Revisiting Affirmative Action, With Help From Kant
Stanley Fish/New York Times blog, January 14, 2007
Whenever I teach the political writings of Immanuel Kant (1724-1804), I always ask my students, would Kant have been for or against affirmative action? I thought of the answers they typically give to that question when University of Michigan spokesperson Theresa A. Sullivan announced last week that the university would comply with Proposal 2, a successful state ballot initiative banning affirmative action programs that give preferential treatment to persons and groups on the basis of race, gender, color, ethnicity or national origin.
On Nov. 8, one day after Proposal 2 was approved by 58 percent of Michigan voters, President Mary Sue Coleman expressed disappointment at the result and vowed to continue the battle by every means possible. The university’s lawyers then requested that the force of the new proposal be stayed until the present admissions cycle was complete. But when a federal appeals court denied the request, the university bowed to the ruling while reserving its right to mount legal challenges to the new law.
No one believes that this is the end of the story. Debates about affirmative action have been going on since the concept was first introduced during the Nixon administration. And in fact the debate goes back at least as far as Kant and his ideas about how to tell the difference between principled and unprincipled policies. Kant says that correct political thinking must begin by affirming two propositions: 1) “the freedom of every member of society as a human being,” and 2) “the equality of each with all the others as a subject.” This emphasis on freedom and equality has led some of my students to conclude that Kant would have been in favor of affirmative action because, they reasoned, it was the denial of freedom and equality to African Americans that produced the injustices affirmative action is intended to redress.
That, I tell my students, is the wrong answer, because it confuses and conflates two aspirations Kant was concerned to keep separate: the achieving of results that many would think good, and acting in conformity with the moral law. In some philosophies – utilitarianism in some of its versions would be an example – morality and the bringing about of a desired social outcome (a more equal distribution of wealth, proportionate representation of minorities in positions of influence and power) would be one and the same. But Kant is, at least philosophically, indifferent to outcomes, in part because, as he puts it, “men have different views on the empirical end of happiness” – that is, different views about what society should look like and therefore different views about the policies that should be pursued.
A state dedicated to morality rather than to happiness will not take sides and choose one end before the others; rather it will protect the right of every man to choose the end he prefers, provided that he in turn accords the same right to his fellows. “Each may seek his happiness in whatever way he sees fit, so long as he does not infringe upon the freedom of others to pursue a similar end which can be reconciled with the freedom of others.” It is the abstract right rather than “the object in relation to which” it might be exercised, and the condition of freedom rather than any action freely performed, that Kant values. His interest is not in the particular life plan an autonomous citizen might wish to pursue, but in the ability of that citizen to pursue it without having the plan preferred by others imposed on him or realized at his expense.
Kant, in short, divorces morality from policy, and he makes the point by contrasting two maxims: “Honesty is the best policy,” and “Honesty is better than any policy.” The first maxim is a strategic recommendation. It says, when you want to accomplish something, you’ll have a better chance if you are honest. The second – and in Kant’s mind, superior – maxim takes no notice of strategy. It says, being true to your principles independently of the result they may or may not produce is the only moral way to go.
The danger the first maxim holds out is that at some point you might decide that your goal would be better served by a little deception and secrecy (just as some have said in recent years that the protection of democracy requires doing undemocratic things), and at that point morality will have given way to expediency and considerations of prudence. “The legislator may indeed err,” says Kant, “in judging whether or not the measures he adopts are prudent,” but he will not err in judging “whether or not the law harmonizes with the principle of right,” because “he has ready to hand an infallible a priori standard.” Honesty is to be followed not because it works or might work (a prudential judgment), but because it is right.
It is because Kant insists on distinguishing what works (at least in the short run) and what is right that he would, I believe, be against affirmative action. He would have said, as many opponents of affirmative action do say, that it is wrong to respond to past acts of discrimination by discriminating in the present, even if your intentions are good. If discrimination – the unequal treatment of inherently free and equal citizens – is to be condemned when the motives behind it (to preserve power or maintain a way of life) are suspect, it is also to be condemned when the motives behind it (to redress an historical injustice or have the student body reflect the diversity of America) are benign. Otherwise the calculation of happiness (at least by someone’s lights) will have taken precedence over the upholding of principle.
This is of course the logic of reverse racism, and it was powerfully articulated by Supreme Court Justice Clarence Thomas in Adarand v. Pena (1995): “It is irrelevant whether a government’s racial classifications are drawn by those who wish to oppress a race or by those who have a sincere desire to help those thought to be disadvantaged. In each instance, it is racial discrimination, pure and simple.”
I do not mean to suggest that because Kant (at least in my account of him) would agree with Justice Thomas, the case against affirmative action has been decisively made. I am just noting that the two actions Kant contrasts – legislating in response to perceived social needs and legislating with an eye always to first principles – have defined the affirmative action debate from its beginning and continue to do so.
The other, non-Kantian, side has had its share of champions, including Justice John Paul Stevens, who declared (also in Adarand v. Pena), “There is no moral or Constitutional equivalence between a policy that is designed to perpetuate a caste system and one that seeks to eradicate racial subordination.” Stevens is saying in effect, come on, the difference between Jim Crow laws and minority set-asides is obvious to anyone; why obscure it in a fog of so-called principle?
The same opposition was played out later in Grutter v. Bollinger (2003), this time between Justice Thomas (he is the justice most addicted to principle) and Justice Sandra Day O’Connor. O’Connor, writing for the majority that upheld the University of Michigan Law School’s admission policy, speculated that “in 25 years from now the use of racial preferences will no longer be necessary to further the interest approved today.” The interest she refers to is the interest in achieving a diverse student body, and in order to further it, she is willing, at least for a while, to allow the use of an otherwise suspect means.
Thomas replies in his dissent that if racial preferences of the kind the law school employs will be illegal in 25 years, they are “illegal now,” for the Constitution, if it means anything, “means the same thing today as it will in 300 months.” For Thomas, what is at stake is the question of whether the Constitution has an unchanging meaning to which we are obliged to adhere, or whether, on the other hand, the Constitution is a dynamic, living document that adjusts to circumstances and the emergence of problems the founders never contemplated. This is a very old question in the field of constitutional interpretation, and it is at heart the very same question raised by affirmative action. Do we judge policies as being more or less likely to have the consequences we seek in the persent, or do we judge policies as being more or less compatible with first principles that know no time but are always applicable? Do we ask, will it work? Or do we ask, is it right?
My interest in these questions is (for once) more than academic, for I have been a participant in the affirmative action debates since the early 1990s. I have written a bunch of essays (some in this newspaper), two books (“There’s No Such Thing as Free Speech And It’s A Good Thing Too” and “The Trouble With Principle”), done radio and television shows, participated in forums, appeared at city council meetings, and all in support of the position enunciated by Justice Stevens: that so-called principled arguments against affirmative action work by evacuating both history and morality – evacuating history by going to a level of abstraction so high that the difference between acts motivated by beneficence and acts motivated by malice disappear, and evacuating morality on the same reasoning.
Now I’m not so sure. Nor am I sure why I’m no longer so sure, although I expect it has something to do with two arguments I have been making with increasing vigor in the past several years. (My wife says that I’m just moving to the right in the manner typical of all old Jews.)
I have been arguing that the answer to the question, what does a text mean? is that a text means what its author intends, and that therefore it is incoherent to speak of a living Constitution with an evolving meaning. An evolving meaning – a meaning that alters with the times – is, I have insisted, not a meaning at all, but a projection of the interpreter’s desires. If interpretation is to be a serious activity rather than a game with no rules, it must have an object, and the only object it can intelligibly have is a meaning that is prior to anyone’s efforts to determine it. In the context of that argument, affirmative action is an appropriate remedy for historical injustices only if it can be brought into line with constitutional and interpretive principles. The question is, can it be made to square with the meaning the founders intended and with the values – equality and equal treatment – the judicial process has enshrined? I’m not saying that the answer to that question is no, only that it is a different question from the one I used to ask, the pragmatic question of whether it will improve a bad situation.
The second argument I have been making is that institutions of higher education (and their faculty) have only two proper tasks: to introduce students to bodies of material and to equip them with analytical skills. Anything else, in my very strong view, is the job of some other industry or institution, and that includes fashioning character, molding democratic citizens, taking moral or political stands, and performing actions designed to make the world a better place. One reason for supporting affirmative action is that it will make the world a better place, a more democratic place. But from the perspective of my severe notion of what universities should and should not be doing, that is not a good reason. It follows then that if affirmative action is to be defended, it must be on the basis of a pedagogical goal it directly furthers. I’m not saying that there is no such goal – several have been proposed – but that it has not yet been identified in a way I find entirely persuasive.
On the other had (or is it the third hand?), I still feel that affirmative action is a noble endeavor inspired more often than not by the best of motives, and I feel too that many who oppose it are the heirs (metaphorically) of those who have stood in the way of every advance in social justice made in the last 60 years. And there I stand, or rather, wobble. I’ve been thinking again and not finding it much fun.