Ratio Juris: The “Truncated Autonomy” of Raz’s Liberal Perfectionism

The most explicit and thorough statement of Joseph Raz’s intriguing theory of “liberal perfectionism” is found in his invaluable book, The Morality of Freedom (1986) wherein “the end of liberal political morality is the good of an autonomous moral life.” In Constitutional Goods (2004) Alan Brudner proffers a succinct formulation of Razian liberal perfectionism:

“Autonomy…is essential to well-being, since personally endorsing one’s goods is essential to one’s deriving fulfillment from pursuing and accomplishing them. Because, moreover, the achievement of autonomy depends on the existence of a public culture offering a wide array of valuable pursuits, it is a common good requiring state action to realize it.”

But Brudner identifies, by my lights, a significant shortcoming in Raz’s formalist conception of autonomy that leaves us with a rather attenuated form of perfectionism: “Raz argues that, although a morally valuable goal is one that truly conduces to well-being, an agent can find valuable only what is socially valued.” Brudner labels this the “social dependency thesis,” as it relies on a formalist rendering of autonomy effectively “empty of specific content” insofar as it is without criteria whereby to independently (i.e., autonomously) assess the notion of well-being. His critique is as follows:

“This thesis implies that the ideal of well-being is attainable only in a well-ordered society where the available options are indeed morally valuable, for only there are the conventional fetters to well-being overcome. However, even in a well-ordered society, the social dependency thesis can avoid moral conventionalism only if it is possible for an agent to value a goal because it is independently valuable and not simply because it is socially valued. Raz affirms this possibility. He argues that goals are adopted because they are thought to serve well-being, and it is possible to have true and false beliefs about whether they do because well-being is distinguished from contentment. Yet that there is a distinction between the valued and the valuable will not by itself save moral autonomy from the clutches of the social dependency thesis. An agent will not be able to distinguish a false social belief about the goals that conduce to well-being from a true one unless he or she has an independent conception of what well-being is—independent, that is, of social opinion. If she does, it is difficult to see why the social dependency thesis should be true; but if she does not, then she will be dependent on social opinion after all. Raz himself offers no such independent conception, or rather he offers a circular and decidedly unhelpful one. Well-being, he says, consists in the successful pursuit of morally valuable ends, and a morally valuable end is one that conduces to well-being. But then the agent, who can value only what is socially valued and who has no way of connecting socially valued ends to a non-circular criterion of the morally valuable is effectively left with a morality of convention.”

Thus, writes Brudner, “Raz’s perfectionism, in failing to generate a rational content and scheme of fundamental ends, is not perfectionist enough given its commitment to autonomy.” Or, in other words, a formalistic conception of autonomy leaves us with an attenuated perfectionism. This may be analogous to, if not perhaps in some ways connected with, Isaiah Berlin’s influential (or infamous) analysis of freedom, at least as discussed by Jon Elster in Sour Grapes: studies in the subversion of rationality (1983). Indeed, it might be the case that Raz is assuming or perhaps influenced by Berlin’s (or something akin to Berlin’s) understanding of freedom:

“What is freedom, freedom tout court, being a free man? We may distinguish between two extreme answers to this question. One is that freedom consists simply in being free to do what one wants to do, irrespective of the genesis of the wants. In a well-known passage Isaiah Berlin argues against this notion of freedom: ‘If degrees of freedom were a function of the satisfaction of desires, I could increase freedom as effectively by eliminating desires as by satisfying them; I could render men (including myself) free by conditioning them into losing the original desire which I have decided not to satisfy.’ And this, in his view, is unacceptable. By this argument Berlin was led to the other extreme in the spectrum of definitions of freedom: ‘It is the actual doors that are open that determine the extent of someone’s freedom, and not his own preferences.’ Freedom is measured by the number and importance of the doors and the extent to which they are open [one might say this is Berlin’s version of the ‘social dependency thesis’]. Disregarding the last clause, which appears to conflate formal freedom and real ability, this means that freedom is measured by the number and non-subjective importance of the things one is free to do [emphasis added]. True, Berlin suggests that the notion of importance should also take account of the centrality of freedoms to the individual, but this would seem to smuggle in preferences again, contrary to his main intention. Importance, in his view, must be divorced from the individual’s own evaluation of importance.”

It seems Raz’s notion of autonomy is predicated upon or at least endorses Berlin’s understanding of freedom although Raz is, to be sure, sensitive to a welfarist conception of the satisfaction of desires or the value of individual preferences. As Elster explains, the degree of freedom depends on the number and importance of the things one (i) is free to do and (ii) autonomously wants to do.” The social dependency thesis appears in the end to make (ii) wholly dependent on or a function of (i) such that, for example, it’s irrelevant that I may “live in a society that offers me a great many important opportunities, which do not at all overlap with what I want to do.” Raz needs a more robust or substantive conception of autonomy that is able to invoke, as it were, independent criteria of “the good” or “the good life” which, in turn, allow for truly free individual judgment and assessment. The social dependency thesis and corresponding lack of an independent criterion or criteria of the good (or self-fulfillment) or what Brudner terms a “rational content and scheme of fundamental ends,” preclude full appreciation of the fact that “it is better to desire things because they are desirable than to do so because they are available,” or at least precludes the would-be autonomous person from knowing whether or not her choices are wholly determined by the latter condition. Perhaps, for example, I have inordinate desires for consumption based on the (possibly false) belief that it promotes happiness or well-being: how do I know that my desires are not shaped by the fact that I live in a society which provides a plethora of opportunities for consumption, the socially regnant or widespread de facto or default belief being that a life of consumption is conducive to or promotes well-being?

As Elster writes elsewhere, liberalism, in our case Razian perfectionist liberalism, “neglects the endogeneity of preferences. Liberalism advocates the free choice of life-style, but it forgets that the choice is to a large extent preempted by the social environment in which people grow up and live,” a social environment today that is defined by the ethos of the latest form of capitalism.* We might consider the possibility that there exists a space that might be carved out betwixt and between the liberalism of a capitalist democracy compatible with the truncated autonomy of Razian perfectionism, and a dictatorially or unduly paternalist (i.e. externally) imposed conception of the good life. Elster argues this space can be defined by a form of individual and collective self-paternalism: “If people do not want to have the preferences they have, they can take steps—individually or collectively—to change them.” Does the apparent widespread preference for (or belief in the good of) a life of consumption rule out the possibility of generating truly independent or substantively autonomous conceptions, and the preference for same, of the good life (Marxist or otherwise)? What steps might we take to generate truly autonomous conceptions of the good life (and preferences for same) beyond conceptions of welfare, well-being or happiness yoked largely to a life of consumption (including aesthetic pleasures and entertainment as well as material goods in the ordinary sense)? What might religious worldviews contribute to our understanding of the contours of “the good life?”

*See Elster's chapter, “Self-realisation in work and politics; the Marxist conception of the good life,” in Jon Elster and Karl Ove Moene, eds. Alternatives to Capitalism. Cambridge, UK: Cambridge University Press, 1989.