Everyone believes in the rule of law. But how do we tell what the law is, especially for new and difficult cases? Can judges and lawyers just look it up—read it out of a book? Many people—politicians especially, when they are debating the confirmation of federal judges—say the answer is yes. Not Ronald Dworkin. Since the early 1970s, Dworkin has been the most distinguished exponent of the view that lawyers and judges cannot say what the law is, on any issue, without engaging their own values and principles, their own conceptions of right and wrong, their own best estimates of what makes a society a good society and what makes it just or unjust.
Professor Dworkin has expounded this position in several books that have had an enormous influence in legal philosophy and constitutional theory: Taking Rights Seriously (1977), A Matter of Principle (1985), Law’s Empire(1986), and Freedom’s Law (1996).1 Like most of these books, Justice in Robes brings together lectures, essays, and other interventions that have been published over the previous eight or nine years.2 But the collection is not leavened, as some of the earlier ones were, with essays on particular topics of public concern, like civil disobedience and affirmative action in Taking Rights Seriously, or abortion and pornography in Freedom’s Law. The only topical discussion in Justice in Robes is an eleven-page appendix to one of the essays, arguing with Richard Posner—the 7th Circuit judge, Chicago jurist, and self-styled “legal pragmatist”—about the style of adjudication that would have been appropriate for Bushv. Gore (2000).3
So Justice in Robes is not an easy read. A lot of it involves quarrels internal to legal philosophy, as Dworkin defends his theory against opponents like Posner, Cass Sunstein, Jules Coleman, Stanley Fish, Joseph Raz, Antonin Scalia, and the late H.L.A. Hart. Still, the underlying question is of the first importance—How do we reconcile the rule of law with judges’ deployment of their own values and their own theories of the good society?—and on that there is enough here that is new and important to make the book well worth reading for anyone interested in the law.
When Dworkin says that judges have no choice but to engage their own values, he is not saying that they should simply decide cases in the way they think would be best for all concerned. He is not calling for judges to substitute their own value preferences for those embedded by the Framers or the legislature or earlier generations of judges in existing legal materials. There are some commentators who talk like that. They call themselves legal pragmatists and they think that deference to existing legal materials makes little sense: they say that our Constitution is antiquated; they complain that our precedents require us to engage in labored and unproductive analogies. As for our statutes—well, they say, judges applying their own values are likely to do much more for the public good than politicians sitting in the pork barrels that we call legislatures. The best-known defender of this brand of legal pragmatism is—again—Richard Posner, who has established himself over the years as Dworkin’s most vociferous opponent.4
According to Dworkin, pragmatists like Posner have a very tenuous grip on the rule of law. Posner will say that if it is worth deferring to the antiquated clauses of the Constitution or to some well-established statutory or doctrinal rule, it is only because a modicum of value can be gained from not disappointing the expectations of those who are more invested than a pragmatist is in the meaning of these materials.5 For Posner, there is nothing more to the rule of law than that. Dworkin, on the other hand, insists that it is the law itself, not our own self-assurance, that requires us to incorporate value judgments into legal reasoning. And he insists, too, that the law requires us to incorporate them in a structured and disciplined way.
Consider the phrase “cruel and unusual punishment” in the Eighth Amendment. Dworkin repeats in Justice in Robes an argument he has made for many years now in favor of a “moral reading” of provisions like this. He rejects the “originalist” view that “cruel” means “what the Framers thought was cruel in 1791.” The Framers were educated men; they knew that standards of cruelty changed over the generations. They did not seek to freeze their own judgments on this matter in the way they did, for example, when they wrote down in Article II of the Constitution that no one under thirty-five was mature enough to be president. There they used a numerical age instead of the value term “mature”; here they used the value term “cruel” instead of a list of punishments that were frowned on in the late eighteenth century. Dworkin is absolutely right—in a chapter criticizing the jurisprudence of Justice Antonin Scalia—to wonder about the consistency of an originalism that refuses to acknowledge historical facts like these.
But it is not necessary to get into this debate about the Eighth Amendment to make the more general point about value judgments. The law is overflowing with terms and doctrines that command the judge to make the best value judgment he can: “unreasonable restraint of trade,” “the best interests of the child,” and so on. As Dworkin says, “Citizens, lawyers, and officials deciding what these provisions require in practice must indeed reflect on some of the same issues that they would consider if they were concerned only to act morally.”
On the other hand, in coming to a judgment about what is unreasonable or what is in someone’s best interests, the judge is often constrained by the decisions of earlier generations of judges. An earlier court may have decided, for example, that judges must not adopt the position in custody cases that a mother’s care is necessarily in the best interests of the child (even if they personally believe that), but that they must consider in each case the rival claims of the particular mother and the particular father. If that obligation has been laid down as a binding precedent, then the judge may be precluded from making the value judgment that he would make if he were coming unencumbered to the case.
But even when he applies a precedent, the judge still has to make value judgments of his own. Suppose he faces a custody case involving the rival claims of a mother and a former stepfather (the child’s father is deceased and the mother is divorcing her second husband). Is he precluded from dismissing the stepfather’s case out of hand? If the judge takes the view that a mother’s care is obviously preferable to that of a stepparent, is he violating the precedent? All cases are similar in some respects and different in others. In the absence of statutory guidance, who is to say whether a peremptory decision for the mother in this case is relevantly like the sort of decision that the earlier precedent forbids? Judgments of relevant similarity are not made by some mysterious intuition of analogy: the judge has no choice but to figure out whether the new case is like the old case in some important respect and he has to meld that value judgment with a value judgment about the importance of adhering to precedent in the first place.
That last point is very important. One of the challenges of legal reasoning is determining when precedents should be followed and when it is appropriate to depart from them in order to allow the law to grow. In the recent Supreme Court decision in Hamdanv. Rumsfeld about military commissions at Guantánamo Bay, Justice Clarence Thomas, writing in dissent, lambasted the plurality for refusing to adopt a flexible “common law” approach to the question of whether “conspiracy” (unaccompanied by any overt criminal act) could be charged as an offense against the laws of war. On the other side, Justice John Paul Stevens was adamant that “an act does not become a crime without its foundations having been firmly established in precedent.”6 Whether you agree with this or not, the matter is clearly one that requires a value judgment—that is, a judgment about the value of caution and fair warning versus the value of flexibility in the development of a fraught and difficult area of the law.
I have used some simple examples of uncertainty about the application and binding force of a precedent. In real life, judges confront not single precedents but tangled masses of precedent. A case that Dworkin uses several times involves a class of plaintiffs suing a number of pharmaceutical companies, who over a long period manufactured a drug the plaintiffs took during pregnancy (to prevent miscarriage).7 The manufacturers do not deny that they continued manufacturing and advertising the drug after it became clear that it had carcinogenic side effects, but each company says that unless a particular plaintiff can actually prove that its batch of pills caused her cancer, she has no remedy against that particular company. Is this right? Or may a court do what the California courts did and award damages against the companies on the basis of their proportionate share of the market without proof of particular causality?
Dworkin does not think the “market-share” theory should be rejected out of hand as unprecedented. (After all, just about the whole of our law of negligence has grown up case by case in ways that extended or went beyond existing precedents.) Nor, however, does he think we should decide the matter, like Posner, simply on the basis of what would be best for society. The question, says Dworkin, is whether the new solution corresponds in a principled way with what we have done already in the law—when we insist, for example, that those who profit from some risky enterprise should be required to bear some of its costs. Whether this principle has standing in the law and whether the market-share theory can be described as a plausible application of it—these are matters of value judgment that are embedded in the dispute over the damage done by the drugs and they cannot be avoided by the judge when he approaches a case as tangled and tragic as this one.
We sometimes talk about value judgments as though they were simply preferences. But there is nothing simple about the value judgments that are required of Dworkin’s judges. In easy cases, they may be uncontroversial; they may seem more like common sense than moral philosophy. But it is in hard cases—cases where the existing law is too meager (or too abundant) or where its precedents and texts are contested or ambiguous—that Dworkin’s blend of moral and legal philosophy really becomes important. That is where the judge has to get in touch with his inner Kant or his inner Rawls. “You cannot think about the correct answer to questions of law,” Dworkin writes, “unless you have thought through a vast over-arching theoretical system of complex principles….”
This has always been a controversial feature of Dworkin’s jurisprudence. He once labeled his model judge “Hercules” because of the labor-intensive character of his assignment.8 In reaching a decision in the pharmaceutical case, for example, Hercules would have to pore through all the relevant precedents, not just in the vicinity of the case before him (tort law) but all over the legal system (in contracts, too, and in regulatory law). He would have to figure out which of these precedents were important, and why. And Hercules would have to arrive at a finding that took account of the principles embodied in all or most of those precedents, in a way that made sense in light of a further theory that he would also have to come up with—a theory about why it is good to defer to preexisting statutes and holdings and constitutional texts, and, in light of that explanation, how the force of particular precedents should be assessed.
Opponents like Cass Sunstein, of the University of Chicago Law School, have argued that this is all too much: judges, they say, can and should get by with much less theorizing than this.9 As a practical matter Dworkin agrees. The reasoning of a real-life judge is necessarily approximate and it is constrained, too, by the conditions of compromise and collegiality on a particular court. As I have said, common sense applied to straightforward legal materials will be fine for most cases. But we still need ideal theory. From time to time a case will suddenly up the ante by challenging a conventional doctrine, or the bearing of a long-held precedent, or the entire way we think about the rule of law. When this happens, we have no choice but to consider what ordinary common-sense legal reasoning is an abbreviation of. And for that we do need a jurisprudence like Dworkin’s. We need to know what legal reasoning would be like at its best, in order sometimes to be able to make it better.
The emphasis on theory is one way Dworkin distinguishes the proper role of value judgment in legal reasoning from the simple imposition of a preference. Another distinction has to do with objectivity. When he says that judges have no choice but to apply their own values, Dworkin does not mean that a judge can simply say “I’m a conservative, therefore I’m entitled to apply my conservative values.” A judge is not like a legislator or a president, elected because of the values he holds and entitled to apply those values just because they are his. Dworkin’s judge takes a more objective attitude than that. He asks himself not “What are my values?” but “What is reasonable? What would be really just in this situation?” And he is supposed to deal with these questions as scrupulously as he deals with any questions that need to be answered in the course of legal proceedings, whether they are questions of fact or questions of law or questions of morality. In the end, the answers he comes up with will have to be given in his own voice (not as quotations from someone else), but he is supposed to be making his best effort to answer them objectively.
Dworkin is surely right to distinguish between a serious attempt to find the objectively right answer to a question of value embedded in the law and the simple knee-jerk application of one’s own preferences. And he is right to castigate those who say that value systems are just rival constructions and there is nothing to choose between them.10 But there is still a problem for jurisprudence.
None of this talk about objectivity, none of this denigration of post-modernism, makes the slightest dent on the fact that different judges asking and answering the objective questions of value that Dworkin’s jurisprudence requires will come up with different answers. There may be just one objective truth in these matters, but as long as it does not disclose itself unambiguously from the skies, our answers here on earth are bound to remain legion. And this matters in law. For there is no telling whose moral views one will be at the mercy of when one comes into a courtroom. We say we are a nation of laws, not men; but it is the men and women on the bench who have to answer the complex array of questions that, on Dworkin’s account, are posed in legal reasoning. And the answers will differ depending on the person, not depending on the law.
Is there any jurisprudence more in keeping with the rule of law than this? Obviously on the pragmatist account, things are even worse: it is Dworkin without the structure of legal reasoning I have described. But what about legal positivism, a view that identifies law with the rules prescribed by a particular authority, rules that can be recognized and applied by judges without engaging in any of these value pyrotechnics?
Legal positivism corresponds to most laypeople’s view of law. Law is something laid down (posited, hence positivism) by those empowered to make it and recognized by judges on the basis of shared criteria of authenticity. There is no particular reason, says the positivist, why law should correspond to what I regard as right. Those who make it may not be intending to do the right thing or, even if they are, their sense of justice may not be the same as my sense of justice. So I might as well put my moral faculties on hold; the alternative is embarrassment or worse. Every so often a defendant argues that a rule enacted by the legislature is too wicked to be law. But as the nineteenth-century positivist John Austin observed, courts are fond of demonstrating the inconclusiveness of this argument by hanging people in pursuance of the very laws whose validity they have impugned.11 That sounds brutal; a more benign version is that positivism helps us retain a healthy skepticism about the moral worth of our laws. The law is just the law; it may be good or it may be bad. If it is bad, it needs to be changed; but until it is changed, it is the law.
Legal positivism may seem comforting from the perspective of the politicians I mentioned at the beginning—debating the confirmation of federal judges. What they think they want in a judge is a reliable law-detector, one who can recognize law when he sees it and not adulterate it with values of his own. But the comfort is meager. Positivism offers no guarantee that the law on the books is complete or unambiguous—everything depends on the foresight and clarity of the lawmakers—and they acknowledge that in the hard cases where Dworkin’s method is most clearly applicable, there probably is no law to guide the judge. Positive law is like Swiss cheese; there are lots of holes in it and if a case presents itself in one of the holes, as hard cases inevitably do, the judge has no choice but to either send the litigants away with the lawsuit undecided or make new law himself to cover the case.
There is an important moral here. Of the three views I have mentioned—pragmatism, positivism, and Dworkin’s theory—not one of them says that judges can decide all or even most of the cases that come before them without relying on their own judgments of right and wrong. The only question is whether the judge’s value judgment is embedded in his legal reasoning or whether it is advanced as a substitute for the law’s inadequacy. Politicians pretend that there is a plausible jurisprudence that would not require judges to make their own value judgments. But they are simply wrong; there is no such jurisprudence. Value-free adjudication is not an option.
Dworkin devotes a lot of attention to legal positivism in this collection—more than he has since his disagreements in the 1960s and 1970s with the arch-positivist H.L.A. Hart.12 The attention he offers, however, will not be welcomed by the current guardians of Hart’s legacy, for they are largely uninterested in the possibility that positivism may offer comfort so far as the rule of law is concerned. That is too political for them. They are interested in legal positivism as an analytic answer to a philosophical question—“What is law?”—and Dworkin has little patience with such a purely conceptual approach to jurisprudence.
The positivism that interests Dworkin most—the legal positivism he wants to get to grips with—is the older jurisprudence of Jeremy Bentham, John Austin, and perhaps even Thomas Hobbes. These were theorists who thought that law could somehow supersede the vagaries and contestation of individual value judgment and give us something determinate that could stand in the name of us all. If there is disagreement in society about an issue like estate taxes and an executor is trying to figure out what to pay, the silliest thing for him to do is to ask the moral question whether the death tax is just; instead he should go to a lawyer and ask him to look up what we as a society have decided for the time being on this controversial question. If he cannot do this, then testators, their heirs, their financial advisers, and the tax collectors will be acting at hopeless cross-purposes as they try to plan their affairs. Positive law, then, is supposed to be a way of securing islands of coordination among individuals with different values, and also a way of securing predictability without reproducing the disagreements that flare up whenever we proclaim our own individual views about justice or the general good.
As well as that argument about coordination, there is also an argument about the work that legal positivism can do to protect democratic rule and reduce the power of unelected judges. Jeremy Bentham is credited specifically in this regard. (As well as teaching legal philosophy at New York University, Dworkin is currently the Bentham Professor of Jurisprudence at University College, London, where the old utilitarian’s remains are mummified.) Bentham, says Dworkin, “hoped to undermine the political power of judges who claimed to have discovered law in natural rights or in ancient traditions beyond what Parliament, as the congress of the people, had explicitly declared.” Law was what an elected Parliament enacted, whether or not that corresponded to the class-bound value judgments of members of the judiciary or not.
As I have said, Dworkin takes these arguments seriously—indeed, he takes them more seriously than most modern positivists in our law schools. But in the end he does not find them convincing. The coordination argument “presupposes…a degree of deference toward legal authority that almost no one shows in modern democracies.” And as for the democratic argument, the ascendancy of purely positive law might have seemed plausible in Bentham’s day, “in an age of simpler and more stable commerce and a more homogenous moral culture.” But it won’t work for us, says Dworkin. The statutes that regulate our commerce and industry cannot function as detailed codes. They consist of general statements of policy that require elaboration by regulatory agencies. Increasingly the judge’s role is to scrutinize the reasonableness of these administrative decisions in the light of the authorizing legislation, and that takes us straight back to the value judgments which assessments of reasonableness involve. Law simply cannot work in a modern society in the way that positivists say it is supposed to work; it cannot provide the sort of crisp, detailed guidance that would be necessary to supplant our own judgments of fairness and reasonableness. It certainly cannot work like that in modern America where our Constitution has been set up to fuse moral and legal questions in exactly the sort of mélange that Dworkin’s jurisprudence attempts to parse.
Throughout his career, Ronald Dworkin has “argued that [legal positivism] is not faithful to the actual practices of citizens, lawyers, and judges in complex political communities: in practice…people who argue about the content of law draw on moral considerations in a way that positivism cannot explain.” We have no choice, he says, but to embrace the complexity and the dissensus that the melding of legal and value questions is likely to generate.
So here is the final question: Can Dworkin really hope to reconcile all this with the rule of law? Certainly, for difficult cases he cannot reconcile it with a conception of legality which requires predictability or fair warning. And it has to be said that, for most theories of the rule of law, predictability and fair warning are the key: that is why the rule of law is understood to prohibit secret laws and ex post facto laws and laws that are so obscure in their terms or change so frequently or vary so much from judge to judge that nobody knows where they stand. That is what the rule of law means for most people. And Dworkin’s jurisprudence is at odds with it. By presenting law as the argumentative and contestable upshot of Herculean consideration of different precedents and doctrines and the values implied by them, a method that demands much abstract theorizing, Dworkin risks turning law into exactly the thing that many doctrines of the rule of law have always said it shouldn’t be: arcane, inscrutable, and changing unpredictably from judge to judge.
Still, we must not insist on what we cannot have. Theorists of the rule of law who celebrate clarity and predictability have yet to demonstrate that these virtues are available for legal reasoning in hard cases. If they are not available, then it is wrong to reproach Dworkin for failing to secure them and we should certainly not be in the business of sacrificing the other virtues that his jurisprudence does secure for the sake of these chimeras. Remember, Dworkin can always play the Posner card: Do you really want a legal philosophy that abandons structure and constraint altogether and dissolves complex argument in the equivocal platitudes of pragmatist “common sense”? Or will you accept a jurisprudence that at least tries to take seriously aspects of the rule of law that are not necessarily discredited by an impracticable emphasis on predictability?
After all, the rule of law has never been just one thing. It is a complex and contestable concept. The factor of predictability is one important part of it. But fairness in the sense of treating like cases alike is also a crucial component in the heritage of legality. Citizens cannot plausibly insist that judges turn themselves into machines, that they reason in exactly the same way, or that they avoid any intellectual exertions that might show up the differences in values between them. But they can insist that any judge addressing their case should try as hard as he can, by his own lights, to judge them by standards that have been applied already to others. That’s what it means to treat like cases alike. They can reasonably expect each judge to make the best effort he can to resolve their case on the basis of coherent principles that also make sense of the way other cases have been resolved in the community they belong to. Dworkin argues that the importance of this aspiration—an aspiration to coherence, to integrity in legal reasoning—is undiminished by the certainty that different judges will approach this exercise with different values and come up with different conclusions. I am not sure whether he is right about that, but I am convinced that Justice in Robes does more to open up new options in our thinking about the rule of law than a whole library full of books by opinionated pragmatists or by the careful conceptual purveyors of a purely analytic jurisprudence.
August 10, 2006
All four books were published by Harvard University Press. ↩
Law’s Empire is the exception; it is a sustained monograph, not a collection of essays, and it presents Dworkin’s mature legal theory as a continuous argument. ↩
Bushv. Gore 531 U.S. 98 (2000). See Richard Posner, Breaking the Deadlock: The 2000 Election, the Constitution, and the Courts (Princeton University Press, 2001). ↩
See, e.g., Richard Posner, The Problematics of Moral and Legal Theory(Harvard University Press, 1999), pp. 240–242. ↩
Hamdanv. Rumsfeld, No. 05-184, decided by the U.S. Supreme Court June 29, 2006. ↩
Sindellv. Abbott Laboratories, 26 Cal.3d 588, 607 P.2d 924 (1980). ↩
See Dworkin, Taking Rights Seriously, pp. 105–130. ↩
See Cass Sunstein, Legal Reasoning and Political Conflict (Oxford University Press, 1996). ↩
This castigation was conducted at much greater length in Ronald Dworkin, “Objectivity and Truth: You’d Better Believe It,” Philosophy & Public Affairs, Vol. 25 (1996), p. 87. For some reason this essay is not included in Justice in Robes. ↩
John Austin, The Province of Jurisprudence Determined (1832), Lecture V (note), p. 158 of the Wilfrid E. Rumble edition (Cambridge University Press, 1995). ↩
For the original disagreements, see Dworkin, Taking Rights Seriously, Chapters 1–4. One chapter of Justice in Robes is a response to a “Postscript” that was appended posthumously to Hart’s book The Concept of Law when it was published by Oxford University Press in a new edition in 1994. Hart had tried in the “Postscript” to settle scores with Dworkin, and Dworkin attempts in this volume to have the last word in that debate. ↩