Liberal Neutrality Conference

Two weeks ago, May 1-3, McGill and the Centre de Recherche en Ethique de L’Universite de Montreal (sorry about my pronunciation) co-sponsored a conference on “Liberal Neutrality: A Re-Evaluation.” Papers are here, and my notes are below. Take this for what it is – impressions and imperfect summaries from an audience member. (I was there for the first two days, but not the third.)

After a brief introduction, Peter De Marneffe started things off with “The Possibility and Desirability of Neutrality.” He distinguished between legislative neutrality, according to which “the government should not adopt a coercive policy unless it can be justified by neutral reasons,” and foundational neutrality, according to which a conception of justice “valid only if it can be justified without making any assumptions about what kinds of life are best for people.” These are logically independent of each other, and he aimed to show that legislative neutrality is possible and desirable. In other words, non-neutral reasons can’t justify coercive legislation that limits liberty.

But what is a non-neutral reason? Well, there are some obvious examples such as: that a person’s conception of the good is false; that some activity would displease God; that pursuing some goal would not realize some higher faculty to the greatest extent possible. In fact, he provides a list of 12 such non-neutral reasons. Can we provide a unified analysis of what makes these reasons non-neutral? No – at least De Marneffe hasn’t been able to do so. There seems to be no morally relevant property that all non-neutral reasons share. The idea of neutrality seems to unify various ideas of liberty, but it does so in name only. At the level of philosophical analysis, then, the idea of neutrality does little work. However, this is consistent with saying that the principle of legislative neutrality is substantively defensible, that is, that those 12 types of reasons fail to justify coercive legislation.

Two additional notes. First, legislative neutrality does not rule out all forms of paternalistic legislation, since in some cases it might be possible to offer justifications that are not listed among the 12 types of non-neutral reasons. This shows that the Neutrality Principle is a superior successor to the Harm Principle. Second, Rawls’s idea of public reason does not provide a unifying analysis of neutral (or non-neutral) reasons because a liberal, political conception of justice may still be non-neutral. Consider, for example, Rawls’s two principles of justice followed by (given lexical priority over) a principle allowing the government to promote certain conceptions of the good through ordinary legislative action based on majority preference. According to Rawls’s criteria, this counts as a liberal conception, and it is not based on any particular comprehensive doctrine, so it is political. Yet, it seems to violate the principle of legislative neutrality (although it depends how this is spelled out precisely). (It seems to me that this suggests the idea of “constitutional neutrality” which may have been closer to Rawls’s position than “legislative neutrality.”)

Next, Jacob Levy asked “Is Neutrality Sustainable?” The question concerns the relationship between the justice of institutions and the virtues of individuals. The challenge goes like this: the stability of just institutions depends on the virtues of individuals, but neutral liberal institutions cannot foster the necessary virtues among citizens. Critics have tended to assume that the relevant virtues can only be promoted through deliberate state action, but this is not at all clear.

After all, historically liberal institutions emerged out of non-liberal institutions, and presumably there were not already liberal individuals pushing for this emergence. Or at least there weren’t already liberal institutions to create the virtues among the relevant virtues among the individuals who created the liberal institutions. Kant, in particular, recognized the possibility that institutions may improve morally without a prior moral improvement among citizens. So, the case has not been made that neutral institutions must be unstable. He ended by suggesting that the state has obligations of neutrality in its domain, but so do other institutions in their own limited domains.

Ruwan Ogien then presented “A Minimalist Justification for State Neutrality.” Although most discussions are concerned with neutrality among controversial conceptions of the good, we should still remain neutral even if there is empirical agreement on (part of) a conception of the good in a given society. Where there is controversy, neutrality can be based on the need to preserve peace and stability. The case for neutrality is harder where there is agreement. The fundamental case for neutrality rests not on a contrast between the right and the good, but between what we do to ourselves and what we do to others. Sometimes it is argued that attempts to promote a specific conception of the good will be self-defeating, because in order for a life to be good, it must be freely chosen. But this assumption is itself a controversial assumption about the nature of the good life. More narrowly, however, it is self-defeating to try to force someone to live an autonomous life. Mill himself defended the Harm Principle on utilitarian grounds, but in fact, it is older – part of the French Declaration of the Rights of Man, and Raz defends it on autonomy grounds. So, the Harm Principle is something like a fixed point in our common sense morality and it can be given different justifications.

Richard Arneson criticized neutrality in “Neutrality and Political Liberalism.” He began by questioning the alleged asymmetry between controversies about the nature of the good, which requires neutrality, and the treatment of controversies about the nature of the right, which does not. An explanation for this asymmetry, perhaps, is that it is only reasonable controversy about the good which disqualifies the use of specific conceptions of the good. But is there reasonable disagreement about whether “same-sex sex” is good? Empirically, there would seem to be, but it doesn’t seem reasonable. If we attempt to base our defense of same-sex sex on a general right to autonomy, this would require the promotion of equal opportunity for sexual goods. If the state can promote racial harmony – and equal opportunity along this dimension – it can also promote same-sex tolerance and equal opportunity along that dimension. But this involves declaring it to be good (or at least not bad).

Political Liberalism fails as an interpretation of neutrality because affirming a particular comprehensive doctrine while recognizing others as rational and reasonable is incoherent. Or at least it would be incoherent once the reasons for and against it are shared and understood. And insofar as the best evidence available is inconclusive, we should be indifferent or unsure. If we relax our standards and require not full rationality but some range of normal human imperfections – so that we require a person to be merely “somewhat responsive to reasons” – then there is no hope of achieving an overlapping consensus. If we have a broad and inclusive account of “reasonable,” we’re not going to be able to generate any specific principles or policies. If we have a narrow and demanding account of “reasonable,” we’re not going to have divergent comprehensive doctrines and we will not be neutral among conceptions of the good.

The next speaker was Alan Patten, “Religious Accommodation and Religious Neutrality: Accommodating Conscience.” Liberty of conscience certainly includes liberty of belief, but what about acting on those beliefs? Clearly two extreme positions are inadequate: liberty of conscience is not only about beliefs and not actions, nor does it require tolerating any and all actions. Contrast two models for drawing the limit. The priority view says that restrictions on conscientious action are acceptable only if there are strong reasons for doing so. The formal view holds that restrictions are permissible only if the restrictions are neutral (in aim) among conscientious positions and must not make accommodations to other comparable concerns. In other words, the formal view does not directly weigh the strength of reasons. The key type of case where they differ will be one in which a non-discriminatory policy conflicts with conscientious action that is adopted for neutral reasons, but these reasons aren’t very strong or important. In such a case, the priority view says that the restriction is not okay while the formal view says that it is. This tells in favor of the priority view, although recent Supreme Court decisions seem to favor the formal view.

The final speaker on the first day was George Crowder, “Neutrality and Liberal Pluralism.” Crowder developed an account of the relationship between value pluralism (along the lines of Berlin) and liberal neutrality. But this limited accommodation is compatible with a moderate perfectionism that also supports autonomy. The pluralist holds: 1. some moral values are objective and universal; 2. there are multiple and irreducible such values; 3. these values often conflict; 4. they are incommensurable. Such a position is not relativism, which denies the universality of any values. How is rational choice (and really the issue seems to be the choice of social policies rather than individual decision-making) possible? Three models: 1. subjectivist – choice is non-rational, so we get a kind of agonistic politics (rather than looking for a consensus); 2. contextualist – rational choices can be made given a context, which can be understood as a culture – this suggests a kind of conservatism; 3. conceptual – value pluralism itself may give reasons for liberalism.

The conceptual account goes as follows: 1. respect for plurality of the goods requires that we recognize the goods that could contribute to a good life, even if we cannot accommodate them all and regret those that we do not realize. 2. although we cannot promote all values (since many are incompatible), there is a presumption in favor of inclusion – liberalism is not neutral in this regard but promotes diversity. 3. there is room for reasonable disagreement since pluralism holds that there is a diversity of reasonable rankings of goods. 4. to choose well under pluralism is to think for ourselves and not simply to rely on tradition or other rules. Autonomy is not just one more of the pluralistic values, on this account, because it follows from the basic commitment to respect the plurality of values. Although autonomy is not always given priority over other values, there is a strong (but rebuttable) presumption in favor of it.

The second day started with Christine Sypnowich, “Human Flourishing: A New Approach to Equality.” The ultimate answer to the question “equality of what?” is flourishing. If people flourish equally, we don’t care what other inequalities there might be. On the one hand, liberals tend not to take flourishing seriously, while on the other hand, perfectionism seems largely uninterested in equality. We need to develop a form of egalitarian perfectionism. Historically this has not been an uncommon position and was exemplified by William Morris. Flourishing has three elements: 1. autonomy; 2. objectively valuable pursuits; 3. happiness (understood subjectively). It is too much to expect that we could equalize flourishing since, among other reasons, people just have different characters which will always result in different levels of happiness.

A major concern is people who have capabilities but choose not to realize them. Providing an unconditional social minimum may be disrespectful toward our capacity for self-discipline. Feeling that one is deserving is itself part of well-being, so if society picks up the tab for bad decisions, this may actually be harmful. Still, we should not assume that all forms of flourishing are productive. Surfing may be an example – think of it as a kind of art – even if surfers are missing something by not contributing to social production.

Another major concern is the charge of paternalism. But perfectionism need not be monolithic or coercive. In fact, coercion is incompatible with perfectionism since part of flourishing is autonomy. If people are coerced, they are not flourishing. Still, sometimes autonomy must give way to objective content. Market relations, in particular, are not neutral but promote certain values, often incompatible with the objective goods of flourishing. Social policies always encourage some values over others, so they should be guided by a proper account of flourishing.

During discussion, Daniel Weinstock clarified an earlier remark of his about the relative availability of pop music (“Britney Spears”) and classical music (“Benjamin Britton”). On his view, it makes sense for the state to subsidize classical music so that people have a genuine choice regarding what to listen to. Sypnowich replied dismissively that this is what she would expect from a liberal – the real reason we don’t subsidize polka music (her example) on the radio is because “it’s crap.”

Next was Andrew Lister, “Public Reason, Liberal Neutrality, and (Same-Sex) Marriage.” Opponents of same-sex marriage say that it is incompatible with God’s will. Sandel distinguishes two replies – the “naïve” view engages with and denies the claim directly, while the “sophisticated” view says that God’s will, whatever it is, is irrelevant to social policy. But marriage is itself a public affirmation of the value of certain type of relationship. Is it possible to defend marriage on the basis of public reasons at all? After all, marriage is not only a private contract – it can impose social costs, for example, in granting immunity from testimony against a spouse. It is possible to make an instrumental argument based on public reason for a two-person family – it serves children better than alternatives, for example. But the empirical grounds for this claim are not so clear. And even if it is true, its truth depends on other social conditions, such as the lack of availability of socialized child-rearing. Any attempt to specify an account of public reason in terms of “reasonable acceptability” will have to specify the precise conditions under which acceptability is assessed.

Steven Wall attempted to develop a limited neutrality within a broader perfectionism in “Pluralistic Perfectionism and Restricted Neutrality.” Like Crowder, he began from an account of value pluralism. If reasons run out in selecting among some values, there will be no rational grounds for choice among models of a good life A, B, and C, and some people may opt for each. Perfectionists allow the state to promotion the (or a) good life. But perhaps the state should be neutral among A, B, and C since each is a form of the good life and exemplifies different and competing values. Restricted neutrality allows the state to promote good lives over mistaken or bad ones, even if the mistaken views have adherents. This account is a non-neutral defense of neutrality. But why be neutral among A, B, and C? If the state were to favor one over the others, it would be arbitrary and this would be unjust discrimination. But favoring these over a mistaken conception D would not be arbitrary or unjust. As Rawls emphasizes, self-worth is important and the social basis of it is important, as well. However, the self-worth of someone wrapped up in an unworthy way of life is problematic.

Stephen Macedo gave a broad defense of public reason against various criticisms in “Neutrality and Public Reason.” He began with a quote from Barak Obama (from 2006) which was remarkably close to Rawls’s idea of “inclusive public reason.” Although the idea of public reason has liberal credentials, fundamentally it is democratic – it’s a way for a polity to engage in collective deliberation and decision-making. Adherents of religion may feel excluded – that the society doesn’t respond to their sincerely felt beliefs in the way that they would want – but this is simply because of the fact of diversity. Public reason does not lead to “sanitized” philosophical argument – there remains plenty of room for the “rough and tumble” of ordinary politics. And there are good grounds for giving special protection to the process of political deliberation itself. A political community is a special kind of community with its own duties and norms, and it is possible that these norms may conflict with those of other communities. There will always be a potential tension between the political norms of a diverse society and the norms of the various religious communities. And in some cases, individuals will follow the norms of their particular communities rather than the political norms.

Anthony Appiah apparently cut short in the interest of time his discussion of “Expressive and Instrumental Neutrality.” We should think of neutrality as an expressive ideal. It prohibits the state from favoring people on the basis of their identification with some group. For example, left-handed people are disadvantaged by some state policies. But this is not due to hostility toward members of that group so is not a violation of neutrality. Rather, it is based on the need for uniformity in certain matters, which is then left up to majority to decide based on considerations of efficiency. So it is unlike a policy that involves discrimination against a group on the basis of their identification with it. Some, like Nagel, argue that neutrality is required where(ever) there is coercion. But this is inadequate. Consider a proposal that bans Sikhs from wearing turbans because they have been subject to bigoted attacks when they do. The protection of Sikhs from attacks is a good public reason and their desire to wear a turban is not based on a public reason. Still, it would be wrong to ban the wearing of a turban on this ground. Toleration is required because of the requirements of equal respect – and this is a matter of expressive neutrality.

Lastly, Joe Heath presented “Political Egalitarianism” – with “political” understood in Rawls’s sense. It is surprisingly common to see appeals to egalitarianism that obviously go far beyond the limits of “the political.” Of course, sometimes this is done as part of a self-conscious effort to generate a comprehensive or partially comprehensive moral doctrine. But it is also common in political discussions. Luck egalitarianism is an example of this. As soon as we start to take seriously the demand for a political answer to the question “equality of what?” we are pushed in the direction of relying on an account of welfare. Rawls (apparently) bases his choice for “narrow scope” egalitarianism – in which only the benefits of cooperation are the object of egalitarian distribution – over “wide scope” egalitarianism – in which the accidents of birth are also compensated for – on the strains of commitment argument. Nobody should be expected to sacrifice everything (or too much) for others. This leaves him vulnerable to Cohen’s reply that we force people to sacrifice for each other all the time. A better defense of narrow scope egalitarianism emphasizes the need for a political argument. The problem isn’t the difficulty of motivating people to sacrifice for one another. It is the justificatory burden in a political context in arguing for wide scope egalitarianism. This type of narrow scope egalitarianism is compatible with endorsing a threshold conception (non-egalitarian) for dealing with radically different capacities.