当前位置: 首页»大师研究»罗尔斯»Ronald Dworkin :RAWLS AND THE LAW

Ronald Dworkin :RAWLS AND THE LAW

添加时间:2005-06-29 16:37    浏览次数: 6000 次

Ronald Dworkin RAWLS AND THE LAW


Copyright (c) 2004 Fordham Law Review

Fordham Law Review

April, 2004

72 Fordham L. Rev. 1387

LENGTH: 9647 words


NAME: Ronald Dworkin*


* This Lecture represents a lightly edited and footnoted version of the Keynote Address Professor Dworkin delivered on November 7, 2003 at the Fordham University School of Law conference, "Rawls and the Law."


... We might speculate about how important it was to the shape of Rawls''s theory of justice that he lived and worked in a political community as dominated by law as ours is, and in which certain crucial political issues - matters of basic liberty and constitutional essentials, if you will - are removed from ordinary politics and made the special concern of courts. ... For there is a strong affinity between a utilitarian conception of justice and a positivist conception of legality: It is no accident that the two founders of modern legal positivism, Bentham and Austin, were arch utilitarians. ... On no conception of law - positivist or interpretivist - can judges in complex pluralistic communities acquit their institutional responsibilities without relying on controversial moral convictions. ... So nothing in Rawls''s general conception of justice as fairness supports the so-called "majoritarian" objection to judicial review in its most comprehensive form. ... Could Rawls have thought that it is indeterminate whether there is a basic liberty to an abortion, or to assisted suicide, or to a prayer-free school? If so, he might then have thought that there is a good case for leaving such issues to politics, because politics is superior to adjudication when quasi-pure-procedural justice is all that is anyway on offer. ... They think that a proposition can be objectively true only if, in addition to the substantive reasons we can offer for embracing the proposition, the proposition also has a ground in some kind of reality that goes beyond these reasons. ...



I. Rawls as Legal Philosopher

It is a mark of John Rawls''s greatness as a political philosopher that my topic, "Rawls and the Law," can be approached in so many ways. Politicians around the world cite his ideas and American and other judges appeal to his work, so we might talk about the impact that he has already had on the law in different countries. Or we might consider the impact that he might have: We might ask what changes in American tax or tort law his famous difference principle would recommend, for example. Those are indeed among the issues that will be explored tomorrow. We might also examine his impact in the other direction. We might speculate about how important it was to the shape of Rawls''s theory of justice that he lived and worked in a political community as dominated by law as ours is, and in which certain crucial political issues - matters of basic liberty and constitutional essentials, if you will - are removed from ordinary politics and made the special concern of courts.

I plan to talk about Rawls and the law in a different way: about Rawls as himself a legal philosopher and, indeed, lawyer. He did not suppose himself a legal philosopher, and though there are several important discussions of law in his work, some of which I will mention, he made his main contributions to legal theory through his political philosophy, because legal theory is a department of political philosophy and Rawls wrote abstractly about the whole discipline. In these keynote remarks, I want briefly to identify but also to evaluate those aspects of Rawls''s theory of justice that speak directly to traditional issues of jurisprudence.

I will begin with a short list of those traditional issues. Any general legal theory must answer the ancient question of what law is. But that ancient question in fact poses two different issues. The first is methodological: What kind of theory counts as an answer to that question? Are general theories of law descriptive theories? If so, [*1388] what do they describe? Are they pieces of conceptual analysis? If so, what makes one analysis of the concept of law better than another? Are they normative political theories? If so, then how does a theory of what law is differ from a theory about what law should be? Legal philosophers disagree about these methodological matters. What view, if any, follows from Rawls''s philosophy?

The second question is then obvious. Once a legal theory has taken a position about the methodological issue, it must try to answer the substantive one. Given what a theory of law should be, which theory of law is the most successful? For some time, theories of law have been divided, by both their authors and commentators, into roughly two groups: positivist theories of law, which insist that what the law of any jurisdiction requires or permits is only a matter of social fact, and anti-positivist theories, which claim that what the law requires sometimes depends not on social facts alone but also on controversial normative issues including moral issues. Rawls did not, so far as I know, explicitly choose one or another of these general positions as his own. But do his theories support one choice rather than the other?

That second question inevitably presents a third one. On either a positivist or an anti-positivist theory, a judge will often face "hard" cases in which what lawyers regard as settled law does not decide the immediate issue. Positivists say that in such a case a judge must exercise discretion to make new law. Anti-positivists describe the same necessity in different ways: A lawyer who thinks, as I do, that judges must aim at integrity in their decisions will agree that what integrity requires will often, perhaps usually, be controversial, so that a fresh judgment is needed. Both sides, or rather all versions of each side, must confront the question of what sorts or sources of argument are appropriate to that judicial responsibility.

What kinds of reasons may or should judges offer to defend their fresh judgments? May they appeal to religious convictions? To their personal moral convictions? To philosophical systems of morality or eschatology? To macroeconomic phenomena - would it be a good argument for a judge to say that he is deciding in a particular way because that will help the dollar on the international currency markets? These crucial questions are relatively neglected in legal theory. But Rawls developed a doctrine, which he called the doctrine of public reason, about the arguments public officials may properly use to justify their decisions, and he said emphatically that the doctrine of public reason applies with particular stringency to judges. We must examine this doctrine. If we find it unsatisfactory, as I suspect it is, then we must ask whether any other part of Rawls''s general theory is more helpful in defining the character of proper judicial reasoning.

The fourth issue is particularly pressing for legal theorists in [*1389] America and in other mature democracies where constitutional courts have the power to invalidate laws adopted by legislators elected by and accountable to the people. Is that power consistent with democratic principles? If not, is it unjust for that reason? Rawls spoke directly to that issue, on different occasions, and we must notice what he said. But he recently spoke to what is in many ways an even livelier and more important issue, a matter, we might say, of constitutional strategy rather than legitimacy. Should a constitutional court decline to decide certain issues - for example, about abortion or assisted suicide - because its nation is not ready for a judicial resolution of the issue? Should the Court stay its hand to allow ordinary politics to reduce the issue''s divisiveness and perhaps reach a compromise more acceptable to the whole community? Several prominent legal scholars have endorsed that suggestion, and Rawls has said that he thinks their argument a "good" one. We should consider why he thinks their argument good.

Finally, I would like to touch on what might seem a much more abstract issue. Can controversial claims about what the law requires be objectively rather than merely subjectively true? This is not an issue that troubles lawyers and judges in their day-to-day practice. It nevertheless has considerable practical importance, because many issues of legal and civic policy turn on it, including whether the rule of law really is different from rule by men and women with power, whether it makes sense to suppose that we have a general moral obligation to obey the law, and whether judicial review of legislative enactments really is legitimate. Some legal theories are in fact built around the supposition that legal practice is essentially subjective: for example, the influential movement called American legal realism, which in our time morphed into the incandescent, though brief, fireworks of critical legal studies. Rawls said a good deal about truth and objectivity, some of it inconclusive and even obscure, but much of it helpful when lawyers turn to these more explicitly philosophical issues.

II. The Nature of Legal Philosophy

Assume, for the moment, as almost all lawyers do, that a proposition about legal rights and duties can be true. If so, then a theory of law should tell us under what circumstances such a proposition is true. What in the world can make it true, for example, that the speed limit around here is 55, or that Microsoft violated the antitrust laws, or that affirmative action is unconstitutional? Legal philosophers defend general theories of law that attempt answers to that question. Legal positivists claim that a proposition of law can be true only in virtue of social facts: facts, for example, about what a legislature has declared or a judge decided in some prior case. I shall say something about the merits of that view in a moment, but we must [*1390] first consider an antecedent question. What kind of a claim are they, the positivists, making? What could make their claim about the truth conditions of law itself true?

Many legal philosophers believe that their theories of law are descriptive theories about the social practices or conventions that the bulk of lawyers follow in making, defending, and judging propositions of law. Of course, lawyers often disagree about which legal propositions are true and which false. They disagree, for example, about the legal position of a woman who has suffered side effects from a drug that her mother had taken many years prior, but who cannot identify the manufacturer of the particular pharmaceuticals her mother had taken at any particular time because the pill was manufactured by several companies and she does not know whose pill she took when. n1 Is she legally entitled to recover damages from all of the companies that manufacture the drug in proportion to their market share? But these legal philosophers assume that if this kind of disagreement is genuine, then the lawyers must agree about a more basic issue. They must agree on the right tests to use to decide if a proposition of law is true; otherwise - if different lawyers used different tests - they would simply be talking past one another. If that assumption is correct, then a philosophical theory of law should aim to describe that background agreement. It should tell us what law is by telling us what tests lawyers actually use to identify true or sound propositions of law.

On this view, legal philosophy is best understood as a descriptive exercise: it is an exercise in legal sociology. But, in fact, it is extremely difficult to account for any general theory of law if we take it to be descriptive in that way. Consider the version of legal positivism developed by H.L.A. Hart. n2 He argued for what he called the "sources" thesis which holds (in substance) that propositions of law are true when, but only when, they can be inferred from explicit decisions taken by legal institutions, like legislatures, that are authorized by convention to make such decisions. n3 If a lawyer can show that it follows from something that the pertinent legislature has said that the woman in our example is legally entitled to market-share damages, then he has shown that she is so entitled. But if that proposition does not follow from anything any authorized institution has said or done, then it is not true.

In a posthumously published postscript to his book, The Concept of Law, Hart insisted that this sources thesis is purely descriptive. n4 But it is mysterious in what sense it could be thought descriptive. Hart did not intend it, he insisted, as a description of how lawyers talk, how [*1391] they use the word "law." For it is plainly not part of the very meaning of "law" that law can only be valid in virtue of positive enactment. Nor could he have intended it as a description of what all lawyers accept as belonging to the very concept of law, as we accept that it belongs to the concept of bachelor that a bachelor is unmarried. For lawyers disagree about whether the sources thesis is right: The lawyers who think that the woman in our example has a legal right to market-share damages plainly reject the sources thesis, because no institution had declared such liability before imaginative lawyers n5 argued for it. If these lawyers are mistaken, their mistake is a legal not a conceptual one. Nor, for the same reason, could Hart have meant his sources thesis as the sociological hypothesis that lawyers everywhere actually claim law on their side only when the sources thesis has been satisfied. That hypothesis, too, is plainly false.

How, then, should we understand a theory of law like the sources thesis? Rawls spoke directly to that issue by example - through his analysis of the concept of justice. He did not suppose that everyone who shares and uses the concept of justice shares some substantial background understanding about what makes an institution just or unjust. On the contrary, he insisted that people have radically different conceptions of justice. They do, he allowed, share some very abstract understanding that makes these all conceptions of justice rather than of some other virtue. But this shared understanding is exceedingly thin, all but empty of real content. What makes disagreement about justice possible is that people sufficiently agree on certain specific instances or examples - everyone agrees that slavery is unjust, that wage exploitation is unjust, and so forth. So Rawls recommended that philosophers of justice engage in the interpretive enterprise he called seeking reflective equilibrium. We try to generate principles of some general scope and to match those general principles to the concrete judgments about what is just and unjust with which we begin, shifting our views about either principles or concrete judgments, or both, as becomes necessary to achieve an interpretive fit.

We can restate this interpretive exercise as a method for legal philosophy. We can identify what apparently goes without saying is part of our law - the speed limit, the tax code, the ordinary everyday rules of property, contract, and so forth, that we are all familiar with. These are, we might say, paradigm instances of law. And then we can construct the other pole of an interpretive equilibrium because we share an abstract ideal that can play the same role in legal theory as [*1392] the concept of justice played for Rawls. This is the concept of law - though sometimes, when we are emphasizing its political character, we describe it in another way, as the concept of legality or the concept of the rule of law. We can then try to provide a suitable conception of legality, that is, a conception of legality that brings our various pre-analytic assumptions about concrete propositions of law into equilibrium with the general principles of political morality that seem best to explain the character and value of legality. In that way we can embed a theory about the truth conditions of propositions of law in a larger conception of value that we find convincing. A positivist theory of law will offer a thesis about the truth conditions of such propositions, like the sources thesis, that is supported by a positivist conception of legality that is in turn supported by an appropriate more general theory of justice. That interpretive design provides the best way of understanding the arguments that leading legal philosophers have actually made. Legal philosophy so conceived is in a way descriptive because it begins with some understanding about what is taken for granted within the community to which it is addressed, but it is in other ways substantive and normative because the equilibrium it seeks is with principles judged for independent appeal. So Rawls''s work is, right from the start, a major contribution to legal philosophy''s self-understanding.

III. What Is Law?

Now let us turn to the substantive side of the ancient question. Which understanding of law - a positivist understanding or some other - is most successful as a conception of the concept of legality? To bring out the implications of Rawls''s ideas for that further question, we might embed the question in his imaginary construction of justice as fairness. Suppose that the representatives in the "original position" n6 he described are asked to choose, in addition to general principles of justice, a conception of legality as well. They are offered, to make the point simpler, a menu of only two choices. They may choose a simplified positivist account of legality, which specifies that judges use a particular test for true propositions of law, or a simplified interpretivist non-positivist account.

On the stipulated simple positivist account, judges are to enforce rules laid down by the legislature so far as these rules are unambiguous or can be made unambiguous by consulting legislative history and other standard sources of legislative intent. But when, as will often happen, rules laid down and interpreted only in that way are insufficient to decide the case, then judges should announce that the law provides no answer, and then legislate themselves to fill the gap so created. They should legislate, however, modestly and marginally, [*1393] and as they believe that the legislature now in power would legislate if seized of the issue. Judges should do, that is, what they think the pertinent parliament would have done. On the rival simple interpretivist account, judges should enforce rules laid down by the legislature, interpreted in the same way, but when confronted by a so-called gap, should not attempt to legislate as the legislature would, but should instead try to identify the principles of fairness or justice that best justify the law of the community as a whole and apply those principles to the new case.

Now suppose that, contrary to Rawls''s supposition, the representatives have settled on a general, all-embracing utilitarian conception of justice. Then they will think that they have a strong case for choosing the simplified positivist conception of law over the simplified non-positivist one. For there is a strong affinity between a utilitarian conception of justice and a positivist conception of legality: It is no accident that the two founders of modern legal positivism, Bentham and Austin, were arch utilitarians. As Bentham pointed out, sound utilitarian legislation must be organized and directed from a single source: The best program for maximizing utility is an integrated program in which different laws and policies can be tweaked and coordinated so as to yield a maximum utility bang. n7 The legislature is the best institution for achieving that maximum bang, because it can survey the whole architecture of law and policy, and because its composition and procedures of election are conducive to providing information about the mix of preferences in the community that is indispensable to sound calculations of the trade-offs necessary to achieve maximum aggregate utility. Judges are essential to the retail enforcement of rules designed to maximize utility over time, but they should be as little as possible architects of policy, because that would be multiply inefficient. So they should be told that when the legislature''s dictum has run out without issuing a decision, they should declare there is no law controlling their decisions that comes from any other source. They should announce a gap, which they then fill as modestly as possible, as lieutenants of their political masters, in the spirit of what those masters would themselves do, going, as the arch-positivist and -utilitarian Oliver Wendell Holmes put it, from the molar to the molecular. n8

That is the positive case for positivism from the utilitarian perspective. There is a corresponding negative case from that perspective against interpretivism: that it is irrational. For utilitarians, moral and political principles are simply rules of thumb [*1394] for achieving maximum utility in the long run, and there can be no independent value, and much mischief, in pursuing a coherence of principle for its own sake. Utility is surely better served by concentrating on the future alone, with no backward glance at integrity with the past, except so far as this is in itself strategically wise.

Now suppose, however, that representatives in the original position do choose as Rawls supposes they would. They reject utilitarianism in favor of the two principles of justice, one of which gives priority to certain basic liberties and the other of which seeks to protect the position of the worst-off group in society. Then it would seem natural for them also to choose interpretivism over positivism, because interpretivism would then be a better bet to achieve justice, at retail as well as wholesale, in the long run. The two principles require implementation at successive levels of detail. They require, first, a constitutional stage at which institutions are designed so as most likely to produce the outcomes that the two basic principles demand. Then they require those institutions to make, at what Rawls calls a legislative stage, more specific decisions about laws and policies guided by more specific principles of justice in service of the basic principles. People who place a lexical priority on equal liberty, and then a further priority on protecting the position of the worse-off group, will be particularly sensitive to the possibility of slippage at this legislative stage. They will worry that a legislature dependent on majority approval will be under great pressure to advance the interests of some groups at the expense of others. They will, therefore, be attracted to the idea of a judiciary with independent powers and responsibilities. They will be drawn to the idea of judicial review of a written Constitution, and I will discuss the implications of Rawls''s arguments on that score later. n9 But they will also be drawn to the idea that judges should also exercise a less potent, but still important, supervision over the application and development of the more quotidian law made by legislatures. And to the further idea that they should exercise that power in the direction of equality before the law, that is, in the direction of insisting that, so far as a reasonable doctrine of legislative supremacy permits, whatever principles are presupposed by what the legislature has done for some groups be available generally to all. They will have that strong reason for favoring an interpretivist conception of law that deems people to have legal rights not only to what legislative institutions have specifically directed, but also to the principled elaboration of those directions. Coherence is the best protection against discrimination. That is, after all, the premise of the Equal Protection Clause of our own Constitution''s Fourteenth Amendment.

[*1395] Rawls did not make this argument for interpretivism; indeed, so far as I am aware, he made no explicit argument for any conception of law. But he did explicitly endorse the principle that I said supports interpretivism, and he endorsed that principle in the course of a discussion of legality or the rule of law. Let me quote:

The precept that like decisions be given in like cases significantly limits the discretion of judges and others in authority. The precept forces them to justify the distinctions that they make between persons by reference to the relevant legal rules and principles. In any particular case, if the rules are at all complicated and call for interpretation, it may be easy to justify an arbitrary decision. But as the number of cases increases, plausible justifications for biased judgments become more difficult to construct. The requirement of consistency holds of course for the interpretation of all rules and for justifications at all levels. n10

Note Rawls''s emphasis on complexity as itself a constraint, and his insistence that consistency hold, as he put it, for "all rules ... at all levels." n11 Citizens are best protected from arbitrariness and discrimination when judges interpreting the law and elaborating it in hard cases are responsible for coherence, not simply with particular doctrines here and there, but, as best as it can be achieved, principled coherence with the whole structure of law.

Now it might be objected that in spite of the historical affinity I cited a positivist need not be a utilitarian. We can suppose, instead, a positivist judge who is not a utilitarian, and who stands ready, in cases in which he supposes he has discretion, to adopt the rule that he believes best comports with justice on some other understanding. Why wouldn''t people who have chosen Rawls''s two principles of justice in the original position also choose that characterization of a judge''s role? Why isn''t that a better choice on grounds of what Rawls called imperfect procedural justice? But this suggestion neglects the fact that judges, even if they aim only at justice, will nevertheless often disagree about what justice is, and that judges may themselves be influenced by preconception, prejudice, or the other enemies of impartial justice. People choosing a conception of law have no reason to think that a decision in their own case will better reflect justice, on any conception of what that is, if judges are free to disregard principled coherence with what other officials and judges have done than if they are asked to respect principled coherence. They might well think that they safeguard themselves better against arbitrariness or discrimination if they do not instruct judges to do justice as they see it, but seek to discipline judges by insisting that they do their best to [*1396] respect principled consistency as they see it. That is, as I said, the assumption of our Equal Protection Clause. n12

IV. The Constraints of Legal Reasoning

Now I turn to a more specific question that must be faced by all conceptions of law, but which is particularly difficult for some. How should judges reason in hard cases? Under the simple package of positivism and utilitarianism that I described, judges must make fresh judgments to fill gaps in the law, but the package dictates the character of that judicial reasoning. It holds that judges should try to do what the legislature would have done. Interpretivism, as well as other legal theories, also supposes that judges must make fresh judgments of political morality in hard cases: It instructs them to seek an interpretive equilibrium between the legal structure as a whole and the general principles that are best understood as justifying that structure. That, as I have argued elsewhere, is in fact the traditional common law method. n13 But are there any constraints on what kind of principles judges can cite in constructing this interpretive equilibrium, that is, in justifying the law''s record as a whole?

It would certainly seem wrong for them to deploy certain kinds of arguments. They must not appeal to their personal interests or to the interests of some group to which they are connected. That obvious constraint seems part of the very idea of a justification. But may they appeal to their religious convictions, if they have any, or to the doctrines of their church, if they have one? After all, some judges think, as a matter of their deepest conviction, that religion provides the most compelling or perhaps the only true justification of political morality and, therefore, the most compelling or only true justification of past legal decisions. In the United States, religious argument might be thought excluded from judicial reasoning by the First Amendment. But what about elsewhere? In a country, for example, like the United Kingdom or Israel, in which there is an established religion? If religion is also an impermissible ground for adjudication even there, is [*1397] this constraint ad hoc to religion? Or does it follow as only one case from some more general principle of political morality? What about arguments of moral philosophy, for example? May a judge properly appeal in his opinions to the philosophical doctrines of Immanuel Kant or John Stuart Mill? May he appeal, as a number of American judges have in fact done, to the philosophical writings of John Rawls? n14 May a judge appeal to macroeconomic policy? May he decide that one principle better justifies the legal structure as a whole because following that principle will help control inflation or promote savings?

Rawls''s doctrine of public reason is devoted exactly to defining the kinds of arguments that are permissible for officials in a politically liberal community, and he insists that the doctrine applies with particular stringency to judges. I find the doctrine of public reason difficult to define and defend, however. I know it will be discussed in a later panel and I look forward to the discussion, but I must try to summarize my difficulties now. There are two ways of stating what the doctrine requires. The first, and more basic, appeals to the important idea of reciprocity. The doctrine permits only those justifications that all reasonable members of the political community can reasonably accept. The second is presumably the upshot of that more basic test. Public reason requires officials to offer justifications that are based on the political values of the community and not on comprehensive religious or moral or philosophical doctrines. The doctrine, therefore, requires judges searching for a justification of the law''s structure to avoid controversial religious, moral, or philosophical doctrines.

I do not see, however, what the doctrine of reciprocity excludes. If I believe that a particular controversial moral position is plainly right - for example, that individuals ought to take charge of their own lives and bear the financial responsibility for any mistakes they make themselves - then how can I not believe that other people in my community can reasonably accept the same view, whether or not it is likely that they will accept it? Perhaps Rawls means that judges should not appeal to ideas that some reasonable citizens could not accept without abandoning their convictions of a certain sort - their X convictions. But we seem to have no basis for stipulating what these X convictions are. I accept that religious convictions are special for several reasons. Certainly someone who believes that the religious truth is only available through divine grace, or some other privileged access, cannot hold that all reasonable citizens could reasonably [*1398] embrace his own religious convictions. But Rawls offers no reason to think that the test of reciprocity excludes any reasonable convictions beyond religious convictions.

I have equally great difficulties with the distinction between political values on the one hand and comprehensive moral convictions on the other. Rawls''s own conception of justice as fairness depends critically on what seem to be controversial moral positions. The difference principle, for example, is generated and defended in reflective equilibrium by a set of assumptions, including assumptions about the fundamental moral irrelevance of effort or responsibility: If the arrangement that best maximizes the position of the worst-off group turns out to reward slackers, that is no objection. Rawls defends this conclusion by supposing that effort is influenced by endowment. n15 So it is, but it is not exhausted by endowment, and the question of how the interaction between the two is to figure seems a mixed question of psychology and morality of just the kind that divides different comprehensive moral views about personal responsibility. Rawls''s position is certainly controversial in our community, and some people reject it in favor of a theory of distributive justice that depends more on personal responsibility.

These difficulties are confirmed, I think, by Rawls''s examples of the idea of public reason in operation. He discusses the abortion controversy on several occasions, though in each case only very briefly. His discussion assumes that the question whether an early fetus has rights and interests of its own, including a right to life, is a question for a comprehensive moral or religious or philosophical position and is not settled by any political value of a liberal community. But how can we take a position about whether American women have a constitutional right to abortion - how could the Supreme Court decide Roe v. Wade n16 or Planned Parenthood of Southeastern Pennsylvania v. Casey n17 - without taking some position on that comprehensive issue? There seems no default position here. The view that a fetus does not have interests and rights of its own is as much drawn from a comprehensive position as the view that it does, and we cannot reach a decision about abortion without adopting one of these two views. The Equal Protection Clause applies to all persons, and any argument that a woman has a constitutional right to abortion in the first trimester of pregnancy must deny that a fetus is a "person" within the meaning of that clause. n18

[*1399] I therefore doubt that Rawls''s doctrine of public reason can help us much in filling out a conception of legality and adjudication. We must look elsewhere. In my view, we can find the necessary constraints on judicial argument in the conception of law that I said Rawls''s general arguments suggest: interpretivism. If we accept an interpretivist conception, we do not need a separate doctrine like the doctrine of public reason. Judges may not appeal to religious convictions or goals in liberal societies because such convictions cannot figure in an overall comprehensive justification of the legal structure of a liberal and tolerant pluralistic community. This interpretive constraint cannot, however, exclude moral as distinct from religious convictions. Judges interpreting a string of cases in tort law can appeal to Rawls''s theory of justice as a ground for rejecting a utilitarian interpretation of past decisions and doctrines in favor of an interpretation more firmly grounded in a conception of equality. n19 Perhaps we''ll have occasion to pursue that suggestion in one of the later panels.

One more point: In his discussion of public reason, Rawls says that in any case judges may not appeal to their personal moral convictions. If that means that a judge cannot argue that one justification of past law is superior because he happens to think so, then it is obviously correct. A judge''s intellectual biography is not a legal argument. But if it means that a judge may not give any place to controversial moral opinions in his judgment, because he would then be citing the moral opinions that he but not others think right, then it states an impossible demand. On no conception of law - positivist or interpretivist - can judges in complex pluralistic communities acquit their institutional responsibilities without relying on controversial moral convictions. n20

V. Constitutionalism

The institution of judicial review, under which appointed judges have the power to declare enactments of legislatures and other representative institutions void because they offend constitutional guarantees of individual rights, is often said to be anti-democratic because it allows a few unelected and virtually unsackable lawyers to override the considered verdicts of elected representatives. Rawls addressed himself to that classic complaint on several occasions. He made plain, first, that according to his favored conception of justice as fairness, the various institutions that a community constructs at what he calls the constitutional level, in the light of principles of justice chosen behind the veil of ignorance, are chosen in the spirit of perfect [*1400] rather than pure procedural justice. They are chosen, that is, with an eye to outcomes. The principles of justice establish the basic liberties and their priority, and the question to be decided at the constitutional stage is an instrumental one: Which scheme of institutions is best suited to protect those liberties?

Of course, among the equal liberties that institutions must be designed to protect are the political liberties, which include the right to vote and to participate in politics. But, as Rawls says in Justice as Fairness: A Restatement, these and the other basic liberties are themselves to be seen as quasi-instrumental. n21 They are justified as essential for the development and exercise of the two fundamental moral powers, that is, the power to form and act on a sense of justice and to form and act on a conception of the good. This means, as I understand it, that though people have a basic right to broadly democratic procedures, because extensive rights to vote and participate in politics are plainly and inescapably necessary to the development of these moral powers, people have no basic right to any particular form of democracy, and therefore no basic right that democratic institutions follow any particular design or have any particular jurisdiction. The question is rather which parliamentary structure and jurisdiction has the best prospects of securing the other mandated or desired outcomes.

So nothing in Rawls''s general conception of justice as fairness supports the so-called "majoritarian" objection to judicial review in its most comprehensive form. But his theory leaves room, at the constitutional level of construction, for the more limited objection that in fact the American structure of constitutionalism and judicial review cannot be justified in that instrumental way, that the basic liberties including the political liberties would be better served by some other arrangement, which might be pure parliamentary sovereignty or a mixed case like that of the United Kingdom after the enactment of the Human Rights Act, which permits Parliament to legislate in violation of the rights the Act specifies if Parliament clearly states its intention to do so. Though Rawls did not attempt anything like a thorough outcome-based case for the American model against such more majoritarian rivals, he makes several arguments that seem to support roughly the American model. He distinguishes, for example, between parliamentary and popular sovereignty, and says that the American model is consistent with popular sovereignty. That model promotes people''s basic moral powers, he says, because the people in general not only endorsed the original Constitution, but have prompted and overseen its cardinal developments since - in the Reconstruction period, for example, and in the New Deal. (In that [*1401] view he follows, as he says, the arguments of Bruce Ackerman.) n22 Second, he points out a further way in which constitutionalism and judicial review help rather than constrain the development of the two moral powers. He says that the fact that the Supreme Court acts as a forum of principle encourages and focuses public political discussion of central moral issues. n23

I can now turn to the different issue I mentioned earlier: not the legitimacy of judicial review, but its proper strategy. The Supreme Court is often pressed to recognize a concrete constitutional right that it has not recognized before, and whose standing as a right is very much in dispute among thoughtful people in the nation. If it recognizes and enforces that new right, its decision will be massively resented, and its own standing and legitimacy may be called into question. The Court faced that situation in Brown v. Board of Education n24 and the other early racial discrimination cases of the 1950s. It faced it in the school prayer cases, in the abortion cases beginning with Roe v. Wade, n25 and in the more recent cases about assisted suicide for terminally ill patients.

It is widely argued that in such situations the Court ought to refuse to recognize the new right in order to allow the political process more time to consider the merits of the issue through local politics and decision, which might vary across states, and which might therefore provide a kind of experimentation in what Justice Brandeis called the laboratories of the several states. n26 It might do that in some cases through its certiorari policy; it might decline to take a case that required it to decide basic issues of individual rights because it thought it wiser to let those issues percolate in politics further. In most such cases, however, one or more lower courts will have spoken to the issue in a way that requires the Supreme Court to decide whether the Constitution grants the claimed right. In that case the passive or cautionary strategy I described would require the court to hold that the asserted controversial right did not exist, as it did hold, for example, in the assisted suicide cases. n27

Rawls had himself urged the Supreme Court to recognize a limited right to assisted suicide: He signed a brief urging that decision as an amicus curiae, along with three others who are at this conference [*1402] today. n28 But Rawls later said that the cautionary argument is what he called a "good" argument for the decision the Court made against his advice, and he also called it a "good" argument against the Court''s 1973 decision to recognize a limited right to abortion in Roe v. Wade, n29 which of course doesn''t mean that he thought it finally a persuasive argument. n30 However, there seems to me a straight-forward and powerful - even knock-down - Rawlsian argument against the cautious view. In these contentious cases a plaintiff or group of plaintiffs claims that some law or practice denies their basic liberties and therefore offends the first principle of justice, which in justice as fairness is given priority over everything else, presumably including civil peace and quiet. Of course, any particular justice might not believe that the targeted law or practice does deny a basic liberty. But we must assume that Justice Rawls would be convinced by the argument he himself put forward, for example, in the Philosophers'' Brief. We can easily see how a utilitarian who calls himself a pragmatist might be attracted to the cautious argument. But why should Rawls be? Why should he think that the cautious argument is even a "good" one?

One answer might be epistemic. Perhaps Rawls thought it arguable that a Supreme Court justice, recognizing the fearsome Burdens of Reason, should accept that his own judgment might well be flawed, that the political process might over some years work out a different compromise that would be very widely accepted, and that this compromise, if it is ever achieved, would be a more accurate or reasonable account of the basic liberty in question than a majority of justices could devise for themselves in advance. There are several obvious difficulties in this answer, however, and we may expose these using the abortion issue once again as an example.

First, it seems unlikely that a non-divisive compromise would soon have been reached in the politics of this country. The Europeans have by and large settled on a position that, with cosmetic formalities, allows abortion on demand. n31 This has not generated continuous controversy there, but that is because Europe is not plagued with fundamentalist religious movements or any serious fundamentalist sensibility. But we are so plagued, as has once again been [*1403] demonstrated by the recently legislated ban on so-called partial birth abortion. n32 The only solution that would still militant objection from fundamentalists here would be a harsh anti-abortion regime, and that would not be tolerable to women''s movements that can be almost equally militant. Second, even if a compromise were reached politically that proved reasonably acceptable to all, there would be no reason to think that this compromise would have more accurately or reasonably identified the basic liberties in question. On the contrary, whatever view one takes of those basic liberties, it seems likely that a compromise would mean injustice to some. Suppose, for example, that most people ceased much to object to anti-abortion laws in their own states, but only because women who wanted an abortion could conveniently travel to a nearby state in which abortion was legal. That would deny the equal value of liberty to people too poor to afford the various expenses of the travel.

Could Rawls have thought that it is indeterminate whether there is a basic liberty to an abortion, or to assisted suicide, or to a prayer-free school? If so, he might then have thought that there is a good case for leaving such issues to politics, because politics is superior to adjudication when quasi-pure-procedural justice is all that is anyway on offer. But it is extremely implausible that Rawls thought that issues like these are matters of indeterminacy, because he himself took positions on several of these issues. Nor could he consistently suppose (as several scholars in effect have supposed) that state-by-state politics is a better vehicle for developing the various public virtues he recognized than is adjudication. His argument that Supreme Court adjudication stimulates the development of the two moral powers seems to apply as thoroughly to divisive cases as to less dramatic ones; indeed, more so.

Might he have appealed to the virtue of civility, arguing that it is better not to take decisions that will seem deeply offensive to some citizens? But these decisions will seem equally offensive to the losing side if legislatures rather than courts impose them. In any case, this kind of civility is aimed at a mere modus vivendi, which Rawls rejected as inadequate, rather than at anything one might defend in principle. Of course, if the authority of the Supreme Court or of the constitutional arrangement as a whole were actually at stake, that would be different. We could understand the wisdom of a cautious counsel in that case: Better to ignore the rights of a few people than to sacrifice the system that protects everyone''s rights in the long run. But, of course, that is not the situation. Contrary to Justice Frankfurter''s worries, the Court''s authority survived Brown v. Board of Education n33 and the miscegenation cases; n34 it has also survived Roe [*1404] v. Wade n35 and the school prayer decisions. n36 It could have survived deciding for a limited right to assisted suicide. Indeed, I''m tempted to think that since it has apparently survived the shame of Bush v. Gore, n37 it can survive almost anything.

VI. Truth and Objectivity

I want finally to notice, though I must do so very briefly, the last aspect of Rawls''s views that I cited as particularly important for legal theory. It is a frequent objection to celebrations of the rule of law that legal judgments, particularly in hard cases, cannot be reports of any objective truth, but simply express the speaker''s psychological state of approval or disapproval. This is a familiar skeptical view about morality and other departments of value, but it is of particular practical importance in law, because it is thought to provide a substantive argument in various controversies: for example, about whether people have a moral duty to obey the law, or whether judicial review of majoritarian legislation is defensible.

In Political Liberalism, Rawls identified a conception of objectivity that he believed suitable to political claims, and much of what he said holds for controversial claims of law as well. He insisted that objectivity, in the sense he defined, does not depend on any assumption that political or legal reasoning is a case of perception, that is, that a political or legal claim can be objectively true only when the belief that it is true is caused by the situation it reports. Legal facts are not in any causal relationship with lawyers'' central nervous system. But why should it follow that a controversial proposition of law - that the manufacturers of a dangerous medicine are legally responsible for injuries in proportion to their market share, for example - cannot be objectively true? Whether a proposition claims objective truth depends on its content. It claims objective truth if it claims that its truth is independent of anyone''s belief or preference: that manufacturers would be liable, on the present state of the law, even if lawyers didn''t think so. That is all the claim of objectivity means. Whether that claim is successful depends on the legal arguments we can offer for it, that is, on our reasons for thinking that manufacturers would still be liable even if lawyers didn''t think so. If we think that our reasons for thinking that are good reasons, then we must also think that the proposition that the manufacturers are liable is objectively true.

Objectivity so understood doesn''t depend on a metaphysical [*1405] assumption that seems popular among some so-called moral realists. They think that a proposition can be objectively true only if, in addition to the substantive reasons we can offer for embracing the proposition, the proposition also has a ground in some kind of reality that goes beyond these reasons. They are wrong: Substantive reasons are enough. But they must not be isolated reasons. Our arguments for objectivity are sufficient only if they are sufficiently systematic and mutually and reciprocally examined. Rawls puts that crucial point this way:

Political constructivism does not look for something for the reasonableness of the statement that slavery is unjust to consist in, as if the reasonableness of it needed some kind of grounding. We may accept provisionally, though with confidence, certain considered judgments as fixed points, as what we take as basic facts, such as slavery is unjust. But we have a fully philosophical political conception only when such facts are coherently connected together by concepts and principles acceptable to us on due reflection. n38

I do not agree with all of Rawls''s discussion of objectivity. Indeed, I think some of it is shown to be unnecessary and unjustified by the rest - his view that we cannot properly claim objectivity for a domain unless we can explain what we take to be error in that domain in a non-question-begging way, for example. But I commend his general discussion of objectivity to lawyers who hope to understand what their arguments are really about.

VII. Confession

Some of you will have noticed a certain congruence between the positions in legal theory I say Rawls''s arguments support and those I have myself tried to defend, and you may think this no accident. So I offer you a confession, but with no apology. The work of philosophical icons is rich enough to allow appropriation through interpretation. Each of us has his or her own Immanuel Kant, and from now on we will struggle, each of us, for the benediction of John Rawls. And with very good reason. As this conference shows, after all the books, all the footnotes, all the wonderful discussions, we are only just beginning to grasp how much we have to learn from that man.


n1. See Sindell v. Abbott Labs., 607 P.2d 924, 936-38 (Cal. 1980).

n2. See infra text accompanying note 4.

n3. See infra text accompanying note 4.

n4. See H.L.A. Hart, The Concept of Law vii (6th prtg. 1972).

n5. These lawyers were themselves inspired with the idea for market-share liability by a student Comment published by the Fordham Law Review, which is coincidentally the institutional sponsor of this symposium. See Naomi Sheiner, Comment, DES and a Proposed Theory of Enterprise Liability, 46 Fordham L. Rev. 963 (1978).

n6. John Rawls, A Theory of Justice 15-19 (rev. ed. 1999).

n7. See generally Jeremy Bentham, An Introduction to the Principles of Morals and Legislation (1879).

n8. See So. Pac. Co. v. Jensen, 244 U.S. 205, 221 (1917) (Holmes, J., dissenting) ("I recognize without hesitation that judges do and must legislate, but they can do so only interstitially; they are confined from molar to molecular motions.").

n9. See infra Part V.

n10. Rawls, A Theory of Justice, supra note 6, at 209.

n11. Id.

n12. I do not mean that the arguments for interpretivism that I have drawn from Rawls match my own arguments for such a conception. I mean only to show the bearing of Rawls''s work on this central question of jurisprudence. But, at least on one interpretation of the basic structure of Rawls''s argument from the original position, his arguments are in fact not very far from my own. I believe that integrity expresses the right view of equal citizenship: Principles applied to one person must be applied to others unless clearly directed otherwise by competent institutions. In my view, some idea of equality of that sort provides at least part of the set of ideas that the heuristic device of the original position is best understood to model and enforce. However, in footnote 19 to his article, Justice as Fairness: Political Not Metaphysical, Rawls considers and rejects my interpretation. See John Rawls, Justice as Fairness: Political Not Metaphysical, in Collected Papers 388, 400 n.19 (Samuel Freeman ed., 1999).

n13. See generally Ronald Dworkin, Law''s Empire 276-312 (1986).

n14. See, e.g., Uhl v. Thoroughbred Tech. and Telecomms., Inc., 309 F.3d 978, 985 (7th Cir. 2002) (referring to Rawls''s "veil of ignorance" from A Theory of Justice); Goetz v. Crosson, 967 F.2d 29, 39 (2d Cir. 1992) (quoting A Theory of Justice); Memphis Dev. Found. v. Factors Etc., Inc., 616 F.2d 956, 959 (6th Cir. 1980) (same); W. Addition Cmty. Org. v. NLRB, 485 F.2d 917, 938 (D.C. Cir. 1973) (same).

n15. See Rawls, A Theory of Justice, supra note 6, at 274.

n16. 410 U.S. 113 (1973).

n17. 505 U.S. 833 (1992).

n18. Some philosophers believe that a moral right to abortion can be defended even if we assume that a fetus is a person, because even on that assumption a woman has no moral responsibility to continue the burdens of pregnancy. For a discussion of that suggestion, see my book, Life''s Dominion 102-17 (Vintage Books 1994) (1993). But even if we accept that view, it does not follow that a constitutional right to abortion can be defended in that way. If a state may properly treat a fetus as a person, it may constitutionally treat it as a person toward whom a mother has a special responsibility that excludes elective abortion.

n19. See Dworkin, supra note 13, at 276-312.

n20. See Dworkin, Un Pontificat Lac, in Les Entretiens de Provence: Le Juge Dans la Societe Contemporaine 83 (Robert Badinter & Stephen Breyer eds., 2003).

n21. John Rawls, Justice as Fairness: A Restatement 112 (Erin Kelly ed., 2001).

n22. See Bruce Ackerman, We The People (1991).

n23. John Rawls, Political Liberalism (1996).

n24. 347 U.S. 483 (1954).

n25. 410 U.S. 113 (1973).

n26. See New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting) ("It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.").

n27. Vacco v. Quill, 521 U.S. 793 (1997); Washington v. Glucksberg, 521 U.S. 702 (1997).

n28. Brief of Amici Curiae Ronald Dworkin et al., Glucksberg (No. 95-1858, 96-110), available at 1996 WL 708956. In addition to John Rawls, the brief was signed by Thomas Nagel, Thomas Scanlon, and myself, who are participants in this conference, and Robert Nozick and Judith Jarvis Thomson.

n29. 410 U.S. 113 (1973).

n30. See John Rawls, Commonweal Interview with John Rawls, in John Rawls: Collected Papers 616, 618 (Samuel Freeman ed., 1999).

n31. For an exhaustive discussion of the status of abortion law in Europe, see Inter-Departmental Working Group on Abortion, Gov''t of Ir., Green Paper on Abortion P 3.02, available at http://www.taoiseach.gov.ie/index.asp?docID=238 (last visited Apr. 14, 2004).

n32. Pub. L. No. 108-105, 3(a), 117 Stat. 1206 (2003).

n33. 347 U.S. 483 (1954).

n34. Loving v. Virginia, 388 U.S. 1 (1967); McLaughlin v. Florida, 379 U.S. 184 (1964).

n35. 410 U.S. 113 (1973).

n36. Sch. Dist. of Abington Township, Pa. v. Schempp, 374 U.S. 203 (1963); Engel v. Vitale, 370 U.S. 421 (1962).

n37. 531 U.S. 98 (2000).

n38. Rawls, Political Liberalism, supra note 23, at 124.

上一篇:陈宜中:罗尔斯的国际正义论与战...      下一篇:哈贝马斯:评罗尔斯的《政治自由...
发表评论 回到页顶
正来学堂版权所有 © 2009 沪ICP备042465号
地址:上海市杨浦区邯郸路220号光华楼东主楼28楼复旦大学社会科学高等研究院 邮编:200433