On Reading the Constitution
By Laurence H. Tribe and Michael C. Dorf
(Harvard University Press, 144 pp., $18.95)
Although the problem of constitutional interpretation has produced an avalanche of complex and sometimes barely digestible theorizing, in one sense the issue is very simple. Everyone agrees that the Constitution is law. And If the Constitution is law, then it stands above politics. All public officials — Democratic, Republican, or something else — must obey. The Constitution does not just mean what particular people want it to mean. If it did, it would not be law.
The problem is that the text of the Constitution is often extremely vague. Reasonable people disagree about what it means. Some people think that a Constitution that guarantees "the equal protection of the laws" requires race-conscious affirmative action policies, in order to remove the effects of past and present discrimination; other people think that a state that engages in affirmative action is violating that same guarantee. Some people think that a Constitution that guarantees "the freedom of speech" has no bearing on, and even argues in favor of, severe limits on expenditures on elections; other people think that such limitations violate that same guarantee. Both groups are perfectly capable of reading English, and both believe that their views are compelled by the Constitution.
On so many of the central constitutional questions, then, the Constitution's words tell us much less than we need to know. And since the words of the Constitution do not resolve hard constitutional cases, it seems undeniable that people who interpret the Constitution have to look to something other than those words in order to do their jobs. This idea seems to many both unavoidable and intolerable, the latter because it endangers the status of the Constitution as law. It means that the interpretation of the Constitution must inevitably be based on principles that are external to the words of the Constitution itself. Those principles have to be created rather than found: the Constitution does not contain the instructions for its own interpretation. And if interpretive principles must be created by the judges, it seems again to many that the Constitution could mean almost anything at all, and is not law after all.
In the last generation, many professors and judges have tried to answer this problem. Robert Bork's response has won a fair number of adherents among citizens and public officials, including important judges. Bork, thinks that the words of the Constitution carry the meaning that was originally understood by those who ratified it. There are many difficulties, however, with this view. Some of them are historical: How do we figure out what the numerous ratifiers, who often disagreed with each other, understood their document to mean? Some of the problems are more broadly interpretive: How do we go about applying the original understanding to problems that the ratifiers never anticipated, like electronic eavesdropping? Or, how do we decide whether the original understanding identifies general concepts whose meaning changes over time, or instead particular conceptions limited to the specific issues that the ratifiers sought to address? History does not answers such questions.
But for those who share Bork's view, the real problems go deeper. Bork and his followers act as if the Constitution and the original understanding are the same thing. In fact, however, they are not. The Constitution does not say that it should be interpreted in accordance with the original understanding. The Constitution does not say that judges should follow the original understanding. The proposal that we should follow the original understanding is itself an interpretive principle. Like any other such principle, it must be justified on substantive grounds.
The popularity of Bork's approach is disturbing, not because of its advocacy of a form of judicial restraint (which may be justified), and not because the original understanding is not relevant (surely it matters a lot), but because his confident presentation disguises the dependence of his own position, at every turn, on judgments of policy and principle that he never defends, or even recognizes. Conservatives, though, have no monopoly on the view that a particular, controversial set of interpretive practices is inevitably part of what it means to be faithful to the founding document. Justice Hugo Black, one of the fiercest defenders of civil liberties in the Court's history, made a career in large part out of denying his actual reliance on his own values and commitments. Black's legacy can be found in the position of many other liberals who act as if the liberal consensus from, say, 1968 to 1980 is, necessarily and without resort to controversial interpretive principles, itself a part of the Constitution.
If there is sometimes no way to interpret the Constitution without relying on external principles, how are we to proceed? Some people think that those principles might be found in our "tradition." Thus, for example, Griswold v. Connecticut (protecting the right of married people to use contraceptives) and Roe v. Wade (protecting the right to have an abortion) might be justified on the basis of the "traditional" protection accorded to marital privacy or control over one's body. But there is a significant problem with relying on tradition as the basis of constitutional interpretation: defining the level of generality at which the tradition is to be described. If we describe the tradition at a low level of generality — is there a traditional right to use contraceptives? to have abortions? — we will decide cases quite differently from how we would if the tradition is described in broad terms (as including, say, control over one's body). How do we decide at what level of generality to read traditions? And why exactly does tradition matter in any event?
In the last decade or so, under the influence of certain new methods of literary study, it has become fashionable to respond to problems with the original understanding and tradition by saying that the Constitution could mean anything at all, that it is "indeterminate," But this extreme claim does not at all follow from the need to resort to interpretive principles. For one thing, the text of the Constitution, accompanied by principles on which there is a consensus, sometimes admits only one answer. For another, what the Supreme Court now thinks, and what it will say about questions not yet decided but likely to arise soon, can be predicted with considerable accuracy.
Moreover, some interpretations of the Constitution are better than others. Some interpretations of ambiguous provisions will produce an unacceptable increase in judicial discretion; others will leave politically weak groups at the mercy of the state; others will make political liberty too fragile; others will offer too little room for democratic experimentation; others will he insufficiently respectful of the claims of history. Where the Constitution does not speak clearly, there is no choice but to evaluate interpretive principles by reference to questions of this sort. In any case, it is surely possible to evaluate the relevant interpretive principles, and to do so by reference to reasons. We might therefore conclude that those interested in making sense of the process of constitutional argument must develop principles of interpretation that are good ones — that is, principles that are based on good reasons. There is simply no escape from this task.
Much of the focus of this book by Laurence Tribe and Michael Dorf is on the extremely controversial case of Bowers v. Hardwick. In Hardwick (argued and lost, by a 5-4 vote, by Tribe himself), the Supreme Court concluded that the Constitution does not protect the right of homosexuals to engage in sodomy. Hardwick had been arrested for participating in an act of oral sex (defined by Georgia law to count as "sodomy") with another man in Hardwick's apartment. As Hardwick described the relevant events, a police officer had seen him leave a gay bar, followed him to his apartment, entered it with an arrest warrant for drinking in public, and opened the bedroom door.
To Hardwick, the Georgia sodomy law violated the right of sexual privacy recognized in Griswold v. Connecticut, Roe v. Wade, and other cases. Hardwick claimed that if the Constitution protects the right to use contraceptives, and to have an abortion, it must also protect the right to sexual freedom, heterosexual or homosexual, within the home. According to the Court, however, the text of the Constitution does not protect the right of homosexual sodomy, which is hardly respected by American traditions, and which is fundamentally different from the privacy rights involved in previous cases centering on issues of family and reproduction. Tribe and Dorf believe that Hardwick was "egregiously wrong." Part of their goal is to explain why the Constitution might protect the right to engage in homosexual sodomy without turning the document into a blank slate for judges to fill in as they wish.
This book amounts to an energetic and often highly illuminating discussion of how constitutional interpretation inevitably involves substantive choices but is not simply a matter of making things up. Tribe and Dorf are quite effective in showing that any system of interpretation depends on value judgments, and that neutrality in interpretation, in the strongest sense, turns out to be a hopeless charade. Thus, for example, they argue, against Bork, that the original understanding is not in the Constitution, and that only its words, and not "the secret, hidden, and unenacted intentions," count as part of the founding document. Thus they effectively criticize the view, recently proclaimed by Justice Scalia, that any relevant tradition should be described at "the most specific level of generality." Tribe and Dorf show that the decision so to characterize the relevant tradition is itself a value judgment, and one that would throw many well-established constitutional rights out the window.
Still, Tribe and Dorf are not concerned only to demonstrate that constitutional decisions require value choices. They want also to demonstrate something more: "how the Constitution channels choice." Here their book is, I think, somewhat less successful. What they have to say seems to boil down to four points. (1) The Constitution should be read as a whole; each provision should be interpreted in light of other provisions. (2) The Constitution is not based on any single theory and is not a unified whole; different provisions look in different directions, and contradictions should be expected. (3) Prior decisions should be read by reference to their rationale as well as their holdings; the reasons given by previous courts count. (4) Courts should create general rights not out of the blue, but with attention to the specific liberties that the Constitution unambiguously ensures.
The problem with these claims is not that they are wrong. Quite the contrary, all of them seem right. The problem is that these points, taken separately or together, are too general, too vague, too open-ended to provide the necessary help for those who are seriously concerned about the risks of judicial discretion or about the development of interpretive principles in a constitutional democracy. Certainly they do not by themselves justify the authors' particular views on disputed cases, including Bowers v. Hardwick. In the end those views must be justified on external grounds, not on the basis of abstract claims of this sort.
Take the proposition that each constitutional provision should be read in light of the others. It offers precious little help in real cases. It would allow highly plausible arguments for and against the constitutionality of affirmative action, the death penalty, federal regulation of state government, and discrimination on the basis of sex and disability. One might argue, for example, that the cruel or unusual punishment clause forbids the death penalty, because it should be read in light of the equal protection clause, and because the death penalty tends to be administered in a racially discriminatory fashion; or that the same clause permits the death penalty, because it should be read in light of the Fifth and Fourteenth Amendments, which operate on the assumption that the government will sometimes take "life."
The notion that constitutional provisions should be read together, in fact, probably excludes no position now held by any serious person. And it becomes even less helpful when it is combined with the authors' second principle, which dramatically reduces its weight. To say (rightly) that we do not have a consistent constitution, and that constitutional provisions point in quite different directions, is to say that any particular provision gives little guidance with which to understand other provisions.
What about precedent? Every first-year law student learns that courts should usually respect the reasoning as well as the result of previous cases. Every second-year law student learns that this principle offers little assistance in hard constitutional cases. In light of the multiplicity of precedents and the difficulty of figuring out "the rationale" of any one of them reasonable people can construct quite different understandings of any line of precedents. Many of the Warren Court's racial discrimination cases, for example, appeared to turn on both an antipathy to measures ensuring white supremacy and a belief that the government may not ever take race into account — ideas that cut in different directions with respect to the constitutionality of affirmative action. How does one apply their "rationale" to that problem?
The answer to such questions depends on something other than previous rationales. Indeed, Tribe and Dorf do not even insist on the requirement of respecting those rationales, especially in light of their disapproval of Bowers v. Hardwick. The actual reasoning of the early privacy cases, taken on their own terms, seemed not to extend to homosexual sodomy at all.
As for their fourth principle, the authors generalize from the First (protecting speech), Third (preventing the quartering of troops), and Fourth Amendments (protecting against unreasonable searches and seizures) a general right to immunity from government of "consensual intimacies in the home." The argument is not implausible. it seems sensible to say that if the Court is to give substantive protection to liberty in general, it should do so by reference to things specifically accorded constitutional protection. Yet it might be equally plausible to construct instead from those very amendments — especially when combined with the Fifth (protecting against the taking of property without just compensation) — a general right against redistributive regulation by government, including, for example, minimum wage and maximum hour laws.
We might say, for example, that the First Amendment shows an aversion to government interference with private behavior; that the Third is rooted in an effort to immunize property from public incursion; that the Fourth reflects the same idea; that the Fifth attempts to ensure that property can be taken only with compensation. In light of all this, perhaps "liberty" should include the right to freedom from any governmental effort to take resources held by one person for the benefit of another. Minimum wage and maximum hour laws would of course run afoul of this principle; indeed, such a right would endanger much of what the federal government has done since the New Deal. But Tribe and Dorf are hardly in favor of that.
All this suggests that the process of generalization by reference to specific rights involves an inescapably large measure of discretion. A decision to infer a right lo homosexual conduct within the home, and not to freedom from redistributive regulation, has to depend in large measure on substantive views about homosexuality and redistributive regulation — not on what the Constitution specifically protects. What is necessary, then, is a full-scale defense of those substantive views, and a full-scale explanation of why they should be implemented by constitutional courts.
I doubt that Tribe and Dorf would disagree with any of this. Their analysis is open-minded, non-dogmatic, and tentative: "More often than not, we have no answers, and those we offer are almost never held with certitude." And I do not believe that any of their largest claims is wrong. I do not even believe that those claims offer no help at all. The real problem is that in order to ascertain the appropriate role of the Supreme Court in the constitutional system, their basic principles tell us far too little.
On Reading the Constitution reminds us of the extent to which our understanding of constitutional interpretation remains in a primitive state. There seem to be several ways to improve the present state of affairs.
Much more empirical work needs to be done on the real-world effects of constitutional decisions. Lawyers (and not a few non-lawyers) seem to think that when the Supreme Court speaks, the country changes course accordingly. But often it isn't so. Ten years after Brown v. Board of Education — the usual symbol of Supreme Court efficacy in changing social norms and practices — no more than 2 percent of black children in the South were attending desegregated schools. Moreover, there is some (though not decisive) evidence that, contrary to the conventional wisdom, Brown did not fuel the civil rights movement, that it played only a small role in bringing about civil rights reforms. There is also some evidence that Roe v. Wade did not substantially increase the number of abortions, did not fuel the pro-choice movement, and indeed helped demobilize the women's movement and create the moral majority (though it did increase the number of safe abortions, a genuinely dramatic change).
Research on questions of this sort is in its earliest stages. At the moment the most that can be said is that an aggressive role for the Court in social reform may have unanticipated consequences and fail to bring about its intended results. There is enormous room for further inquiry. What is the actual effect of the First Amendment (as opposed to, say, culture) in protecting political dissent in the United States? What are the consequences of desegregation remedies for students and racial equality? What effect, if any, has Bowers v. Hardwick had on public attitudes toward the movement for gay rights?
Those who study and discuss constitutional law almost never bother with such empirical questions; but what they say almost always assumes answers to them. We need to know whether those answers are correct. Although they will not compel any particular approach to the Constitution — to say that Brown was only partly effective is not to say that it was wrong — they might well help show when Supreme Court involvement is most likely to accomplish its purposes.
The nature of constitutional debate in this country should also undergo a change. Constitutional law needs to be much more candid and concrete about the substantive reasons that are doing the real work in hard cases; and it must simultaneously be able to be confident that good reasons, rather than mere preferences, support certain views about the relationship between state and society. Here it is necessary for constitutional law to develop interpretive principles by confronting squarely, and in the context of the broad frameworks set out by constitutional commands, the question of what sorts of consequences the legal rules under attack have for constitutional liberty and equality (including the constitutional commitment to a form of democratic self-government).
If Roe v. Wade is correct, far example, it cannot be because there is some general right to "privacy" or to control of one's own body. Such a claim provides no answer at all to those who believe in the importance of protecting the fetus. Without further elaboration, moreover, such a claim would mean that there would automatically be a constitutional right to use crack, to participate in surrogacy arrangements, to commit incest, and perhaps to eliminate much of the work of the FDA and OSHA, even the EPA.
As Tribe and Dorf themselves point out, if Roe is right, the reason must have something to do with sex discrimination, and thus with the equal protection clause. It is crucial, after all, that laws restricting abortions apply only to women; that they may well owe their origins to illegitimate and discriminatory views about the appropriate role of men and women; that they represent the selective co-optation of human bodies for the protection of others, which is certainly unique in American law, and specific to sex; and that they tend in practice not to protect fetuses at all, but instead to produce dangerous abortions threatening women's lives and health without significantly decreasing the abortion rate.
Similarly, if Bowers v. Hardwick is wrong, the reason is surely that run current social disapproval of homosexuality is at least roughly analogous to past social disapproval of racial intermarriage, and illegitimate for at least roughly similar reasons. If Hardwick is wrong, it is because neither form of disapproval can be allowed unjustly to subordinate an entire class of Americans. An effort to challenge the decision in Hardwick will therefore have to say something empirical and extensive about the actual history, well-springs, and consequences of laws directed against homosexuals.
It is tempting to think that a judicial inquiry of this sort would not be law at all. But we should not think of law, and certainly not of constitutional law, in so sterile or constricted a way. In fact, an assessment of the purpose and consequences of discrimination is an indispensable part of the enterprise of figuring out what it means to deny someone "the equal protection of the laws." In Roe and Hardwick, arguments of this kind would have to be far more elaborate, of course, than I have been able to suggest here; but it is impossible to avoid resort to them in assessing the two cases.
Most important, the development of interpretive principles should proceed with far more attention to the values and the possibilities of democratic government. In this respect, Tribe and Dorf's book counts as an unusually articulate contribution to the large number of recent works attempting to justify, to preserve, and to extend the work of the Warren Court. But the Warren Court is long gone. From the standpoint of the 1990s, that Court increasingly appears to be a historical anomaly, indeed an unprecedented exception to American political traditions: it was an adjudicative body willing to use the Constitution as an engine of social reform in the interest of civil rights and civil liberties.
This is not to deny that the Court accomplished considerable good. Still, the effort to build on its legacy gives off the distinctive and perhaps not altogether pleasant aroma of nostalgia. At least under current conditions, it would be more valuable to attend to the possible use and reform of democracy itself. Any set of interpretive principles for the Supreme Court must be centrally concerned about the potentially undemocratic character of judicial intrusion into political processes.
A set of principles that focuses judicial controls on correcting defects in American democracy will therefore have much to be said in its favor. In view of the fact that judges are unelected, ambiguous constitutional provisions should usually be read to allow maximum scope for representative processes — except (as John Hart Ely has argued) when political processes are not open to self-correction, and when judicial protection is sought by groups that are without sufficient political power to protect themselves.
Of course, it remains to explain exactly what "sufficient" means. But attention to questions of this sort may actually provide an important ingredient in a defense of Roe and a criticism of Hardwick, at least if those cases are conceived as raising issues of equality as well as privacy. Both Roe and Hardwick involved groups facing important disabilities in the political process. These disabilities argue in favor of a less cautious judicial role.
Finally, and even more fundamentally, we might in the future see, and even attempt to spur, an increase in constitutional deliberation, and political or moral argument, outside of the courts and inside the democratic process. The judges are not the only figures in government charged with constitutional and moral responsibilities, and sometimes the focus on the Court can distract attention from other possibilities. The recent deliberations of Congress and the president with respect to the Persian Gulf war were distinguished by close reference to constitutional commands — and without the involvement of the Supreme Court at all.
Consider also a striking set of recent events: the passage of the most important law in history protecting handicapped people, the Americans With Disabilities Act, enacted in the face of Supreme Court decisions refusing to use the Constitution in this area; the extraordinary citizen response to Supreme Court decisions restricting the abortion right, a response that has dramatically energized the women's movement; the new Clean Air Act, in the face of Supreme Court decisions taking a relatively passive role in protecting the environment; and the likely passage, soon, of a new civil rights bill broader than anything that the Supreme Court might have required. Whatever the merits of any one of these developments, they attest to the enormous potential of moral and political and sometimes even constitutional discussion outside the courtroom.
It may not be entirely implausible, if all this is so, to think that the recent retreat of the federal judiciary from social reform, a retreat made in the name of the Constitution, might ultimately count as one of many steps in the revival of democratic processes in America and, not incidentally, in the spurring of involvement of heretofore weak or passive groups in those processes. Such steps would represent a major redirection of constitutional energies. They might be far more fruitful than our customary obsession with the occasional decisions of the nine people who sit on the Supreme Court.
Cass R. Sunstien teaches at the University of Chicago and is the author most recently of After the Rights Revolution (Harvard University Press).