Review - Justice, Luck, and Knowledge
by S. L. Hurley
Harvard University Press, 2003
Review by Mark Rigstad, Ph.D.
Aug 24th 2005 (Volume 9, Issue 34)

In her latest book, Justice, Luck, and Knowledge, Susan Hurley has done a great service to ethical and political philosophy by exploring important connections between theories of responsibility and theories of distributive justice. She begins by defending particular conceptions of responsibility and luck, and then proceeds to examine the roles which such notions can and cannot play within egalitarian political philosophy. Many philosophers -- let's call them constructivists about responsibility -- will think this way of proceeding wrongheaded on grounds that what responsibility essentially is ultimately depends upon the social norms, including principles of distributive justice, according to which we hold people responsible for what they do. Yet constructivists who follow Hurley's line of thought from start to finish may be compelled to revise or restate their assumptions. It is in many parts a difficult book to understand, but well worth the effort. 

Following the seminal work of Harry Frankfurt, Hurley rejects the notion that responsibility depends upon an actor's ability to do something other than what she actually does. Frankfurt famously discredits the notion that alternate sequences of possible action must be available to responsible actors by constructing hypothetical cases in which some outside agent or force is prepared to guarantee that Actor will do X, though Actor ends up doing X by choice, without outside intervention. In such cases, Actor could not have done otherwise, but is responsible for X just the same. Hurley moves beyond Frankfurt-style science fiction scenarios involving such "counterfactual intervenors" and draws upon intuitions about, for example, weakness of the will and action in a deterministic world in order to lend further support to the conviction that responsibility does not depend upon the ability to do otherwise. The upshot is that conditions of responsibility reside, instead, in the actual sequence of causes that determine action.

Hurley presents an exceedingly lucid account of actual sequence theories of responsibility. She is especially concerned to show the error of supposing that Actor's responsibility for X requires not only choice or control of X but choice or control of the actual causes of X. This "regression requirement" would make responsibility impossible because the actual causes of X will invariably extend back beyond Actor's lifetime. Either (1) responsibility is regressive and therefore impossible, or else (2) it is possible but not regressive. In defending option (2), Hurley devotes her most methodologically trenchant chapter (chapter 3) to a thorough and cogent critique of the eliminativist alternative favored, for example, by Galen Strawson. In her positive view, responsibility is possible, even in a deterministic world, if it is conceived (following Susan Wolf and John Martin Fischer) as hinging upon Actor's responsiveness to reasons rather than her capacity to control regressive causal sequences.

More philosophically tentative is Hurley's "thin" conception of luck as the "inverse correlate" of responsibility, which she defends against the thicker senses of the term employed by Bernard Williams and Thomas Nagel in their accounts of "moral luck." If actions and outcomes due to luck are by definition those for which Actor is not responsible, then the very idea of responsibility for matters of luck obviously makes no sense. We do not, however, ordinarily think of luck simply as whatever does not result from responsiveness to reasons; and Hurley's plea for conceptual tidiness is unlikely to sway philosophers with even modest ordinary language leanings.

Although Hurley's special sense of 'luck' avoids the ambiguities of ordinary language, it generates the confusion of conceptual cross-purposes. The second half of Justice, Luck, and Knowledge is a sustained argument against the "luck-neutralizing egalitarianism" of Gerald Cohen and John Roemer. Yet Roemer's theory of equal opportunity, for instance, only purports to neutralize 'luck' in a very different sense of the term. It would take considerable unraveling to figure out how much of what Hurley says about 'luck' in her special sense impugns what Roemer says about 'luck' in his special sense. Hurley's conceptions of responsibility and luck are given prior to substantive principles of distributive justice, and she is at pains to argue that no such principles can be derived from them or explicated by means of them. "Giving people what they are responsible for is what would neutralize luck, not giving them something else they are not responsible for. Therefore, there is no inherent connection between equality and neutralizing luck" (153). Roemer constructs equal opportunity as a principle which aims to reward effort. Inequalities that are not due to unequal efforts are then "defined" as being due to luck. Thus, to reward people in proportion to their efforts is to give them something they deserve which, in the stipulated sense, is not a matter of luck. It is true, of course, that Roemer "does not show that someone's position after redistribution would not be a matter of luck for her" (195 my emphasis). For Roemer, this judgment is stipulated and constructed out of a substantive principle of distributive justice.

Conceptual cross-purposes aside, Hurley may be on to something. It does seem wrongheaded to suppose -- as Gerald Cohen supposes in a statement which Hurley takes to be emblematic of luck-neutralizing egalitarianism -- that "anyone who thinks that initial advantage and inherent capacity are unjust distributors thinks so because he believes that they make a person's fate depend too much on sheer luck" (141). The practice of determining distributions in accordance with luck, on the basis of a coin toss for example, is sometimes ideally fair, especially when the affected parties are relative equals meeting on a level playing field. From a Rawlsian perspective, what makes native endowments an unfair distributor is not the fact that they are products of luck but the fact that they mark initial positions of relative inequality. And our intuitive sense of what constitutes an ideally fair situation from which to determine the basic terms of social cooperation is one in which parties are conceived as equals.

Yet, Hurley also claims to detect a luck-neutralizing theme in John Rawls' A Theory of Justice. After all, Rawls famously states that it is "one of the fixed points of our considered judgments that no one deserves his place in the distribution of native endowments any more than one deserves one's initial starting point in society." It would take more showing to convince the reader that Rawls' theory of justice as fairness aims to neutralize the element of luck in general. Hurley is right to point out, however, that this important fixed point in Rawls' theory gives the "concept" of responsibility a basic role in explaining why it is appropriate to cast a veil of ignorance over one's native endowments when thinking about distributive fairness. But it is not clear that Rawls needs to draw upon any particular conception or philosophical doctrine of responsibility at this point in the construction of his theory. Even the radically regressive conceptions that make responsibility impossible would support Rawls' intuition that we are not responsible for how we initially enter into the world. The positive account of the conditions of responsibility that would flesh out Rawls' conception of justice as fairness would be constructed in terms of principles of the rule of law (Rawls' gestures towards H. L. A. Hart's Punishment and Responsibility, which Hurley does not engage). In short, Rawls conceives of responsibility as accountability, which is political, not metaphysical. Although Hurley is, for the most part, strangely silent about (or dismissive of) pragmatically, ethically or politically constructive approaches to responsibility, it is evident that her response to Rawlsian principles of responsibility would parallel her response to Williams' theory of moral luck. We may have pragmatic reasons for holding Actor accountable for certain matters of luck, but we can have no cognitive reasons for seeing her as responsible for them.

For Hurley, although a theory of responsibility cannot tell us how to distribute social goods, it can and ought to do more than serve as a fiction of legal compliance. In advancing her bias-neutralizing alternative to luck-neutralizing egalitarianisms, she makes a compelling case for supposing that a pre-institutional conception of responsibility must be delineated prior to the full epistemic justification of distributive principles. Some sense of what agents are responsible for must be presupposed in specifying what incentives they can reasonably expect as rewards for their productive efforts. Until this "incentive parameter" has been set, in part by means of a theory of responsibility, the social implications of any proposed principle of distributive justice cannot be adequately determined. This argument presents a significant challenge to constructivists, from Rawls to Scanlon, who suppose that what people can be said to deserve, on the basis of actions for which they are responsible, must await the determination of principles of distributive justice. In the new direction that Hurley envisions for the theory of distributive justice, the Rawlsian device of the veil of ignorance is retained as a way of filtering out biasing influences which undermine knowledge. But the element of Rawlian "voluntarism," which attempts to separate ethico-political justification from metaphysics, epistemology and philosophy of mind, is discarded. Political ethics can only be modestly autonomous because it can neither be built up from exogenous foundations nor insulated from other fields of knowledge.

Hurley's way of specifying the auxiliary role that a theory of responsibility should play within a "cognitivist" approach to distributive justice is sure to inspire new research which attempts to bring recent developments in the philosophy of neuroscience and cognitive science into contact with moral and political theory.

© 2005 Mark Rigstad

Mark Rigstad, Ph.D., Department of Philosophy, Oakland University, Rochester, Michigan.