Yet more on Hart and his 1973 paper on John Rawls’s A Theory of Justice. Hart put forward five queries; in this post I’ll be covering the fourth and fifth.
- Liberty or liberties?
- Restricting liberty for liberty’s sake
- Restricting liberty for harm’s sake The third post in this series focused on what seems to be a blind spot in Rawls’s argument: the idea that liberties may sometimes need to be limited for harm-related as well as liberty-related reasons.
- Choosing (more) liberty Rawls argues that subjects in the original position would, in their own interest, tend to choose more rather than less extensive liberties. Given the potential adverse effects of liberties extended to the whole of society, is this valid?
- The priority of liberty Following on from the previous point: Rawls appears to believe that, all else being equal, subjects in the original position would choose a quantum of liberty over a quantum of material benefit. Is this an unstated presupposition on Rawls’ part? How is our perception of his model affected if this is granted?
Choosing (more) liberty
Hart’s argument on the fourth point, above, hinges on the difference between enjoying a liberty and being affected by other people’s enjoyment of it – a consideration which may cast a different light on Rawls’s assumption that it is rational, from the standpoint of the original position, to want as large a share of liberty as possible. “Even if we assume with Rawls that every rational person would prefer as much liberty as he can get … it does not follow that a liberty which can only be obtained by an individual at the cost of its general distribution through society is one that a rational person would still want.” If the question is whether I want to be free to have a symphony orchestra in my back garden, the answer is Yes, of course – why wouldn’t I? If the question is whether I want my neighbours to have that same freedom, the answer is No. Or rather (crucially) the answer may be No. It may not: I may have a high tolerance for noise; I may have an overpowering fascination with the sound of a symphony orchestra, or an academic interest in discord. Different people will make different choices; in many cases, “whether it would be rational to prefer liberty at the cost of others having it too must depend on one’s temperament and desires” – which, of course, cannot be referred to in the original position.
There are two issues here. One is whether liberties can be generalised in any straightforward way, given that each person’s actions will have effects on other people: a generalised liberty-to-X is also a generalised right-to-protection-from-interference-in-Xing, or else a liberty-to-X-subject-to-interference (which in some situations will barely deserve the name of ‘liberty’). We’re in familiar Kantian or post-Kantian territory here: from my right to X we cannot simply read off a prohibition on some other action Y, even where your doing Y makes my Xing difficult or impossible. (Where X = ‘maintain my bodily integrity’ and Y = ‘swing your fists wildly’, the argument for liberty to be accompanied by protection works well enough; where X = ‘run my business as I think best’ and Y = ‘strike for higher pay’, however…) On Hart’s account Rawls has basically overlooked this.
The second issue is the irreducible fact of human variety. Hart sometimes seems to mean this in a relatively weak sense – preference for quiet vs tolerance of noise – which is vulnerable to Rawls’s ‘best worst case’ argument: since subjects in the original position wouldn’t know what preferences they had, they would work on the basis that they might prefer quiet (or privacy, harmony, etc) and reluctantly forgo the possible pleasures of noise (or intrusiveness, strife, etc). (Something interesting surfaces here in terms of the kind of values likely to be selected in the original position; I’ll return to this later.) However, Hart’s argument also takes a stronger form. When it comes to human variety, Hart stresses repeatedly – and I think correctly – that it applies at every level. I’ll discuss this below, when I come to consider Rawls’s answer to Hart on the second point (and on Hart’s fifth question).
On the first point, Rawls argues that the basic liberties themselves can do the job. Or rather, he argues that a society committed to the general effective exercise of the basic liberties would ipso facto be a society in which their exercise was limited. Not only is an extensive scheme of liberties the only justification for limiting individual liberties; it is, Rawls argues, the only thing necessary. He notes “that the basic liberties not only limit one another but they are also self-limiting”, adding that “[t]he notion of significance shows why this is so”. His explanation merits quoting at length:
while we might want to include in our freedom of (political) speech rights to the unimpeded access to public places and to the free use of social resources to express our political views, these extensions of our liberty, when granted to all, are so unworkable and socially divisive that they would actually greatly reduce the effective scope of freedom of speech. These consequences are recognized by delegates to a constitutional convention who are guided by the rational interest of the representative equal citizen in a fully adequate scheme of basic liberties. Thus, the delegates accept reasonable regulations relating to time and place, and the access to public facilities, always on a footing of equality. For the sake of the most significant liberties, they abandon any special claims to the free use of social resources. This enables them to establish the rules required to secure an effective scope for free political speech in the fundamental case.
The basic liberties are means to the end of securing a general freedom to co-operate reasonably (where reason implies justice) and to deliberate rationally (where rationality includes an idea of the good). Whether “rights to the unimpeded access to public places and to the free use of social resources to express our political views” would in fact prove to be unworkable and socially divisive is a secondary, political question (and one on which I’m inclined to disagree with Rawls). The point here is that the basic liberties are, so to speak, a weapon that can only be used for good: freedom of speech extended to the point where it undermines its own object ceases to be desirable and hence will, in Rawls’s model, be curtailed. The question of unintended consequences does not arise; the most extensive possible scheme of basic liberties is one which has no unintended consequences (any more extensive would be too extensive).
This is certainly an answer, but I’m not sure how useful it is. At this point Hart and Rawls seem to be talking past each other; certainly the question of how liberties could justly be restricted by and in a society of people committed to the most extensive possible scheme of basic liberties isn’t one that we can imagine detaining Hart for long. But we’re moving now onto the terrain of the second point above, and of the fifth question, which I’ll introduce here.
The priority of liberty
In the final section of his essay Hart comments on a curious detail of Rawls’s model: the ‘priority rule’. It is envisaged that, when subjects behind the veil of ignorance adopt a scheme of basic liberties, they do so on the understanding that the implementation of the scheme may be postponed until the material wealth of the society has developed sufficiently to support it. (I’m personally not sure about the assumption that a scheme of basic liberties requires a certain material substrate, but let’s take it as read.) When society’s material development reaches a certain point, Rawls argues, liberty (or liberties) will take priority: any quantum of material progress which might have been used to make unfree citizens a bit richer will instead be used to develop the institutions required to support the basic liberties. Now, the subjects behind the veil of ignorance have no idea what stage of material development their society has reached, so the question of when to invoke a priority rule – and, more importantly, whether to invoke it at all – will be a live one. Rawls compares different hypothetical social orders by reference to the ‘least worst’ position (in which society would the worst-off individual be least deprived?); in this case, Hart argues, we have a choice between A, an impoverished individual who would rather have a bit more money than the free institutions her society is intent on building (society with priority rule) and B, her counterpart who desires forbidden political freedoms even more than an escape from poverty (society with no priority rule). Rawls’s argument implies (according to Hart) that rational subjects in the original position would consistently think it worse to be B than A.
But is this the case – or rather (a more important as well as an easier question) is this self-evidently the case? Hart suggests not: “[w]hen the veil of ignorance is lifted some will prefer A to B and others B to A”. Rawls’s ideal subjects do not have the variability of human beings, in other words – variability which (as Hart insists) operates at every level. I myself prefer strawberries to raspberries, folk songs to pop songs, tranquillity to bustle, democratic accountability to executive efficiency, freedom of religion to compulsory observance, agnosticism to atheism, wisdom to knowledge, debate to certainty and a Fullerian liberal idealist reading of the rule of law to the disaggregative scepticism of contemporary legal positivism – but it’s perfectly legitimate to take the opposite position on any or all of these points, and plenty of people do. As the 25-year-old Karl Marx pointed out, the law works by treating unequal people equally, but it can only do so by taking a partial view – treating them from a particular angle, and an angle (Hart might have added) which may vary from one legal system to another. (On first consideration I thought that these two challenges – the problem of subjecting human variety to a uniform rule and the possibility of multiple approaches to doing so – were also a score against legal positivism, but on reflection that’s the wrong way round: these are questions which positivists leave unanswered, but the possibility of a coherent theory which leaves them unanswered is a point in positivism’s favour.)
If Rawls’s model relies on there being a single answer on which everyone can agree – at any level – that’s a score against it; at least, it suggests that there’s something different and more idealistic going on than Rawls appears to acknowledge. Hart:
I think the apparently dogmatic course of Rawls’s argument for the priority of liberty may be explained by the fact that, though he is not offering it merely as an ideal, he does harbour a latent ideal of his own, on which he tacitly draws when he represents the priority of liberty as a choice which the parties in the original position must, in their own interest, make as rational agents choosing from behind the veil of ignorance. The ideal is that of a public-spirited citizen who prizes political activity and service to others as among the chief goods of life and could not contemplate as tolerable an exchange of the opportunities for such activity for mere material goods or contentment. It is, of course, among the chief ideals of Liberalism, but Rawls’s argument for the priority of liberty purports to rest on interests, not on ideals, and to demonstrate that the general priority of liberty reflects a preference for liberty over other goods which every self-interested person who is rational would have.
We seem to have come a rather long way round to end up back with Schutz and the (social) scientist as puppeteer. One sign that there may be a bit more going on here is Hart’s use of the word ‘rational’ to refer to self-interested means/end rationality; as we have seen, the word has a specialised and somewhat teleological meaning for Rawls. So too does ‘reasonable’, a word which features heavily in Rawls’s response to Hart on this point. Rawls acknowledges that Hart was correct “in supposing that a conception of the person in some sense liberal underlies the argument for the priority of liberty”. But:
this conception is the altogether different conception of citizens as free and equal persons; and it does not enter justice as fairness by imputation to the parties. Rather, it enters through the constraints of the Reasonable imposed on the parties in the original position as well as in the revised account of primary goods. This conception of the person as free and equal also appears in the recognition by the parties that the persons they represent have the two moral powers … This conception of the person can be said to be liberal (in the sense of the philosophical doctrine) because it takes the capacity for social cooperation as fundamental and attributes to persons the two moral powers which make such cooperation possible.
The point is not that everyone in Rawls-world is good and public-spirited. The point is, rather, that Rawls assumes a society of free and equal persons, each of whom is capable of two things: social co-operation, subject to the demands of fairness and promise-keeping which can be called ‘reasonable'; and ethical deliberation, within the framework of logic and value which can be called ‘rational’. In terms of entry requirements for the world of his model, Rawls has set the bar surprisingly low. To derive the priority of liberty – or any other of Rawls’s apparently idealistic or counter-intuitive formulations – we may not need to assume a world of model citizens; perhaps all we need to do is to assume that everyone is capable of working together and valuing one set of ideas more highly than another, and then take those assumptions seriously.
Enough for now; I’ll conclude next time with a bit of summing-up and some thoughts on Rawls as libertarian, as bourgeois liberal, as conservative… as quite a variety of unpleasant names.