Equality, you might think, is the more or less universally shared value of the modern world, or the West, or anyway these United States. “We strive to ensure that the values upon which our country was built, including our belief that all people are created equal, are reflected in everything our nation does.” That’s from the 2016 Democratic Party platform.1×1. Democratic Platform Comm., 2016 Democratic Party Platform 45 (2016), http://www.presidency.ucsb.edu/papers_pdf/117717.pdf [https://perma.cc/L5S9-9RAZ]. And look! “We continue to encourage equality for all citizens and access to the American Dream.” That’s from the 2016 GOP Platform.2×2. Republican Platform Comm., Republican Platform 2016, at 9 (2016), https://prod-cdn-static.gop.com/media/documents/DRAFT_12_FINAL-ben_1468872234.pdf [https://perma.cc/Y6KV-3GT4]. Of course, the parties disagree deeply on the demands of equality. If they share an abstract concept, as Ronald Dworkin might put it, they have sharply divergent conceptions.3×3. Ronald Dworkin, Law’s Empire 70–72 (1986). Or, as someone more skeptical might put it, “equality” is just the name of an empty vessel into which partisans pour whatever contents they like. Or, more skeptically yet, maybe one of the parties is lying.
Bad enough, but that underplays the problem. Unless you’ve been held prisoner in a faculty lounge for an awfully long time, you will have noticed that equality is not in fact all that widely embraced. It is, we are endlessly instructed, the gauzy mask barely disguising base envy and resentment, the repulsive desire of social inferiors to pull down the talented and accomplished to their own vile level. Recall Tocqueville’s baleful warning against “a debased taste for equality, which leads the weak to want to drag the strong to their level.”4×4. Alexis de Tocqueville, Democracy in America 57 (J.P. Mayer ed., George Lawrence trans., HarperCollins Publishers 2000) (1835). Tocqueville of course also thinks there’s a “mâle et légitime” passion for equality, Alexis de Tocqueville, De la démocratie en Amérique 86 (Michel Lévy Frères 1864) (1835) — gotta love the gendered imagery — that makes people want to be strong and honored. That putatively hopeful possibility doesn’t do a lot of work in his work or conservative thought more generally, or so I’d argue. Zip forward a century and some, cross the ocean, jazz up the imagery, but still the underlying thesis remains the same. Here’s Kurt Vonnegut’s chilling portrait of a future dystopian America:
THE YEAR WAS 2081, and everybody was finally equal. They weren’t only equal before God and the law. They were equal every which way. Nobody was smarter than anybody else. Nobody was better looking than anybody else. Nobody was stronger or quicker than anybody else. All this equality was due to the 211th, 212th, and 213th Amendments to the Constitution, and to the unceasing vigilance of agents of the United States Handicapper General.5×5. Kurt Vonnegut, Harrison Bergeron, in Welcome to the Monkey House 7, 7 (2010).
George, who’s bright, has to wear “a little mental handicap radio” that assails him with different bursts of noise every twenty seconds, so he can’t think better than other people.6×6. Id. If you’ve got a nice face, you have to wear a mask. You get the idea.
I’ve never understood why so many conservatives find this picture so compelling. It’s blatantly, hilariously, false. Egalitarians marveling at the wonders of Oxford University did not bomb the medieval buildings or try to stop the students from mastering Latin and differential calculus. Instead they worked to build other great universities that would accommodate people without blue blood. Egalitarians struck by the gap between the grimy East End of London and the wonders of the Lake Country did not trash the rolling hills or try to build tenements there. Instead they fought to extend the railroad so that workers with a bit of time off could immerse themselves in natural beauty. (Wordsworth, then a curmudgeonly conservative, harrumphed at what a grotesquely bad idea this was: it would despoil the pristine beauty and serene social practices of his favored haunts; and lacking fine aesthetic sensibilities, the lower orders wouldn’t get anything out of the experience anyway.7×7. William Wordsworth, Kendal and Windermere Railway (1845), reprinted in 3 The Prose Works of William Wordsworth 337 (W.J.B. Owen & Jane Worthington Smyser eds., 1974). ) The actual history of egalitarian politics is leveling up, not down.
Still, the critique of equality is longstanding, ardent, seductive, and very much ongoing. Like the idea of equality, it is richly complex and diverse. I don’t want to offer a potted summary or overview of the tradition here. But I do want to invoke one sadly neglected strand, because I’ll draw on it later and you might as well let your thoughts about it start percolating now. It’s the thought that social inequality is jointly beneficial. Here’s James Boswell in a 1779 newspaper column:
Subordination is in my mind not only necessary for order, but conducive to the felicity of society. I consider society like a grand musical composition, in which there must be a wide compass and gradation of notes to produce pleasure. The equality of men, for which some have argued, would be a dull monotony, a wearisome repetition of the same notes, varied only by the sharps and flats of natural tempers and dispositions. Whereas in a monarchy with all the gradations of nobility, gentry, citizens, in short, all the numerous ranks of society, there is a delightful entertainment, while infinite changes of melody and harmony are continually perceived, and where the pleasure of hope may be freely indulged in the possibilities of rising to wealth, splendour, and honours. In a republick, men grow selfishly lazy in the consciousness of their independency. Whereas in a monarchy there is a reciprocation of active benevolence from the highest to the lowest. The great have the pleasure of humane condescension and respect, their inferiors have the pleasure of receiving kindness and exerting gratitude.8×8. James Boswell, 1 The Hypochondriack 245–46 (Margery Bailey ed., 1928) (1779). See too, just for instance, Monarchy, or Mob-archy, in Notes upon Paine’s Rights of Man no. V, 14–15 (1791); Samuel Parr, Charity-School Sermon, in 6 The Works of Samuel Parr, LL.D. 3, 16 (John Johnstone ed., 1828).
Startled by that closing thought? Condescension here isn’t a vice. It’s the virtue of a social superior who graciously lowers himself to deal with an inferior on apparent or as-if terms of amiable or anyway civil equality. But both parties know perfectly well that it’s a game of pretend, and woe to those underlings who forget it. It becomes a vice when egalitarians undercut the social and conceptual resources that invited people to see themselves as superior in the first place. And that has to change the very experience of what it is like to condescend — and to be condescended to. Likewise, insolence and impudence have changed dramatically. Insolence used to be the vice of a social superior who swaggered or was rude or abusive in exercising his authority; impudence was the vice of the social inferior who mouthed off or was insubordinate. They become rough synonyms for rudeness when we lose confidence in the everyday grammar of superiority and inferiority. To realize that we can still detect the older usages in action — not that we use the words, but that we still do in fact experience and contrast those emotions, whether we shrink from them or embrace them — is to realize that we still hang on, however furtively, to inequality.9×9. For more, see Don Herzog, Poisoning the Minds of the Lower Orders 206–19 (1998). So political controversies about equality cut as deeply as what emotions one can properly experience and how to understand them.
T. M. Scanlon’s latest10×10. T. M. Scanlon, Why Does Inequality Matter? (2018). contributes to our understanding of equality in moral and political theory. It boasts his signature strengths: with great lucidity and patience, he works up elegant distinctions and presses them home. And it systematically defangs some of these time-honored indictments of equality. I learned from it and I liked reading it, too. No, that’s not the sort of false geniality reviewers utter before snatching their daggers. But yes, I have some criticisms. I suspect the last book I didn’t want to criticize was one of those journals with (still) blank pages.
You might think of Scanlon as treating the concept of equality as an umbrella or arch, beneath which cluster disparate or at least independent ideas with some loose family resemblance. So immediately emerges the helpful idea that there are always tradeoffs internal to equality itself (to equality themselves, if you’d like an ungrammatical way of underlining the point), let alone between equality and liberty or other values. Or you might think of him as showing that the same abstract commitment shows up in different domains with different force. I’m not sure what, if anything, hangs on the difference. Regardless, Scanlon is interested in humiliating status differences, the rich dominating the poor, equality of opportunity, fair economic and political institutions, and the equal concern the state must manifest in allocating benefits (pp. 8–9, 152–54). Scanlon suggests — it is just one of many illuminating distinctions motoring the book — that it is one thing to object that some are badly off. So, for instance, it is distressing to learn that in some places life expectancy is low or that some are so poor that they can’t afford a minimally decent way of life. That’s not yet a worry about equality. Knowing that others live on to wrinkled venerability, or that they spend thousands on gleaming home espresso machines, might only remind us that those worse off could in fact be better off. Then, what seems like an appeal to equality has epistemic force alone. But it instantly becomes a worry about equality if you imagine unjust institutions, say, in the economic case, governments administering the legal regimes of job markets, foreign trade, taxation, and so on, but administering them with contemptuous disregard for the interests of the worse off. Now the complaint isn’t just that the poor have low welfare. It’s that the difference between their low welfare and others’ prosperity is unfair, so something could and should be done about it. Scanlon says that his view isn’t necessarily limited to the nation-state (p. 9). Sometimes — it’s an open question — there are institutions of the right sort that have international or even global scope. But here it takes such institutions to give the concept of equality a grip, or so Scanlon thinks.
Sometimes, I’m inclined to add, it’s enough that there could and should be such institutions, at least if the reason there aren’t is that the better off squelch their development lest those institutions redistribute their goodies. I don’t doubt that current institutional arrangements make a normative difference. So do all kinds of other facts. But I do doubt that we are normative captives to whatever those arrangements happen now to be. Scanlon recognizes that possibility. The argument, he says, would proceed in two steps: first, that there needs to be such an institution; second, that if there were one, we would properly see an objectionable inequality (pp. 24–25). “[E]ven if this is correct,” he adds, “it would remain the case, I believe, that under present circumstances what is objectionable about international disparities in life expectancy is not the inequality involved” (p. 25). The thought is that we need an actual institution, not a hypothetical one, to invoke inequality. It’s not even enough if we have good reason to build such an institution. That reason might of course be overridden. To take a hoary concern, we might fret that a world government would be tyrannical or would gut the possibilities of meaningful democratic participation in the nation-state. (Insert here a snatch of the Brussels and Brexit Blues, not particularly melodious and in a minor key.) But Scanlon’s view is that even if our all-things-considered best course of action is to build such an institution, until and unless it’s around we don’t have an inequality.
What divides Scanlon from his imagined interlocutor (let’s call him “Herzog”) here? Apparently just a linguistic point: whether we should now use the term inequality. Elsewhere in the book, though, the current facts have actual practical bite. They dictate what we should do, not merely how we should talk. Scanlon argues that affirmative action for women and minorities at, say, medical schools might be perfectly compatible with his conception of equality of opportunity (pp. 48–49). It doesn’t stigmatize anyone, he thinks, to give a boost in the admissions processes to minorities and women, if, say, they will practice in currently underserved communities, or if their emergence as doctors will help demolish the belief that only white men are truly able — and if these are proper goals for medical schools to pursue. In this way, and as a general matter, Scanlon thinks, our notions of desert are “institution-dependent” (p. 57).11×11. Cf. T. M. Scanlon, What We Owe to Each Other 44–46 (1998) (arguing that notions of desire are similarly constructed by aggregating disparate reasons for action). Everything hangs not just on what meritorious traits an individual might possess, but what makes them meritorious: the way they equip the individual to perform a task or fulfill a role well (p. 57). (The central thrust of current constitutional law, by contrast, does not allow a public institution to depart from race-blind treatment solely in order to help overcome “societal” discrimination.12×12. See, e.g., Shaw v. Hunt, 517 U.S. 899 (1996); City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989); Wygant v. Jackson Bd. of Educ., 476 U.S. 267 (1986); Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978). Croson is the most sustained and important treatment of the point here. See 488 U.S. at 498–506. It would be facile to think that Fourteenth Amendment jurisprudence ought simply to track our best understanding of equality: what the Constitution dictates and what courts can enforce can’t conceivably exhaust the demands of equality. As if legislatures have nothing to do; as if some pressing demands of equality, like how we treat each other in everyday settings, weren’t largely beyond the state’s ken. It would also be facile to imagine one has nothing to do with the other.) Given the work these institutions are properly doing, it makes sense for admissions officers to pay attention to such traits. Though Scanlon doesn’t put it this way, we might say that if one goal of a medical school is to train doctors who will be role models for minorities, paying attention to a candidate’s minority status isn’t some regrettable departure from merit, to be justified, if at all, as the unpleasant price of trying to move toward justice or equality. It is a kind of merit: it is a trait that helps get the institution’s work done. Then too we might say, though Scanlon also doesn’t put it this way, that in such circumstances affirmative action isn’t merely permitted. It’s required. Why should an institution be free to ignore factors in the admission process that help it pursue its proper goals?
And now the crucial bit, crucial anyway for my purposes here. Thinking about conquering discriminatory attitudes by showing that women and minorities perform well, Scanlon writes:
This rationale for affirmative action depends on the empirical claim that such a policy of preference will have the intended effect of undermining discriminatory attitudes (rather than just triggering resentment, or leading its intended beneficiaries to be seen as unqualified because they have been given this preference). It also justifies a policy of affirmative action only as a transitional measure. After a period of time it will either have had its intended effects, and will thus no longer be needed, or have been shown not to do so, in which case it cannot be justified in this way. (p. 49)
I want to focus on the way contingent facts work here. “After a period of time” doesn’t bite the bullet that Justice O’Connor did for the Court: “We expect that 25 years from now, the use of racial preferences will no longer be necessary to further the interest approved today.”13×13. Grutter v. Bollinger, 539 U.S. 306, 343 (2003). As a member of the committee that drafted the admissions policy challenged in that suit, I was deposed by Grutter’s lawyer. Good times. (Tick, tick: that was 15 years ago. Yes, litigators will pounce to treat that expectation as a holding. I bet some judges — and Justices — will, too, even if that 25-year line seems yanked from judicial thin air — not the sort of thing conservatives claim to like in their constitutional law.) But the instinct is the same. If the policy is supposed to be justified by its consequences, either you realize them, so you don’t need the policy any more, or you never do, which makes the policy seem causally inefficacious. Either way, at some point you have to scrap it.
One could quibble over whether that pair exhausts the alternatives. But instead look at the prior empirical claim. Maybe we’ll learn that the policy doesn’t work because no matter how good black doctors are, affirmative action redoubles (some or enough) whites’ conviction that blacks are less qualified. (“Oh,” they silently groan — or smirk — when an African American doctor walks into their office for their appointment. “I know how you got here. I wonder if I can still get a real doctor.”) We could construct fanciful scenarios about how that might happen, but let’s stick with the depressingly obvious one: it’s because whites’ racism is more or less evidence-proof. I’ll turn up the stylistic volume, though not I think to hysterical levels: Scanlon is implying that the justifiability of affirmative action is held hostage by white racism. If that racism is obdurate enough, we ought not pursue affirmative action, at least as far as the proffered rationale goes.
If you’re a certain kind of consequentialist, you won’t see anything worrisome here. We have to take the facts as they are, you’ll think. And we could dress up that view in more sophisticated and plausible garb.14×14. Consider the treatment of excluding gays from the military — you can decide how sophisticated and plausible it is — in Richard A. Posner, Sex and Reason 314–23 (1992). For that matter, if you’re a certain kind of Kantian, you will have an easy time adopting a different but equally simple view: what’s decisive is the counterfactual world in which your maxim is universalized, so that everyone (similarly situated) acts as you propose. Or we could shift the idiom. Politics, we might agree, is the art of the possible. It is not an arena for utopian dreams; it is the awkward business of stumbling around and trying to make the world just a bit better, or just a bit less awful. We must act on the basis of people as they are, not as we might like them to be.
Still I want to suggest there is something to worry about here, a worry that neither textbook consequentialism nor textbook Kantianism illuminates all that well. What’s undercutting the efficacy of affirmative action is that some hold an unreasonable belief, to put it softly — or, aloud — a belief that’s not just indefensibly stupid but also outrageously unjust, indeed, itself part of the problem we’re trying to solve. It’s as though racism becomes not just self-perpetuating but self-justifying.
It’s not that the facts never matter. Of course they do, often unexceptionably so. (I’ve long thought, and even said, and sometimes written! that political theorists are people who want to mock their colleagues in political science as barefoot positivists who believe in some scarecrow version of a fact/value distinction, but themselves want to disdain facts on the ground that their work is normative. Surely something here has to give.) But when we’re pursuing equality, we often run roughshod over perfectly sincere anti-egalitarian commitments. Antidiscrimination law does not permit an employer to defend racially discriminatory hiring policies by pleading that she’s merely giving the customers what they want.15×15. See especially Diaz v. Pan Am. World Airways, Inc., 442 F.2d 385 (5th Cir. 1971). You might think that’s a prophylactic rule: we want to prevent the racist employer from hiding behind an apparently benign pretext. But the law doesn’t finally care about the motives of the employer. And the law is cheerful about overriding bromides about consumer sovereignty, or taking preferences as they come, or Kaldor-Hicks efficiency, or anything of that sort. Does the defensibility of this legal regime hang on the claim that it will rid the market of racist preferences, whether those of employers, employees, or consumers? Well, surely it hasn’t accomplished that. I’m sometimes inclined even to doubt it’s made much of a dent. But that dour skepticism doesn’t begin to undercut my commitment to the project of antidiscrimination law. And I don’t think that makes me a fanatic, as say a would-be prudent consequentialist might suggest.
You might approach this question as a matter of equality of opportunity. So, you could try to shrug off the worry about the consequences of antidiscrimination law: it’s enough, as a matter of equality, that it gives individual job candidates a fair shake. That sounds reassuring, but it doesn’t squarely meet the worry I’m gnawing away at — and I doubt it’s a move available to Scanlon. Remember that Scanlon holds that affirmative action isn’t at odds with equality of opportunity — not, that is, where the relevant traits we are focusing on are relevant to the goals the institution serves. It won’t do to say that affirmative action depends on the consequences, but antidiscrimination is just constitutive of what equality of opportunity is. Given Scanlon’s commitments, that looks like an optical illusion. Suppose for instance we learn that members of some racial group — call them Xs — don’t do as well as their individual applications might otherwise suggest. If they systematically underperform, they won’t be as deserving as they otherwise look. So the institution will justifiably pursue its goals by giving them short shrift. It isn’t acting on the basis of irrational stigma; it is according Xs equal concern. Likewise for boosting the credentials of Ys once we learn they systematically overperform. And that goes whether we’re considering the relation between admissions credentials and performance within the institution, or admissions credentials and performance later, in the roles the institution prepares people for.
Now consider Scanlon’s position on what he calls substantive equality of opportunity (pp. 53–73). He is worried — so am I16×16. See Don Herzog, Happy Slaves: A Critique of Consent Theory 215–47 (1989). — about glib appeals to what individuals choose. In the debates over equal pay for equal work, about which I cheerfully refuse to provide a footnote for the insatiable editors of the Law Review, people sometimes shrug and say, “As long as firms aren’t discriminating in the hiring process, it’s neither here nor there if women choose to pursue lower-paying careers: if they’d rather be nurses than doctors, secretaries than bosses, grade-school teachers than college professors.” Not so fast, urges Scanlon. We want to be able to put pressure on how and why individuals came to see themselves as actually having certain choices, or to see those choices as potentially valuable for people like themselves. If you’re a girl growing up in a culture saturated with misogyny, just for instance — or, um, not exactly just for instance — you may well come to think that women shouldn’t be doctors, or you wouldn’t be any good at it anyway, or it would get in the way of having a family, where that last depends on not even beginning to think critically about the current constellation of gender scripts and job markets. Or as Scanlon puts it:
The value of having a choice is undermined when one is uninformed about the nature of the alternatives, or when conditions make it unlikely that one will consider certain valuable alternatives or take them seriously. So one thing that individuals have strong reason to want is to have what happens to them depend on how they react when given the choice under sufficiently good conditions for making such choices. This is particularly true in the case of important features of their lives, such as what careers they will pursue. (p. 62)
And again, “the requirements of substantive opportunity are not satisfied if young women fail to strive for positions for which they would be qualified because their families believe, and encourage them to believe, that these careers are not appropriate for women” (p. 64). But suppose such a young woman says, “Look, maybe I’d be better off, more autonomous, with the upbringing you’d have given me. But like it or not I am now the person I am, and I am in no position to rework my identity so fundamentally. Don’t tell me my choices don’t count. Don’t even tell me they aren’t fully sufficient. They’re all I can plausibly make and I refuse to be overridden by my counterfactual self, or imagine that that self has more dignity, more standing to assert herself, than I do.” I’m unsure what would count as a persuasive response.
Let’s step back and contrast the way Scanlon handles facts in these three examples. One: suppose there are no institutions that could handle issues raised by global variations in life expectancy, but there could be. That gives us good reason to work toward such institutions, but it doesn’t change the fact that even sharp differences in life expectancy don’t now properly qualify as inequality. Two: suppose increasing the pipeline of women or minority doctors would not in fact help serve underserved communities or help undercut stigma and prejudice about the abilities of women and minorities. Then we ought not increase the pipeline, period. Three: suppose women don’t try to become CEOs because society or their families have drummed sexist idiocies into their heads. Then their choices on the job market don’t reflect substantive equality of opportunity, even if all hiring decisions are fair. Depending on which example you choose, the current facts of the matter are either conceptually decisive for the applicability of equality talk, or actually decisive for what we should do, or shunted aside for counterfactuals with practical force.
I have no interest in quibbling with Scanlon’s examples. And I do not contrast them to suggest that his position on the role of facts in normative argument is incoherent or even unconsidered. I don’t myself see any interesting differences between the role of the current facts in affirmative action and in substantive equality of opportunity, but I’m open to argument. The point I want to press is rather this: I doubt there is a simple blanket rule for how social facts and counterfactuals properly work in normative argument. I have up my sleeve nothing like an architectonic theory, carving up the domain and explaining why different sorts of issues properly feature different treatments. The problems here are real. They are large, amorphous too, and I suspect the best we can do, at least for now, is to muck along by delving deeper into particular issues and trying to make reflective sense, shuttling in the now familiar way between tentative principles and concrete applications, of how we argue and how we believe we should argue.
Here’s one punchline: uncertainty about how the facts should matter generates a distinct kind of normative uncertainty. (Try scrambling those three normative issues and the way Scanlon deploys the facts in each. See how much changes?) One reason our normative convictions are insecure, that is, is that we don’t know quite how to use the facts and counterfactuals at hand. This would remain true even if we somehow commanded rock-solid agreement on what facts and counterfactuals actually obtain, agreement too on our normative principles; the point I am pressing here is then very much in the spirit of Rawls’s burdens of judgment, if not the same as anything on his list.17×17. See John Rawls, Political Liberalism 55–57 (expanded ed. 2005). It means too that critics always have another powerful if indirect way of criticizing our arguments. They can assail not whatever normative principles we state, but how we happen to enlist the facts. Even those of us committed to the concept of equality — even those of us who share particular conceptions of what it entails — are likely to go on debating what is to be done in various settings, just because of these multiple ways of enlisting the facts. And you thought normativity itself was the mysterious problem.
Time to return to Boswell’s paean to inequality. If or insofar as he’s imagining a world where the fact of birth dictates your social status, there’s an obvious affront to equality of opportunity. (Assignment for the reader: identify what’s wrong with the suggestion that whether you’re born a prince or a pauper is random, that everyone has the same chance of occupying any position, so inherited status is perfectly egalitarian. No, sorry, the answer is not that there are more paupers.) And surely monarchy works that way. But Boswell is explicitly imagining people happily contemplating “rising to wealth, splendour, and honours.” Let’s suppose that that competition is genuinely open. (No, it wasn’t all that open in Boswell’s England. But how open is it in today’s United States? Any viable conception of equality of opportunity is going to have to say something about equality of starting points, and when it comes to public education, we are risibly far from meeting that standard.18×18. Here, I’m essentially agreeing with Scanlon (see pp. 65, 71–72, 89, 116). ) Do the “inferiors” have a complaint? Boswell wants us to see them as not stigmatized. There is nothing, he thinks, contemptible about them, and nothing contemptuous in their treatment by “the great.” This sentiment was in fact readily underwritten by Christianity, with assiduous reminders to the superiors that all were equal in God’s eyes, along with Jesus’s declaration that “it is easier for a camel to go through the eye of a needle, than for a rich man to enter into the kingdom of God.”19×19. Matthew 19:24. Let’s eavesdrop on Queen Charlotte sternly instructing her loathsome son, the Prince of Wales and future George IV, on his eighteenth birthday:
Be charitable to everybody, not forgetting your meaner servants. Don’t use them with indifference, rather pity them that are obliged to serve, and do unto them as you would be done by. I mean by that you should not think yourself above doing good to them. The contrary will make you appear vain, and vanity is the root of all vice and a sure proof of ignorance. For what is man to man? We are all equal and become only of consequence by setting good examples to others, and these must be given with a view of doing our duty but not with the idea of superiority, for then the action loses its merits.20×20. Letter from Queen Charlotte to the Prince of Wales (Aug. 12, 1770), in 1 The Correspondence of George, Prince of Wales 1770–1812, at 5, 5–6 (A. Aspinall ed., 1963). Her last thought is intriguingly close to Kant’s position on moral worth.
I would assure you, after endless absorption in the relevant sources, that it’s way too easy to dismiss Boswell’s sketch out of hand as threadbare ideology. Some of “the great” really cared for the “inferiors,” and condescended not just to talk to them but to offer aid. Some of the “inferiors” really did bask in the pleasures of gratitude. Some, not all: I wouldn’t enlist his sketch as terrific social history, either. Regardless, those grumbling and agitating for change were often dismissed, not just from Parliament and the pulpit but also by their neighbors, as malcontent and — you guessed it — envious and resentful.
If you think the scheme is still hopelessly and outrageously unequal, why? Is it that the rich have “unacceptable forms of power over” the poor (p. 8)? (Social status is different from income and wealth, but I set that aside.) Or that the scheme undercuts “the fairness of political institutions” (p. 8)? Well, yes: theirs was a world in which — if male workers had the franchise; franchise requirements varied widely around England; and don’t imagine women were voting, because precious few contemporaries even glimpsed the possibility — voting was public, so their landlords knew how they voted, and reprisals were at the ready.21×21. Indispensable here is E. Anthony Smith, Earl Fitzwilliam and Malton: A Proprietary Borough in the Early Nineteenth Century, 80 Eng. Hist. Rev. 51 (1965). Though I’ve barely mentioned these strands of his argument, Scanlon can readily explain what’s wrong in these ways.
But shift the setting. An anthropologist from some hazily Scandinavian paradise of your most fervent egalitarian dreams (or nightmares) joins you in strolling through an American factory. “Who are those people?” she asks, pointing at some overweight men, mostly white and some in suits, walking around but not obviously doing anything productive. “Foremen and managers,” you say. You’re alarmed: Does she not know this already? Does her country not have such characters? And she looks quizzical: “And they get to tell the workers what to do? and when to do it? and how to do it? and discipline them for even trivial infractions?” “Sure,” you say. Given today’s English locutions, you’re not likely to rhapsodize over the “grand musical composition” and “delightful entertainment” presented by the factory hierarchy.22×22. Boswell, supra note 8, at 245–46. But you are likely to think that the setup is jointly beneficial, that it is not nutty to hope that managers and foremen will take pleasure in treating the workers well and that the workers will take pleasure in being so treated. (You needn’t think that pleasure or desirable consciousness or preference satisfaction is the only game in town to think both that it matters on its own and that it is often evidence that other and more important values are being realized. If you take pleasure in being treated with respect, for instance, that’s in part because you know what is true: that it is a good thing to be treated with respect.) So how different is your view from Boswell’s? No one is born a foreman or a manager — though we should worry about nepotism and inheritance — and let’s set aside worries about whether hiring practices are genuinely open, or people have been put in positions where their education and socialization gives them a valuable range of choices they can genuinely consider as open to them.
The anthropologist is still baffled. “Why do the foremen get to tell the workers what to do?” Stymied temporarily, you happily recall Engels’s trenchant rejoinder to the anarchist thought that we could dispense with authority. You can’t run a factory without some hierarchy, you say.23×23. Friedrich Engels, On Authority, in The Marx-Engels Reader 730, 730–33 (Robert C. Tucker ed., 2d ed. 1978). “That’s probably right,” she says. “But think about how the foremen and managers came to exercise authority. They just signed contracts with the company. And the workers came to be subject to authority by signing contracts, too. But the foremen and managers never even contracted with the workers. The foremen and managers have no duty to be responsive to the workers. You can say they all consented, though before I agreed I’d want to know whether they had reasonable alternatives. But I don’t care if they consented. Even if there has to be authority here, shouldn’t it be exercised democratically? Shouldn’t the managers have to be responsive, at least in large part, not just to the company or its stockholders, but also to the workers they supervise? Remember company towns? Workers signed a contract permitting the company to run not just the workplace but also the town they lived in. And surely that’s unacceptable. If you can’t hire a mayor and get people to sign contracts consenting to his rule, why can you hire a foreman and get people to sign contracts consenting to his rule?”24×24. Consider Elizabeth Anderson, Private Government: How Employers Rule Our Lives (and Why We Don’t Talk About It) (2017); Michael Walzer, Spheres of Justice 291–303 (1983). For government as not just the institution of the state, but also an activity stretching across the social landscape, see Don Herzog, Household Politics 123–25 (2013).
I’ve said nothing about the size of the managers’ and workers’ paychecks or bank balances. I confess that even though I care a lot about equality, Gini coefficients put me to sleep, and not because I have any difficulties with statistics. They might be evidence of something that ought to concern us: say that the political process has been captured by actors who deal themselves unjustifiable regulatory and tax benefits. But standing alone? Zzzzzzzz. Yet when Scanlon writes about economic inequality, he mostly thinks about income. He declares that it’s “troubling” that in today’s America, some earn vast multiples of what others do (pp. 134, 150).25×25. Scanlon says that his view, like Professor Elizabeth Anderson’s, is “relational” (p. 9 n.10). See Elizabeth S. Anderson, What Is the Point of Equality?, 109 Ethics 287 (1999). But either he has a very different understanding of relationality or he’s thinking of an entirely different set of relations. And then he backs up and tries to figure out whether there is good reason to structure markets and especially the regime of taxation to enable such inequalities of income (pp. 133–51). But he doesn’t think about the content of the relationship between the manager and the worker. Yes, factories aren’t nearly as prominent fixtures in the American workplace as they used to be. I’m all in favor of being sociologically concrete, so I’d want to think too about the lives of those sometimes labeled the precariat; and those “contractors” who drive for Uber and Lyft; and temporary workers who bounce around from workplace to workplace, in part because firms find it useful to avoid paying benefits; and so on. In all these cases, there’s way more at stake for equality than the size of a paycheck or a bank balance. There’s always a question of how supervision, management — government — works, and whether we should see it as legitimate. And one plausible intuition about equality is that even if all the workplace roles are genuinely open in a fair competition, and even if every individual voluntarily chooses the job she takes, there is something objectionably unequal in the terms of life on the job. Workers, one might well think, should be citizens, not subjects: actors entitled to engage in self-government, not the passive recipients of rule from above.
There’s lots to say about this, but I want to use the scenario first to touch again on the difficulties of how we use facts in normative arguments, second to turn to some worries about Scanlon’s contractualism.
Surely there is an institution that could change the rules of the workplace: American government. (If you’re worried that somehow states would run into dormant commerce clause objections, have Congress do it as regulation of commerce.) So we’re not in the position Scanlon thinks we are in pondering life expectancy around the globe. Still nothing now stops workers and companies from contracting on democratic terms, and apparently they choose not to. How much deference is that fact entitled to? We could investigate the socialization that might make workers — and bosses! — think it simply out of bounds or inappropriate for people like them: remember that we have to meet some background conditions for choice to be valuable or to have legitimating force. Here our egalitarian will point an accusing finger at Americans’ longstanding ideological anathema against socialism. She will suggest that we could at least distinguish state socialism, with a centralized authority dictating prices across the board, from market socialism, where prices float freely but we’ve changed governance relations in the firm.26×26. On the epistemic failures of the former, the canonical text remains F. A. Hayek, The Use of Knowledge in Society, 35 Am. Econ. Rev. 519 (1945). Collectivist Economic Planning (F. A. Hayek ed., 1935) and Oskar Lange & Fred M. Taylor, On the Economic Theory of Socialism (Benjamin E. Lippincott ed., 1938) remain of more than purely historical interest. And — recall the complaint of the woman whose occupational choices are indicted as reflecting poor socialization — here Americans may object that “maybe we’d be better off had we not been ideologically conditioned in that way. But we are who we are, and we can’t turn our backs on our traditions, our identities, now.”
But now consider a much broader range of facts: how a differentiated society is arranged. State, church, market, family, university, sports team, private club, and so much more: that’s the top-down view. (For the bottom-up view, think about the various roles an individual moves in and out of.) Our social landscape is nothing like a unified hierarchy or one of Goffman’s total institutions.27×27. See generally Erving Goffman, Asylums (1961), especially chapter 1. It’s a ramshackle collection of largely autonomous institutions with messy but real jurisdictional divisions. Any institution ought to be selectively blind and pay attention only to contextually relevant facts. That immediately gives equality a social-structural form. That’s the force of the iconic representation of justice as blindfolded, and the source then of painfully obvious indictments of our criminal justice system for paying attention to race and wealth. To be equal under law is just to have the law ignore such irrelevant facts. (The same exact thought is at the center of modern thinking about freedom, which provides a useful corrective when people summon up some apocalyptic conflict between freedom and equality. Recall Locke: “the Community comes to be Umpire, by settled standing Rules, indifferent, and the same to all Parties.”28×28. John Locke, Two Treatises of Government 324 (Peter Laslett ed., Cambridge Univ. Press 1988) (1690). ) Or again, it’s weird if you get a better grade in your philosophy course because your professor shares your religious faith, or a worse grade if he doesn’t: he has to treat his students as equals, which here means in part simply ignoring whatever religious commitments they have. There’s an important sense in which that’s true even in a theology course.
Plenty of theorists have asked the question, By virtue of what feature of persons are we equal? The usual candidates include rational agency, the ability to frame and pursue a conception of the good, creation by God, and so on. Never forget Hobbes’s sinister and amusing suggestion that we’re equal in part because anyone can kill anyone else,29×29. Thomas Hobbes, Leviathan 74–75 (Edwin Curley ed., Hackett Publ’g Co. 1994) (1651). or, as Shakespeare’s Nym puts it, “Men / may sleep, and they may have their throats about / them at that time, and some say knives have edges”;30×30. William Shakespeare, Henry V act 2, sc. 1, ll. 20–22 (David Bevington ed., Bantam Books 1988) (1623). and in part because everyone is vain enough to prefer his own wisdom to others’, and “there is not ordinarily a greater sign of the equal distribution of anything than that every man is contented with his share.”31×31. Hobbes, supra note 29, at 75. But here the right question is, By virtue of what feature of society are we equal? And the answer is social differentiation.
Social differentiation also enables us to pursue a range of different values: salvation in church, wealth on the market, athletic excellence on a sports team, and so on. And surely we do not now arrange every one of these institutions in democratic ways. It is incongruous, in some cases pernicious, to imagine doing so. Must egalitarians reject Catholicism and embrace Congregationalism? Must they insist that students decide, or even help decide, on course syllabi and requirements for the major? And so on. Even if you think equality requires democratic governance across the social landscape, you need to remember that we sensibly might trade off equality against other values.
Armed with this broader factual picture, you return to our lefty anthropologist and ask, “Why should the workplace be democratic? You’re just threatening us with another kind of monotony, if not the one Boswell had in mind. I prefer the diversity of values we can achieve when we don’t democratize everything in sight.” I don’t propose to settle this dispute in a review essay (or anywhere else); I don’t see any point in tipping my hand on what my own druthers or even considered judgments are. Here I want only to underline again one kind of difficulty that the debate will inevitably spawn: not just what the facts of the matter are, but which facts are relevant or decisive as they stand, which should be undercut by critique, which should be shunted aside in the name of counterfactuals, and so on.
But I mostly want to use this scenario to dig in and explore two worries about Scanlon’s contractualism. That view, which he’s been developing for some time now, says at its core that “when we address our minds to a question of right and wrong, what we are trying to decide is, first and foremost, whether certain principles are ones that no one, if suitably motivated, could reasonably reject.”32×32. Scanlon, supra note 11, at 189. Scanlon repeatedly invokes the view (see pp. 105–06, 113, 116). See also, e.g., T. M. Scanlon, Being Realistic About Reasons 96 (2014) [hereinafter Being Realistic] (“If it would be reasonable to reject any principle that would permit a certain action, then that action would be morally wrong.”). T. M. Scanlon, Moral Dimensions 89–105 (2008), helpfully compares and contrasts his views to facets of Kant’s. For Scanlon’s recent thoughts on how he might now recast the view, see T. M. Scanlon, Contractualism and Justification (2014) (unpublished manuscript) [hereinafter Justification], http://www.law.nyu.edu/sites/default/files/upload_documents/Scanlon%20Contractualism%20and%20Justification.pdf [https://perma.cc/J56N-G7VX]. There’s a lot to unpack here: Scanlon of course has spent many thoughtful pages unpacking it himself. We could put pressure on the role of principles, or suitable motivation, or reasonableness, or the centrality of rejection. (That we could put pressure on any of these is compatible with acknowledging that of course they hook up with each other.) But here I want mostly to put pressure on reasonableness.
Reasonableness isn’t the same as rationality, especially if we take the latter in the formidable instrumental sense of economists and decision theorists, with constrained maximization, quasi-concave utility functions, expected utility as a way of collapsing possible outcomes with different probabilities into one value, and so on. Rawls’s strategy was to characterize the original position so that the instrumentally rational choices of parties behind the veil of ignorance would be reasonable.33×33. See, e.g., Rawls, supra note 17, at 52–53 (providing an especially concise explanation of the issue). One worry is that the original position is an amplifying device: tiny adjustments in how we characterize it (or, to be more terminologically accurate, other ways of characterizing the initial situation, Rawls’s name for any such hypothetical contract setup34×34. See John Rawls, A Theory of Justice 11–16, 105 (rev. ed. 1999). ) could lead to huge swings in the principles of justice selected. If that’s right, it undercuts the justificatory force of seeing the principles emerge from such a procedure, unless you have a lot of confidence in the precise features of the original position.
Scanlon’s approach, by contrast, is to appeal head-on to reasonableness, and to imagine us, here, now, not hypothetical agents behind a veil of ignorance, proposing principles and making objections. So Scanlon is emphatically not imagining a world of strategic bargaining, with the economist’s usual if covert background principle of justice, “to each according to his threat advantage.” It might be good for you if we adopted the principle, “everyone should do whatever you want, whenever you ask.” But it is obviously reasonable for others to reject that view: it is a severe incursion on their autonomy and serves no purpose that would make sense to others. Similarly, you can’t reasonably reject a proposed principle just because you think it might be bad for you: to cost you money, say. It might be rational, in the narrow self-interested sense, to fight the adoption of such a principle. But unless there’s more to say — for instance, that people like you are asked to pay too often or too much — it wouldn’t be reasonable.
That’s the core of the view, enough for my purposes to generate two connected worries. One: what, if anything, remains of the individual point of view in proposing principles and considering objections? Two: what if there are no reasonable objections to competing principles, or reasonable objections to all principles?
On the first, consider again the contrast with Rawls. Rawls writes about multiple parties behind the veil of ignorance, but of course they’re all identical, endowed with the same knowledge, the same lack of knowledge, the same motivations. So “multiple” isn’t doing any work and “contract” is a misleading category. We might as well think of the original position as modeling an individual choice. Because Scanlon is thinking about us, here, now, he has plenty of room to explore what individual differences will be salient.
But on this matter, to the best of my knowledge anyway, Scanlon is surprisingly brief. Here’s the core claim:
[A]n assessment of the rejectability of a principle must take into account the consequences of its acceptance in general, not merely in a particular case that we may be concerned with. Since we cannot know, when we are making this assessment, which particular individuals will be affected by it in which ways (who will be affected as an agent required to act a certain way, who as a potential victim, who as a bystander, and so on), our assessment cannot be based on the particular aims, preferences, and other characteristics of specific individuals. We must rely instead on commonly available information about what people have reason to want. I will refer to this as information about generic reasons.
Some examples: We commonly take it that people have strong reasons to want to avoid bodily injury, to be able to rely on assurances they are given, and to have control over what happens to their own bodies.35×35. Scanlon, supra note 11, at 204.
The move to such generic reasons makes it seem plausible that anyone can work out the relevant reasons by herself, just sitting and thinking it through, and that she can do so even for those very differently situated. Scanlon readily offers, for instance, that “[s]everely disabled humans have reason to want those things that any human has reason to want, insofar as these are things that they are capable of benefiting from. These will include, at least, protection and care, affection, and those enjoyments of which the person is capable.”36×36. Id. at 186. And Scanlon works up elaborate arguments to defend surprisingly intricate principles. Here’s one:
Principle F: If (1) A voluntarily and intentionally leads B to expect that A will do X (unless B consents to A’s not doing so); (2) A knows that B wants to be assured of this; (3) A acts with the aim of providing this assurance, and has good reason to believe that he or she has done so; (4) B knows that A has the beliefs and intentions just described; (5) A intends for B to know this, and knows that B does know it; and (6) B knows that A has this knowledge and intent; then, in the absence of special justification, A must do X unless B consents to X’s not being done.37×37. Id. at 304.
The picture of anyone working up a view on her own might be reinforced. “In order to decide whether a principle could reasonably be rejected,” Scanlon tells us, “we need to consider it from a number of standpoints”38×38. Id. at 213. : beneficiaries of it, those constrained by it, people who’d fare better under some other principle, and the like. But any suitably motivated individual trying to propose a principle that can’t reasonably be rejected already has to think through all those standpoints. (I wonder whether “suitable motivation” adds anything at all to reasonableness and rejectability, but leave that aside for now. Parsimony is the least of my worries about this theory.) This is just another way of underlining the importance of the fact that we are not playing “to each according to his threat advantage.” You’re not behind a veil of ignorance; you may know perfectly well that some principle you’re considering would be wonderfully good or horribly bad for you; but unless those facts turn out to be the appropriate sorts of considerations for you to venture and others to think through, they’re neither here nor there. In this way, actual individuals in actual social positions seem to be vanishing from the argument, or, better perhaps, turning into hazy cartoon versions of themselves.39×39. I am reminded of an oddity in Professor John Harsanyi’s utilitarianism. He writes, “The social welfare function W of individual j must be of the following mathematical form . . . .” John C. Harsanyi, Essays on Ethics, Social Behavior, and Scientific Explanation 81 (1976). But j appears nowhere on the right-hand side of the equation; k and l and . . . have the identical function. So it doesn’t matter who the individual is; in invoking an “individual social welfare function,” Harsanyi adds nothing with “individual.”
Still, it would lampoon Scanlon’s project to suggest he thinks he can singlehandedly spin out a set of principles that no one could reasonably reject. He notes for instance that what we think about generic reasons “is subject to modification under the pressures of moral thought and argument.”40×40. Scanlon, supra note 11, at 206. And surely that means actual argument in the actual world, not just further chatting silently to oneself. So not just intricate Principle F, but any and all of Scanlon’s suggestions for principles no one can reasonably reject are up for grabs in moral argument. We should take them as candidates: plausible, well-considered for sure, but ineluctably one person’s point of view on these matters. So the turn to generic reasons doesn’t mean we can trust Scanlon, or anyone else, to think hard and reveal the right set of principles. If you don’t think generic reasons might be hard to grasp, ask yourself what you think of the view that it doesn’t matter whether gays or the poor or blacks or women show up in legislative chambers, because straight rich white guys can grasp and represent their interests.
What, then, about the worry that on many issues maybe multiple principles can’t be reasonably rejected, so that different and contradictory principles can each be exhibited as right? or that all principles can be reasonably rejected, so that nothing is right? On our everyday understanding of reasonableness, in any setting plenty of views and actions are reasonable. Tort law adopts that everyday understanding. Some courses of conduct don’t offer reasonable care for others: think of the homeowner who never shovels his sidewalk after a snowstorm, or worse maliciously lubricates the sidewalk to make it more slippery. But plenty of lines of conduct are entirely reasonable. You could shovel within twenty-four hours of snowfall’s end. You could hire a neighborhood teenager or a company with a little truck equipped with a big rotary brush to do it. You could painstakingly shovel the entire width of the sidewalk or just enough for people to walk safely in single file. You could sprinkle salt or sand or gravel on the ice. And so on. Surely all of those are reasonable. Imagine how bizarre it would be for someone to say, “It is unreasonable to hire a company; the only reasonable policy is to do it yourself.” That’s bizarre not just because some homeowners are too feeble or disabled to do it, but also because there’s nothing obviously wrong with contracting out here. You might think it better to do your own shoveling: it’s good for your health, and you fret about commodifying one everyday responsibility after another. And you might be entirely correct about that. But that’s not enough to show that patrons of other approaches to snow removal are unreasonable. Any approach that clears some threshold is reasonable.
So how does Scanlon’s theory work? Take the basic issues of political economy, central to current debates about equality and prominent on Scanlon’s radar screen. (I suppose each is properly understood as a cluster of principles, but no matter.) Let’s make a list of different possible schemes:
- Boswell’s melodious subordination;
- Current American market and taxation schemes;
- Current German practices, with high-level collective decisionmaking by labor, business, and government officials;
- Libertarianism à la Nozick;41×41. Yes, that would be Robert Nozick. See Robert Nozick, Anarchy, State, and Utopia (1974).
- A decent minimum and robust equality of opportunity, with no regard to the shape of the income distribution or wealth accumulation; and
- Social democracy and sharply redistributive taxation, in whatever version our Scandinavian anthropologist would endorse.
Did I leave out your favorite contender? Sorry, you can add it, but it won’t make a difference to the point I want to press. On the everyday understanding of reasonableness, one might think that one can indeed reasonably reject any and all of these — or that no one can reasonably reject any of them.
Scanlon doubts there are good arguments for natural rights of property that entail libertarian policies. He casts market wages as “predistribution” and taxation as “redistribution” and, following Murphy and Nagel,42×42. See Liam Murphy & Thomas Nagel, The Myth of Ownership: Taxes and Justice (2002). he urges that there is no way to redeem the intuition that taxation is some illegitimate interference with the property rights we really do just plain have (pp. 102–09). I agree. So I think libertarianism is misconceived. But I’m not inclined to say that libertarians therefore can’t reasonably reject other schemes. They have a well-considered view, in the sense that they’ve thought about it, and more important they can adduce real and weighty reasons to support it. So it isn’t a daffy or indefensible view. If it turns out not to be the best view in town, still it is reasonable for them to hold it. And — put a flag on the field for the umpires to review this play, as I promise they will, I will, shortly — it seems to follow immediately that it’s reasonable of them to reject competing views. And then patrons of any candidate on my list can reasonably reject all the others. In no time at all, no candidates for the right principles of political economy are standing. If you plug this outcome back into Scanlon’s contractualist formula, you end up with the baffling inference that everything is wrong, because everything can be reasonably rejected. Call this the permissive approach to rejectability.
Or you might run the argument in the other direction. You might say that no one can reasonably reject any of the other contenders. After all, everyone should understand that all the contenders plausibly purport to offer reasonable terms of social cooperation. If any one of them has flaws, so do they all: if libertarianism is objectionable for the pros-pect of leaving poor and unemployed people shuddering in the cold, current American schemes are objectionable for offering unjustifiable rewards to the well-off, and Scandinavian social democracy is lamentably inefficient, and . . . . That a scheme has flaws, that some reasons count against it, can’t be enough to reasonably reject it. We have to consider all the reasons, pro and con, and come to an overall assessment. But that’s enormously difficult, and different people will reasonably weigh the pros and cons differently — and, remember, enlist facts and counterfactuals differently — and come to different conclusions. Epi-stemic modesty or caution here threatens to teeter into paralysis of judgment on the theory’s core question of what can be reasonably rejected. “I think the best all-things-considered judgment is that the reasons weigh against libertarianism, just because one or more contenders defeat it. Given the generic reasons that people like me can adduce, it imposes unacceptable risks and burdens on me.” But you can think that and still think it would be unreasonable of you to reject libertarianism, because you’re well aware that a parallel line of reasoning, even if you think it’s finally mistaken, would enable libertarians to reject your favored view. And you know that can’t be right. In no time at all, a distressingly wide range of candidates are left standing. Plug this picture back into the contractualist formula and now you get an equally baffling inference: every one of these sharply conflicting views is right. Call this the restrictive approach to rejectability.
Time to call in the umpire. Think of these two approaches as different responses to the intriguing question of whether there is space between (1) what is reasonable for you to believe and (2) what is reasonable for you to reject: that is, whether it follows that once it’s reasonable for you to believe something, it’s reasonable for you to reject other views. Or, to take a closely connected formulation, the question is whether there is something important at stake between thinking about principles that it’s reasonable to accept and thinking about principles that it’s reasonable to reject.43×43. Scanlon usually focuses on rejectability, for reasons he makes perhaps clearest in T. M. Scanlon, Contractualism and Utilitarianism, in The Difficulty of Tolerance 124 (2003) [hereinafter Utilitarianism]:
“Consider a principle under which some people will suffer severe hardships, and suppose that these hardships are avoidable. That is, there are alternative principles under which no one would have to bear comparable burdens. It might happen, however, that the people on whom these hardships fall are particularly self-sacrificing, and are willing to accept these burdens for the sake of what they see as the greater good of all. We would not say, I think, that it would be unreasonable of them to do this. On the other hand, it might not be unreasonable for them to refuse these burdens, and, hence, not unreasonable for someone to reject a principle requiring him to bear them. If this rejection would be reasonable, then the principle imposing these burdens is put in doubt, despite the fact that some particularly self-sacrificing people could (reasonably) accept it. Thus it is the reasonableness of rejecting a principle, rather than the reasonableness of accepting it, on which moral argument turns.”
Id. at 133. But see, e.g., Scanlon, Justification, supra note 32, at 2 (“Contractualism . . . makes the justifiability of principles depend on the reasons of certain kinds that individuals have to accept or reject them.”). If the suitable-motivation provision is doing work not already being done by reasonableness, this is where it has to pick up some slack: reasonable rejectability is from your (generic-reason) point of view, and suitable motivation is sensitivity to what others have generic reason to reject. Still I can’t see how there’s room for any space that will provide a way out of the dilemma I’m probing here. You are considering your own generic reasons, not surely your own idiosyncratic reasons or threat-advantage possibilities, and you’re considering others’ generic reasons, too. Others are doing the same. Your own generic reasons may give you reasons to reject; maybe you don’t yourself reject a principle because others have generic reasons that would lead them to reject it. But once you see they do, you know you can’t properly advance that principle, anyway. So it all comes out in the wash. That ends the annoying suspension of play while the umpires confer. We return to our regularly scheduled programming. Except that the whole damned program is nothing but relentless umpiring.
Anyway I think the permissive approach — everything can be reasonably rejected, so everything is wrong and nothing is right — is a more sensible way to take the dictates of the everyday understanding of reasonableness. Regardless, we have a problem. The theory was supposed to illuminate how to make sense of what’s right and what’s wrong. Scanlon has long conceded that particular normative judgments might not have any truth value.44×44. See, e.g., Scanlon, Being Realistic, supra note 32, at 2. But that’s very different from thinking that on one central issue of moral and political theory after another, the theory breaks down.
I know of one place where Scanlon has squarely addressed this problem. He writes: “It seems likely that many nonequivalent sets of principles will pass the test of nonrejectability. This is suggested, for example, by the fact that there are many different ways of defining important duties, no one of which is more or less ‘rejectable’ than the others.”45×45. Scanlon, Utilitarianism, supra note 43, at 133. (Here Scanlon takes the restrictive approach: plenty of alternatives can’t reasonably be rejected.) He goes on to propose that if actual social conventions embody one of the acceptable principles, that should be the winner; and he acknowledges that this move introduces an element of relativism to the theory. An easy rejoinder is that the problem remains daunting in cases where our actual conventions are indefensible; and Scanlon surely believes that when it comes to equality, in one domain after another, they are. But take the cases where convention does pick out one scheme that can’t be reasonably rejected from a larger set of such schemes. It’s not at all clear why critics of the current convention, with reasonable alternatives of their own, should be told, “Sorry, we’re prisoners of the status quo.” The salient worry might not be relativism, as familiar a bugbear as that is in moral philosophy. The salient worry might be a kind of conservatism. We want normative theories to have critical bite. Imagine this prefatory gloss on contractualism: “If and only if the status quo is simply unreasonable, we are free to canvass alternatives in the following way . . . .” Not for nothing did Marx complain that “[t]he tradition of all the dead generations weighs like a nightmare on the brain of the living.”46×46. Karl Marx, The Eighteenth Brumaire of Louis Bonaparte, in The Marx-Engels Reader, supra note 23, at 594, 595. You domesticate the complaint if you say it applies only to traditions that can reasonably be rejected.
Another line of argument is available, and I think Scanlon sometimes adopts this line, too. Maybe the problem flows from the everyday understanding of reasonableness, on which again plenty of alternatives can be reasonable — and on which you can be mistaken, though not grotesquely or transparently, but still reasonable. Maybe we should take reasonable as meaning what all the relevant reasons taken together and weighed appropriately finally counsel. That’s obviously a departure from ordinary usage. If the advocate of shoveling your own snow turns out to be right — if it is all things considered best to save money, not commodify, and pursue your health — then we have to swallow hard and agree that all the other approaches are unreasonable. Clearing a threshold isn’t enough.
Of course different people are differently situated. So you can hang onto the thought that just one course of action is reasonable, but insist on taking separate cases as they come. If you’re a gym rat, you don’t need the exercise as much, and maybe you have plenty of money and lots of childcare responsibilities: so for you it might be best, all things considered, to hire someone. But if my exercise consists in waddling my stertorous way to the kitchen, ripping open jumbo bags of Doritos, opening bottles of beer, and plodding back to the couch, and my budget is tight, and I live alone, well then for me it’s best, all things considered, to shovel by myself. (But what about the risk of a heart attack?) Fair enough, but the point remains. One and only one course of action is in this sense reasonable. Sure, one can imagine an occasional tie. But that’s nothing like worrying that we regularly face whole families of views that cross a threshold of reasonability.
Sometimes Scanlon seems to adopt this alternate understanding of reasonableness. He writes: “A claim about what it is reasonable for a person to do presupposes a certain body of information and a certain range of reasons which are taken to be relevant, and goes on to make a claim about what these reasons, properly understood, in fact support.”47×47. Scanlon, supra note 11, at 192. On the issue at hand, the sentence is perhaps ambiguous, but it is sensible to read it as saying that only the best all-things-considered position is reasonable. More striking, Scanlon writes:
If the objections to permission are strong enough, compared to the objections to prohibition, to make it reasonable to reject any principle permitting doing X in C, then one would not expect the objections to prohibition to be strong enough, compared to the objections to permission, to make it reasonable to reject any principle that forbids doing X in C.48×48. Id. at 195.
He goes on to discuss the case of two people, gone overboard from a sinking ship, who find one life jacket. On the everyday understanding of reasonableness, one would surmise that plenty of options — first come, first served; whoever’s strong enough to wrestle it away gets it; whoever’s younger; whoever’s life is of more social value; and plenty more you can fill in during your sadistic leisure time — are reasonable; so again either none of them can reasonably be rejected or they all can. Or you might think that the desperate scarcity here puts us outside what we now think of as Hume’s circumstances of justice and makes the quest for a moral solution quixotic or worse. But Scanlon holds:
A principle permitting each to struggle for the jacket at least has the merit of recognizing the symmetry of their claims and the need for some decisive solution. It would be reasonable to reject this principle if, but only if, there were some alternative that did this better (such as a principle requiring them to take turns or, unrealistic as it may seem, to draw lots). Similarly, a principle forbidding the use of force could not reasonably be rejected if there were some other (nonrejectable) method for resolving the matter.49×49. Id. at 196.
The best reading of these passages, I think, is that Scanlon is deploying not the everyday understanding of reasonableness, on which multiple options might be perfectly reasonable, but the alternate understanding of reasonableness, on which only the best option is.
The alternate picture gets rid of the worry that either all positions can be reasonably rejected or many can’t be. In principle, it gives us the kind of determinacy we’re looking for, at least as a general matter. But it comes with two whopping price tags attached. Not the departure from ordinary usage: ordinary usage has no special claim in theory-building. The first price is that it threatens to corrode our epistemic confidence in controversial normative claims. It’s one thing for a libertarian or a Scandinavian social democrat to think, “At the very least I have a serious contender.” That’s easy enough. It’s another to think, “Actually I have the right view.” It won’t do to say that there is no logical space between “I believe p” and “p is true” or “p is right,” that all of them simply assert p. There is nothing incoherent or dicey about thinking, “I’ve thought hard about political economy, and I think social democracy is the best view, but I could well be wrong about that.” This threatens to land us back in the vexing problem we faced with ordinary reasonableness: you just don’t know if you can reasonably reject another’s view, because it is exceedingly hard to say with a straight face that you’re confident that your view really is the best one.
The second price is that it leaves Scanlon’s contractualism teetering dangerously close to a truism. Remember the core proposition: a principle is right if no one can reasonably reject it. With the alternate understanding, that now becomes, a principle is right if all the relevant reasons, weighed correctly, finally counsel it. That’s not quite a truism: it tells us that the currency of normative argument is reasons, and that we have to take all the relevant ones and consider them. (“Wait: not just reasons, but reasons to reject”? On the alternate understanding I’m pursuing here, it seems to follow trivially that it’s reasonable to reject any view which is not in fact finally best supported by all the relevant reasons.) So it rules out, say, the view that a principle is right if it accords with your inarticulate intuitions. Even if intuitions are at some level an inescapable part of the story, in one argument after another we can do better than assert them and then stick out our tongues at each other. I quite agree. But — your mileage may vary, as the kiddies say — I confess I don’t see a whole lot of theoretical oomph in that.
So here’s my diagnosis. (The reviewer as quick-change artist: umpire’s uniform off, doctor’s coat on.) I suspect the “no one can reasonably reject” formula trades for its intuitive appeal on the everyday picture of reasonableness. It would be nice to imagine that on any issue there’s just one normative principle that it is goofy or indefensible or obviously out of line to reject. But the everyday understanding makes that unattainable, at least on a distressingly wide range of issues. The move to the alternate picture of reasonableness, on which only the single best view counts as reasonable, solves that problem. But it surrenders the intuitive appeal of the contractual formula. The dilemma here is a frontal challenge to Scanlon’s contractualism. You can get normative plausibility or you can get normative closure. You can’t get both — in this approach. I don’t see a satisfactory way to deal with this dilemma. And if I’m right in thinking Scanlon’s contractualism founders, it can’t supply any justificatory basis for his account of equality. Either it’s too easy to say it can’t be reasonably rejected, because neither can tons of rival views, or it’s too easy to say it can be reasonably rejected, because so can everything else in the terrain. Here Goldilocks either likes one bowl of porridge after another or she disdainfully turns up her nose at all of them; either she’s a shameless gourmand or an impossibly demanding gourmet; nothing turns out just right.
Where are we? Scanlon’s contractualism raises some puzzles and worries. They threaten his account of equality: we don’t know that it can’t reasonably be rejected, and we don’t know how to begin evaluating it if we jettison the contractualist machinery. But I don’t think these puzzles and worries have a philosophical solution. They have a practical solution. Its name is democracy.
In democratic politics — I don’t mean just voting or elections or legislative affairs — we don’t have to conjure up imagined others with imagined standpoints and try to figure out what they might say about our proposals. We can talk to actual others with real standpoints, people with different sorts of life experiences and different areas of knowledge, and listen to what they have to say. They can enrich or correct our stylized grasp of generic reasons; they can bring into sharper focus what the pros and cons of some principle are. If we think more than one policy option is reasonable, we can try collectively to figure out which one seems best. We can think group deliberation is epistemically more reliable than any one person’s arguments, however carefully considered and ingenious. That remains true if we think we are looking for the actual single best view. We have real equals in the real world, chock full of views of their own. They may know a whole lot more about the problems and possibilities of being an undocumented immigrant or a middle-aged laid-off salesman or a patient with a chronic illness struggling to make ends meet than we do. The collision between our arguments and theirs should improve our proposals. That is the promise of democratic debate or “government by discussion.”50×50. The phrase is initially from Walter Bagehot, Physics and Politics 115 (Beacon Press 1956) (1872), but it captures a central insight of liberal democratic theory. The debate can unfold among citizens, among their elected representatives, and between those two groups too. And we can have epistemic divisions of labor — think for instance of the roles of congressional subcommittees, administrative agencies, economists, scientists, and other experts — as long as they are accountable to the broader democratic public. As the discussion goes on, we provisionally adopt policies that currently seem justified. We can constantly revise our positions and our practices as we learn from experience and argument. Democracy after all is (in)famous for its irreverence to the past. Here’s John Stuart Mill, urging in Parliament that women be given the vote:
[T]he despotism of custom is on the wane; we are not now satisfied with knowing what a thing is, we ask whether it ought to be; and in this House at least, I am bound to believe that an appeal lies from custom to a higher tribunal, in which reason is judge.51×51. 28 John Stuart Mill, The Admission of Women to the Electoral Franchise (May 20, 1867), in Public and Parliamentary Speeches, 151, 153 (John M. Robson & Bruce L. Kinzer eds., 1988).
Now that is surely a ludicrous description of today’s politics in these United States. The wealthy do in fact have an outsized voice, even control. (Moaning and groaning about corporations, especially after Citizens United,52×52. Citizens United v. FEC, 558 U.S. 310 (2010). is misguided. I’m a lot more worried about Sheldon Adelson than I am about Greenpeace, and that’s not a covert way of advancing the left: I’m also a lot more worried about George Soros than I am about Massachusetts Citizens for Life.) Sometimes we practice brute-fact preference aggregation and mechanical interest-group pluralism, two obviously repellent views of democracy championed (in the name of realism! sigh) by roughly 13,284 economists and political scientists. More and more of us live in epistemic bubbles, where we read and watch and listen only to like-minded sources and adopt threadbare caricatures of our opponents. Slogans and sneers with the most tenuous connection to the actual landscape command the day. The quintessential ideas of reasonable (in the everyday sense!) disagreement and loyal opposition seem to be the stuff of maudlin nostalgia.
But principled arguments on the merits still comprise a real strand of our democratic politics, and we can always reform the practice to improve it.53×53. John Dewey, The Public and Its Problems (1927), still best read as in part a response to Walter Lippmann, Public Opinion (1922) and Walter Lippmann, The Phantom Public (1927); and, more narrowly but more lucidly, 14 John Dewey, Creative Democracy — The Task Before Us (1939), reprinted in The Later Works, 1925–1953, at 224 (Jo Ann Boydston ed., 1988). (That doesn’t mean eliminating interest-group pluralism or even preference aggregation. I don’t think democratic politics should be modeled on a boring graduate seminar, where everyone is trying to figure out the best argument but no one cares or has any particular stakes. But that’s another matter.) So yes, ours is a democratic system in considerable disrepair. But only a blind philistine could shrug off what we are doing as of no theoretical interest. And we can helpfully fan out to a broader set of collective and discursive practices. We can mine the common law of tort and consider what claims a wide range of actors have pressed over many issues and many decades, and try to make critical sense of what the law has done in response. We can think about the unfolding constitutional constraints on majoritarianism and what they tell us about principles of right.
An unvarnished pragmatist, I think we’re always working to improve our web of beliefs and practices, or anyway we always could and should be. (Okay, fine, we get time off to daydream and the like.) So we can change the ground rules for democratic politics, and we can revise our understanding of reasonableness, and so on across the board. Scanlon’s version of equality might turn out to be a big winner, but like absolutely everything else it is corrigible. Still the best way to test its merits is not to explore it in review essays or philosophy graduate seminars. It’s to articulate and pursue it in the world, in democratic politics, and see what happens. Not just the brute fact of whether or to what extent it wins or loses, but what we make on reflection of how things unfold. Will some of that reflection appear in law review pages or university press books? Sure. But that’s not where the action is.
* Don Herzog teaches law and political theory at the University of Michigan. He’s the author, most recently, of Defaming the Dead (2017) and is working on a proposal that we retire the concept of sovereignty. He thanks Liz Anderson, Scott Hershovitz, and the editors of the Harvard Law Review, especially Steve Schaus and ImeIme Umana, for comments on an earlier draft.