Before today, it had never really occurred to me to wonder what Hart had made of Nozick. The answer, according to a 1976 address collected in Hart’s Essays in Jurisprudence and Philosophy, seems to be “not a lot”. Hart cites, with qualified approval, Bentham’s attack on the notion of absolute and inalienable rights as making any form of government impossible:
nothing that was ever called government ever was or could be in any instance exercised save at the expence of one or other of those rights … in as many instances as Government is ever exercised some one or other of these pretended unalienable rights is alienated
In Anarchy, State and Utopia, Hart argues,
Nozick raises precisely Bentham’s question and asks “How much room do individual rights leave for government?” What is astonishing is that Nozick also gives Bentham’s answer: No room except in an imaginary world. … he argues that granted a set of natural rights – such as the right not to be killed, assaulted, coerced, not to have property taken or destroyed, and not to be limited in the use of property – only a minimal form of state, the so-called “nightwatchman” state, whose functions are limited to the punishment of violations of such rights, can be legitimate. Moreover, given those natural rights, even that minimal form of state could be justified only under conditions which Bentham never considered. Yet Bentham might be forgiven for failing to do so, for they are conditions produced out of Nozick’s lively imagination which are highly unlikely to be satisfied in the real as contrasted with the imaginary world.
The conditions in question are that the state should have arisen through individuals voluntarily joining a private protection association which might eventually achieve, without infringing any natural rights, dominance in a limited territory even if not everyone joined it. But all this seems indeed imaginary and and irrelevant in a world where states do not arise in that way.
Bentham was wrong, Hart argues, to dismiss all talk of rights as either utopian or trivial, and irrelevant either way – rights being either overriding constraints (which cannot possibly obtain in the real world) or interests to be balanced against others (and hence deserving no special consideration). But, Hart concludes,
I do not think we yet have a satisfactory theory showing how respect for such rights is to be combined with the pursuit of other values. Some theories seem to me to throw out the baby – that is basic rights compatible with each other and with government – with the bathwater of excessive rigidity. Other theories – perhaps Professor Nozick’s among them – do worse; they throw out the baby and keep the bathwater.
Perhaps it’s unfair to castigate Nozick for producing a model that was both utopian and inapplicable to the real world. It could equally be argued that, once having identified principles of justice from which an ideal model of society could be elaborated, declaring that the model was in fact a map of the world would be an anti-climax at best. There would be more critical mileage in going the whole utopian hog, the better to measure the distance between that model and our world as it is.
Ultimately this passage is probably less valuable as a knock-down critique of Nozick than as a demonstration of the importance of one’s starting-point. For Hart, a model of justice was first and foremost a model of justice as it was administered in the real world: if such a theory pointed us in the direction of greater, less compromised or better-distributed justice, so much the better, but its first hurdle was to fit the reality of justice as we knew it. In Hart’s view, by defining justice in terms of principles which could only be realised in Utopia, Nozick had succeeded only in severing his own ideal of justice from the common-or-garden justice about which other theorists wrote. There is a parallel here with Hart’s insistence on the existence of law in slave-holding regimes or in Nazi Germany (although these are separate arguments, clearly; Hart didn’t deny that those regimes were characterised by great injustice). For Hart, a theory of law which only covered the law in peaceful, egalitarian, democratic regimes – while excluding other contexts in which people recognised as lawyers practised something recognised as law – would not be stronger but weaker for it: any gain in coherence would be decisively outweighed by the loss of breadth.
Unlike Nozick, Hart didn’t define justice in terms of a more-or-less unrealised end-state – or, perhaps it would be more accurate to say, he wasn’t very interested in looking at justice in those terms. Law, still less: he didn’t think in terms of law as its own self-description and its own regulative ideal – law consisting at once of the carrying-out of law-like practices in law-like ways, and as an ideal which was the more fully realised the more law-like that process was. In other words, he didn’t see law (or didn’t find it interesting or useful to see law) as providing the resources for its own immanent critique. But more on that another time.