• Email
  • Single Page
  • Print

Law from the Inside Out

Ronald Dworkin contributed over one hundred articles, reviews, and comments to these pages starting in March 1968. The following was originally given as a talk at the Palazzo del Quirinale in Rome in November 2012, when he received the Balzan Prize for his “fundamental contributions to Jurisprudence, characterized by outstanding originality and clarity of thought in a continuing and fruitful interaction with ethical and political theories and with legal practices.” It is presented here with small changes. Ronald Dworkin died on February 14 of this year.

dworkin_1-110713.jpg
John Earle
Ronald Dworkin, 2005

Law is at the cutting edge of many different disciplines and I am going to try to illustrate this point by talking about my own career, not because I believe that everything I think is right, but because my career has illustrated a marked trajectory from the very concrete to the very abstract.

My last book, Justice for Hedgehogs (2011), offers a panorama of the work I have done over half a century: not just a survey but an integration, trying to explain how it all fits together. I began my professional life as a young lawyer in a grand Wall Street law firm. My work could not have been more detailed and less abstract then. I wrote elaborate bond indentures and studied the balance sheets of giant corporations, helping them to satisfy the laws that allowed them to raise more money and grow greater still.

Since then, in an academic career at several institutions, my interests have grown steadily more abstract. But in each case the intellectual pressure I felt developed from the bottom up, not the top down. I took up steadily more abstract philosophical issues only because the more practical and political issues that first drew my attention seemed to me to demand a more philosophical approach to reach a satisfactory resolution. I will try to illustrate that process of philosophical ascent here.

When I left Wall Street to join a law school faculty, I took up a branch of law—constitutional law—that is in the United States of immediate and capital political importance. Our Constitution sets out individual rights that it declares immune from government violation. That means that even a democratically elected parliament, representing a majority opinion, has no legal power to abridge the rights the Constitution declares. But it declares these individual rights in very abstract language, often in the language of abstract moral principle. It declares, for example, that government shall not deny the freedom of speech, or impose cruel punishments, or deprive anyone of life, liberty, or property without due process or law, or of the equal protection of the law.

The Supreme Court has the final word on how these abstract clauses will be interpreted, and a great many of the most consequential political decisions taken in the United States over its history were decisions of that Court. The terrible Civil War was in part provoked by the Supreme Court’s decision that slaves were property and had no constitutional rights; racial justice was severely damaged, after that war, by the Court’s decision that racially segregated public schools and other facilities did not deny equal protection of the law; a good deal of Franklin Roosevelt’s progressive economic legislation was declared unconstitutional because it invaded property rights and so denied due process. These were the bad decisions that everyone now regrets. There have been very good decisions, too: in 1954 the Court, reversing its earlier bad decision, declared that segregated schools were inherently unequal, and therefore did deny equal protection of the law.

It is therefore a crucial question how courts should interpret the abstract constitutional language: What makes a particular reading of that language correct or incorrect? But the Constitution is law—it declares itself to be the most fundamental law—and so the question how it should be interpreted is really only one form of the ancient jurisprudential question: What is law? How should judges decide what the law of some nation really is on some particular subject? Over centuries philosophers have disagreed. In recent times one theory of law—it is called legal positivism—has been particularly influential. This declares that what the law is on some subject in no way depends on what the law ought to be. What the law is depends, according to positivism, not on morality, but only on history: on what people given the appropriate authority have declared it to be. We discover what the law on any subject is, according to positivism, by identifying those in authority and finding out what they have said.

In most cases this is a relatively easy matter. There are books that set out civil codes and record the other decisions of parliaments and the past rulings of judges. Lawyers know where these books are kept, and how to read them. But what if lawmakers speak in abstract language and what they say can be read in different ways? The United States Constitution forbids “cruel” punishments. Does that rule out capital punishment—the death penalty?

Two answers are open to legal positivists. They can say that since the “framers” who made the Constitution didn’t say one way or the other, there is no law on the subject, and the Supreme Court is therefore free to make up its own mind. But that sounds profoundly undemocratic. Or a positivist can say that the answer depends on what the framers intended, or expected, or would have declared if they had thought of the question. In this case that strategy decides the issue. Since capital punishment was widely used in the United States in the early eighteenth century, when the framers published their clause, they could not have intended or expected to declare it unconstitutional. Therefore it is not, even now, unconstitutional.

So it is of great practical consequence whether legal positivism is a sound philosophy of law; that is important not only in the United States but now in all the many other countries, including Italy, that subscribe to abstract constitutional rights. One can’t properly understand or participate in constitutional law without taking a stand on that subject. So I did. From the 1960s until the present, I have argued in a variety of books and articles that it is very much unsound. It provides only an incompetent description of the actual practice of law. Lawyers and judges typically make claims about what the law actually is that cannot be thought to be grounded just in what authoritative bodies have previously declared.

Even more fundamentally, legal positivism is based on severe misunderstandings in the philosophy of language. It assumes that we share the concept of law the way we share the concept of a triangle, that is, that we all agree on the tests to use to decide whether a legal claim is true or false. But we do not. For us, the concept of law—like other political concepts such as the concept of justice—is an interpretive concept: a theory of law is a normative claim about the tests that we should use to judge claims of law. A theory of law is a special kind of political theory, and so what law is cannot be separated completely from what it should be.

Once we reject positivism, however, we need another, different theory of law. In a series of articles and then in Law’s Empire, in 1986, I offered a theory. A claim of law should be understood as a claim about the best interpretation of past and contemporary legal and political practices in the nation in question. We need a theory of what interpretation is—about what counts as a successful interpretation—in order to fill out this interpretive theory. I proposed what I came to call a “value” theory of interpretation. An interpretation must fit the data—it must fit the practices and history it claims to interpret—but it must also provide a justification for those practices. It must, as I sometimes put it, show the practice in its best light. The first requirement—of fit—is not enough on its own, because more than one interpretation of a complex set of legal data may fit well enough to count. The second requirement—of justification—is therefore crucial. It shows why positivism must fail as a general theory of law. We cannot identify law without assuming some justification, however weak, in political morality.

This account of what interpretation is, and of what makes one interpretation succeed and another fail, cannot hold only for legal reasoning, however. It must hold for interpretation in general, and so I was required to explore other domains of interpretation. In one important chapter of Justice for Hedgehogs, I concentrated on artistic and literary interpretation. I tried to show how the “value” theory of interpretation illuminates the agreements and conflicts among critics in all these domains.

The interpretive theory of law raises deeper questions about moral philosophy. It supposes that one claim about the law—that capital punishment is unconstitutional, for instance—can be true and its opposite false, even though its truth depends on a moral theory about the best interpretation of the Constitution as a whole. But for many decades now most influential moral philosophers have denied this. They insist that claims about morality—that capital punishment is morally wrong, for example—are not really judgments and so cannot be either true or false. Since judgments about moral rights and duties cannot be empirically tested, these philosophers say, it makes no sense to think that they are either true or false. We must either concede that such judgments are, strictly speaking, nonsense, or we must suppose that they are not really judgments at all, but only expressions of emotion or recommendations for conduct.

Now this “anti-realist” theory about morality, as it is called, would cause great trouble for the interpretive account of law I defend. We would have to say something that seems crazy, which is that nothing we say about the law is true—or false either. That may be an attractive conclusion to reach in a philosophy seminar room, but not in a court of law. A judge who sentenced a defendant to jail while admitting that the judge’s own view of the law is only an emotional expression would probably be sent to jail himself.

So I had to take sides about this deep issue of moral theory. In an article in 1991, and then in a fuller version of that article in the first part of Justice for Hedgehogs, I argued that the “anti-realist” view is in fact incoherent. Consider the proposition that rich people have no moral duty to help the poor of their own community. If that proposition is not true, then it is not true that rich people have that duty, and that is itself a moral claim. If no moral claim can be true or false, then that one can’t be true either, so anti-realism is self-defeating.

Now that simple statement may seem too quick an argument, and in fact it took me many pages to explain it. I had to consider a great variety of metaphysical arguments, proposed by a great many very distinguished philosophers, for the position I called incoherent. I cannot attempt to summarize those arguments now, but only to emphasize that this excursion into metaphysics was not for me optional. It was required by way of defense of a theory of law, which was in turn required by politically very important positions I wished to defend in constitutional law. We cannot pick apart these various philosophical issues and call some of practical and others of only theoretical importance. Politics and philosophy are much too closely integrated for that.

  • Email
  • Single Page
  • Print