“ Ideal Theory ” as Ideology

individual subsume the workers, women, and nonwhites who are also persons—even if, admittedly, they were not historically recognized as such? I think the problem here is a failure to appreciate the nature and magnitude of the obstacles to the cognitive rethinking required, and the mistaken move— especially easy for analytic philosophers, used to the effortless manipulation of variables, the shifting about of p’s and q’s, in the frictionless plane (redux!) of symbolic logic—from the ease of logical implication to the actual inferential patterns of human cognizers who have been socialized by these systems of domination. (This failure is itself, re" exively, a manifestation of the idealism of ideal theory.) To begin with the obvious empirical objection: if it were as easy as all that, just a matter of modus ponens or some other simple logical rule, then why was it so hard to do? If it were obvious that women were equal moral persons, meant to be fully included in the variable “men,” then why was it not obvious to virtually every male political philosopher and ethicist up to a few decades ago? Why has liberalism, supposedly committed to normative equality and a foundational opposition to ascriptive hierarchy, found it so easy to exclude women and nonwhites from its egalitarian promise? The actual working of human cognitive processes, as manifested in the sexism and sometimes racism of such leading ! gures in the canon as Plato, Aristotle, Aquinas, Hobbes, Hume, Locke, Rousseau, Kant, Hegel, and the rest, itself constitutes the simplest illustration of the mistakenness of such an analysis. Moreover, it is another familiar criticism from feminism that the inclusion of women cannot be a merely terminological gender neutrality, just adding and stirring, but requires a rethinking of what, say, equal rights and freedoms will require in the context of female subordination. Susan Moller Okin argued years ago that once one examines the real-life family, it becomes obvious that women’s exit options from marriage are far more restricted than men’s, because of the handicaps of sacri! cing one’s career to childrearing (Okin 1989). So a commitment to fairness, equal rights, and justice in the family arguably requires special measures to compensate for these burdens, and reform social structures accordingly. But such measures cannot be spun out, a priori, from the concept of equality as such (and certainly they cannot be generated on the basis of assuming the ideal family, as Rawls did in A Theory of Justice). Rather, they require empirical input and an awareness of how the real-life, nonideal family actually works. But insofar as such input is crucial and guides theory (which is why it’s incorrect to see this as just “applied” ethics), the theory ceases to be ideal. So either ideal theory includes the previously excluded in a purely nominal way, which would be a purely formal rather than substantive inclusion, or—to the extent that it does make the dynamic of oppression central and theory-guiding—it is doing nonideal theory without calling it such. (Compare the conservative appeal to a super! cially fair “color-blindness” in the treatment of people of color, whose practical effect is to guarantee a blindness to the distinctive measures required to redress and overcome the legacy of white supremacy.) Charles W. Mills 179 Similarly, it cannot be claimed that from the possibility of the extension of ideal theory to previously excluded populations that this shows the ideal theory is really not exclusionary. The extension (at least in a society where these populations are subordinated, so that hegemonic concepts and argumentative patterns have accommodated to their subordination) is precisely what requires the work, and marks the transition out of the realm of the ideal. If Kant says all persons should be treated with respect, but arguably de! nes his terms so that being male is a prerequisite for full personhood (Schröder 1997), it is not a minor change to remove this restriction. A Kantian polity where women can only be passive citizens and a polity where this stipulation is removed are not the same: the latter is not “contained” in the former as a potential waiting to be realized. When Okin uses the original position, a Rawlsian construct, to take the nonideal family into account from behind the veil, the result is not (somehow) Rawls’s “real” view—certainly not the Rawls who did not even mention sex as something you do not know behind the veil! What is doing the work are the real “general facts about human society”—the nonideal facts about gender subordination that Rawls apparently did not know. Nor, as I have had occasion to observe elsewhere, did either he or his followers apparently know the nonideal facts about imperialism, slavery, Jim Crow, segregation, and their ilk, that have shaped the United States and the modern world so profoundly and that constitute an ongoing and central injustice yet to be tackled by Rawlsians. How is this possible? Haven’t they noticed they’re living in one of the most race-conscious societies in the world, with a history of hundreds of years of white supremacy? Again, how can one resist the obvious conclusion that it is the factand reality-avoidance of ideal theory that underwrites such ignorance? In A Theory of Justice, as earlier cited, Rawls argues for ideal theory on the grounds that while the injustices of partial compliance are the “pressing and urgent matters,” we need to begin with ideal theory as “the only basis for the systematic grasp of these more pressing problems” (8). But then why, in the thirty-plus years up to his death, was he still at the beginning? Why was this promised shift of theoretical attention endlessly deferred, not just in his own writings but in the vast majority of his followers? My colleague Tony Laden has given me the following two remarkable pieces of information: (1) in a 1999 ! ve-volume collection of eighty-eight essays from three decades of writing on Rawls (Richardson and Weithman, 1999), only one of the included essays deals with race, that being an article by the African-American philosopher Laurence Thomas; (2) in the recently published Cambridge Companion to Rawls (Freeman 2003), not a single one of the fourteen chapters has race as either a theme or even a subtheme. What does this say about the evasions of ideal theory? Is it that the United States has long since achieved racial justice, so there is no need to theorize it? Or consider another example, where the opening for a discussion of race is actually explicitly part of the text, rather than perennially postponed to