公 法 评 论

et revelabitur quasi aqua iudicium et iustitia quasi torrens fortis


Posner’s Charges: What I Actually Said

by Ronald Dworkin

In a recent issue of The New York Review of Books,[1] I reviewed Judge Richard Posner’s book, Problematics of Moral and Legal Reasoning.[2] That book (and his Holmes Lectures of the same name on which much of the book is based[3]) misreport my own opinions on a surprising number of important subjects, but that fact is of no general interest, and I did not think it appropriate to use my review to attempt any corrections. It does seem appropriate, however, to use the Internet for that purpose, since that medium can provide information to the few readers who might wish it without occupying valuable magazine or journal space.

Many writers believe that they are sometimes misrepresented, and I have myself been accused of misrepresenting others.[4] Nevertheless the number and character of Posner’s inaccuracies merits notice, particularly since in many cases the opinions he reports are the exact opposite of my own. Some of his descriptions are taken from other writers, though he embraces them as his own. That is a dangerous practice: it is all too easy for an author to find other critics of a view he wishes to criticize, including some who have misrepresented that view. A good search engine is no substitute for a careful reading of what one criticizes, and for one’s own arguments.

1. “Dworkin claims that his views on the merits of the cases he discusses are generated not by his personal ideology, which is left-liberal, but by impartial reflection on the principles that are seen to be part of law once positivism is rejected.”[5]

I have steadily insisted on the opposite. In Law’s Empire I argued that legal interpretation inevitably brings to bear the interpreter’s own political convictions. In the introduction to Freedom’s Law (a book Posner refers to repeatedly) I said, “Of course my constitutional opinions are influenced by my own convictions of political morality…I not only concede but emphasize that constitutional opinion is sensitive to political conviction… Of course the moral reading encourages lawyers and judges to read an abstract constitution in the light of what they take to be justice. How else could they answer the moral questions that abstract constitution asks them? …This book does indeed offer a liberal view of the American Constitution.”

Posner uses my supposed “claim” to support his further statement that my views are “in the line of descent from Herbert Wechsler’s influential article on ‘neutral principles’”[6] of 1959, and therefore that “In nearly every respect, Dworkin’s ‘moral reading’ of the Constitution was anticipated by the scholars of the later 1950s and early 1960s.”[7] I do not wish to claim originality for my work: that is the most foolish claim an author can make. But I have never accepted the idea of neutral principles, and Posner’s choice of the views that supposedly anticipated mine “in nearly every respect” is unfortunate.

(2) “[Dworkin holds] that judges who conceive of their function more narrowly than Dworkin thinks they should and so decline to appeal to a broad range of principles in deciding new cases, or who appeal to what Dworkin considers mere policies rather than principles, are lawless.”[8]

My view is, once again, the exact contrary. In Law’s Empire I distinguish between the concept of law and a variety of conceptions of law, that is, ways of conceiving what the law requires. One of these, which I called “conventionalism,” denies the relevance of moral principles in deciding what the law is. Another, which I called “pragmatism,” insists that judges should be guided entirely by “policies.” I argued for a different conception from either, but I recognized that my arguments it were not, because they could not be, “demonstrations”,[9] and I later said that “A critic might disagree with [my views] at many levels. He might reject law as integrity in favor of conventionalism or pragmatism or some other conception of law.”[10] There is no suggestion in any of this that those critics who do disagree with me are “lawless” because they are drawn to a different conception of law.

(3) “Dworkin’s polemic against the appointment of Robert Bork to the Supreme Court accused Bork, an influential constitutional theorist, of having ‘no constitutional theory at all’ … By ‘theory’ [Dworkin] means his own approach to law.”[11]

I devoted many pages to explaining that this is not what I meant.[12] I meant that Bork has no theory. Posner supports his charge by saying that anyone has a theory if his opinions have “some degree of generality or abstraction, and a demand for consistency.”[13] But my argument was an attempt to show that Bork’s various statements about constitutional adjudication defy generality or abstraction, and are deeply inconsistent. His opinions lack generality, I said, because though he makes constitutional interpretation rest on a choice among the various levels of abstraction at which the broad constitutional clauses might be read, he fails to supply any method of choosing which level is the appropriate one. His statements are inconsistent, I said, because he picks different levels of abstraction for different constitutional issues. Posner may find my argument weak or fallacious. But he should not accuse me of having no argument without reporting the argument I made.

(4) “[Dworkin] says that once Roe v. Wade was decided, early-pregnancy abortions imposed no moral cost comparable to the cost to pregnant women of being prevented from having abortions, because the Court’s decision diminished the moral entitlement of the fetus by depriving it of its rights. It seems to me that Roe v. Wade left the moral issue exactly where it found it. To think otherwise is to suppose that the Dred Scott decision made a positive contribution to resolving the issue of the morality of slavery or that Plessy v. Ferguson made a positive contribution to resolving the issue of the morality of racial segregation. We can see in these examples where thinking of the law as a branch of moral philosophy can lead.”[14]

The immediate cause of Posner’s mistake lies in failing to read three words in the sentence he refers to. I said, “The Court’s ultimate decision means that, in its view, an early abortion does not involve any violation of rights, and therefore that deciding earlier rather than later does not involve any moral cost of that kind.” (emphasis added.) I said, that is, not that the Court’s decision changed the moral facts, but that its decision took a position about what the moral facts are. But Posner’s overly hasty reading of one sentence does not fully explain his mistake. Throughout the book he refers to me as a “moral realist” who thinks that moral facts are independent of anyone’s belief or decision about them and he several times cites an article in which I have defended that view at great length. Why didn’t this suggest to him that the absurd view he here ascribes to me, which contradicts these philosophical positions, was an unlikely one for me to hold?

(5) Posner quotes the following statement of mine: “Anyone who is convinced that slavery is wrong, and knows that his view is now shared by almost everyone else, will think that general moral sensibility has improved, at least in that respect, since slavery was widely practiced and defended.”[15] He insists that this view is a “conflation of public opinion with moral opinion” and says that “this style of argument could have been used in the 1950’s to demonstrate” that moral progress had been made from the time of the Greeks who accepted homosexuality to the present day when most people condemn it.[16]

He has plainly misunderstood my statement, which doesn’t pretend to offer a “demonstration” of moral progress or even an argument for it. It simply states the obvious: that whether someone thinks there has been moral progress depends on his own moral convictions, so of course a homophobe would think there has been moral progress from the antique Athens to the Georgia one. Again, the context should have made Posner read the sentence more carefully. I was arguing that certain forms of skepticism are incoherent: I said that the familiar view that there is no such thing as moral progress was itself a moral claim, and that people who believe that their own moral convictions are popular cannot consistently deny that there has been progress since a time when their convictions were widely rejected.

(6) “Over the course of his career, Dworkin has endorsed as the legally ‘right answer,’ not just Brown without delay and racial quotas, but civil disobedience [and] nonprescution of draft card burners.”[17]

The syntactically unfortunate statement that “civil disobedience is the legally right answer” requires interpretation. Presumably the statement that Brown is the legally right answer means that the case was rightly decided, and that racial quotas are the legally right answer means that in some circumstances such quotas are not unconstitutional. But what can it mean to say that civil disobedience is the legally right answer? I suppose that it means that it is a good legal defense, in any criminal prosecution, for the defendant to claim that he broke the law out of conscience. Or that civil disobedience does not count as law-breaking at all. Or something similar.

I have written two articles about civil disobedience, and in neither did I say anything even remotely like that. In the first, I discussed a much more specific question: what attitude should citizens, prosecutors and judges take toward those who disobey a law — like the draft laws in the Vietnamese War — that is widely regarded as both immoral and unconstitutional. I offered reasons why, in such circumstances, criminal prosecutors should sometimes exercise a discretion not to prosecute, and why sometimes a court “ought to acquit on the ground that before its decision the validity of the draft was doubtful, and it is unfair to punish men for disobeying a doubtful law.”[18] (That reason for acquittal has nothing to do with civil disobedience; it appeals to a much more general principle.) I then added that “If acts of dissent continue to occur after the Supreme Court has ruled that the laws are valid, or that the political question doctrine applies, then acquittal on the grounds I have described is no longer appropriate.” In the later article I said, “We must not say that if someone is justified, given what he thinks, in breaking the law, the government must never punish him. There is no contradiction, and often much sense, in deciding that someone should be punished in spite of the fact that he did exactly what we, if we had his beliefs, would and should have done.”[19] This is the set of views Posner reports by saying that over the course of my career I have endorsed civil disobedience and the non-prosecution of draft card burners as the legally right answer.

(7) “Dworkin claims that ‘we have no choice but to ask [judges] to confront issues that, from time to time, are philosophical. The alternative is not avoiding moral theory but keeping its use dark.’ Substitute ‘teachers’ for ‘judges,’ ‘pedagogical’ for ‘philosophical,’ and ‘educational’ for ‘moral,’ and the fatuity of Dworkin’s claims becomes evident.”[20]

This supposed demonstration of my “fatuity” is a flop. Of course teachers confront issues that are the subject of pedagogical theory— about, for example, the best way to teach the multiplication tables. If some teacher were to announce that he was not relying, even tacitly, on any such assumptions, then (unless he were wholly incompetent) he would only be keeping his reliance dark. Perhaps Posner thinks that academic pedagogical theory is bad, and that teachers would do better to rely on their own experience and common sense. Even so, they would be relying on their own instincts or convictions about the very same matters that the educational theorists study, and Posner’s claim is not that judges are typically better at philosophy than academic philosophers are, but rather that judges do not need to have opinions about philosophical issues. The parallel claim would be that teachers do not need opinions about how to teach.

(8) “There is a considerable overlap between law and morality. Only it is too limited an overlap to justify a project of trying to align these two systems of social control, the sort of project that Islamic nations such as Iran and Pakistan have been engaged in of late. (So Dworkin and his allies are the Taliban of Western legal thought.)[21]

I assume that Posner does not mean, by bracketing me with the Taliban, that I decapitate people who disagree with me. But what does he mean? That I believe that every religious or moral obligation or responsibility should be enforced by law? I have never suggested such a mad project; on the contrary I have repeatedly argued that individuals have rights against the state that permit them to do what others might think the wrong thing without legal interference. That position is central, for example, to my argument for abortion rights in Life’s Dominion.

(9) “Dworkin appears to confuse slavery in ancient Greece with American Negro slavery. See [Objectivity and Truth: You’d Better Believe It] at 121 (reference to ‘the biological humanity of races they enslaved’). Greek slavery was not racial.”[22]

I did not confuse the two slaveries, and I did not say that Greek slavery was racial. I was discussing the reasons we have for wanting to explain why people have held political convictions different from ours. In one sentence I used our disagreement with the Greeks about slavery as an example of such a disagreement. Two sentences later I wrote, “We might be able to show, for example, that people who defended slavery held false empirical beliefs about the biological humanity of races they enslaved, or that slave owners were subject to special economic stringency that blinded them to slavery’s immorality, or that they lacked pertinent information of some other kind or were subject to other influences known to distort judgment.” There is no suggestion, in that sentence, of which of these explanations — or that any of them — apply to the Greek case. (In fact, however, the Greeks did rationalize slavery by supposing that slaves belonged to a biologically distinct order or race. See Aristotle, Politics I.13 [1260 a12] ("The slave has no deliberative faculty at all") and III.9 [1280 a32-4] "Slaves and brute animals... cannot [form a state], for they have no share in happiness or in a life of free choice" (Ross translation, Oxford: Clarendon Press, 1921).) I assume that Posner has not made the mistake of limiting “race” to pigmentation.

(10) “Yet Dworkin continues to insist that cases in which facts or consequences matter to constitutional decision-making are ‘rare.’”[23]

No sane lawyer could think that cases in which facts or consequences matter are rare: they are important in every case. In the passage Posner cites I was discussing a very different question. How often can constitutional cases be resolved simply by pointing out a fact one side to the controversy had missed? I said, “Of course, in some circumstances, pointing out that a doctrine will have surprising consequences – that a welfare program designed to help a particular group will actually harm that group, for example -- is obviously immensely helpful. But these circumstances are rare: most often controversy is not about what means will in fact achieve an agreed end, but about what end should be agreed – about how high efficiency should rank, for example, against social or distributive goals or the protection of rights or goals of integrity.”

(11) “On this construal of ‘democracy,’ every policy the activist wants can be sought from the Supreme Court in the name of democracy. Dworkin makes a similar point even more brazenly, ‘The American conception of democracy is whatever form of government the Constitution, according to the best interpretation of that document, establishes.’”[24]

My statement, particularly when read in context, does not suggest that every desirable policy can be demanded in the name of democracy; on the contrary, when read in context, it rejects that idea. It is not enough, I said, to find some conception of democracy that supports one’s demands in order to justify an interpretation of the Constitution that finds those demands in that document, because “‘Democracy’ is itself the name of an abstraction: there are many different conceptions of democracy, and political philosophers debate which is the most attractive.” That sentence precedes the one that Posner quoted.

(12) “See Ronald Dworkin, Taking Rights Seriously 198 (1977), explicitly rejecting the idea that social cost should influence the definition of rights.”

It is not easy to see what I said, on the page Posner cites, that prompted him to that report. In any case, his description is amply contradicted in my writings. Of course social costs figure in a legislature’s creation of legal rights when moral rights are not in play: I emphasized that fact throughout Taking Rights Seriously and Law’s Empire. I also emphasized that consequences and social costs can figure in the definition of moral rights. There is an important difference between appealing to social costs to justify compromising people’s moral rights, which I have opposed except in emergencies, and taking social costs into account in deciding what moral rights people have. A long section of Taking Rights Seriously is devoted to making that distinction, and illustrating the role that social costs properly play in that latter decision.[25]

(13) “Dworkin is a high rationalist with a weak sense of fact.”[26] “Dworkin rarely ventures outside the highly publicized domain of constitutional rights, and when he does so the results are unimpressive.”[27] “[Dworkin’s view] that philosophy can be helpful by telling us to reflect on the fairness of imposing liability without proof of causation [reveals] an ignorance of the relevant terrain and the thinness — the essentially rhetorical character — of Dworkin’s invocation of ‘theory’.”[28] “Dworkin has shown that he lacks a clear understanding of what ‘statistical discrimination’ means. Yet it figures prominently in the debate over affirmative action which he has defended in print.”[29]

These are examples of a different form of misrepresentation: hit and run insult. The first three of these accusations originally appeared in a 1997 law review article.[30] I replied, in the same issue, pointing out that the support he offered for each of the charges was contemptuously inadequate.[31] He had cited only a single example of my inability to read common law cases — my account of Cardozo’s decision in Buick v. MacPherson — and had misunderstood my claim, which was not about Cardozo’s opinion but about the best constructive interpretation of his decision. He had cited only two examples of my supposed inability to grasp facts. In one he charged me with a mistake of law, not fact. In the other he charged me with not having paid enough attention to a certain speculative sociological thesis for which he gave no citation, attribution or defense. It seems odd to rest so serious a charge as that I am “weak” on facts on so slender a body of purported evidence. He did not support his further charge — that I am ignorant of “the relevant terrain” of market-share liability cases — at all. In my reply, I suggested that it might be he who is unfamiliar with that legal terrain: I cited an abundance of cases and law review articles discussing market share liability that explicitly referred to “fairness” and “justice” in the moral, rather than any imagined special legal, sense of those terms. Posner simply lifted these various insults from his earlier article, however, and inserted them into his new book, unchanged, with no mention of my responses and with no attempt to answer those responses. The final charge — about statistical discrimination — is new in this book, but it is even more bewildering. He cites a footnote of mine that does not use the term “statistical discrimination,” or mention or use the concept it names, as evidence that I do not understand that concept. He does not explain how the footnote could possibly provide such evidence.

(14) “[Dworkin] runs moral relativism, moral subjectivism, and moral skepticism together, treating them as different names for what he calls ‘external [moral] skepticism.”[32]

I do not run these various theses together, nor — this is much more important — do I describe them as forms of “external” skepticism. The central argument of the article Posner cites was exactly to the contrary: that these positions can make sense — can even be intelligible — only as global forms of what I called internal skepticism, that is, as forms of skepticism that themselves rest on hidden positive moral beliefs or assumptions. Once again, his comment is a repetition: he made it in his Arizona Law Review piece, and in my response to that piece I pointed out the mistake. I said that if my actual argument were correct it would undermine many of his own opinions about morality and moral theory, and I therefore invited him to read my article more carefully and to explain if and why he thought it mistaken.[33] Instead, once again, he has simply repeated the mistake. This is particularly regrettable, because it has contributed to his making a dog’s breakfast of much of his essay.

(15) “Dworkin [calls pragmatism] an intellectual meal fit only for a dog (and I take it he does not much like dogs.)”[34]

This one is my fault. Long ago I said that pragmatism was philosophically a dog’s dinner.[35] My comment troubled Posner, who has referred to it repeatedly over the intervening years, taking it to have anatomical, then scatological, and now caniphobic implications. I thought the expression a familiar one; perhaps it is more familiar in British than in American usage. The Dictionary of Slang defines “dog’s dinner” as “Noun. 1. A mess. See also 'dog's breakfast'.” Under the latter entry we find: “Noun. A mess. E.g. ‘I've made a bit of a dog's breakfast of that essay.’” As it happens, I like dogs very much.


[1] March 9, 2000.

[2] Belknap Press/Harvard University Press (1999.

[3] 11 Harvard Law Review 1637 (1998).

[4] So far as I am aware, no one has given particulars of any misrepresentation of mine in the way I try to do in this note. Perhaps this example will encourage others to use the Internet for that purpose.

[5] Problematics, 287

[6] Problematics, 117.

[7] Problematics, 117.

[8] Problematics, 97

[9] Law’s Empire, 86.

[10] Law’s Empire, 262

[11] Problematics, 116-7.

[12] The most comprehensive statement of my argument is in Freedom’s Law, Chapter 14.

[13] Problematics, 117

[14] Problematics, 136

[15] The quotation is from my article, “Objectivity and Truth: You’d Better Believe It,” 25 Philosophy and Public Affairs 87 (1996).

[16] Problematics, 24

[17] Problematics, 76.

[18] Taking Rights Seriously, 221

[19] A Matter of Principle, 113-4.

[20] Problematics, 113. The quotation is from my article, “In Praise of Theory,” 29 Arizona State Law Journal 353, 375 (1997).

[21] Holmes Lectures, 1695

[22] Problematics, 19.

[23] Holmes Lectures, 1700. He cites my “Reply” to critics of my lecture, “In Defense of Theory,” , 29 Ariz. St. L.J. 432, 433. (1997).

[24] Problematics, 150. The reference is to Freedom’s Law, 75.

[25] Taking Rights Seriously, 294-330.

[26] Problematics, 253

[27] Problematics, 119

[28] Problematics, 120

[29] Problematics, 253.

[30] Posner, “Conceptions of Legal Theory: A Reply to Ronald Dworkin,” 29 Arizona State Law Journal 377 (1997).

[31] “Reply,” at 431.

[32] Holmes Lectures, at page 1643. Posner cites my article, “Objectivity and Truth: You’d Better Believe It,” 25 Phil. & Pub. Aff. 87, 88-89, as his source.

[33] For an explanation of these claims, see my reply to Posner’s Holmes Lectures, under the title, “Darwin’s New Bulldog”, 111 Harvard Law Review 1718 (1998). I have been accused of intending an insult in that title. It is a reference to Huxley, and therefore hardly an insult.

[34] Problematics, 240.

[35] “Pragmatism, Right Answers, and True Banality,” in Pragmatism in Law and Society (Brint and Weaver, eds. 1991) at 359.