How Law Claims, What Law Claims
Abstract and Keywords
Many people think that law, wherever it may be found, makes certain characteristic assertions, claims, self-presentations, or promises. In recent times, such an idea has been endorsed and relied upon by writers as otherwise diverse as Drucilla Cornell, John Finnis, Philip Selznick, and Jacques Derrida. But it has come to be particularly associated with the work of Joseph Raz and Robert Alexy. Both Raz and Alexy believe that it is part of the very nature of law that all law makes a moral claim. They disagree about what exactly the content of the moral claim is. Raz says it is a claim to moral authority. Alexy says it is a claim to moral correctness. This chapter assesses the thesis on which Raz and Alexy converge, namely that the law claims some moral standing for itself. Is this thesis true? Is it even intelligible? It is argued that this thesis is not only intelligible but true. The chapter attempts to allay various doubts and identifies problems with how the thesis has been presented that may have contributed to the spread of those doubts.
Many people think that law, wherever it may be found, makes certain characteristic assertions, claims, self-presentations, or promises. In recent times, such an idea has been endorsed and relied upon by writers as otherwise diverse as Drucilla Cornell,1 John Finnis,2 Philip Selznick,3 and Jacques Derrida.4 But it has come to be particularly associated with the work of Joseph Raz and Robert Alexy. Both Raz and Alexy believe that it is part of the very nature of law that all law makes a moral claim. They disagree about what exactly the content of the moral claim is. Raz says it is a claim to moral authority. Alexy says it is a claim to moral correctness. I will say something shortly about the differences between these views, and their respective attractions. My first objective in this paper, however, will be to assess the thesis on which Raz and Alexy converge, namely that the law claims some moral standing for itself. Is this thesis true? Is it even intelligible? I think it is not only intelligible but true. In what follows, I will try to allay various doubts, but also to identify problems with how the thesis has been presented that may have contributed to the spread of those doubts.
A. Law's Capacity to Claim
A natural first question is this. Is law even capable of making claims? Is it the right kind of thing to do so? A prominent doubter is Ronald Dworkin.5 He directs his (p.30) doubts mainly at Raz's treatment of the topic, although the same doubts could be raised about Alexy's writings. As Dworkin notes, whatever sense we give to Raz's talk of ‘law's claims’, it must be consistent with Raz's thesis that claiming moral authority is not the same as having moral authority.6A claim, for Raz, has to be capable of being true or false. That is already enough to show that Raz uses the word ‘claim’ advisedly, intending to invoke at least part of its literal meaning. It is not, for him, a mere figure of speech. On the other hand, argues Dworkin, Raz's talk of ‘law's claims’ must be partly a figure of speech. For a claim surely requires a claimant, a person who advances it; and law is not a person. It may be a practice; it may be the set of all norms of a certain type, or all normative systems of a certain type; it may be a mode of social organization; it may be an ideal. But it is plainly not a person, and any personification of it must therefore be figurative. So in Raz's talk of ‘law's claims’, we can read ‘claims’ literally only if we read the attribution of claims to law figuratively.
Or can we? Maybe—Dworkin concedes—Raz's attribution of claims to law is not so much figurative as elliptical. For Raz often puts law's supposed claim to moral authority in the mouth of law-applying officials, who are admittedly persons:
So perhaps, when he talks of law's claims, Raz means no more and no less than claims that law-applying officials advance about the legal system of which they are officials. This avoids the personification of law itself. But it faces the new problem, says Dworkin, that not all law-applying officials make the claim that Raz ascribes to all law.8 Dworkin gives the example of Oliver Wendell Holmes. Holmes, he points out, believed that law is incapable of creating moral obligations. So he could not have believed that law is capable of having moral authority, which is a way of creating moral obligations. So he could not, as a Justice of the Supreme Court, have joined in with the claims that Raz says law-applying officials must make for law. More fundamentally, says Dworkin, there is nothing that unites the ‘actual beliefs and attitudes’ of law-applying officials such that any uniform claim—be it a claim to moral authority or otherwise—is made by them all alike and could be elliptically ascribed to law.9 (p.31) Dworkin makes several mistakes in this argument. I will mention three. His first mistake is to think that what an official claims for the law depends on what ‘beliefs and attitudes’ the official has about or towards the law.10 It is true that Raz speaks in the above passage of how officials ‘regard’ the law. But this is misleading. As Raz makes clear elsewhere, claiming that law has moral authority is consistent with believing that it does not. It is also consistent with having no morally favourable attitude towards law.11 It is consistent with regarding law as a joke, or a racket, or a scam. That is because, as well as being capable of being true or false, claims are capable of being sincere or insincere. A charity worker and a confidence trickster may equally claim to be helping the poor. Perhaps neither of them actually helps the poor; perhaps both of them make a false claim. But only one of them makes an insincere claim. The same cross-cutting distinctions may be drawn where the claims of law-applying officials are concerned. As well as classifying their claims as true or false, we may classify their claims as sincere or insincere.12 So it is no objection to Raz's view that Holmes does not believe that law ever creates moral obligations, or does not have a positive moral attitude towards any law. This leaves the possibility that Holmes brazenly (because insincerely) claims otherwise when he sits on the bench. If asked why, he might say: ‘That's how they pay me to talk. I'm only doing my job.’
The claims that law makes for itself are evident from the language it adopts and from the opinions expressed by its spokesmen?.?.?.?The law's claim to authority is manifested by the fact that legal institutions are officially designated as ‘authorities’, by the fact that they regard themselves as having the right to impose obligations on their subjects, by their claims that their subjects owe them allegiance, and that their subjects ought to obey the law as it requires to be obeyed.7
There is a further possibility. Holmes can claim that the law creates moral obligations without even believing that this is what he is claiming, never mind whether he believes that what he is claiming is true. Even if sincere, he might be confused. People often claim things without realizing that they are claiming them, because they are not fully aware of the meaning of what they are saying or doing. So one need not believe that Holmes was insincere in order to admit that he made claims for law that were at odds with his beliefs about law (i.e. claims that attributed to law properties that Holmes believed law not to have).
Dworkin's second mistake is to think that the question whether law-applying officials make a uniform claim for law can only be an empirical one, so that it could be answered by a study of the behaviour of law-applying officials (once they have been independently identified as such).13 Not so. Suppose Dworkin is right that Holmes, in certain of his pronouncements, declined to make moral claims about law. The only conclusion we should draw, according to Raz's analysis, is that (when he made those pronouncements) Holmes was not acting in his capacity as a law-applying official. He was not acting on behalf of the legal system. This chimes with what we know of Holmes. His conceptually revisionist work in the law journals, attempting to debunk the ordinary legal discourse of obligations and rights and so (p.32) on,14 did not prevent him from participating fully in the ordinary legal discourse of obligations and rights and so on when he sat in the Supreme Court.15 In the former role he was not an official and did not speak for the law; in the latter role he was and he did. You may say that the criteria by which we determine this are independent of any claims that Holmes made in the two contexts. If Holmes was not writing in his capacity as a law-applying official when he was debunking law, that is because at the time he was not wearing his robes, or not sitting in his courtroom, or not dealing with a dispute between real people, or not being paid by the US Treasury, or such like. Yet none of these indicators of official activity is any more than casual. In principle—and in many legal systems in practice too—a senior judge can exercise her authority on the phone, wearing pyjamas, outside working hours, and in anticipation of a possible future dispute. By what criterion is this activity official, now that the casual indicators are gone? Says Raz: by the criterion, inter alia, of the official's claims as she acts. And this is a conceptual, not an empirical, proposition, a proposition that no survey could displace (for there is now no independent way to identify the class of behaviour that would need to be surveyed).
Dworkin's third error lies in his thinking that the claims of law-applying officials can be attributed non-elliptically and non-figuratively to law only by an implausible personification of law. What makes a personification of law implausible, I suggest, is its attribution to law of what I will call ‘concerted agency’. There are two types of concerted agent.16 There are those, such as football teams and symphony orchestras, that come into existence simply by virtue of the mutually responsive actions of their members; and then there are those, such as companies and legislatures, that come into existence by virtue of constituting rules which ascribe actions to them on the occasion of certain of their members’ actions (whether those in turn be individual or concerted actions).17 The hallmark of concerted agency, either way, is that the actions of a concerted agent are a function of the actions of (one or more of) its members, and are intended by its members, yet are not identical with the actions of its members. When a legislature legislates, for example, its doing so is a function of the voting of those of its members who intend the legislature thereby to legislate, but the legislature does not thereby do what its members do. The members vote; the legislature legislates. And when an orchestra plays a symphony, its doing so is a function of the playing of various instrumental parts by its members, who intend the orchestra thereby to play a symphony, but again the orchestra does not do what its members (p.33) do. The members play their parts; the orchestra as a whole plays the symphony. In the same vein, the CEO signs her name intending that the company should thereby enter into a contract, but it is the company that enters into the contract. And the striker scores a goal intending that his team should thereby win the game, but only the team wins the game. We could capture this point by saying that the intentional actions of concerted agents are logically, but not materially, autonomous.
Now law as a whole is not a concerted agent. Nor, I would add, is the law of a single jurisdiction. Nor, for that matter, is a single legal system. Yet it is easy to see why one might be tempted to think otherwise. Legal systems do perform logically autonomous actions. For example: a typical modern legal system regulates almost every area of its subjects’ lives, even though no legal official regulates almost every area of its subjects’ lives. However, such logically autonomous actions of legal systems—often ascribed elliptically to law itself—are not intentional. Like the actions of markets, societies, and other highly complex webs of human activity they are the unintentional by-products of many actions performed with other (more modest) intentions. So they do not represent any kind of concerted agency.
This point is irrelevant, however, to the truth of Raz's thesis that law claims moral authority. Claiming cannot but be an intentional action. But the law's action of claiming moral authority is not autonomous, even logically autonomous, of the actions of law-applying officials. Law makes claims only insofar as law-applying officials make those very same claims at the very same time and place. The claims of law are identical to certain claims of its officials. And these claims must be non-elliptically ascribed to law, not because of any mutual responsiveness among the law's officials, nor because of any constitutional rules that make law itself the agent of anything in virtue of what its officials do. Rather, they must be non-elliptically ascribed to law because the only way to unpack the idea that they are claims made by law-applying officials is as follows. Some people (be they dressed in robes or in pyjamas) make these claims on behalf of law, and making these claims on behalf of law is part of what makes them law-applying officials. It is an irreducible part of this explanation that the claims in question are made on behalf of law. One cannot omit, from any adequate explanation of what a law-applying official is, the fact that law-applying officials serve as law's representatives or spokespeople, identified by law to do law's bidding. So one cannot explain the nature of the action performed by the official without ascribing agency (albeit not autonomous agency) to law itself. I tend to think that persons are agents whose intentional agency is at least logically autonomous, so that ascribing non-autonomous agency to law is not a personification of law at all.18 But if there is some kind of personification here, it is only a very attenuated and not at all disturbing personification, falling well short of what befits concerted agents such as teams or companies or indeed specific institutions of law (e.g. legislatures and courts).19 (p.34)
B. Law's Claims as Moral Claims
So that, with no more than the barest of personifications, is how law makes claims. Armed with this capacity to make claims, what claims, if any, does law indeed make? The place to begin, nobody doubts, is with the language that law-applying officials use. In explaining the law, they cannot but use the language of obligations, rights, permissions, powers, liabilities, and so on. What they thereby claim—and they cannot say it without claiming it—is that the law imposes obligations, creates rights, grants permissions, confers powers, gives rise to liabilities, and so on. The question is: what do these claims amount to?
Some people hold that the full necessary extent of the claims made by officials who use this language is that there are legal obligations, legal rights, legal permissions, legal powers, legal liabilities, and so on.20 Now it is certainly true that there are such things, and that their existence can be and routinely is claimed by law-applying officials. But this claim cannot be law's claim. Why? Because a legal obligation or right is none other than an obligation or right that exists according to law. And an obligation or right that exists according to law is none other than an obligation or right, the existence of which law claims. So the claim that there is a legal obligation or right—whether made by a law-applying official or by anyone else—is a second-order claim, a claim about what law claims. Now it is true, of course, that law could make a second-order claim about its own claims. But not this one. For as we already learned, a claim has to be capable of being true or false. It is not a claim unless there is logical space for its falsity; and it makes no sense to attribute to law a false claim about these legal obligations and rights, for there is no criterion of legal truth and falsity that is independent of law. So an error about, and hence a claim about, what legal rights and obligations there are can only be attributed to a particular law-applying official, not to law itself.21 I am not suggesting, of course, that the official makes the error in her personal capacity. For in other respects—for example in making law's moral claim—she may continue to represent the law even while she is making her error of law.22 All I am saying is that her error of law, and her claim about what legal rights and obligations there are, are hers and not the law's, albeit they are an error and a claim that she makes in the course of her work as a law-applying official.
It is against this background that the proposal emerges that law's own claim is a moral one: that when, according to law, there are obligations and rights and so (p.35) on, law's claim is that these are moral obligations and rights and so on, not merely legal ones. Unless accompanied by some clarification, this way of presenting the content of law's claim courts confusion. Notoriously, the sense of the word ‘moral’ shifts depending on what it is contrasted with.23 Two of its many senses are important here. We sometimes use the word ‘moral’ in a broad sense simply to draw the contrast already drawn in the previous paragraph. There are obligations and rights, such as legal obligations and rights, or the obligations and rights accorded by tradition or convention, which are merely claimed or supposed obligations and rights. And then there are obligations and rights that are not merely claimed or supposed. They are the very ones that the claimed or supposed ones are claimed or supposed to be. If we use ‘moral’ to mark out this entire group of NMCS (‘not merely claimed or supposed’) considerations, it is an innocent tautology that law claims its obligations and rights, and indeed all other legal considerations, to be moral ones. However we sometimes use the word ‘moral’ more narrowly to refer to just some of the considerations that qualify as ‘moral’ in the broader sense just mentioned. We contrast them with other NMCS considerations. We contrast them with aesthestic considerations, say, or with considerations of prudence or self-interest. We might disagree about which considerations qualify as NMCS, and we may doubt the sustainability of some of the familiar contrasts drawn among even what we agree to be NMCS considerations. But we can still agree that it is a familiar use of the word ‘moral’ to pick out just some of these NMCS considerations, so that the proposition ‘law makes moral claims’ is no longer tautological.
Or is it? I shifted here from talk of obligations and rights to talk of ‘considerations’ because otherwise tautology quickly seems to return. Which subset of NMCS considerations are we to designate by the word ‘moral’? Here we are faced, no doubt, with a subsidiary fragmentation of the possible senses of ‘moral’. But it is tempting to think that, in all the relevant senses of the word ‘moral’, the subset of NMCS considerations that qualify as moral includes all those considerations that are obligatory, which in turn includes all rights-based considerations. There may be self-interested and aesthetic considerations, but there are no self-interested or aesthetic obligations. As soon as we encounter obligations, we are thereby crossing the line into morality. Thus, if the law's ultimate concern is with obligations—if the law's concern with permissions and powers and liabilities and rights and so on is ultimately a concern with the obligations to which they give rise—then the law's claims turn out to be tautologically moral even in the narrower sense of ‘moral’. Or so it is tempting to think. In fact the matter is not quite so simple. First, it is not so clear that the law's ultimate concern is only with obligations.24 Second, there are (p.36) some NMCS obligations which resist the designation ‘moral’, such as obligations of etiquette. When we are subdividing the NMCS domain, we usually reserve the title ‘moral’ for important NMCS obligations, those NMCS obligations, breach of which has important consequences for somebody. So even if we limit our attention to NMCS considerations ‘moral obligation’ is not quite a tautology.
It follows that the idea that law claims its obligations to be moral obligations is equally not quite a tautology. Yet we can restore our confidence in the idea that all law makes a moral claim for itself, even in this narrower sense of ‘moral’, by noting the following. The mere fact of their being embodied in law already lifts what would otherwise be non-moral NMCS obligations—such as obligations of etiquette—from a relatively unimportant to a relatively important position. Advertisement of the law will cause some people to alter their daily pursuits. Enforcement of the law will put some people under stress or cost them money or freedom. Even if the law is not advertised or enforced, this itself raises moral issues about the behaviour of those responsible for advertising or enforcing it. Every legal issue, however superficially technical, is a moral issue, for its resolution inevitably has important consequences for someone.
You may wonder why we should be interested in whether law makes a moral claim, in the narrower sense of ‘moral’. Here is one suggestion. We use this distinction between moral claims and non-moral claims to distinguish legal systems from, for example, games and recipes. Let me focus on games.25 Games include such things as obligations, permissions, rights, powers, and liabilities. In Monopoly, for instance, there is the right to receive £200 as one passes ‘Go’, the power to buy an unowned property when one lands on it, the obligation to pay rent when one lands on an unmortgaged property owned by another player, and the permission to leave jail with a ‘Get Out of Jail Free’ card. It is a normative system, so far indistinguishable from a legal system. Where it differs is in the claims it makes on its players. It is not that Monopoly makes no claims—for it too has a rule-applying official who speaks for it, in the form of the banker—nor that it does not claim to create NMCS considerations. It is merely that these claims are not moral claims, in the narrower sense of ‘moral’. For adding an obligation, or power, or permission, or right, or liability to Monopoly need not have important consequences for anyone (assuming that it does not affect the overall playability of the game). This difference between games and legal systems has important consequences of its own. Since law is not a game, nobody—least of all law-applying officials—should take a playful or light-hearted attitude towards it. As we saw, a law-applying official might conceivably take such an attitude, speaking with apparent earnestness for law but all the time laughing to himself about law's stupidity. This attitude does not stop one from being a law-applying official but it does call into question one's fitness for the role. By contrast, one may be fit for the role of banker in Monopoly even though one takes a light-hearted attitude to both the role and the game.
(p.37) In a way more fundamental, though, is a feature that law shares with Monopoly. There are two opposite errors about what is known as the ‘normativity’ of law that constantly recur in the literature, and it is hard to keep them both at bay at once. On one hand there is the error of thinking that legal obligations (and rights and permissions and so on) are themselves one family of NMCS obligations (etc.), and hence are not merely claimed to be. Indeed they are not even claimed to be. It cannot be a claim because there is no logical space for it to be false. On this view, when the law says ‘jump’ one has without further ado a NMCS obligation to jump. It does not even need an argument from the legal to the NMCS obligation; for if there is no NMCS obligation there is by that token no legal obligation. On the other hand we find the opposite but sadly even more familiar error of understanding law as the ‘gunman situation writ large’.26 Legal considerations are prudential considerations—threats and other incentives—and inasmuch as law makes claims, the only claim it makes is a claim on the prudential attention of those who are subject to it. If one takes the latter line one must read the law's talk of obligations, rights, permissions, powers, etc. as a smokescreen, for such categories have no place in the alternative discourse of threats and incentives.
One might sum up the two polarized alternatives here by saying, as Matthew Kramer does, that ‘morality and prudence exhaust the realm of reasons’.27 If legal obligations, rights, permissions, powers, etc. are not moral obligations, rights, permissions, powers, etc. they must, on this view, be dictates of prudence, which are merely labelled as ‘obligations’, ‘rights’, ‘permissions’, ‘powers’, etc., using all this moralistic language in a technical (and, we should add, euphemistic) legal sense.
The way to avoid both errors at once is to understand law to be making a moral claim for itself, in the sense of a claim to be made up of moral obligations, rights, permissions, and so on (i.e. obligations, rights, permissions, and so on that are not merely claimed or supposed to be such). This interpretation is consistent, as we saw, with individual, law-applying officials advancing the law's claims insincerely or in the grip of confusion. Unlike the ‘gunman’ interpretation, however, this interpretation does not require insincerity or confusion on the part of law-applying officials when they talk of obligations, rights, permissions, and so on. As a law-applying official, one might also be a sincere and clear-headed moral supporter of the law. And there are numerous options in between. The challenge, as H.L.A. Hart saw, is to explain the following facts about law:
How is this possible? Hart experimented fruitlessly with the idea that there is some belief or attitude on the part of officials that makes it possible. He most often characterized it as an attitude of acceptance, which is consistent with moral disapproval.29 His successors, including Raz and Alexy, have helped us to see the source of the fruitlessness in this line of thought. Legal officials must make a moral claim on behalf of the law, but they need not believe it and it need not reflect their own attitudes to law—even the barest acceptance of law.30 For legal officials do not, or at any rate need not, speak for themselves. Librarians advocate literacy but they may be TV-loving philistines. Recycling officers agitate to reduce waste but they may be gas-guzzling slobs. Judges make moral claims for law but they may be anarchist subversives who are trying to bring law down from the inside. Or, less racily, they may just be people who need to hold down their jobs to pay their over-ambitious mortgages. When they speak of the law as being made up of obligations, rights, permissions, powers, and so on, they speak as officials of law and it is the moral claim of law that they express, whether or not this is a claim that they themselves believe or even so much as accept.
[A]llegiance to the system may be based on many different considerations: calculations of long-term interest; disinterested interest in others; an unreflecting inherited or traditional attitude; or the mere wish to do as others do.?.?.?.?[Officials] express their sense of [law's] requirements in internal statements couched in the normative language which is common to both law and morals: ‘I (You) ought’, I (he) must; I (they) have an obligation’. Yet they are not thereby committed to a moral judgement that it is morally right to do what the law requires.28 (p.38)
C. Correctness, Justice, Authority
Raz and Alexy, as I said, converge on the thesis that law makes a moral claim. Alexy defends his version of this thesis with two famous examples, each of which is said to reveal the impossibility of law without its distinctive claim. The first example is that of a constitutional provision according to which ‘X is a sovereign, federal, and unjust republic’.31 The second is that of a judge who rules: ‘the accused is sentenced to life imprisonment, which is an incorrect interpretation of prevailing law’.32 We can all see that these statements are awkward. Alexy argues that they are also conceptually incoherent because they defy law's claim.
But do they? The claim that is defied—or rather the counterclaim that is made—by the judge in the second example is a claim about what the law has to say on a certain point, about what legal powers and obligations the judge has in sentencing an offender. As we saw already, claims to the effect that such-and-such is the law (that it is a legal power, or is a legal obligation, or a legal right, or such like) are not law's claims. Why? Because a legal power or obligation or right is no more and no less than what the law claims to be a power or obligation or right. The judge in Alexy's first example is not making a first-order claim on behalf of the law but making his own second-order claim about whatever it is that the law independently claims. It is, of course, an interesting further question whether judges and other law-applying officials necessarily claim to be applying the law, or at least (a weaker requirement) to be not defying the law in their official activities. In my (p.39) view, the answer is no. Personally I find a pragmatic but not a conceptual problem in the judicial utterance in Alexy's second example. But that is irrelevant to our topic here. For the claim in the second example is not law's claim and does not help to show the content of law's claim.
Alexy's first example is different.33 It includes, not a claim about legal obligations and so on, but a claim about their moral standing. They are claimed to be unjust. Is it intelligible for such a claim to be made on behalf of law? Alexy thinks not. He says that law necessarily claims to be just and so cannot claim to be unjust.34 He also thinks that law necessarily claims to be morally correct, and the example is also supposed to support that view. How are these two theses about the content of law's claims supposed to be related? They are certainly not equivalent. It is possible for something to be morally correct yet unjust, or vice versa. This is because, although all considerations of justice are moral considerations, not all moral considerations are considerations of justice. Not even all moral obligations are considerations of justice. There are also moral obligations of humanity, mercy, honesty, prudence, tolerance, etc. It follows that a certain rule or ruling found in the law, or indeed a whole legal system, may be morally correct but unjust, or just but morally incorrect. This being so, the law may claim to be unjust without claiming to be morally incorrect, or claim to be just without claiming to be morally correct. Indeed, the constitution imagined by Alexy may continue ‘?.?.?.?for injustice is the price we rightly pay for our tolerant, humane and merciful civilization’. Would this, according to Alexy, be conceptually incoherent? It would not, after all, defy the claim to be morally correct.35
Possibly, according to Alexy, justice is the special kind of moral correctness suited to legal systems, so that, in his view, a claim to moral correctness on the part of a legal system must resolve into a claim to justice. I think Alexy would be wrong to embrace this line of thought. The connection between law and justice is more complex.36 Even law-applying officials should sometimes sacrifice justice to mercy or prudence, and do so in the name of the law. But be that as it may, law does not make either of the moral claims that Alexy ascribes to it—either the claim to be just or the claim to be morally correct—and so it is unnecessary to consider in any (p.40) detail how it could be thought to make both claims together. To illustrate, let us compare Alexy's first example with the following, a real-life judicial comment of a kind not unfamiliar in common law systems:
What does Goff LJ claim here? He does not claim that the law is just or that it is morally correct. Indeed, he expressly says that he finds the law either ‘unjust or inappropriate’ in its application to the case before him. He does not make clear which: maybe both. But this does not matter, for he does not need to claim that the law is unjust in order to avoid claiming that it is just; nor does he need to claim that the law is morally incorrect to avoid claiming that it is morally correct. To avoid making the positive claims it is sufficient for him to claim, as he does, that the law is either unjust or morally incorrect. Is this claim inconsistent with his acting as a law-applying official? Is he a secret follower of the extrajudicial Holmes, denying the moral obligatoriness of the law?
I feel?.?.?.?that I would be lacking in candour if I were to conceal my unhappiness about the conclusion which I feel compelled to reach. In my opinion, although of course the courts of this country are bound by the doctrine of precedent, sensibly interpreted, nevertheless it would be irresponsible for judges to act as automatons, rigidly applying authorities without regard to consequences. Where therefore it appears at first sight that authority compels a judge to reach a conclusion which he senses to be unjust or inappropriate, he is, I consider, under a positive duty to examine the relevant authorities with scrupulous care to ascertain whether he can, within the limits imposed by the doctrine of precedent (always sensibly interpreted), legitimately interpret or qualify the principle expressed in the authorities to achieve the result which he perceives to be just or appropriate in the particular case. I do not disguise the fact that I have sought to perform this function in the present case.?.?.?.?I have considered anxiously whether there is any other interpretation which the court could legitimately place on Lord Diplock's statement of principle in Caldwell, which would lead to the conclusion which I would prefer to reach, that the respondent was not reckless whether the shed and contents would be destroyed by fire. I have discovered none which would not involve what I would regard as constituting, in relation to the relevant offence, an illegitimate departure from that statement of principle.37
No. For Goff LJ expressly holds himself to the standards of interpretative ‘legitimacy’ which, in his view, the law places upon him. He regards the rule previously set out by Lord Diplock as constraining him, and not merely as claiming to constrain him. In other words, he does not merely report what he takes to be his legal obligations, but holds these out as his moral obligations too. Does he really believe that they are his moral obligations? Is he a morally committed judge? Nothing in the passage tells us either way. All that we can discover from the passage are his claims. And they clearly include a moral claim on behalf of the law, viz. a claim that the law is morally binding whether or not it is just and whether or not it is morally correct. He claims that it would be ‘illegitimate’ for him to attempt to render the law just or morally correct because, as he puts it in an adjoining passage, he is ‘constrained to [not] do so by authority’.38
So we can see at once why Raz renders the moral claim of law as a claim to moral authority, not a claim to moral correctness, nor a claim to justice. It is because legal (p.41) officials often speak as Goff LJ speaks, and accept that they are morally bound by some prior exercise of the law's authority—a statute or a previous judicial decision—while challenging the justice or other moral merit of the exercise of authority in question. There is a sense, of course, in which a claim to moral authority always incorporates a claim to moral correctness. It always incorporates a claim that those subject to the authority would be acting morally correctly if they were to submit to the authority and do its bidding. Thus Goff LJ naturally represents himself as acting with moral correctness in following Lord Diplock's rule. This, however, is a red herring in the present context. For it is consistent with Goff LJ's denying that Lord Diplock acted with moral correctness in creating the rule. In other words it is consistent with denying the moral correctness of the law itself, and hence denying law's claim to correctness. Indeed, that is the point: moral authority is such that abiding by it is morally correct even though the exercise of it be morally incorrect. Authority may bind one morally to do certain things that one should never have become morally bound to do, for of any moral power it is true that its valid exercise does not depend entirely upon its correct exercise.39
This may lead us to return our attention to Alexy's first example. Why would it seem conceptually challenging, as it certainly seems to me, for a constitution to announce the moral incorrectness of the legal order that it constitutes? The answer is not that law makes a claim to moral correctness. The answer is that law makes a claim to moral authority. But the constitution, of course, contains those rules of the legal system that allocate ultimate (non-delegated, or if delegated then irrevocably delegated) authority. So a confession of law's moral incorrectness in the provisions of the constitution is quite different from a confession of law's moral incorrectness in, for example, the decision of Goff LJ. A confession of moral incorrectness in the provisions of the constitution, unlike a confession of moral incorrectness in Goff LJ's decision, contradicts law's claim to moral authority. And that, it seems to me, is its only conceptual problem. For law makes a claim to moral authority, but law makes no claim to moral correctness or to justice.
D. A Necessary Connection
That law makes a moral claim, Raz and Alexy agree, constitutes a necessary connection between law and morality. No surprise there, since there are numerous necessary connections between law and morality. Nobody has ever got close to defending the view that there are none. Alexy seems to regard ‘legal positivism’ as defending this view. But inasmuch as anyone, ‘legal positivist’ or otherwise, ever set out their view in this way, they always had to introduce caveats and provisos. In particular, they always had to point to ways in which law and morality are (p.42) related in order to get enough purchase for meaningful distinctions to be drawn between law and morality. Central to this endeavour, and of common concern to ‘legal positivists’ and their opponents alike, has been the fact that the law shares a core conceptual, or at least linguistic, apparatus with morality, the apparatus of obligations, rights, powers, and so on. This already marks a necessary connection between law and morality. The question is: how do we interpret it? Interpreting it as the making of a moral claim by law is one attractive way to make sense of it.
In two ways this interpretation of the law's apparatus is conducive to the historic project of so-called ‘legal positivists’. First, on this interpretation of legal discourse there can be immoral laws. The law claims to be moral, so it must be the case that the law can fail to be moral, for nothing is claimed unless there is logical space for it to be false. Second, for something to make claims it must, minimally, be capable of making claims. It must be a person or at least have persons (‘officials’) who act and speak on its behalf. If law is like this, that is another respect in which it differs from morality at large. Morality has no officials and cannot make claims. It is no accident that Dworkin, whose main mission is to strike out against the ‘legal positivist’ tradition, also opposes the thesis that law makes moral claims. What is not so clear is why Alexy regards this thesis as pointing in the opposite direction. He makes it part of ‘A Reply to Legal Positivism’ when, in reality, it is much more comfortably understood as part of legal positivism's reply to him.
Appendix: On Holmes’ Claims
In a previous discussion of Dworkin's views on the claims of law, I issued the warning, repeated above, that the beliefs of Oliver Wendell Holmes, the noted legal intellectual, should not be confused with the claims of Justice Holmes, the celebrated judge who sat on the United States Supreme Court.40 Dworkin rebuked me on that occasion for not citing any of the relevant judicial writings; and he ventured a guess that, were I to have cited any, they would have confirmed Holmes's adherence, even on the bench, to his famous extrajudicial stance as a sceptic about moral obligation, or at least about the moral obligatoriness of law.41 I must confess that my failure to cite came of the belief that Holmes's judicial work is universally well known, and that just about any example of it taken at random would have borne out my point. But Dworkin's view that, on the contrary, I could not have supported my point with any example of Holmes's judicial work makes me wonder whether that work really is as well known as I supposed. So, to bear my point out in some detail, I propose we consider the following sample passages, from two of Holmes's best-known dissents:
(p.43) In this case, sentences of twenty years’ imprisonment have been imposed for the publishing of two leaflets that I believe the defendants had as much right to publish as the Government has to publish the Constitution of the United States now vainly invoked by them.?.?.?.?Only the emergency that makes it immediately dangerous to leave the correction of evil counsels to time warrants making any exception to the sweeping command, ‘Congress shall make no law?.?.?.?abridging the freedom of speech.’ Of course, I am speaking only of expressions of opinion and exhortations, which were all that were uttered here, but I regret that I cannot put into more impressive words my belief that, in their conviction upon this indictment, the defendants were deprived of their rights under the Constitution of the United States.42
This case is decided upon an economic theory which a large part of the country does not entertain. If it were a question whether I agreed with that theory, I should desire to study it further and long before making up my mind. But I do not conceive that to be my duty, because I strongly believe that my agreement or disagreement has nothing to do with the right of a majority to embody their opinions in law.?.?.?.?Every opinion tends to become a law. I think that the word liberty in the Fourteenth Amendment is perverted when it is held to prevent the natural outcome of a dominant opinion, unless it can be said that a rational and fair man necessarily would admit that the statute proposed would infringe fundamental principles as they have been understood by the traditions of our people and our law. It does not need research to show that no such sweeping condemnation can be passed upon the statute before us.43
Throughout these two passages Holmes speaks of rights and obligations (a.k.a. duties), his own and other people's. Some of them he explicitly presents as rights and obligations conferred or imposed by law (see the second italicized remark in the first passage). Others he does not (see the first italicized remark in the second passage). One might insist that even when he does not explicitly present them thus, he is still referring only to legal rights and obligations, and makes no claims for them other than that they are legal rights and obligations—in particular, no moral claims for them. But this interpretation sits ill with the way in which Holmes integrates the claimed legal rights and obligations into his lines of argument: for he uses moral lines of argument in support of these claimed legal rights and obligations. Consider, for example, the first italicized remark in the first passage and the second italicized remark in the second passage. In the former Holmes relies on the ‘danger’ of some things and the ‘evil’ of others to make out his argument for the scope of the legal right to free speech under the First Amendment. In the latter, he relies on the imagined judgements of a ‘rational and fair man’ about ‘fundamental principles’ to defend his interpretation of the scope of the right to due process of law in relation to deprivations of liberty under the Fourteenth Amendment.
Some, including Dworkin in other work, read argumentative moves of this kind as breaking down the distinction between legal rights and obligations on one hand and moral rights and obligations on the other.44 But this radical reconstruction goes far beyond what is needed to make sense of what Holmes is saying. It is pos (p.44) sible to be faithful to exactly what Holmes says while preserving the possibility that Holmes is legally correct about these rights and obligations and simultaneously morally misguided about them, or morally correct and simultaneously legally misguided. Perhaps he is wrong about what the law of the Constitution says but right about what it would say if only it were morally upstanding. Or perhaps, more salient to our present purposes, he is right about what the law of the Constitution says but wrong to endow it, as he does, with moral credibility. Objections of both these types might intelligibly be advanced by a judge who sides with the majority against Holmes’ dissent. To keep these objections intelligible we should understand Holmes as making only a moral claim for the law as he presents it. He is claiming that the said legal rights and obligations (as he claims them to be) also have moral standing, that they are moral rights and obligations too. The ‘also’ and the ‘too’ here presuppose, rather than eliminating, the distinction between legal rights and obligations and their moral counterparts. They presuppose that at least some legal rights and obligations could fail to be moral rights and obligations, for (as we saw) nothing is a claim unless there is logical space for its falsity.
For completeness, we should note which moral claim it is that Holmes makes, i.e. what content he gives to the claim. Like Goff LJ, he clearly does not claim law's moral correctness. He claims only law's moral authority. He argues that his ‘agreement or disagreement has nothing to do with the right of a majority to embody their opinions in law’. Or to put it another way: where law has moral authority—perhaps because of its democratic credentials—it has the ability to create moral obligations whether or not it took the morally correct path in doing so. Of course Holmes reserves his own right, working with his fellow judges, to issue a ‘sweeping condemnation’ of the democratically created law and thereby to remove it from ‘the law’. Judges often do this, although different legal systems give them different scope to do it. Alexy, like Dworkin, tends to read this kind of judicial remark as a sign that grossly immoral rules cannot form part of the law.45 But that is not what such remarks signify. Here, as elsewhere, the remark is a sign that Holmes claims the moral authority of the Supreme Court, speaking on behalf of law itself, to review the moral authority of a democratically elected legislature. How do we know that this is what Holmes claims? We can be fairly confident about it because we can be fairly confident that, even if Holmes had felt able to issue the ‘sweeping condemnation’ called for by the final sentence of the second passage, he would not have regarded this condemnation as having legal effect—as determining the rights and obligations of the parties to the case, let alone of other courts in the United States in similar or related cases—unless his condemnation were an operative part of the ruling of a majority on the Supreme Court—unless, as well as being a sweeping moral condemnation of the law, it also met the conditions for itself being a binding authority.
(1) D. Cornell, Beyond Accommodation (1999) 122. (‘To enforce law?.?.?.?is to reinforce the male viewpoint, in spite of law's claim to do the exact opposite.’)
(2) J. Finnis, ‘The Authority of Law in the Predicament of Contemporary Social Theory’ (1984) 1 Notre Dame Journal of Law, Ethics and Public Policy 115, 120. (Law ‘presents itself as a seamless web’).
(3) P. Selznick, The Moral Commonwealth (1992) 444. (Law ‘promise[s] justice’.)
(4) J. Derrida, ‘Force of Law: the “Mystical Foundations of Authority”‘ in D. Cornell (ed), Deconstruction and the Possibility of Justice (1992) 22. (Law ‘claims to exercise itself in the name of justice’.)
(5) R. Dworkin, ‘Thirty Years On’ (2002) 115 Harvard Law Review 1655, 1665–8. Similar doubts have been expressed by K. Greenawalt, ‘What Does “the Law” Claim about Trivial and Extremely Broad Legal Norms?’ (2005) 50 American Journal of Jurisprudence 305, 307; and by K. Himma, ‘Law's Claim of Legitimate Authority’ in J. Coleman (ed), Hart's Postscript (2001) 271, 277, 279. Himma's treatment is more thorough than Dworkin's, and perhaps more deserving of a detailed response. The two authors differ in various ways (see n 17 below for an example). However Himma shares with Dworkin the view that a moral claim is made by or on behalf of the law only if there are persons (legal officials or law-subjects) who have morally favourable attitudes towards the law. So at least the first of my rejoinders to Dworkin below is equally, and equally decisively, a rejoinder to Himma.
(6) Dworkin, ‘Thirty Years On’, 1666.
(7) J. Raz, ‘Authority, Law, and Morality’ (1985) 68 The Monist 295.
(8) Dworkin, ‘Thirty Years On’, 1666.
(11) J. Raz, Practical Reason and Norms (1975) 147–8; J. Raz, ‘Legal Validity’ in J. Raz (ed), The Authority of Law (1979) 146, 154–7.
(12) Contrast Postema's statement that, for Raz, ‘self-identified participants in legal practice must believe that legal norms?.?.?.?have some sort of moral justification’. G.J. Postema, ‘Law's Autonomy and Public Practical Reason’ in R. George (ed), The Autonomy of Law (1996) 79, 84 (emphasis added). In the passage relied upon by Postema, Raz is talking about what must be ‘adduced’, not believed: J. Raz, ‘The Purity of the Pure Theory’ in R. Tur and W. Twining (eds), Essays on Kelsen (1986) 79, 92.
(13) Dworkin, ‘Thirty Years On’, 1667.
(14) Notably ‘The Path of the Law’ (1897) 10 Harvard Law Review 457.
(15) Notably Abrams v US 250 US 616, 629–31 (1919); Lochner v New York 198 US 45, 75–6 (1905). See the appendix below for further discussion.
(16) I have written about the first type in my ‘Reasons for Teamwork’ (2002) 8 Legal Theory 495. For more about the second type, see my ‘Some Types of Law’ in D. Edlin (ed), Common Law Theory (2007).
(17) In ‘Law's Claim of Legitimate Authority’, Himma seems to think that the best prospect for rescuing the idea that law makes claims lies in thinking of it along these lines, i.e. as an artificial (he says ‘fictional’) concerted agent. Here he differs from the many who have tried, with considerably less plausibility, to ascribe to the aggregated law-applying officialdom of each legal system a (more or less) orchestral kind of agency. See J. Coleman, The Practice of Principle (2001) 96–9; S. Shapiro, ‘Law, Plans, and Practical Reason’ (2002) 8 Legal Theory 387; C. Kutz, ‘The Judicial Community’ (2001) 11 Philosophical Issues 442.
(18) The connection between personhood and autonomy is, of course, widely discussed, although most discussions focus exclusively on human persons and their material autonomy. A classic example is H. Frankfurt, ‘Freedom of the Will and the Concept of a Person’ (1971) 68 Journal of Philosophy 5.
(19) Compare Dworkin's own ruminations on the agency of concerted agents, such as legislatures in Law's Empire (1986) 169–71. And note his willingness to extend the same ‘working personification’ to ‘communities’: ibid, 172–5. Here he helps himself to various moves that he later denies to Raz, including that of distinguishing ‘officials in their official capacity’ as ‘agents of the community’. If of the community, why not of law?
(20) S. Perry attributes this view to J. Finnis. See S. Perry, ‘Law and Obligation’ (2005) 50 American Journal of Jurisprudence 263, 289, relying on scattered remarks in J. Finnis, Natural Law and Natural Rights (1979).
(21) On related grounds, S. Shapiro says that the attribution of such a claim to law is ‘banal’ yet also ‘deeply paradoxical’: ‘On Hart's Way Out’ (1998) 4 Legal Theory 469, 469–740.
(22) These considerations, incidentally, begin to reveal what was wrong with the movement in English public law after Anisminic v Foreign Compensation Commission  2 AC 147 to treat all errors of law as jurisdictional, i.e. as taking an official outside her official capacity.
(23) It is what J.L. Austin calls a ‘trouser word’. Compare the word ‘real’: ‘a definite sense attaches to the assertion that something is real, a real such-and-such, only in the light of a specific way in which it might be, or might have been, not real.?.?.?.?I don't know just how to take the assertion that it's a real duck unless I know just what, on that particular occasion, the speaker has it in mind to exclude.’ J.L. Austin, Sense and Sensibilia (2nd edn, 1962) 70.
(24) Rights can be understood as grounding obligations. Permissions can be understood as cancelling obligations or as cancelling some of their obligatory force. H.L.A. Hart argued that powers, although they are logically distinct from obligations, exist to enable variation of obligations: The Concept of Law (1961) 78–9. Liabilities, in turn, are best understood in terms of powers. Does all this add up to yield the conclusion that obligations are law's ultimate concern? The obscurity of ‘ultimate’ makes it hard to be sure.
(25) I discussed recipes in more detail in ‘Nearly Natural Law’ (2007) 52 American Journal of Jurisprudence 1, 5–8.
(26) As Hart famously put it in The Concept of Law, 6–7.
(27) M. Kramer, ‘Requirements, Reasons, and Raz: Legal Positivism and Legal Duties’ (1999) 109 Ethics 375, 379.
(28) Hart, The Concept of Law, 198–9.
(30) See Raz, ‘Legal validity’, 155: officials ‘normally’ accept the law but—be that as it may—they cannot but claim to endorse it.
(31) R. Alexy, The Argument from Injustice: A Reply to Legal Positivism (2002) 36.
(33) Or is it? Sometimes Alexy explains law's claim as a claim to correctness simpliciter, not as a claim to legal correctness coupled with a claim to moral correctness. The problem with this suggestion is that all claims, or at least all claims with a propositional content, are necessarily claims to correctness. ‘I claim that P is correct’ means the same as ‘I claim that P’ in much the same way that ‘I assert that P is true’ means the same as ‘I assert that P’. Thus ascribing to law a bare claim to correctness leaves us none the wiser regarding the content of the claim—the relevant P—except that it is propositional. I am assuming in the text that Alexy is interested in explaining something about the content of law's claim(s). See J. Raz, ‘The Argument from Justice, or How Not to Reply to Legal Positivism’ in G. Pavlakos (ed), Law, Rights, and Justice: The Legal Philosophy of Robert Alexy (2007) 17, 29–30. Raz goes further than I do and suggests that all intentional actions trivially include the generalized claim to correctness.
(34) Another writer who believes that law claims to be just is P. Soper. See his ‘Law's Normative Claims’ in George, The Autonomy of Law, 215, 247. Soper also criticizes Raz's alternative proposal.
(36) I have examined some aspects of it in ‘The Virtue of Justice and the Character of Law’ (2000) 53 Current Legal Problems 1.
(37) Elliott v C  2 All ER 1005, 1010, 1012.
(39) The moral power to promise illustrates the same point. A promise to do something positively immoral is not morally binding. But a promise to do something stupid or vapid usually is morally binding, even though it is not only an act that should never have been done apart from the promise, but also an act, the doing of which should never have been promised.
(40) Gardner, ‘Law's Aims in Law's Empire’ in S. Hershovitz (ed), Exploring Law's Empire: The Jurisprudence of Ronald Dworkin (2006) 207, 215.
(42) Holmes J Abrams v US 250 US 616, 629–31 (1919). Emphasis mine.
(43) Holmes J Lochner v New York 198 US 45, 75–6 (1905). Emphasis mine.
(44) This was an implication of Dworkin's very first argument against Hart in ‘The Model of Rules’ (1967) 35 University of Chicago Law Review 14. The argument was designed to establish that there is no test (and we must therefore assume, no criterion) for distinguishing legal norms from non-legal norms.
(45) Alexy, The Argument from Injustice, 28. I have not dwelt on this thesis of Alexy's, which I regard as conflicting with his other views.