that the theory does entail that we do not owe persons as such our impartial concern. They are owed concern only to the extent that it can be shown that a policy of treating them with concern somehow accrues to our benefit. Even on a deontological version of libertarianism, though, it is not at all clear that the view can plausibly claim to enshrine an impartial respect for persons as such. The reasons are threefold, and have to do, paradoxically, with the very strong conception of individual rights definitive of libertarianism. So literal is the standard libertarian conception of persons as their own property that it seems to allow that a person has the right to sell himself into slavery. (Nozick is explicit about this, though in fairness it must be noted that Barnett rejects the idea.[xxiii]) As Samuel Freeman points out, this would entail that any agreement to become another person’s slave is one that a libertarian government would have to enforce just as dutifully as it enforces any other contract.[xxiv] But it is surely odd to think of a government that forces someone to become another’s slave as one that is impartial between persons. The fact that the agreement was entered into voluntarily does not seem to mitigate this point. Surely only someone who was stupid, insane, or in a state of total desperation would make such an agreement, and most of them would inevitably come deeply to regret having done so. Any political system that would enforce such agreements thus seems partial to the intelligent, sane, and fortunate. Moreover, it is hard to see how the considerations that lead us to think persons worthy of impartial concern in the first place – the fact that they are rational, possess free will, are capable of formulating a life plan, and so forth – could plausibly be thought to entail a polity on which these very attributes could be practically nullified as a matter of justice. A second and related point concerns the libertarian commitment to absolute property rights in external resources, which leaves open the possibility that some person or persons could justly acquire all those resources that are necessary for bare survival and refuse to allow others access to them even in return for payment or labor. The chilling upshot for these others is (as Herbert Spencer put it) that “save by the permission of the lords of the soil, they can have no room for the soles of their feet.”[xxv] Of course, libertarians argue, not implausibly, that such a consequence is extremely unlikely to occur in the actual world and that the market can adequately guarantee almost all persons the means to provide for themselves at well above subsistence level. The point, though, is that the theory does allow that in principle someone could justly be allowed by government to starve to death simply because he was too stupid, mad, or just plain unfortunate not to get hold of enough resources to sustain himself and could find no one willing voluntarily to help him. Once again, a libertarian polity would seem partial to the intelligent, sane, and fortunate. And once again, it is at least questionable whether the considerations that lead us to think persons entitled to impartial concern could entail property rights that are so strong that the very small level of taxation required to fund even temporary assistance to those destitute through no fault of their own is ruled out as a grave injustice. Finally, and as Freeman again points out, the libertarian conception of political power is one on which such power is merely a special case of the furtherance of private economic interests. The minimal state advocated by Nozick is essentially a private security firm which has (for morally legitimate reasons, in Nozick’s view) taken it upon itself to decide whether and how its clients can be punished for alleged offenses committed against others, and which compensates those others for this inconvenience. It owes its clients protection only insofar as they have contracted for it, and only at the level at which they have agreed to pay for it; it owes those who accuse its clients of wrongdoing and whom it does not allow to exact justice on their own initiative only protection against said clients and not against anyone else who might seek to violate their rights; and it owes nothing to anybody else. The libertarian minimal state is clearly not impartial to all those who live under its jurisdiction, then, any more than any other private corporation is impartial to all those who might benefit from its services. It favors paying customers, and favors most those who can afford to pay the most. Those who do not pay get no protection at all, and this is, on the libertarian view, perfectly just, even if the reason they do not pay is that they are stupid, mad, or just cannot afford to do so. Once again, the intelligent, sane, and fortunate have the advantage in a libertarian polity.
V. All told, libertarian claims to impartiality prove, upon analysis, to be devoid of interesting content. A libertarian theory of justice is neutral between various comprehensive political philosophies – as long as they more or less share the same premises as the libertarian writer who happens to be presenting said theory of justice. A libertarian polity is neutral between the various comprehensive moral and religious doctrines that prevail in a pluralistic society – as long as they are willing to incorporate into themselves a basically libertarian conception of rights and justice. A libertarian society treats all its citizens equally – but the intelligent, sane, and fortunate are more equal than the stupid, mad, and unfortunate. As J. L. Austin famously put it, ‘there’s the bit where you say it and the bit where you take it back.’[xxvi]
[i] John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971). [ii] John Rawls, Political Liberalism (New York: Columbia University Press, 1996). [iii] Though for a different interpretation, see Samuel Freeman, ‘Illiberal Libertarians: Why Libertarianism Is Not a Liberal View’, Philosophy & Public Affairs 30, no. 2 (2001): 105-51. [iv] Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974). [v] G. A. Cohen, Self-Ownership, Freedom, and Equality (Cambridge: Cambridge University Press, 1995), p. 12. [vi] Though Nozick is not clear about exactly what he takes the relationship between these principles to be. See Edward Feser, On Nozick (Belmont, CA: Wadsworth, 2003), pp. 43-45 for discussion. Some libertarians are less inclined to regard the principle of self-ownership as a useful way of formulating or grounding libertarianism. See e.g. Douglas B. Rasmussen and Douglas J. Den Uyl, Norms of Liberty (University Park, PA: University of Pennsylvania Press, 2005), ch. 9. [vii] Nozick, Anarchy, State, and Utopia, pp. 297 and 312. [viii] See e.g. Loren Lomasky, ‘Libertarianism at Twin Harvard,’ Social Philosophy and Policy 22, no. 1 (2005): 178-199. [ix] In Peter Berkowitz, ed., Varieties of Conservatism in America (Stanford, CA: Hoover Institution Press, 2004). See Barnett’s The Structure of Liberty: Justice and the Rule of Law (New York: Oxford University Press, 1998) for a more thorough statement of his conception of a libertarian theory of justice. [x] Libertarian theorists Rasmussen and Den Uyl acknowledge that the glib assimilation of morality to rules for avoiding interpersonal conflict is a common foible of liberal theorizing, a foible they strive to avoid. See Norms of Liberty, ch. 3. [xi] ‘The Moral Foundations of Modern Libertarianism,’ p. 70. [xii] Ibid., p. 64. [xiii] For a critique of this conception, see Bob Brecher, Getting What You Want? A Critique of Liberal Morality (London: Routledge, 1998). [xiv] ‘The Moral Foundations of Modern Libertarianism,’ p. 69. [xv] Ibid., pp. 70-71. [xvi] Ibid., p. 73. [xvii] Ibid., pp. 72-73, 74. [xviii] Some libertarians allow that abortion can legitimately be outlawed under certain circumstances, but the majority of libertarians seem to take a ‘pro-choice’ view. [xix] But see Cohen, Self-Ownership, Freedom, and Equality, pp. 238-43 for problems with the first strategy, and Jeremy Waldron, God, Locke, and Equality (Cambridge: Cambridge University Press, 2002) for problems with the second. [xx]The Structure of Liberty, chapter 1. [xxi] Barnett does tentatively suggest at p. 24 of The Structure of Liberty that his theory might be ‘deontological’ to the extent that he thinks the analysis he gives provides the necessary conditions for each individual’s being able to pursue happiness in society with others. But since he does not claim to show that we ought to pursue this goal, the fact (if it is a fact) that his theory shows us the only way to attain it does not make the theory a deontological one. [xxii] See e.g. James Buchanan, The Limits of Liberty (Chicago: University of Chicago Press, 1975) and Jan Narveson, The Libertarian Idea (Philadelphia: Temple University Press, 1988). [xxiii] See Anarchy, State, and Utopia, pp. 58, 283, and 331 for Nozick’s endorsement of the possibility of voluntary slavery and The Structure of Liberty, pp. 78-82 for Barnett’s rejection of it. [xxiv] ‘Illiberal Libertarians,’ p. 132. [xxv]Social Statics (New York: Robert Schalkenbach Foundation, 1995), p. 104. [xxvi] J. L. Austin, Sense and Sensibilia (Oxford: Oxford University Press, 1962), p. 2.