Navigation
Newsletters Index (06:2)
apaOnline Home Page

APA Newsletters

Spring 2007
Volume 06, Number 2

Newsletter on Philosophy and Law

Articles

Razian Concepts

Liam Murphy
New York University

In much of his work in political and legal philosophy, Joseph Raz has been concerned to explain certain concepts—such as those of law, authority, and rights. He has always insisted that these explanations are not investigations into the meanings of the words used to express the concepts. But they also seem to be distinct from classical substantive political theorizing. The Morality of Freedom is about freedom, but not, I think, about the concept of freedom—at least not in the same way in which its second and third chapters are about the concept of authority and its seventh is about the concept of rights.

Raz writes that "moral and political philosophy has for long embraced the literary device (not always clearly recognized as such) of presenting substantive arguments in the guise of conceptual explorations."1 If that device is employed, any political theory can be cast as an account of one or more concepts, and The Morality of Freedom can be said to be about the concept of freedom after all—as Raz himself notes.2

There is thus a vexing ambiguity in the meaning of "concept." H. L. A. Hart called his book The Concept of Law despite the fact that very little of it was—in what seems to me to be the current ordinary sense—about the content of a concept at all. The three key questions he associates with interest in the nature of law are: "How does law differ from and how is it related to orders backed by threats? How does legal obligation differ from, and how is it related to, moral obligation? What are rules and to what extent is law an affair of rules?"3 The second of these questions involves conceptual issues, but the others, especially the third, do not. The question of whether law is an affair of rules is one that can be answered by looking, once you know where to look, and most of The Concept of Law is about this question; its answer constitutes Hart’s "improved analysis of the distinctive structure of a municipal legal system."4 Though it proceeds on the assumption of a positivist conceptual view about the boundary between law and morality, most of that analysis is descriptive in the sense that Hart claims for his enterprise.5 We can call this descriptive core of the book an account of the concept of law only if we embrace a wide sense of "concept" close to Raz’s "literary device." It is now more typical to understand conceptual questions as ones of sense and categorization; only about five pages of The Concept of Law were devoted to such questions.6

But more has changed since 1961 than literary conventions. Moral and political philosophy is today also less concerned with issues of sense and categorization than it once was. Perhaps this is due to the publication of A Theory of Justice in 1971—which makes it all the more interesting to be reminded that, as late as 1964, Rawls continued to insist that he was providing an analysis of the concept of justice.

What such an argument might show is that, if certain natural conditions are taken as specifying the concept of justice, then the two principles of justice are the principles logically associated with the concept when the subject is the basic structure of the social system. The argument might prove, if it is correct, that the principles of justice are incompatible with the principle of utility. The argument might establish that our intuitive notions of justice must sometimes conflict with the principle of utility. But it leaves unsettled what the more general notion of right requires when this conflict occurs. To prove that the concept of justice should have an absolute weight with respect to that of utility would require a deeper argument based on an analysis of the concept of right, at least insofar as it relates to the concepts of justice and utility.7

From the perspective of students of A Theory of Justice, there is an other-worldly quality to this passage. The position of A Theory of Justice seems roughly to be this. Let’s take for granted that we all want our social institutions to be just; we all also have some rough sense of the subject matter of justice—of the political and moral questions to which a theory of justice is expected to offer answers. Where we disagree is in the content of our substantive theories of what justice requires. Two theories of justice are particularly significant—justice as fairness, drawn out of the social contract tradition, and utilitarianism. A big part of the argument of the book is then devoted to showing the superiority of the former. But that superiority is not grounded in its better match with the alleged content of our existing concept of justice.

In the opening pages of A Theory of Justice, Rawls writes: "The concept of justice I take to be defined, then, by the role of its principles in assigning rights and duties and in defining the appropriate division of social advantages. A conception of justice is an interpretation of this role."8

This is Rawls’s expanded version of Hart’s idea that all accounts of justice tell us to treat like cases alike, while different accounts of justice vary in their criteria for likeness.9 Rawls and Hart held that so much is fixed by the very concept of justice. But neither took that to be controversial or in need of sophisticated philosophical defense. The important point, for both, is that theories of justice offering very different criteria of distribution are equally compatible with the concept of justice and that what is fixed by the concept is, as Hart put it, "incomplete" and cannot "afford any determinate guide to conduct."10 As Rawls writes: "Clearly this distinction between the concept and the various conceptions of justice settles no important questions. It simply helps to identify the role of the principles of social justice."11

So we get a book that its author later abbreviated as Theory, rather than Concept; and this difference from the earlier articles is not just a matter of terminology.12 It is partly that, I think, because much of the argument in the earlier papers to the effect that the concept of justice is best analyzed in terms of the two principles of justice does look very much like substantive political theorizing. But if Rawls were merely employing the literary device, presenting a theory of justice under the name of an account of the concept, it is not at all obvious that he would need a further argument to show the priority of justice (as fairness) to utility.

I have raised this issue about the development of Rawls’s views because it will help illuminate, I believe, the question I want to raise about Raz’s political theorizing. In his most recent methodological writings, Raz appears to embrace the wide sense of "concept" he once described as a literary device. He is happy to follow Hart and see account of the concept of law as an account of the nature of law.13 And an account of the concept of authority will evidently answer the obviously substantive question of the "possible justification of subjecting one’s will to that of another."14 But then Raz also believes that his accounts of the concepts of law and authority are explanatory, in the sense that they bring out something we already all have in common. For example, he writes that it is "a major task of legal theory to advance our understanding of society by helping us to understand how people understand themselves."15 Since there is clearly widespread disagreement about the nature of law and of rights, and about when subjecting one’s will to that of another is justified, there would seem to be a puzzle here. But then Raz has long signaled this feature of his approach to the analysis of concepts by giving it the label "normative-explanatory."16 His accounts are not just explanatory, capturing common ground; the normative dimension takes a stand on matters in dispute—such as the "precise connections" between various concepts17 and also, I assume, such matters as the exact conditions for justified subjection of one will to that of another.

Raz has recently characterized his approach in a somewhat different way. He distinguishes between the conditions for "minimal possession" of a concept—"those essential or nonessential properties of what the concept is a concept of, knowledge of which is necessary for the person to have the concept at all"—and the conditions for the "knowledge involved in complete mastery of the concept, which is the knowledge of all the essential features of the thing it is a concept of."18 It seems to me that the right way to read this is to match "minimal mastery" with what I have said is the current, ordinary sense of "concept"—as a shared practice of fundamental categorization. But in defining complete mastery of a concept as he does, Raz suggests that he has in mind, as basic, the wide, "literary device" sense of "concept." One interesting aspect of this way of looking at things is that it admits no clear distinction between concept and theory. Some believe it is one thing to fix concepts (in the narrow sense having to do with sense and categorization), and another to offer a theory (descriptive or normative) of the thing or subject in moral or political theory that the concept picks out. This way of looking at things runs into familiar philosophical objections. In response, many have followed Quine in holding, in effect, that "it’s all theory." One might say that Raz agrees with this, but, like Hart, simply uses "concept" to mean what others mean by "theory."
But this will still leave the question of how to understand the explanatory parts of Razian conceptual accounts and how to assess the role such claims may play in his legal and political theory. To keep things terminologically tractable, let me now just stipulate that a conceptual question is a question of basic categorization.19 The most natural way to approach the question of the content of a concept is by addressing the meanings of the words typically used to express it.20 This also brings out most clearly that the conceptual, on this usage, is not a matter of substantive theory. It is not a conceptual question whether Rawls’s two principles of justice are the proper moral principles to guide institutional design. It is not even a conceptual question whether they provide the best account of what justice requires when applied to the basic structure of society. Since, like all concepts, the concept of justice is indeterminate, the most it can do is specify the rough subject matter of theories of justice. This narrow sense of the conceptual thus matches the later rather than the earlier Rawlsian view.

However, the very idea of the conceptual, in this narrow sense, does not commit us to any philosophically significant distinction between analytic and synthetic claims. Though claims about proper categorization do feel like claims about the proper use of words, there seems to be no reason why we could not understand them as just the most fundamental descriptive or normative commitments we have, the shared background that is required for disagreement to be possible. The label "conceptual" could be understood just to mark out positions on the nature of law or authority or rights that are beyond the pale, not worth considering, at least for the time being. Since such commitments are not the subject of reasoned argument but, rather, taken for granted, they would be revealed in the same way truths of meaning have traditionally been thought to be revealed—by intuitive responses to cases. And there would be no reason to insist that such commitments are immune to revision in light of further experience.

My understanding, then, is that the explanatory aspect of a Razian account sets out (at least) what he calls the conditions of minimal possession of the concept and that all this is equivalent to a conceptual analysis understood along the lines I have just given.

I doubt that conceptual analysis, thus understood as the attempt to uncover shared practices of basic categorization, can play any significant role in political or legal theory. But I also think that it plays very little role in Raz’s political philosophy. Here, I just want to raise doubts about its role in one famous argument in Raz’s philosophy of law.

Raz obviously does not believe that his accounts of authority or rights are conceptual in my narrow sense. That sense of the conceptual seems to correspond to what Raz calls the "purely linguistic." He writes:

A purely linguistic account of authority claims to yield a simple explanation of what people believe who believe that someone has legitimate authority. Had the above account been a linguistic account, an explanation of the meaning of "legitimate authority," it would have followed that anyone who believes of a person that he has legitimate authority believes that that person satisfies the condition set by the justification thesis. This implication does not hold for a normative-explanatory account. In being normative it avows that it does not necessarily conform to everyone’s notion of authority in all detail.21

In other words, it is clearly not common ground among competent users of the concept of authority that Raz’s normal justification thesis sets out the (normal) conditions under which the subjection of one person’s will to another is justified.
All this seems as clear as can be, yet it makes Raz’s celebrated argument for the "sources thesis"—the idea that moral considerations cannot be among the grounds of law—in "Authority, Law, and Morality" rather puzzling. It is clear enough, and too familiar to need reviewing here, that a system of governance in which moral considerations in part determine what the rules of conduct are cannot meet the conditions of Raz’s theory of authority. We can also simply grant Raz’s assumption that "necessarily law. . .either claims that it possesses legitimate authority or is held to possess it, or both."22 The tricky part is to know how to understand Raz’s inference to the conclusion that, therefore, the sources thesis is an essential feature of law:

A legal system may lack legitimate authority. If it lacks the moral attributes required to endow it with legitimate authority then it has none. But it must possess all the other features of authority, or else it would be odd to say that it claims authority.23

What is puzzling is that insofar as Raz’s service conception of authority is normative, it does not seem in the least bit odd to say that law claims something that it could not, as it turns out, have. Raz considers this worry:
Can it not be objected that my argument presupposes that people know the normal justification thesis, and the others which go with it? To be sure such an assumption would not be justified. Nor is it made. All I am assuming is that the service conception of authority is sound, i.e., that it correctly represents our concept of authority. It is not assumed that people believe that it does.24
But we know that, for Raz, "our concept of authority" is not something that we all share fully, even implicitly. It is not a shared practice of categorization that can be discovered by people’s intuitive reactions to questions about particular cases. For Raz, "our concept" is something we partly share and partly disagree about. But since there is no agreement about the normative part of Raz’s account of authority, and it is the normative part that does the work in the argument for the sources thesis, I am at a loss to understand why we should expect that claims to authority will be consistent with that account. It would be odd to see a universal claim to authority that was inconsistent with a practice of categorization that we all share. That would amount to what is commonly called a "performative contradiction" in contemporary German philosophy; but Raz is not making this kind of argument.

Raz also notes that the claim to authority is made by legal officials, and so it is appropriate to rule out the possibility that it is normally based on a "conceptual mistake."25 But by "conceptual mistake" Raz clearly does not mean here a failure to appreciate the truth of the service conception of authority, since that is presumably almost the universal condition of legal officials. He must rather mean something like failure in the minimal conditions for possession of the concept, and minimal possession of the concept is not enough for the argument to go through.

All this suggests misunderstanding on my part. Perhaps, like the early Rawls, Raz has in mind a sense of "concept" that is more complex than the simple combination of category and theory I have attributed to him.

Raz writes that "there is an interdependence between conceptual and normative argument."26 Perhaps it is significant that we can say the following: though we don’t necessarily already share, even implicitly, the correct account of "our concept of authority," the account, as an account of our concept nonetheless is an account of something we all share. When Rawls offered justice as fairness as an account of the concept of justice, he was not claiming that everybody already implicitly knew the content of the two principles. But he did believe that those principles were an attractive development of some normatively significant features of what we did share by way of a practice of categorization, and that this mattered. It mattered that he was offering a theory of justice, rather than something else—such as the political morality relevant to the design of social institutions. Similarly, perhaps it is important that Raz is offering a theory of authority, and not, say, of the conditions under which one person or group can impose obligations on others. As he often writes, accounts of concepts are important because they are accounts of how we understand ourselves (in terms of such concepts). Perhaps we here can understand the contrast with his account of freedom, which is not presented as a theory about any particular concept of freedom. He writes that if The Morality of Freedom is to be characterized as defending a concept of freedom, it is "only important to remember that that concept is a product of a theory or a doctrine consisting of moral principles for the guidance of and evaluation of political actions and institutions."27 The whole theme of self-understanding seems to be missing here.

I agree with Raz that self and social understanding is important and that we cannot approach that without reflecting on our conceptual practices. But I am doubtful about the connection between this project and the kind of positive normative theorizing Raz offers with his accounts of authority and rights. I don’t think it is important to link our political and legal theory closely to particular concepts. So long as we share enough, conceptually, to understand what we are arguing about, there seems to be no particular reason to offer political theory category by category—fixing each component in this way before turning to consider how they are related within the category of the right. If it is important that these are the concepts we have employed to understand our social world, the factors that make it so will not be lost in any adequate paraphrasing of our political and legal discourse. Reflection on concepts is extremely important if we are to understand what we and others have been saying and to be in a position to subject our political discourse to critique. But there is no philosophical reason why what is uncovered through critique and the search for self-understanding should be preserved.

In any case—returning now to Raz’s argument for the sources thesis—even if a theory of authority is a theory of a concept we all in a sense share, it still won’t be odd to claim something about law and authority that makes no sense if seeing that it makes no sense requires us also to share the correct full account of the concept of authority, which we do not. Absent some claim of implicit knowledge of the correct full account of authority, this step in Raz’s argument remains puzzling to me. A comparison case may be helpful. Suppose that on the correct accounts of the concepts of justice and authority, there is a "precise connection" between them: the issue of justice cannot arise other than in a social setting where a de facto coercive power claims legitimate authority. It is an implication of this view that, as our world is currently set up, there cannot be a question of global justice. Would there be anything odd about people making claims about global injustice?

Insofar as my remarks here are critical, they apply to Raz’s theory of law only. His accounts of authority and rights remain foundational contributions to political philosophy whether one thinks of them as all theory or all concept (not to mention the accounts of the rule of law, reason, well being, freedom, and so on). But unlike debates about these issues in moral and political theory, I believe that the debate about the eligibility of moral considerations in determinations of the content of the law in force is, in the end, just a conceptual question in the narrow sense I have outlined. As such, I believe, it has no resolution since the concept of law is simply indeterminate on the issue; there are, among us, several different fundamental understandings of law as a category.28 If this is right, it is no surprise that Raz’s broader and more sophisticated understanding of what it is to offer an account of a concept cannot move an argument for positivism forward.

Endnotes

1. Joseph Raz. The Morality of Freedom (Oxford: Oxford University Press, 1986), 16.

2. Id.

3. H. L. A. Hart. The Concept of Law, 2nd ed. (Oxford: Clarendon Press, 1994), 13.
4. Id., 17.

5. For further discussion of this point, see Liam Murphy, "The Political Question of the Concept of Law" in Hart’s Postscript, edited by Jules Coleman (Oxford: Oxford University Press, 2001), 371-409.

6. Specifically, pp. 207-12.

7. John Rawls. "Legal Obligation and the Duty of Fair Play" (1964). In Collected Papers, edited by Samuel Freeman (Cambridge, MA: Harvard University Press, 1999), 125

8. Rev. ed. (Cambridge, MA: Harvard University Press, 1999), 9.

9. See The Concept of Law, 159-60.

10. Id., 159.

11. A Theory of Justice, 5.

12. In "Justice as Fairness" (1958), Rawls is at pains to note that he has "been dealing with the concept of justice" (Collected Papers, 71).

13. "Two Views of the Nature of the Theory of Law," in Coleman, 8.

14. "The Problem of Authority," Minnesota Law Review 90 (2006): 1003.

15. "Authority, Law, and Morality," in Ethics in the Public Domain, 221.

16. See The Morality of Freedom, 63-6.

17. Id., 63.

18. "Can There Be a Theory of Law?" In Philosophy of Law and Legal Theory, edited by Martin P. Golding and William A Edmundson (Blackwell, 2005), 326.

19. I follow Frank Jackson, From Ethics to Metaphysics (Oxford: Oxford University Press, 1998).

20. Raz rejects the relevance of an inquiry into the meaning of the word "law "to the study of the concept of law on the ground that "law" is used in a variety of contexts that have nothing to do with legal systems (see, e.g., "Two Views," 7). I don’t understand this point, since it would seem to apply equally to the concept of law. In any event, though Raz insists that "law" is not ambiguous, in fact it seems to be. Hence the joke in the following lines from a song by Billy Bragg:

The laws of gravity are very very strict
And you’re just bending them for your own benefit.

Perhaps there is a common meaning to "law" in all contexts where it is appropriately used. But, in the first place, Raz’s account of what that is seems wrong: "The word is used in all these contexts [legal, religious, mathematical, etc.] to refer to rules of some permanence and generality, giving rise to one kind of necessity or another" ("Can There Be a Theory of Law?" 325). For classical natural lawyers, the natural law was at least in part a matter of following natural inclinations in the right way; it is hard to understand this in terms of rules. And when Dworkin rejects the model of rules for law—essentially arguing that the law is the outcome of a certain weighing of values as applied to a particular case—his position doesn’t seem to run afoul of the meaning of "law." Last, many of us don’t recognize necessity of any kind in the mere existence of a legal rule. But even if this definition were correct, I can’t see why it would be incompatible with the view that there are (say) four different senses of the word "law," appropriate for different contexts. In particular, there seems to be all the difference in the world between a law of physics, which is not in any sense practical, and the rest. That’s why it’s funny to say that the laws of gravity are strict.

Raz also notes that we need not use "law" to make use of the concept of law (id.). That seems right, but since the way we usually make use of the concept of law is by using the word "law" or some translation of it, the place to look if we are trying to figure out what kind of categorization the concept of law recommends is by looking at the usage of the words that express it.

21. The Morality of Freedom, 65; see also, "The Problem of Authority," 1006-7.

22. "Authority, Law, and Morality," 199.

23. Id.

24. Id., 203-4.

25. Id., 201

26. The Morality of Freedom, 63.

27. P. 16.

28.I defend this position in an unpublished manuscript. Some first thoughts appear in "Concepts of Law," Australian Journal of Legal Philosophy 30 (2005): 1-19.

Previous Article | Index | Next Article

APA Newsletters

Spring 2007 Volume 06, Number 2

Spring 2007
Volume 06, Number 2