i) Pragmatist Commitments

In Dworkin’s debate with Rorty, he refers to his own views as commonsense, non-metaphysical claims, and, as noted above, this fierce disagreement with Rorty about metaphysics is in many ways mysterious, given that they seem to agree that metaphysical speculation is unproductive. I now want to unpack this more carefully. I will start by providing a brief outline of some pragmatist commitments,^{Footnote 5} against the background of which we will be able to see both their similarities and their differences more starkly.

The foundational idea of pragmatism is that truth is connected to human experience. Peirce’s pragmatic maxim is as follows: “Consider what effects, that might conceivably have practical bearings, we conceive the object of our conception to have. Then, our conception of these effects is the whole of our conception of the object.”^{Footnote 6} James’s more famous statement is: “Whenever a dispute is serious, we ought to be able to show some practical difference that must follow from one side or the other’s being right.”^{Footnote 7} We gain our knowledge through experience, through the real world as people engage with it, not via a priori reasoning.^{Footnote 8} Hence, pragmatists are skeptical about metaphysical claims that cannot be investigated through our experience and that don’t make a difference to the world as we experience it. Because truth is connected to experience, and there may always be more experiences that could shed additional light on a matter, we can never say that whatever the community happens to believe is in fact true—we can rarely say anything is true with any certainty. This fallibilism is a further fundamental aspect of the pragmatist outlook. We can say that the best view on the evidence we currently have points to X; X is therefore reasonable to believe. But we can’t know for sure if we have ever gotten to the truth; it is a continual process of discovery. So our views remain open to revision.^{Footnote 9}

Finally, perhaps one of pragmatism’s most important commitments is its insistence that the human perspective is the only one available. “The trail of the human serpent is…over everything.”^{Footnote 10} This is James’s poetic way of expressing the idea that we inevitably interpret our experiences through a human lens. If the relevant inputs for truth are human experiences, then what we get out will necessarily be colored by a human perspective. We can’t stand outside ourselves to see how things ‘really are’. There is no archimedean point^{Footnote 11} that we can access in order to discover truth—all we can possibly aim at is what is within the grasp of human minds.^{Footnote 12} Pragmatists are thus suspicious of metaphysical absolutes, which assume that we can attain some external perspective on the way the world really is.

So, in brief, “[t]he central thought of pragmatism is that our philosophical theories must be connected to experience and practice.”^{Footnote 13} Pragmatism takes seriously the fact that we are human beings, with a necessarily limited viewpoint that we cannot transcend. Experience of the world, as mediated through our senses, is all we have, and our theories must recognize that. We can’t say that something just is essentially a certain way.^{Footnote 14} We must appeal to experience to explain why it is best understood in that way.

ii) Metaphysics

Dworkin’s main argument against Rorty and the pragmatists is that, while they claim to object to talk of the ultimate structure of reality, it is they who have created this ‘external’, metaphysical perspective. In Dworkin’s view, Rorty mistakenly distinguishes between two levels of discourse: the internal, where people talk about how things actually are, and the external, where philosophers can step outside the enterprise and make claims about these disciplines (such as law, science, or morality).^{Footnote 15} Rorty, Dworkin argues, incoherently occupies the very external level that he attempts to disavow.^{Footnote 16}

In later work, still fighting over the same issue, but putting it in a different way, Dworkin says that Rorty imagines two ‘language games’: the one we normally play, and a further, archimedean language game, which asks:

Dworkin says “pragmatists use scare-quotes and italics like confetti: They say that the bad philosophers think not just that things really exist but that they “really” or really exist, as if the quotes or italics change the sense of what is said.”^{Footnote 18} He continues: “These metaphors are meant to suggest, as it were, that the bad philosophers are claiming a new, different, metaphysically special kind of reality, reality beyond the ordinary, a new, supernatural, philosophical level of discourse. But it is only the pragmatists who, in fact, ever talk that way. They have invented their enemy or, rather, tried to invent him.”^{Footnote 19}

Humorous as Dworkin’s critique here is, he misses what the pragmatist means by emphasizing such words. The emphasis on what is ‘really’ true can be understood as indicating a claim about what is essentially or necessarily the case. The question of what entities the world really contains is precisely what ontology is about. Theorists do have debates about whether there are really tables or whether there are only particles arranged table-wise.^{Footnote 20} In trying to figure out the ‘fundamental structure of reality’, the question of whether mountains really exist is something metaphysicians might plausibly debate.

If we see the scare quotes as indicating claims about the very nature of reality, we see that there are many people—call them metaphysical realists—who do make claims phrased just like this. Importantly for present purposes, analytic legal philosophers are especially fond of such claims. Theorists like Joseph Raz, Julie Dickson, and Scott Shapiro make claims about law’s nature, or how law necessarily is.^{Footnote 21} The idea that we can know about the very essence or nature of something necessitates positing an external perspective, and arguing that from that perspective we can understand what the thing is essentially like—exactly what Rorty objects to when he criticizes theorists who speak this way. So there is a genuine difference of meaning captured by Rorty’s emphasis on the real.

In other words, Rorty’s objection to metaphysical realism is not an objection to some invented mode of doing philosophy. It is an objection to a kind of metaphysical realism that is widespread. It seems that Dworkin and Rorty misunderstood each other, and each assumed the other was committed to a mysterious ontology that ought to be rejected. But they both agree that it should be rejected.^{Footnote 22} So, if the metaphysical claims are not just nonsense invented by the pragmatist, but are something some people (including analytical legal philosophers) take themselves to be engaged in, then it seems that Dworkin and the pragmatists should be united against this metaphysical project.

Posner, another frequent pragmatist interlocutor of Dworkin’s, also argues against metaphysics, saying that the pragmatism he favors “dislikes distinctions that make no practical difference—in other words, dislikes “metaphysics””.^{Footnote 23} Specifically with respect to ontological questions about law, he says that it “has no nature, no essence”.^{Footnote 24}

As suggested above, this opposition to metaphysics is a core idea in pragmatism. It is worth quoting Peirce’s delightful way of putting the point: “[M]etaphysics is a subject much more curious than useful, the knowledge of which, like that of a sunken reef, serves chiefly to enable us to keep clear of it”.^{Footnote 25} Pragmatists of both strands resist metaphysical approaches that suggest that something could be true not just because our experience shows it to be, but because the thing in itself simply is a certain way, in its essence. Once we allow that something can be a certain way just in and of itself, without exhibiting any outward signs that experience could attend to, we are talking nonsense. Peirce expounds this through the example of the belief in transubstantiation. All we can mean by wine is that which has a particular effect on our senses. Thus, to hold that the wafer-cakes and wine are really something else, though they have all the outward qualities of wafer-cakes and wine “is senseless jargon”.^{Footnote 26} The idea of a thing just is, for the pragmatist, the idea of the effects it produces in us, and so we cannot speak of an essence of something that is not reflected in any difference in our experience of it. Views without any practical upshot are not eligible truth-candidates.

So although Dworkin thinks he is objecting to the pragmatists, they are really both objecting to metaphysics. It seems that on this important issue, Dworkin and his main pragmatist interlocutors were actually in agreement. But there is something about which Dworkin genuinely disagreed with both Rorty and Posner, and the next section turns to that issue.

iii) Skepticism & Relativism

Here is where the differences between the two strands within pragmatism become clear. Pragmatism’s close association of truth and experience raises the worry that truth cannot be objective; there is no standard beyond our subjective experiences. Some pragmatists, such as Rorty, embrace this and suggest that we should be relativists about truth.^{Footnote 27} For Rorty, there is no one way things really are, there are simply various descriptions of the world, which are more or less useful for different purposes;^{Footnote 28} the real question is not what is true, but what our peers will “let us get away with saying.”^{Footnote 29} This approach finds its ancestor in James. While James’s formulation of the pragmatic maxim is similar to Peirce’s, there is a significant difference between the two: “His way of distinguishing himself from Peirce is to highlight the consequence a belief might make to a particular individual.”^{Footnote 30} This generates a theory of truth that is much more subjective: by including whatever makes a difference to any individual, it opens the door to a great deal of problematically subjective experiences. James’s view, in short, is that beliefs are true if they ‘pay’—if they work for us in the very broad sense that can include satisfying our desires. Posner too belongs on the relativist side of things. He argues that truth is a ‘problematic concept,’ and that “[t]he pragmatist’s real interest is not in truth at all but in belief justified by social need.”^{Footnote 31}

On the other hand, Peirce’s brand of pragmatism is decidedly anti-skeptical. He insists that we must start from where we find ourselves; we cannot doubt everything all at once. We must never claim to be too certain, but we cannot doubt everything—universal, Cartesian doubt is unpragmatist in denying a huge swathe of our everyday experiences. For Peirce, there must be “real and living doubt”.^{Footnote 32} It is not that anything is in principle immune from doubt—we must remember our fallibilism—but that we cannot doubt everything at once. Starting from this anti-skeptical position, and holding onto the commitment to prioritizing experiences that are accessible to all, Peirce builds a more objective theory of truth. On Peirce’s view, there is an objective truth out there toward which we aim. Seeking truth and settled beliefs is a process he terms “Inquiry”.^{Footnote 33} Inquiry aims at settling beliefs, and we obviously prefer true beliefs, as they are more reliable for getting us what we want.

These same skeptical and relativistic tendencies manifest with respect to questions of moral truth. Both Posner and Rorty believe that there is no objective truth about moral value. They argue that, given this, we should simply push and hope for change on the basis of ‘human needs’ or ‘desires’. Rorty has at times described himself as a relativist or as an anti-foundationalist.^{Footnote 34} For Rorty, the idea “of ‘clarifying our unconditional moral obligations’” is objectionable to pragmatists, because it presupposes “the existence of something nonrelational, something exempt from the vicissitudes of time and history, something unaffected by changing human interests and needs.”^{Footnote 35} Rorty argues that we should understand moral progress not as a matter of transcending appearance to access reality, but as “a matter of wider and wider sympathy”,^{Footnote 36} or greater sensitivity.^{Footnote 37} He often says things like “A liberal democratic utopia, on the pragmatists’ view, is no truer to human nature or the demands of an ahistorical moral law than is a fascist tyranny. But it is much more likely to produce greater human happiness.”^{Footnote 38}

Posner is declaredly a relativist about moral truth.^{Footnote 39} He argues that there are “rudimentary principles of social cooperation”,^{Footnote 40} but these are too abstract to be genuine moral standards. “If moral relativism means that the criteria for pronouncing a moral claim valid are local, that is, are relative to the moral code of the particular culture in which the claim is advanced, so that we cannot call another culture ‘immoral’ unless we add ‘by our lights,’ then I am a moral relativist.”^{Footnote 41}

The relativism and skepticism widely associated with pragmatism has seemed unsatisfactory to many people: there must be something more that can be said about a description than that it is preferable or useful, or that our community won’t challenge it. In the next section I will argue that, for this very reason, Dworkin found the skeptical brand of pragmatism unappealing, but can be understood as a Peircean pragmatist.

iv) Dworkin as an Anti-Skeptical, Peircean Pragmatist

We have seen that there are two camps within the pragmatist tradition, and that while Dworkin agreed with Rorty in his opposition to metaphysics (though he failed to realize this), they did have a genuine disagreement about skepticism, objectivity, and truth. We should put Dworkin in the Peircean camp. And indeed, this is validated by some of Dworkin’s last thinking on the matter. In Justice for Hedgehogs, he engages directly with the philosophical nature of truth, and in the process he endorses Peirce’s view, albeit in a somewhat limited way.

Dworkin starts by understanding truth in relation to other ideas, such as inquiry. Truth is the aim of inquiry—we always have it as our goal when we inquire.^{Footnote 42} It is worth noting immediately the similarities here to Peirce, for whom inquiry and truth are deeply interwoven.^{Footnote 43}

Dworkin claims that we can make sense of arguments about truth by understanding truth as an interpretive concept.^{Footnote 44} We treat truth and its related concepts (such as authenticity, accuracy, belief, inquiry, sincerity, etc.^{Footnote 45}) as values. Therefore our theories of truth should be understood as attempts to interpret this practice of truth-seeking and see it in its best light.^{Footnote 46} For example, we would reinterpret a given theory of truth as a way of weaving together all our truth-related concepts so that it makes interpretive sense to treat truth as that theory demands.

Dworkin thinks that the right approach is to develop a high-level, abstract concept of truth that would lead to different accounts of truth appropriate for different areas of inquiry.^{Footnote 47} Different domains, such as morality and science, have such different methods of justification that if we take a monolithic approach and say that one theory exhausts the concept of truth, we will have to conclude that some things are not truth-apt, in particular morality. This leads to the skeptical position Dworkin rejects. His strategy allows us to tailor our theories of mathematical, moral, and other truths to their respective subject matters. The abstract theory of truth would need to show us why all these different standards still count as standards of truth, and it would have to fit and justify our practices around truth and inquiry.^{Footnote 48}

Here, in proposing his “tentative and incomplete suggestion” for such a theory, is where Dworkin engages directly with Peirce’s views:

Different accounts of truth for different domains would then flow from different ideas of success. Dworkin continues: “These sketchy remarks are reminiscent, at least, of much of what Charles Saunders Peirce said about truth.”^{Footnote 50}

However, Dworkin objects to Peirce’s position, arguing that “we must not say, as Peirce once said, that truth is always or just what enables us to satisfy some desire we have.”^{Footnote 51} It is here that we can see the impact of years of debating with Rorty and Posner on Dworkin’s conception of pragmatism; he seems to have internalized the more skeptical view that those thinkers adopt. Peirce said and believed no such thing; he does not hold Rorty’s ‘truth is whatever works’ picture.^{Footnote 52} Peirce’s claim is that there is an objective truth, and that it is only by conforming to that truth that we will get beliefs that will help us satisfy our aims. Otherwise we will bump up against the hard reality that Peirce agrees with Dworkin is out there.

Peirce defines ‘the real’ as “that whose characters are independent of what anybody may think them to be.”^{Footnote 53} And what is real influences our beliefs. We seek true beliefs through the scientific method, which Peirce believes will result in us converging on an answer if inquiry is pushed far enough.^{Footnote 54} This leads to his definition of truth: “The opinion which is fated to be ultimately agreed to by all who investigate, is what we mean by the truth, and the object represented in this opinion is the real.”^{Footnote 55}

When Peirce says that truth “is distinguished from falsehood simply by this, that if acted on it should, on full consideration, carry us to the point we aim at and not astray”,^{Footnote 56} we can see the seeds of the mainstream understanding of pragmatism. It sounds as though it rests truth on what someone might find useful to believe. But it is not right to say that Peirce holds a view according to which truth is a matter of desire-satisfaction. He identifies a strong connection between the truth and what works for us, but he articulates it thus: “It is certainly best for us that our beliefs should be such as may truly guide our actions so as to satisfy our desires; and this reflection will make us reject every belief which does not seem to have been so formed as to insure this result.”^{Footnote 57} Hence, what Peirce is saying is that it is good for our beliefs to “work”—we want beliefs that serve us in our various aims—but whether or not they will work is constrained by reality. This is very different from saying that whatever happens to “work” is in fact true.

Dworkin is hesitant to fully embrace Peirce, and thinks pragmatism has to operate only in a more abstract role, helping us to choose our theory of truth for different domains.^{Footnote 58} But he need not be so tentative in his endorsement. The better way of understanding Peirce is that he holds a view very like Dworkin’s and very much opposed to Rorty’s. Dworkin is right that Rorty’s pragmatism would tell us that truth is whatever seems good to us. But when we apply Peirce’s, the story is different. Peirce’s view agrees with Dworkin’s: truth is not about what is delightful or useful. It is about, as Dworkin puts it, “the uniquely successful solution to a challenge of inquiry.”^{Footnote 59} We find that solution by testing what works, what guides our actions given the conditions of the external world.

Dworkin’s real concern is that adopting pragmatism will lead to skepticism, especially in the scientific realm, so he thinks he cannot fully sign on to Peirce’s account. But Peirce’s pragmatism is less skeptical than Dworkin seems to think, and Dworkin should be willing to adopt it. In light of Dworkin’s hesitance to fully embrace a Peircean pragmatist stance, and also because his comments about Peirce were limited and came at the end of a long career, I want to now spend a little time showing that they were not mere throwaway comments, and that a careful reading of Dworkin’s views throughout his career, paired with a thoughtful understanding of Peirce, validates placing Dworkin in the Peircean pragmatist camp.

Dworkin has long argued against skepticism, distinguishing between two kinds.^{Footnote 60} There is internal skepticism, which must be internal to morality—it makes claims about the way things morally are (that there is in fact no right answer), but these are positive claims that must still rely on an anti-skeptical view further down. This is at least a coherent position, though one Dworkin argues against in the case of morality. He rejects even the possibility of the alternative—external, archimedean skepticism.^{Footnote 61} The only question on the table is which moral views are right, not whether any can be. Any claim that there is no right answer or that there is no ‘truth of the matter’ about what is right has to be understood as a substantive claim about what we should believe, morally speaking. It must be backed by a moral argument that there is no truth of the matter. It cannot be an external moral claim about what is the case with respect to moral claims generally.

An important point to understand about Dworkin’s argument here is that it is not only an argument against the external skeptic; it is also an argument against the external realist.^{Footnote 62} He objects to everyone who grounds morality on some sort of external data, either to deny its truth, or to say that we can access it. This is important: he’s denying the metaphysics of realism, not embracing moral realism. He is a moral realist in the sense that he believes there are objective truths about value, but not in the commonly understood metaethical sense of one who thinks that moral truth can be grounded by something in the world, something other than value claims. His argument distilled is this: “Philosophy can neither impeach nor validate any value judgment while standing wholly outside that judgment’s domain.”^{Footnote 63} “There is no neutral scientific or metaphysical plane on which we can stand finally to adjudicate which of different views about equal concern or liberty or democracy or any other opinion about right or wrong or good or bad is the best or true one.”^{Footnote 64}

Dworkin’s insistence that there is no archimedean point from which to access truth is precisely the pragmatist’s view. We cannot be external skeptics, because this means stepping outside experience to talk about what can ground truth. Dworkin and the pragmatists agree that we must drop the metaphysics that most philosophers think underpins moral theory—the metaethical disagreement. Moral skepticism, then, can only be internal, and it is then itself a moral position. It makes claims that compete with other substantive moral claims. The claim that there are no reasons is itself a moral claim.^{Footnote 65} Dworkin’s view that there is no escape from the making of moral arguments, using the experiential data that we are given, embodies several pragmatist impulses—the idea that truth must be responsive to experience, the anti-archimedeanism, and the anti-skeptical stance that insists that we must start where we find ourselves.

Part of the anti-skeptical stance shared by Dworkin and Peirce is the idea that we cannot question everything at once, and so we must take for granted, at least in the course of inquiry and argument, some set of views, in order to examine a particular subset. This is Peirce’s idea of ‘regulative assumptions.’^{Footnote 66} The thought is that we need to have a belief that a truth exists, and we must proceed on the basis of what we take to be true, keeping in the background the possibility that we might be proven wrong. At one point, Dworkin makes a very similar point. He argues that truth is so important in the political realm that we cannot be value skeptics: “Politics is coercive: we cannot stand up to our responsibility as governors or citizens unless we suppose that the moral and other principles on which we act or vote are objectively true.”^{Footnote 67} It is easy to bristle at this argument: just because it would be bad if there were no objective values doesn’t mean there are any. But we must be careful: Dworkin is talking about the philosophical attitude we should take. This is a point about how humans think, not about what there is. We cannot be effective citizens unless we do take for granted certain things. We cannot help but suppose that there is a truth about value. Dworkin makes this kind of claim in other works, too. In Law’s Empire he uses the famous metaphor about repairing a boat: “We can only inspect and reform our settled views the way sailors repair a boat at sea one plank at a time, in Otto Neurath’s happy image. We must hold constant certain parts of our attitudes and convictions about law, as not under present study, in order to evaluate and refine the rest.”^{Footnote 68} This reads like a restatement of Peirce’s idea of regulative assumptions: we don’t actually conclude that such assumptions are true; rather, since we cannot doubt everything all at once, we hold certain things steady while we examine other parts of our thought.

Dworkin has long been insistent about the objectivity of both moral and non-moral truths. One way of understanding objectivity is as a sort of externalist, god’s-eye view that can see how the world really is, freed of our subjective perspectives. But if Dworkin means this kind of objectivity, it will conflict with his anti-archimedeanism, which, as discussed above, is the cornerstone of his extensive treatments of truth and objectivity over the years. So it cannot be this sort of objectivity that he insists on. Dworkin’s idea of objectivity is better understood as the view that truth is independent of what people think about the matter. What is wrong does not depend on our views. When we say our moral views are objectively true, we mean to say that “[t]hey would still be true…even if no one but me thought them true—even, indeed, if I didn’t think them true.”^{Footnote 69} He says that people who use the term ‘objectively’ when talking about their moral views are not positing some external realm from which we can determine objective truth, but are merely clarifying their opinions, distinguishing them from other opinions which they regard as subjective, or just a matter of taste.^{Footnote 70} It is a way of emphasizing that the thing in question is “just plain wrong, not wrong only because people think it is.”^{Footnote 71} Dworkin’s objectivity preserves our ability to say, for example, that slavery was wrong even when widely practiced. But we make that claim from an internal, human perspective, not from a position of metaphysical objectivity. Dworkin’s objectivity does not depend on a god’s-eye view. It is not about a perspective beyond the human that the pragmatist would insist is inaccessible to us.

Similarly, Peirce seeks a method of settling belief that is determined by something external to us, “something which affects, or might affect, every man.”^{Footnote 72} This is one way in which Peirce distinguishes himself from James, who wanted to allow private experiences, such as divine revelation or mystical experiences, into the relevant category of experience on which we build our beliefs.^{Footnote 73} Peirce was a strong believer in the scientific method, and in his view, this was its core demand: we must appeal to things accessible to all in answering our questions.^{Footnote 74} In this sense, Peirce is deeply committed to a principle of objectivity. His method of fixing belief, he says, is the only one of those he considers that actually has an idea of right and wrong, a sense of objectivity.^{Footnote 75}

Though objectivity of this sort implies that it is not any one person’s opinion that determines what is true, it is still in an important way tethered to the human perspective. “[R]eality is independent, not necessarily of thought in general, but only of what you or I or any finite number of men may think about it…though the object of the final opinion depends on what that opinion is, yet what that opinion is does not depend on what you or I or any man thinks.”^{Footnote 76} In other words, truth is still a matter of how humans see the world, but any given human may be mistaken. What we aim at is not some god’s-eye picture of the world—we could not possibly aim at that—but the best, most well-informed, human understanding of the object under consideration. In this way, pragmatism makes truth something that is at least in principle accessible to us, though we may never reach it on a given topic in our lifetimes, or even in the lifetime of the human race.

For Dworkin, the truth of a moral claim “is a matter of moral judgment and argument.”^{Footnote 77} He notes that most philosophers deny this: they think the truth of moral propositions is a metaphysical matter, about the presence or absence of “chimerical properties or entities ‘in the world’”.^{Footnote 78} We talk in an ordinary, commonsense way as though there are moral facts, he says. But when philosophers take these comments about facts to indicate some further metaphysical assertion, they invent “entirely bogus philosophical projects.”^{Footnote 79}

Again, we should note Dworkin’s denunciation of metaphysics. These ‘bogus philosophical projects’ are the projects of the metaphysical realists, who insist we can uncover the nature of morality (or law, or tables). What Dworkin wants to defend is what he calls the “ordinary view” of moral truth: that moral opinions can be objectively true.^{Footnote 80} Just as with his objections to Rorty mentioned in the opening of the paper, he insists he is simply talking about truth in an ordinary sense, not a metaphysical sense. He aims to validate what seems incontrovertibly true: that there are moral truths. And he wants to do so without metaphysics. This is a Peircean pragmatist project.

Dworkin’s view of moral truth, in brief, is something like this. Moral claims are made true by good arguments for their truth. We are justified in thinking a moral judgment is true “when we are justified in thinking that our arguments for holding it true are adequate arguments.”^{Footnote 81} This might all sound circular, he says, but it is no more so than science, in its reliance on the scientific method, a standard internal to science.^{Footnote 82} All we can do is aim at a comprehensive account, and argue responsibly. The only guidelines we can provide for seeking moral truth are about the way in which we form our beliefs. We must act with moral responsibility.^{Footnote 83} This shares a deep structure with Peirce’s argument in “The Fixation of Belief”, where Peirce focuses on different methods of establishing belief, as opposed to the content of the belief in question.^{Footnote 84}

Dworkin’s approach doesn’t guarantee truth. “But when we find our arguments adequate, after that kind of comprehensive reflection, we have earned the right to live by them. What stops us, then, from claiming that we are certain they are true? Only our sense, confirmed by wide experience, that better interpretive arguments may be found.”^{Footnote 85} This is exactly Peirce’s message about the distinction between truth and certainty. We can aim at objective truth, but we can never be certain that we have reached it. “The most that can be maintained is, that we seek for a belief that we shall think to be true.”^{Footnote 86} This may sound like a disappointing redirection of our efforts, but careful reflection will show that it is not a redirection at all, but a more accurate description of what we in fact do: “nothing out of the sphere of our knowledge can be our object”,^{Footnote 87} so it is plain that, given the limitations of our human minds, the most we can seek is what we understand to be true.^{Footnote 88}

For pragmatists, truth is connected to experience, and humans are the arbiters of what is true. But the claim is not that whatever humans happen to believe is in fact true. This is a subtle point, but an important one to get right. It is that there is nothing outside of us that could be the judge: we must assess what’s right and wrong, true and false, from within the human perspective. We might get it wrong for a while, perhaps forever.^{Footnote 89} But that is the crucial point: we might be wrong. The claim is not that whatever we happen to believe is true for that reason.^{Footnote 90} It is important to distinguish between truth and what is reasonable to believe. What is reasonable to believe at a given time, on the basis of the evidence so far accumulated, is relative to that evidence, and can change over time. But what is true does not change. Our only test of whether a view is wrong is if it fails to stand up to the human experiences we have so far collected. But this does not make truth relative.

So, we have seen that Dworkin rejects pragmatism in his debates with Rorty and Posner, but he rejects it only because he equates it with two claims that are not obviously core tenets of pragmatism. First, the idea that it entails positing a higher metaphysical plane from which to make claims about the nature of reality. Second, a strong skepticism about truth. The former is clearly no part of any pragmatist view. The latter is part of only one tradition within pragmatist thought. Dworkin’s association of skepticism with pragmatism generally is what explains his hostility to it. It is this skepticism within Rorty and Posner’s strand of pragmatism that is the target of Dworkin’s forceful objections.

But I have argued that Dworkin’s views share a great deal with pragmatism, in particular Peircean pragmatism. I think it is plausible to call Dworkin a Peircean pragmatist.^{Footnote 91} The Peircean pragmatist shares his objections to both the ‘gods-eye view’ and to skepticism about truth. As I have shown, Dworkin is deeply committed to taking experience seriously. Experience is the only thing to which we can appeal in deciding what is true or right. But that is the crucial point: we appeal to it in search of truth, not to vindicate our own preferred stance, as the more relativistic pragmatist might think.

I have also tried to demonstrate that Dworkin’s pragmatist commitments are not just passing remarks, but are deeply held views, and are part of his position throughout his career. These same ideas were fundamental in his other disagreements, with analytic legal philosophers. Once we see the centrality of his anti-metaphysical views and his anti-archimedeanism, we can understand how foundational his disagreement with analytic legal positivism really was. I explore this in more depth in the final section.

i) Legal Pragmatism

Posner thinks there are a number of consequences that flow from accepting a pragmatist approach. Some of these I am either in agreement with or ambivalent about. For example, Posner believes that pragmatism encourages us to permit free speech so as not to foreclose inquiry on a subject—it suggests that we should maintain a marketplace of ideas.^{Footnote 92} And he says it encourages us to be critical of “mysterious entities” in law, such as intent, free will, and causation.^{Footnote 93} A pragmatic jurisprudence encourages us to redescribe these ideas in less mysterious ways. Neither of these are particularly objectionable conclusions, but nor are they distinctly pragmatist.

However, there is one main upshot that Posner argues is a result of his pragmatism which I strongly reject. This is his particular attitude to adjudication, which he calls pragmatic adjudication. Pragmatic adjudication counsels us to prioritize economic and utilitarian calculation in deciding cases.^{Footnote 94} Posner says that we should reason backwards, from what works, what is socially good or efficient, not from what rule already exists and whether it is analogous to the current situation.^{Footnote 95} This applies to statutory and constitutional interpretation too; pragmatists are not interested in whether something is the ‘real’ interpretation of the framers’ views. “They are interested in using the legislative or constitutional text as a resource in the fashioning of a pragmatically attractive result.”^{Footnote 96} Posner argues further that there is a connection between pragmatism and economic analysis of law.^{Footnote 97}

The first point I want to make here is that, whatever we think about the merits of Posner’s view, it simply doesn’t follow from a Peircean pragmatist account of truth. One might think that metaphysics is out, and objectivity and truth are matters of human inquiry—one might adopt a wholly Peircean pragmatist stance—and yet think that the upshot of that inquiry is that, in the context of deciding cases, consequentialism and economic analysis have no priority as a way of looking at the decision at hand. So Dworkin as I have interpreted him need not adopt a pragmatic approach to adjudication.

But we can go further than that: even on the skeptical pragmatist view, it is not obvious that pragmatic adjudication of this sort follows. Indeed, Dworkin has criticized this position as internally incoherent. Posner tries to substitute consequential reasoning for moral reasoning when that itself is another form of moral reasoning, and relies on moral judgments. As Dworkin says, “Lawyers and judges must appeal to (or in any case make assumptions about) moral or political principles in order to decide whether the projected consequences of one decision are better than those of another.”^{Footnote 98} Both Rorty and Posner make the mistake of inconsistently embracing both skepticism (there is no moral law to appeal to) and realism (we should do what produces the most happiness).^{Footnote 99} Dworkin argues that Posner repeatedly appeals to moral argument in making his claims, undermining his claim that morality should be rejected: “[T]hough he aims to show that both ordinary people and judges can dispense with moral theory, his own arguments again and again fall back on just such theory.”^{Footnote 100} Similar points can be made with respect to Rorty. There is no escape from making substantive moral claims. Thus, this attitude of skepticism supplemented by hope for the future and some form of consequentialist reasoning is inconsistent.

Posner’s legal pragmatism—with its insistence on consequentialist reasoning without any foundation of moral truth—is unstable and untenable. Further, it is pitched at a very different level than philosophical pragmatism as a theory of truth. Whatever we think about Posner’s pragmatic adjudication, it has no necessary relationship to philosophical pragmatism. Peircean pragmatism does not commit us to legal pragmatism, as understood by Posner, nor indeed to any particular claims about how judges ought to decide cases. It is a theory about how we establish truth, but it is open to any number of different discoveries about what is, in fact, true. Thus we cannot say in advance that any one view of what should be done flows from this framework.

ii) Eliminativism

A further upshot for legal philosophy is the rejection or reformulation of certain questions.^{Footnote 101} I want to suggest that Dworkin’s unique outlook on legal philosophy was a result of his pragmatism, and that much of the debate between him and other legal philosophers can be understood as stemming from different ideas about what questions are answerable.

The most substantial change—or what was perceived as a change, though I will argue it was continuous with his earlier views—Dworkin introduces in Justice for Hedgehogs is the claim that we have to abandon what he calls the ‘two-systems view’, where law and morality are understood as separate realms, in favor of the ‘one-system view’, where law is a branch of morality. Dworkin argues that on the two-systems view, “there is no neutral standpoint from which the connections between these supposedly separate systems can be adjudicated.”^{Footnote 102} Whether we treat the question of the connection between law and morality as a legal question or a moral one, we will beg the question in one direction or the other, by supposing that morality is in or out of the material properly appealed to. To understand this objection, we must keep in mind his arguments about truth and morality. There is no metaphysical plane we can access to see what law really, intrinsically is like, so any standpoint we take to assess it must be a human one. As Dworkin points out, the problem is that if we try to treat it as a legal question, we can’t answer it without building in positivism or non-positivism—we can’t interpret the legal material without a theory in hand about whether morality is in or out. We must either suppose that the material properly appealed to includes principles as well as pedigreed rules, or that it doesn’t. And either way, we will beg the question in favor of positivism or nonpositivism. But turning to morality also means begging the question in the opposite way. Thus, Dworkin concludes, “The two-systems picture therefore faces an apparently insoluble problem: it poses a question that cannot be answered other than by assuming an answer from the start.”^{Footnote 103}

How should we understand this ‘shift’ to the one-system view? Given his opposition to metaphysics, it seems clear that we should not read this as making a claim about the true essence of law or legal rights. It is not the metaphysical question Dworkin is concerned with. Indeed, the whole point of his anti-archimedeanism is to deny that such questions are answerable: there is no way to step outside of ourselves and see what law really is.

Yet Dworkin has long insisted that there are right answers about what the law is, and it would certainly seem implausible to suggest that he is denying that such truths exist. The key is to see that these truths aren’t metaphysical truths—they aren’t truths that demand an archimedean perspective. They are truths accessible from within our human perspective. The whole problem with the ‘two-systems’ view is that it requires an archimedean perspective that is impossible. So the point Dworkin is making in saying we must adopt the one-system view is that we have to frame our questions about law as questions that are answerable.

What questions does the pragmatist think are answerable? Peirce, like Dworkin, holds the view that morality is truth-apt. So there are questions of what we should do, morally speaking. And there are questions that we can understand as legal. Again, Dworkin would not want to abandon his long-held view that judges seek a right answer. They want to know what the law demands of them.

But we should think carefully about how he now frames the question of what the law demands. On the Justice for Hedgehogs view, he argues that law just is that which is properly enforced by courts.^{Footnote 105} ‘Properly’ is important here: the inquiry is a moral one. What should a judge do, given our complex political history? What matters here is the making of moral arguments in the course of adjudicating a case. The moral question arises in a particular factual context, and those concrete historical and political facts influence the answer to what is the morally right action, so it is not a matter of doing morality from first principles.^{Footnote 106} But the inquiry is ultimately a moral one. The question of what is morally right in this particular political circumstance is still, at bottom, a question of what is morally right.

The question which Dworkin thought of as the reformed, one-system version of the question of the doctrinal concept of law, i.e., “Under what conditions do people acquire genuine rights and duties that are enforceable on demand in the way described?”^{Footnote 107} remains an interesting and viable question, and, as Dworkin argues, a question of normative political theory. Because it depends on answering the question of what genuine normative rights we have, it is a question of normative theory.

But not everyone has thought that legal philosophy is a matter of such internal, normative questions. Many legal philosophers see the debates in legal philosophy as questions about what law really is—as metaphysical questions. They say that they are concerned with the very nature of law, with its necessary features, with its essence.^{Footnote 108} We should recall Dworkin’s dismissiveness of the idea of truth that exists “out there” as part of the “fabric” of the universe.^{Footnote 109} He says that people do not make such claims, that the skeptic invents these terms as a straw man. But theorists who argue about law’s nature, about how law necessarily is, really are making claims of this sort. They make claims about the very essence of law. This demands an archimedean perspective, and is a claim Dworkin, with his pragmatism, should best be seen as denouncing. The theory of truth shared by Dworkin and the pragmatists thus seems to suggest a rejection of the idea that we are searching for the essential nature of law. Indeed, his position in Justice for Hedgehogs does seem to reject such questions.

In other words, there are no ‘taxonomical’ questions, which is one way Dworkin characterizes the concerns of analytic positivists.^{Footnote 110} They are occupied with trying to draw a precise line around what properly counts as law. This, I think, is a good way to think about the analytic legal philosophers’ project: the idea that there is something that ‘properly’ or ‘really’ counts as law, divorced from experience or from normative concerns, assumes the possibility of correctly mirroring an independent reality, as the metaphysical realist attempts to do.

If we are to take the pragmatic attitude seriously, we cannot understand any of Dworkin’s claims—such as that we have to adopt the one-system view, or that law must be understood through the interpretive methodology—as assertions that seeing things that way somehow reflects reality as it is in its essential nature. It is not that we need to do normative inquiry to get at the essence of law.

Rather, the pragmatist methodology sees that debates about essences are unresolvable, and counsels us to reject them. When Dworkin makes claims about the one-system view, these are not claims about the essence of law, but suggestions for what we can do instead. He is making the point that normative questions are the kind of thing human inquiry, broadly construed, can in principle resolve. So we replace metaphysical questions with moral questions, some of which can be helpfully seen as ‘legal’, though they are still just a subset of what is really a moral inquiry. We are answering the moral questions that stem from the particular political circumstances in which we find ourselves.

This begins to look like what some have recently called “eliminativism”. As Liam Murphy puts it, eliminativism is the view that we need not answer the question of ‘what the law is’. Law talk “plays no important role in legal practice and social life generally—it is a wheel spinning on its own. We can get on perfectly well by discussing a range of other questions.”^{Footnote 111} Others understand eliminativism as the view that there is no “distinctively legal domain of normativity, or even quasi-normativity.”^{Footnote 112} Scott Hershovitz adds that his view is “ontologically spare”,^{Footnote 113} and refers to it as “a kind of eliminativism, since it denies the existence of an entity…that more traditional pictures presuppose.”^{Footnote 114} Mark Greenberg also puts forward a view that can be understood as eliminativist, arguing that “[t]he content of law is that part of the moral profile created by the actions of legal institutions in the legally proper way.”^{Footnote 115} In other words, the law just is the moral upshot of the political circumstances; moral inquiry is therefore primary.

Several writers have suggested that in his later work Dworkin shifted to a more eliminativist view. Greenberg mentions in a footnote that Dworkin’s Justice for Hedgehogs position could be read as “a version of the Moral Impact Theory that restricts legal rights and obligations to those that should be enforced by courts.”^{Footnote 116} Jeremy Waldron has recently given a similar interpretation of Dworkin’s position: we skip any intermediate ‘what is the law?’ step and move directly to our moral obligations.^{Footnote 117} He argues that Justice for Hedgehogs is more radical than the Justice in Robes position, “because Dworkin’s position now is not just that arguments and judgments in legal theory are moral in character, but that legal judgments are moral judgments and legal arguments are moral arguments.”^{Footnote 118} The judge is faced with a moral question: what should I do? Morality is really the fundamental issue here. Dworkin, on this interpretation, seems to take the radical view that there are only moral questions to be answered.

As I suggested above, however, Dworkin does not want to let go of the idea that there is something called law, and we can strive towards right answers about what the law is. This may not amount to eliminativism as some people understand it, since it doesn’t involve abandoning talk of the law. But it does entail rejecting metaphysical talk about law, and eliminating the idea that there is a metaphysical entity called ‘law’, with an essential nature we can discover. We let go of metaphysical questions and address those remaining questions that are answerable. As Dworkin puts it: “The substance of the old confrontation between positivism and interpretivism would remain, but…in a political rather than conceptual form.”^{Footnote 119} The remaining questions are moral questions, a subset of which can be understood as legal questions. If eliminativism demands that we give up on being able to identify a certain subset of our moral obligations as legal ones, then Dworkin is not an eliminativist. But if it instead asks us to reject the invented metaphysical entity of law, in favor of moral questions, some of which are ‘legal’ questions, then Dworkin is an eliminativist. I am not concerned with whether Dworkin really merits the label ‘eliminativist’. What is interesting is that he has long insisted that something that has been treated as a fundamental question in legal philosophy—the idea of searching for law’s metaphysical nature—has to be rejected.

And the fact that this is a long-standing conviction of Dworkin’s is important. I think the one-system view is not a new position for Dworkin, but just a clearer elaboration of views he had long held, but struggled to articulate in an environment fixated on conceptual analysis of the nature of law. Dworkin himself recognizes the problem when he says that he mistakenly tried to present his ideas in terms of the two-systems view.^{Footnote 120} However, both Greenberg and Hershovitz believe that this is a recent shift.^{Footnote 121} Hershovitz believes that “for most of his career, Dworkin was buzzing around the fly-bottle with the rest of us.”^{Footnote 122} I think this is wrong; though Dworkin may have occasionally put things in a misleading way, his impulse was always towards the anti-metaphysical, quasi-eliminativist view. As early as 1977 he said “My point was not that ‘the law’ contains a fixed number of standards, some of which are rules and others principles. Indeed, I want to oppose the idea that ‘the law’ is a fixed set of standards of any sort.”^{Footnote 123} The idea, all along, was to deny the possibility of describing law as a metaphysical entity whose nature could be captured. His denial wasn’t only of the descriptiveness of legal philosophy, but of the idea that there is any fixed entity to describe. What is interesting is how long it took for people to see the radicalness of the view; the fact that the later work is interpreted as a shift indicates the extent of the misunderstanding of Dworkin throughout his career.

We can recognize Dworkin’s eliminativism underlying earlier fights with pragmatism, too; ironically, Dworkin sounds most pragmatist and most eliminativist when he is arguing against the pragmatists. Take this quote from 1991, twenty years earlier than the ‘radical’, ‘new’, view in Justice for Hedgehogs:

This is Dworkin arguing for the abandonment of any ‘externalist’ questions—metaethical questions about the nature of morality, or metaphysical questions about the nature of law—which require an archimedean perspective that is inaccessible. We answer, instead, the genuine moral questions we are faced with. The view in Justice for Hedgehogs, whether or not it properly counts as eliminativism, is consistent with Dworkin’s pragmatism, and a coherent continuation of his earlier thoughts, not a radical departure.

What I want to emphasize here is the deep level at which his pragmatism operates, and the way it drives much of what he said in his debates with other legal philosophers. It has often been a struggle for people to even engage with Dworkin’s views, because they are so profoundly different from the prevailing preoccupations of legal philosophy. There has therefore been a tendency to understand him as entering the same metaphysical debate, or to see him as engaged in the limited project of developing a theory of adjudication.^{Footnote 125}

Seeing his pragmatist commitments helps us to understand just how different Dworkin really was. He is not doing either of these two things. Rather, he has all along been calling for a different kind of philosophy: a kind that denies that we can have an archimedean perspective, and does away with any questions that demand it.