Vol. 9, No. 2; Summer 2003
The Politics of Good Fortune
Justice, Luck, and Knowledge. By S. L. Hurley. Harvard University Press, 2003. Viii + 341 pgs.
Susan Hurley has written a book of fundamental importance. Although she is by no means a libertarian, and uses no distinctively libertarian assumptions, she eviscerates the egalitarian theories most influential in contemporary political philosophy. Not content with her critical triumph, she advances a new approach to justice; but this fails to break sufficiently with the egalitarian theories whose customary rationale she has challenged .
A line of thought present in John Rawls's A Theory of Justice (1971) has shaped much of the subsequent discussion of distributive justice. Rawls claimed that people do not deserve to benefit from their natural assets. Michael Jordan possesses enormously more skill at playing basketball than I do; but why should this unfortunate state of affairs enable him to earn an income somewhat higher than mine?
But why should we not profit from our talents? Why do we not deserve them? And even if, in some plausible sense, we do not deserve them, why are we not entitled to benefit from them? One way of understanding Rawls's answer is this: our natural abilities, whether good or bad, are the product of luck. We are not responsible for them and hence should neither benefit nor suffer from them.
As Hurley notes, a problem arises if one reads Rawls in this way. This interpretation makes responsibility central to justice, but Rawls sometimes denies that concepts of responsibility play a central role in the theory of justice: "If you judge that no one is responsible for his natural assets, you make a negative judgment, true. But it is still a judgment about responsibility. . . If significant consequences for distributive justice flow from such negative applications, then this concept does indeed play a fundamental role in the theory of justice." (p.135)
Later authors who have been influenced by Rawls, including G.A. Cohen, Ronald Dworkin, and John Roemer, manifest no such ambivalence. Here the aim to cancel luck assumes central importance. Hurley raises against these theories a devastating question. These authors are egalitarians: like Rawls, they reject the free market in large part because it permits large disparities of wealth to arise. But why should we assume that attempting to counter luck has anything to do with promoting equality?
At first sight, Hurley's challenge seems absurd. Let us return to the sad case of Michael Jordan and me. Suppose one grants that Jordan is not responsible for his superior athletic ability. If one's aim is to "correct for luck", is it not obvious that Jordan should have to give me a large part of his money, so long as no other people are taken into account?
Not at all, Hurley argues. We start with the premise that the existing distribution of wealth between Jordan and me arises from luck. This premise does not imply that some other distribution would manifest the influence of luck to a lesser extent. Suppose that someone divides between us the sum of Jordan's and my assets on an exactly equal basis. Why would this egalitarian distribution count as one that reduced the influence of luck?
Hurley states her point with characteristic precision: "Equalities can be just as much a matter of luck as inequalities. The fact that people are not responsible for a difference does not entail that they are responsible for nondifference. There is no more a priori reason to assume that equalities are not a matter of luck than there is to assume that differences are not a matter of luck; people may not be responsible for either."(pp.151-152)
But, one is inclined to object, is not Hurley ignoring an obvious fact about the way we speak of people as being lucky and unlucky? Surely it is Jordan, and not I , who is lucky, given the immense sums of money his talent enables him to command. Why does Hurley raise so much fuss about what is involved in correcting for luck? A redistribution from him to me takes from a lucky person and gives to an unlucky one; what is the problem?
Our author readily acknowledges that this is a legitimate way to speak of luck. But on this conception, being lucky means having more, or much more, than others. It assumes that one starts from a position of equality and judges departures from that position to be due to luck. But then one does not have an argument that inequality results from luck: one has simply made this true by definition. Hurley concludes that the "aim to neutralize interpersonal bad luck begs the question of justification and just helps itself to the goal of equality."(p.157)
Hurley distinguishes another sense of responsibility, which she terms the counterfactual. "In the counterfactual reading, I compare my actual situation with other possible situations I might have been in. I have bad luck when what I have is a matter of luck and I am worse off than I might have been."(p.156) It is this sense that underlies Hurley's claim that correcting for luck leads by itself to no egalitarian outcome.
I can here give but a sample of her subtle discussion: "It is hard enough to say whether people are responsible for what they actually have. But to neutralize luck understood counterfactually, we need to know more than this. We would have to be able to say, when people are not responsible for what they actually have, what they would be responsible for instead, if factors for which they are not responsible were eliminated. . . it is highly doubtful that we have any general, nonarbitrary basis for answering this further question. In many cases the answer is simply indeterminate."(p.162)
Hurley has an approach of her own that she proposes to substitute for the version of egalitarianism she has so effectively dispatched. To my mind, she gets off to a good start. She favors a cognitive view, in which the aim is to discover the true principles of justice. She rejects conventionalism, in which the sole question about justice is the rules on which people can be expected to agree. "Cognitivism in political philosophy is not a doctrine or thesis, but a category. Cognitivist accounts are cast primarily in terms of truth and knowledge rather than choice or preference."(p.256)
We wish, the, to attain truth about justice; how are we to proceed? As Hurley sees matters, we should try to eliminate factors that might bias our decisions. If Michael Jordan, to return to him yet again, were to deliberate on principles of justice while knowing of his superior abilities, might he not be tempted to select a view that would benefit people like him?
To eliminate bias, Hurley suggests, we must imagine ourselves in something like Rawls's original position. We assume, like Rawls, that we lack any information about our abilities or preferences. Will not the principles we choose under such conditions be as immune as we can make it from the influence of bias?
Hurley's specification of her original position follows Rawls closely at a crucial juncture. John Harsanyi and other philosophers of a utilitarian bent agree with Rawls about the imperative need to avoid bias. But, they say, why assume that people are completely ignorant about which social position they will occupy, once the veil of ignorance is lifted? Why not, instead, assume that each person has an equal chance of winding up in any position in society? On this assumption, utilitarians argue, people will choose to maximize average utility rather than conform to Rawls's ideas.
Hurley will have none of this. The utilitarian approach just sketched once more injects bias into our deliberations. "[T]he aim to neutralize also argues against the idea of deciding about justice on the basis of calculations of your chances of gain. But this point holds even if these calculations derive from an assumption that everyone has equal chances, Biases can distort beliefs even if they apply equally to everyone."(p.264, emphasis removed)
The suspicion of bias, then, must at all costs be avoided; and to help secure this aim, deliberators are to kept ignorant of key facts. Given this state of affairs, the way in which Hurley arrives at her principles of justice strikes me as more than a little surprising. It transpires that people do not want to be ignorant: they are at least moderately "averse to uncertainty". They want to know, though this preference is not absolute, what everyone will wind up getting.
People want to know what everyone will get; but, owing to the need to curtail bias, they cannot know particular facts about themselves. How are these imperatives, seemingly in conflict, to be reconciled? In essence, Hurley maintains that people will favor equal distribution of the "basic goods" needed to flourish. Under equal distribution, people will not be left in the dark as to what they will get. By contrast, in a system where "you get what you earn", people will not know how they will fare, since by hypothesis they do not know their places in society. Hence people will prefer equality to a meritocratic rule of distribution. Since their aversion to uncertainty is not absolute, though, they will allow inequalities that increase the total stock of goods, "so long as the level of the worst off is kept as high as possible"(p.272). We arrive at something like Rawls's difference principle. This outstanding critic of egalitarian theories has, at least to her own satisfaction, vindicated a strongly egalitarian view.
I cannot think that Hurley's ingenious labors have resulted in an acceptable account of justice. She seems to me correct that many people are moderately averse to uncertainty. But this is, for all she has shown, a mere preference; she has not argued that people ought, in an objective sense, to shun uncertainty.
What then has happened to her cognitive program? Hurley has given us no reason to think that her theory of justice is true. Rather, she introduces certain conditions designed to make the search for truth as unbiased as possible and proceeds to abandon the quest altogether. Why should one think that what people will prefer under conditions of uncertainty will give us truth?
Hurley's entire approach strikes me as misguided. If we wish to find out the truth about justice, why not proceed as we do with any other philosophical problem, i.e., adduce relevant arguments? When, in a part of her book I have had here to neglect, Hurley discusses issues of responsibility, she follows just this course. She considers various arguments that philosophers have raised about the topic, and arrives at her own conclusions. She does not endeavor to set up special conditions under which analysis of responsibility can take place in an unbiased fashion. Why not then in like fashion reason directly about justice?
Hurley deserves great credit for her outstanding critical account of egalitarian theories that aim to counteract luck. Murray Rothbard grasped her key insight many years ago. In Power and Market, published in 1970, he writes: "[T]here is no justification for saying that the rich are luckier than the poor. It might very well be that many or most of the rich have been unlucky and are getting less than their true DMVP[ discounted marginal value product], while most of the poor have been lucky and are getting more. No one can say what the distribution of luck is; hence there is no justification here for a 'redistribution' policy."(pp.234-235)
Robert Nozick thought very highly of Hurley’s earlier book, Natural Reasons (1989); but, in commenting on her later work in philosophy of mind, he also noted that she tends to express herself with unnecessary complexity. This problem, I regret to say, is present in Justice, Luck, and Knowledge.
Rawls’s contention may have come from Frank Knight. See my discussion in The Mises Review, Fall 2000, pp.3-4.