Few words name a quality so central to Americans’ self-conception as “fairness”; few idioms in American English are so immediately intelligible, so widely resonant, as “fair and square,” “fair dealing,” “a fair chance,” “fair game,” or “a fair fight.” No appeal flies so straight from one American heart to another as “It’s not fair!” Fairness anchors the American democratic ethos the way honor anchored the European feudal ethos – in effect, fairness is an American’s honor. So when a slick literary critic turned deconstructionist law professor rides into town and calls fairness “impossible,” “incoherent,” “a sham and a cheat” … well, those are fighting words.

Or would be, if liberals were not largely defined by an unwillingness to fight over mere words. In a liberal polity, everyone has a right to any opinion. A liberal constitution (i.e., one that contains some version of the First Amendment) protects free speech against state interference; academic freedom protects it in the schools; public sentiment supports it strongly. Truth will emerge, liberal citizens firmly believe, from unfettered competition in the intellectual marketplace. All true liberals will defend to the death anybody’s right to say anything, however obnoxious; in contemporary terms, however racist, homophobic, pornographic, religious, or sacrilegious. Among us, ideas are sacrosanct; the law governs acts, not ideas.

Tosh, says Stanley Fish. There’s no such thing as free speech. Obviously one cannot say just anything. One can’t publish military secrets; can’t libel; can’t advertise falsely; can’t direct a blind person into the path of a speeding car. Such utterances have prompt, harmful, easily foreseeable consequences – that is, they are for all practical purposes acts – so they are uncontroversially prohibited. But, objects Fish, all speech has consequences, weighty or trivial, proximate or remote; in this sense, all speech shades into action. The consequentiality of speech is after all the very premise of First Amendment doctrine. We permit (in principle) all speech because we cannot know in advance, and the state is therefore not allowed to prejudge, which argument will eventually produce the most desirable consequences – i.e., the truth. This implies that if some instance of speech – say, pornography or hate speech – causes grave harm and has nothing much to do with the search for truth -- i.e., has little or no relation to the purpose for which the First Amendment protects speech -- then there is in principle no reason not to prohibit it. And sometimes we do, as I’ve noted.

Of course the First Amendment enjoins the state from promoting as well as proscribing opinions. The state must be, according to a standard formulation, “neutral among competing visions of the good life,” that is, among religious or moral views. No public resources should further (or hinder) any religious purpose. In educational policy (e.g., curriculum design and textbook selection), health policy (sex education, contraceptive availability, abortion funding), civic architecture and ceremony, and everything else it does, the state is required to be strictly impartial toward religion.

Again impossible, says Fish. Leaving religion out of the common life is something many strong believers are unwilling, indeed are forbidden by their beliefs, to agree to. Nor can schools avoid having an effect on a child’s religious development. If the schools are, as the First Amendment requires, indifferent to the claims of sacred authority, the child’s spontaneous deference to those claims – in other words, its faith – will thereby be undermined. This is exactly what conscientious religious parents cannot permit. Liberals may reply that the highest purpose of education in a free society is to train children to reason well and to judge for themselves. But the sovereignty of reason and the supremacy of individual judgment are anathema to believers. Secular liberals’ highest purposes are emphatically not their highest purposes.

Yet another prong of Fish’s attack on neutral principles and procedural fairness is his answer to arguments against affirmative action. In their simplest (and most popular) form, these arguments equally condemn all race-conscious policies as discriminatory and therefore invidious. The proceduralist assumption that “any action tinged with race-consciousness is equivalent to any other action tinged with race-consciousness” gives rise to the principle: “favor no group no matter who the group is or what it has done or what has been done to it” – no matter whether it has, until recently, endured generations of devastating exploitation and neglect or has escaped that misfortune. In its complete (and not at all innocent) disregard of history and context, this allegedly neutral principle becomes, Fish charges, a “device for erasing the difference between oppression and the amelioration of oppression.”

That’s the trouble with principle: it holds politics to an unattainable standard. This standard – neutrality, impartiality, fairness – is unattainable because no speech is without consequences, no purposes are nonmoral, and no policy can equally affect groups with vastly different histories and circumstances. And whenever this standard seems to have been met, it’s because some people have succeeded in getting their own values – secular rationality, individual autonomy, sexual privacy, economic productivity – designated as the general good and therefore beyond politics.

So far Fish’s argument is not, as he readily acknowledges, entirely original. (Except, it’s worth noting, in presentation. The Trouble with Principle is not your usual deconstructionist sludge. Fish is one of the most penetrating, agile, witty, and elegant writers around. It is hard to imagine another book on law, politics, and philosophy – or anything else, really – appearing for quite a while that will give as much sheer pleasure to (impartial and objective) readers of all ideological persuasions as this one does.) But while others also claim to have shown that the philosophical foundations of liberalism rest on sand, Fish’s view of the consequences of this demonstration is unique. He thinks it has no consequences. The appeal to principle, he observes, is an ubiquitous and inevitable rhetorical strategy. Moreover, it’s a legitimate strategy. It’s not that principles don’t exist or that they do no work; they just don’t do the work liberal theorists claim they do: i.e., adjudicate impartially among conflicting substantive views. They’ve done good rhetorical work in the past: e.g., helping to eliminate slavery and to raise the status of women. But at present they’re doing bad work, at least some of the time: e.g., helping to rule out in advance most measures for preventing speech-related harms and at least some measures for remedying racial inequality.

So … ? “What’s a liberal to do? My answer is simple: forget about the principle (and therefore stop being a liberal), which was never what you were interested in in the first place, and make an argument for the policy on policy grounds, that is, on the grounds that you think it is good and right. Argue that civil rights supporters were not working for a color-blind society (even though it may have been rhetorically effective to use that language) but for better conditions for African Americans, and that today the achieving of better conditions might involve practices of voluntary segregation. Argue that in your view the presence of Marxists on campus is beneficial to education and the presence of bigots and racists is not, and that’s all there is to it.”

Now many readers will, as I did, find Fish’s critique of liberal theory wholly convincing. Liberals should by all means forget about principle and argue for policies on policy grounds. But they should by no means make the arguments Fish seems to be inviting them to make here and elsewhere in The Trouble with Principle. Practicing voluntary segregation is about the least useful way imaginable of achieving better conditions for African Americans, followed closely in uselessness by racial redistricting, minority set-asides, hiring and admissions quotas, and special protection against offensive speech. (The most useful ways are: huge and targeted investments in K-12 education; a rise in the minimum wage; basic literacy and numeracy training for welfare recipients who need it; job training, after-school activities, and subsidized summer camp or travel for inner-city youth; and a wealth tax to pay for it all.) The presence of Marxists on campus is indeed beneficial to education (the presence of pseudo-Marxist cultural studies mavens is another matter), but the presence of bigots and racists is even more beneficial, above all to the education of bigots and racists. The latter are, after all, fellow citizens and confused human beings. (And if they are students, they are paying – through the nose, probably – to be enlightened rather than told to shut up.) If such people are stifled, they will not become less dangerous and (eventually) harmful; on the contrary. Liberals owe them, and their potential victims, patient attention and powerful counterarguments.

Of course Fish too is owed more of a counterargument than the above paragraph, but I at least have the excuse of a lack of space. Fish ticks off and disposes of “the nifty nine arguments against affirmative action” with extreme cursoriness – a paragraph each – and sniffs in conclusion: “Sometimes the principled reasons people give for taking a position are just window dressing, good for public display but only incidental to the heart of the matter, which is the state of their hearts.” As a response to the very substantial arguments of Shelby Steele, Orlando Patterson, Jim Sleeper, Tamar Jacoby, Stephen and Abigail Thernstrom, and others – historical and statistical arguments, not abstract, philosophical ones – this is unworthy of Fish and even suggests a lack of political seriousness.

Similarly with free speech. Fish’s attempted demolition of free-speech absolutism is wholly successful: his analyses of Collin v. Smith (the Skokie case) and American Booksellers v. Hudnut (the Indianapolis antipornography ordinance case) are masterly, as is his review of the First Amendment literature. But his political conclusion, that “the notion of … free speech is empty and is thus, at the very least, a questionable counterweight to the harms and injuries permitted in its name,” is not thought through. The issue is thought through more deeply and satisfyingly, without recourse to principle, in a recent essay by Ellen Willis, who concludes:

Symbolic expression, however forceful, leaves a space between communicator and recipient, a space for contesting, fighting back with one’s own words and images, organizing to oppose whatever action the abhorred speech may incite. Though speech may, and often does, support the structure of domination, whether by lending aid and comfort to the powerful or frightening and discouraging their targets, in leaving room for opposition it falls short of enforcing submission. For this reason the unrestrained clash of ideas, emotions, visions provides a relatively safe model – one workable even in a society marked by serious imbalances of power – of how to handle social conflict, with its attendant fear, anger, and urges to repress, through argument, persuasion, and negotiation (or at worst grim forbearance) rather than coercion. In the annals of human history, even this modest exercise in freedom is a revolutionary development; for the radical democrat it prefigures the extension of freedom to other areas of social life.

Willis does here what I wish Fish had done at least occasionally: she explicates and reconciles our conflicting intuitions – in this case, that ideas matter but that words are only words – rather than merely pointing out the conflict.

Debunking is urgently necessary; but liberalism is not pure bunk. Here is Fish in full debunking mode: “What, after all, is the difference between a sectarian school which disallows challenges to the divinity of Christ and a so-called nonideological school which disallows serious discussion of that same question? In both contexts something goes without saying and something else cannot be said (Christ is not God or he is). There is of course a difference, not however between a closed environment and an open one but between environments that are differently closed.”

Having attended both kinds of school, I’m reluctant to leave the matter there. When the subject of the divinity of Christ comes up in a sectarian school, a student who expresses the wrong view is rebuked the first time, sent to the principal’s office the second time, and expelled the third time. When the subject comes up in a nonsectarian school (it is of course not “disallowed”), everybody has a different view or no view, and the student who persists in returning to the subject after everyone else’s interest in it is exhausted is asked to organize a discussion outside of class, attendance optional. Thus are these two environments “differently closed.” In the nonsectarian institution, unlike its sectarian counterpart, dissenters are frustrated but not punished, patronized but not stigmatized, ignored but not silenced. This may not be perfectly fair, but it’s pretty fair; and that – in this time and this place, to most of us, including most dissenters – is fair enough.

There’s something to every intuition, as Fish concedes, indeed insists. And fairness is the mother of all (18th-, 19th-, 20th, and) 21st-century American political intuitions, probably including (he’s so canny it’s hard to tell) Fish’s. I suspect that if Fish ever gets around to laying out his own moral intuitions rather than upbraiding liberal theorists for running away from theirs, he will find himself, on the whole, justifying the ways of liberalism to man.