A. Argument One

Here's the first argument:

We think there's something to this argument; and indeed, we think it's especially promising when we keep in mind, as we noted above, that there are different models for what it would take for a domain to be unified: it could be a genus, be a determinable, form an interrelated network, have an analogical structure, and so on.

But we still think the argument is far too quick. For one thing, the mere fact that we taxonomize reality in a certain way does not entail that the categories we use are in any deep sense “joint-carving” or unified. Consider Jorge Luis Borges's taxonomy of animals in the Celestial Emporium of Benevolent Knowledge: those belonging to the Emperor, embalmed ones, trained ones, suckling pigs, mermaids, fabled ones, stray dogs, those included in this classification, those that tremble as if they were mad, innumerable ones, those drawn with a very fine camel hair brush, miscellaneous, those that have just broken the vase, and those that from afar look like flies.Footnote ⁴¹ We can coin names for the categories in this taxonomy. But we don't think that suffices to make them unified in any interesting or important sense. So to demonstrate that legal reality is unified requires work: one can't suppose from the onset that it must be unified simply because it's part of a taxonomy we use. After all, we could use the taxonomy for reasons other than it picking out a deeply unified part of reality.

For another thing, as we noted earlier, on some models the unity of a domain is not binary, but gradable, and if we think of unity gradably this first argument has even less force. Perhaps something must keep legal reality unified to some degree for it to form a category at all. But the degree of unity that entails may be minimal, and perfectly consistent with the possibility that (for example) positivism is true of some part of legal reality and antipositivism is true of some other part of legal reality. After all, there is obviously something that the category of animals included in Borges's classification all have in common (namely, all being included in the classification). So there is some minimal sense in which that category is unified. But that degree of “unity” is minimal, and can hardly be taken to reflect any deep metaphysical similarities between the things in that category.

B. Argument Two

Here's the second argument:

Unlike the first argument, this argument doesn't insist that the fact that it's all “legal” thought and talk guarantees anything about the unity of legal reality. Instead, it aims to establish that unity by starting with different links between different parts of legal reality, and then generalizing from there.

We are sympathetic to the idea that there will be connections between some different parts of legal reality. For example, with Shapiro in Legality, we think there is strong pressure to think that what makes something a legal norm is tied to what makes something a legal institution.Footnote ⁴² We also think, along with many others involved in the debate over positivism, that there is strong pressure to think that there are important explanatory links between facts about the nature and grounds of legal norms and legal obligations, rights, permissions, etc.Footnote ⁴³ But, even if there are indeed such connections between some parts of legal reality, we think this argument moves too quickly. For one thing, one must defend the auxiliary theses that link the different parts of legal reality. Such theses can't just be assumed. They need to be argued for. For another, such theses may link some portions of legal reality (legal institutions to legal norms, for example), but not all portions of legal reality (legal norms to the behavior of legal officials, for example). In other words, the broader the scope of “legal reality,” the more ambitious one's set of auxiliary theses would need to be to link it all together, and the less plausible it is that we can assume some complete set of theses like that from the get-go in theorizing about legal reality.

C. Argument Three

Here's the third argument:

We think that there's something to this argument too. Indeed, one of us (Wodak) has argued that there is some pressure to think that terms like ‘obligation’ are univocal, and this generates problems for a number of views about legal thought and talk (including Toh's expressivism and the now-popular “moralized” semantics for ‘legal obligation’).Footnote ⁴⁴ But we think this argument also moves too quickly, for three reasons.

First, the costs of giving up on univocality depend on which alternative is offered: appealing to polysemy and/or the idea of “metalinguistic negotiation” may capture much of what makes univocality an attractive thesis.Footnote ⁴⁵ To put this otherwise, we think there are good reasons to resist certain ways of denying the univocality of legal terms; but those reasons may not require us to assume univocality.

Second, many philosophers of law are not well placed to offer this argument about univocality—at least not across the board for all of our legal terms, let alone all of our central ones. Consider Raz's and Shapiro's views about legal language. On their views, what does ‘legal’ mean? That is, what does it contribute to the meaning of a phrase like ‘legal obligation’ or ‘legal institution’? There is no single answer; ‘legal’ is not univocal on their views. When ‘legal’ modifies ‘obligation’, its contribution is to take us to a perspective or point of view, which can be mistaken. On their views, that you have a legal obligation to act just means that according to a certain point of view you have a moral obligation to act, and since this point of view is fallible, you may have no obligation to act whatsoever.Footnote ⁴⁶ What has not been sufficiently appreciated, however, is that neither Raz nor Shapiro thinks ‘legal’ modifies ‘institution’ in the same way. If it did, then “there is a legal institution” would mean that there is an institution according to a certain point of view. And further, since this point of view is fallible, it would not follow from the fact that there is a legal institution that there is any institution whatsoever. This view is implausible, so it's no surprise that Raz and Shapiro don't adopt it. But in another sense, this is surprising, for it entails that ‘legal’ is at least polysemous, and that matters a great deal, for what could be more central to legal thought and talk than the term ‘legal’ itself?

Third, we're not sure that the univocality of legal terms entails the unity of legal reality, at least not in any interesting sense. The terms in Borges's category “innumerable animals” are plausibly univocal. But the category of things that the term ‘innumerable animals’ is about might only be unified to a fairly minimal degree. So while establishing the univocality of legal terms may undercut some of the pressure toward thinking that legal reality is disunified, it doesn't generate much pressure toward thinking that legal reality is unified.

D. Argument Four

Here's the fourth argument:

We have two main responses to this argument. The first is that this argument is invalid as stated. After all, people might well have rival theories about a part of a reality that is disunified. So even if one could vindicate the idea that all of legal thought and talk fits this model, this wouldn't show that legal reality is unified.

Second, we want to emphasize that the model on the table here shouldn't be taken to be a good model of all of legal thought and talk. We grant that many people (including many legal philosophers) have rival theories about the same part of legal reality. Nothing in our arguments for the prima facie plausibility of the thesis that legal reality is disunified requires rejecting that. And, indeed, we don't reject it. However, the observation that this model is correct some of the time doesn't support thinking that it must always be true. In other words: this may be a good model for some aspects of legal thought and talk, without being a good model for every aspect of legal thought and talk. To support this line of thought, consider the following. With some parts of thought and talk, there is strong pressure to think that two people are thinking and talking about the same part of reality. For example, suppose one person says that “the shoes that Nico is wearing are size 9” and another says “no, the shoes that Nico is wearing are size 12,” and they are both looking at the same person (Nico) who is wearing one pair of shoes. In this case, at least prima facie, it seems there is good reason to think these speakers are talking about the same part of reality—and doing so explicitly, via the literal content of what they say. But it's not as if every part of our thought and talk best fits that model. There is not the same kind of pressure to think that when one person says “legal institutions are X” and another says “legal content is Y” they are explicitly talking about the same thing. Moreover, it might be that in some cases the best model for disputes that (at least to some theorists, or some participants) appear to express straightforward disagreements about the same part of reality—and to do so via the literal semantic content of the utterances of what speakers say—turns out to best be one on which speakers are interpreted as talking about different things.Footnote ⁴⁷

E. Summing Up

In this section, we have canvassed four arguments in favor of the unity of legal reality. These four arguments don't exhaust the possible reasons why people might find the unity of legal reality to be an attractive hypothesis. But we think two important points are worth keeping in mind about how these arguments bear on our goals. The first is that once you need an ambitious argument to support the hypothesis that legal reality is unified, you're playing our game. Our goal, after all, is to show why the possibility of disunity is worth taking seriously. And second, the common fate of these arguments suggests that defending the unity of legal reality will require substantial work. That some might want legal reality to be unified does not absolve them from the task of vindicating this hypothesis.

A. The Positivism/Antipositivism Debate

First, the disunity of legal reality provides potentially fruitful ways of combining strands of the positivist and antipositivist traditions in general jurisprudence. Here's a simple way to put this point: if positivism and antipositivism can be true of different parts of legal reality, then they need not be in conflict with each other, as they need not be trying to occupy the same “real estate.” Positivists and antipositivists have little reason to be hostile to the possibility that legal reality is disunified—at least not solely in virtue of their commitment to positivism or antipositivism. This is easiest to see for antipositivists. If one is an antipositivist about some part of legal reality (e.g., legal obligations), one holds that facts about legal obligations depend in part on moral facts. However, one need not think this is true of all parts of legal reality (e.g., legal institutions). The same basic idea is also true of positivists. Some positivists are error theorists or skeptics about morality.Footnote ⁴⁸ As such, they may view accepting antipositivism about any part of legal reality as a cost. But many positivists are nonskeptical realists about morality (which is a stance we recommend, but won't defend here). As such, they don't face the same general pressure to resist the idea that moral facts play some role in determining some part(s) of legal reality.

How, exactly, might combining strands of the positivist and antipositivist traditions be fruitful? Consider the following illustrative example. In Legality, Shapiro argues for a positivist view about both legal institutions and legal content.Footnote ⁴⁹ Put roughly, on his Planning Theory of Law, legal institutions are a particular kind of organization involved in the activity of shared planning, and laws are the plans (or “plan-like norms”) that those institutions have. This is a form of positivism insofar as Shapiro endorses these theses: (a) that facts about both the existence of planning organizations and the existence or content of plans ultimately depend on social facts alone, and not moral facts; and (b) that which of the relevant plans are laws is independent of their moral merit. Shapiro provides an argument to link his positivist account of legal institutions and legal norms; this is why his positivist account of legal institutions helps support his positivist account of legal norms. Perhaps due to Shapiro's confidence in that argument, he defines the terms ‘positivism’ and ‘natural law theory’ (which, in Legality, he uses broadly to mean what we mean by ‘antipositivism’) as theories about legal institutions and laws. Many positivists take a different route, in which they define ‘legal positivism’ as solely about laws or legal content, and not about legal institutions. For example, this is how Gardner understands positivism in his influential account in “Legal Positivism: 5 ½ Myths,” as well as how Greenberg defines positivism and antipositivism in “How Facts Make Law.”Footnote ⁵⁰

Whichever definitions of ‘positivism’ and ‘antipositivism’ one is working with, now consider how the issues here intersect with the possibility of the disunity of legal reality. If one takes seriously the possible disunity of legal reality, we should be open to the idea that positivism and antipositivism cover some parts of legal reality, but not others. We should be open to arguments like Shapiro's that link them, but we shouldn't take the unity of these (or other) parts of legal reality for granted. That might well mean taking more seriously the possibility of a split view between legal institutions and legal content, in terms of whether positivism is true of each. For example, suppose one thought that designating something as a specifically “legal” institution was a sort of honorific, such that ‘legal’ functions partly as a kind of “thick” term akin to ‘brave’ with a mixture of normative and descriptive application conditions.Footnote ⁵¹ Depending on how one develops this idea, it might well yield the result that the existence of specifically “legal” institutions is grounded in moral facts (e.g., facts about the moral achievements of those institutions, that warrant use of the honorific), as well as social facts. This idea could then be combined with the further thought that whatever norms those institutions produce count as the law, which would yield antipositivism across the board. But one could also think that “legal institutions” are just whatever institutions play a certain kind of functional role in society (whether they deserve an honorific or not), but that then calling certain things produced by those institutions the “law” involves the use of a kind of honorific (e.g., reserving it only for those norms that meet a certain standard of moral merit). In such a case, we might then end up with a kind of positivism about legal institutions, but a kind of antipositivism about legal content.

Perhaps more interestingly, take discussions of “legal content” in particular. On one way of thinking about legal content, it consists of the full set of legal norms (i.e., laws) that obtain in a given jurisdiction, at a given time.Footnote ⁵² On another way, it consists of the full set of (general) legal obligations, rights, privileges, etc. that obtain in a given jurisdiction, at a given time.Footnote ⁵³ There are many reasons to think these two parts of legal reality that one might use the term ‘legal content’ to pick out are closely linked. After all, one might think that for them to be genuinely legal obligations that obtain in a given jurisdiction (at a given time), these must in some way be tied to facts about legal norms (as opposed to moral norms, epistemological norms, etc.), such as by being grounded in the existence of those norms.Footnote ⁵⁴ But still, prima facie, these seem to be conceptually distinct parts of legal reality. So taking the disunity of legal reality seriously should leave us open to the idea that they are less closely linked than initially appears to be the case—or perhaps linked in ways that yield quite different accounts of each of their grounds. Perhaps laws (which one thinks of as a kind of “legal norm”) are fully grounded in social facts, but legal obligations are not. Perhaps legal obligations are those genuine moral obligations that obtain because of the laws (along the lines of Greenberg's Moral Impact Theory), but laws are fully grounded in social facts.Footnote ⁵⁵ Or, with a nod to some of our earlier discussion, perhaps that is one kind of obligation some people pick out when using the term ‘legal obligation’.

If we start by taking the possible disunity of legal reality seriously, one begins to see how establishing this Greenberg-style thesis about “legal obligations” is potentially compatible with a range of core things many positivists want to say about other parts of legal reality. To see this, start by conceptually distinguishing three things: (1) what moral obligations (and rights, powers, etc.) we have in virtue of the law (or in virtue of legal institutions), (2) what moral obligations (and rights, powers, etc.) we have according to the law, and (3) how possible actions (or other things) stand in relation to the norms that make up the law, which we might describe using the terminology of ‘obligations’ (and ‘rights’ and ‘powers’) in some way other than the first two options. Most positivists tend to go for something akin to the second or third options here in thinking about our actual thought and talk about “legal obligations” (and “legal rights,” “legal powers,” “legal duties,” etc.). But does the positivist need to take a stand on which of these three options is correct, given her commitment to positivism and the core concerns that motivate her to accept positivism? Plausibly not. Suppose that a Greenberg-style view is correct and that some people sometimes—or even all of us, all of the time—use the terminology of ‘legal obligation’ to refer to what moral obligations (and rights, powers, etc.) we have in virtue of the law (or in virtue of legal institutions). By itself, that would not thereby show anything in particular about the nature (or normative status) of laws or the legal norms themselves—for example, whether they are “authoritatively” normative, or bear the same kind of connection to authoritative normativity that many take morality to have. To illustrate: we might use the term ‘fashion obligations’ to refer to the moral obligations people have in virtue of the norms of fashion, without thereby endorsing the thesis that there is some kind of intimate relationship between the norms of fashion and authoritative normativity (or morality), or anything about how normatively important the norms of fashion are. This illustrates the following: insofar as the core concerns driving positivists concern issues about the nature of laws or legal norms, it's not at all clear why they can't take on a wide range of views about the nature of a range of other things, including things picked out by our current ways of using bits of legal terminology such as ‘legal obligation’.

The broader lesson here is simple. If we look at arguments for positivism or antipositivism on their own terms (about the particular part of legal reality they concern), we might find that they are on point about that part of legal reality, but not about other parts of legal reality. This opens up space to potentially reconcile arguments and insights from a range of discussions by positivists and antipositivists. Exploring that space might of course not allow us to do much reconciliation at all, once we get down to the serious work of exploring it in detail. Our point is just that, if we take the possibility of the disunity of legal reality seriously, we have good reason to take seriously the activity of exploring that space—in terms of our thinking both about existing views (and their possible combinations) and new ones.

In some respects, this point has precursors in the existing literature. As we noted at the start of this paper, the idea that certain theories in legal philosophy might not be in direct conflict, despite initial appearances to the contrary, is something that comes up in many discussions in legal philosophy. This is certainly true in discussions about legal positivism and antipositivism. To take a salient recent example, consider the following. In “The End of Jurisprudence,” Scott Hershovitz argues against “the supposition that there is a single entity called the law to which all such talk [about what the law requires] refers”Footnote ⁵⁶ and emphasizes that “we characterize the legal domain in different ways for different purposes.”Footnote ⁵⁷ Drawing on this, Hershovitz partly aims to reconcile seemingly rivalrous theories within the debate over positivism. As he puts it, “[t]he ambition to characterize just one set of things as the law of our community requires that we cast lots with Greenberg, or Dworkin, or someone yet to come.”Footnote ⁵⁸ Without that ambition, he argues, there might be more compatibility between these theories than we initially think. These points all resonate with ideas we have been discussing in this paper.

That being said, we can appreciate the importance of the generality of our discussion by comparing it to the intervention that Hershovitz makes (as an illustrative example of similar interventions in debates within general jurisprudence). This is, as we noted in the introduction, one of the things that makes our discussion different from standard ways in which possible forms of disunity in legal reality come up in discussions in legal philosophy. To underscore this point, consider the following four differences between our discussion of disunity and Hershovitz's.

First, Hershovitz's claims are primarily directed to possible disunity within certain parts of legal thought and talk—roughly, those about legal obligations, legal rights, legal permissions etc., and associated ideas about “what the law is” tied to those. We have a broader focus, including (for example) considering thought and talk about such things as legal institutions and legal actors.

Second, Hershovitz chiefly focuses on possible disunity tied to different antipositivist theories (and ones of a certain kind). By contrast, we are emphasizing possible disunity that would result in positivism being true of some parts of legal reality and antipositivism being true of others.

Third, Hershovitz casts his claims as part of a broader, “anti-metaphysical” stance with which he aims to dissolve the positivism/antipositivism dispute. Our discussion of disunity does not require such a stance. We focus on the “object-level” of legal reality (which is compatible with more or less metaphysically “inflationary” ways of understanding what it consists in), and do not seek to dissolve metaphysical debates between positivists and antipositivists.

Fourth, important parts of Hershovitz's discussion of disunity in “The End of Jurisprudence” involve normative claims about how we should think or talk, such as in abandoning talk about “legal content.”Footnote ⁵⁹ Our focus is on a descriptive question about the parts of reality that actual legal thought and talk is distinctively about, rather than normative questions about what our thought and talk should be like.Footnote ⁶⁰

B. Situating the Positivist/Antipositivist Dispute Within Legal Theory More Broadly

We now want to suggest that the thesis that law is disunified offers fruitful ways of reconsidering the relationship between, on the one hand, the debate over positivism and antipositivism, and, on the other, debates within other important traditions in legal theory. The other traditions we'll focus on are legal realism and critical legal theory.

We'll start by exploring how the key insights of these debates can be fruitfully reconciled. As we have discussed, there are ways of viewing positivists and antipositivists as being concerned with different parts of legal reality. But insofar as they are concerned with the same part of legal reality, it most plausibly includes legal content. And insofar as questions about the nature and grounds of legal content are key questions in the debate over positivism, we think there's much to be learned from that debate about the nature of legal norms (where “norms” are understood in a broad sense to include all standards to which things can conform or not).Footnote ⁶¹

We also think there's a way of viewing the core tenets of legal realism and critical legal theory such that they intersect with, but do not compete with, these insights. Here's a tentative sketch of this interpretation. Legal realism and critical legal theory can be construed as ultimately concerned with the extent to which law operates as the kind of rule-governed activity that it purports to be (or which at least some people take it to be, or which at least certain legal officials present it as being). Legal realists hold that the legal norms (or at least, the “paper rules” from familiar sources such as statutes and precedents) often massively fail to determine legal outcomes.Footnote ⁶² Critical legal theorists also accept widespread legal indeterminacy, though for a different reason: they hold that the legal norms conflict on so fundamental a level that these norms offer no coherent answer in many important legal cases.Footnote ⁶³ These are obviously rough glosses of core ideas from rich and nuanced intellectual traditions. But, we think, they help get across some ideas that animate each tradition, and set up our discussion.

How do these core ideas intersect with positivism and antipositivism? Consider the question: What explains legal outcomes? That is, what explains which judicial decisions are in fact made, which then alter our legal rights, duties, powers, etc.? There's no single answer according to legal realists or critical legal theorists, obviously, but theorists within both traditions offer views on which there is a significant departure from how law is presented (i.e., judges learning about and applying legal rules determines legal outcomes) and how law operates (i.e., legal outcomes are explained by something very different from judges learning about and applying those rules, such as judges’ political ideologies, social biases, or personal idiosyncrasies).

We think this discussion reveals two points of intersection. The more obvious one is whether and to what extent law is indeterminate. Prominent figures in the positivism/antipositivism debate—especially Hart and Dworkin—have taken stances on the extent to which law is indeterminate. So too have many in the literature in jurisprudence around law and language (such as, for example, Hrafn Asgiersson in recent work).Footnote ⁶⁴ But a crucial point is that endorsing neither positivism nor antipositivism as such commits one to a view on this issue, or associated issues about adjudication. After all, just by saying that the law is (or is not) grounded in certain kinds of facts doesn't itself settle how determinate those facts are—or even commit one to anything about how determinate the grounding facts are (e.g., how determinate moral facts are). Furthermore, as many positivists have noted, positivism is not best construed as a theory of legal adjudication.Footnote ⁶⁵ Neither is antipositivism, when it is understood as a theory of what ultimately grounds the facts about what the law is (in a given jurisdiction, at a given time). To illustrate, consider Dworkin's influential antipositivist view of the nature of law in Law's Empire. On that view, put roughly, facts about what the law is (in a given jurisdiction, at a given time) are grounded in facts about what best justifies the full set of relevant legal practices.Footnote ⁶⁶ That view is arguably consistent with the kind of indeterminacy that critical legal theorists defend. Dworkin did not embrace such indeterminacy. But Dworkin's reasons go beyond his core antipositivist view of legal content that we just glossed.Footnote ⁶⁷

We want to suggest an additional, less obvious point of intersection: whether and to what extent the way law operates departs from its manifest image. This framing hinges on thinking that positivists’ and antipositivists’ insights about the nature of legal content—and tied to that, the nature of legal norms—may be compatible with legal realists’ and critical legal theorists’ insights about how those legal norms operate (especially in adjudication), given what legal actors actually do with those norms. We think this point of intersection matters a great deal. It is deeply theoretically and politically important to know whether, and how, the way legal norms operate differs from how they are made manifest.Footnote ⁶⁸ And in one sense, it shouldn't be too surprising to suggest this point of intersection, since both realists and critical theorists are often in the business of arguing that how things operate is very different from how we take them to operate. But we want to emphasize one way in which taking legal reality to be disunified opens up an attractive way to accept this intersection. If we accept or are open to the hypothesis that legal reality is disunified, then these traditions need not compete for the same real estate. We need not assume that (say) if how law operates departs from its manifest image, then its manifest image is chimerical. Instead, we can think that what the legal norms are like is one part of legal reality, and how the norms operate is another. Each tradition is analyzing something real. And this allows the distinct traditions to be complementary, rather than competing.

Opening up space for these traditions to be potentially complementary can also thereby help us avoid bad arguments. For example, it can help us avoid arguments that take a form of antipositivism to be defeated solely in virtue of knocking down a disconnected, optional thesis (e.g., views about the nature and content of morality, or about the extent of legal indeterminacy) that is also endorsed by the relevant philosopher who endorses that form of antipositivism. Appreciating the space here can also help us avoid arguments that tacitly slide between different uses of the term ‘the law’ (or other legal terms) that are (at least plausibly) used to refer to different parts of legal reality.Footnote ⁶⁹

We also think this way of framing the relationship between positivism and antipositivism, on the one hand, and the ideas animating legal realism and critical legal theory, on the other, can help us identify potential background disagreements about what matters in legal reality. For many philosophers working in general jurisprudence, positivism and antipositivism take center stage. But if these theories concern only one part of legal reality, an important question becomes salient: Why focus mostly or solely on these questions about that part of legal reality? Two obvious reasons would concern explanatory importance and normative importance. But, for either, we think the importance of these questions about that part of legal reality might be context-specific. For example, for most people who interact with the law, the question of whether legal outcomes are rule-governed may seem more explanatorily and normatively important than the question of what the content of the law really is, let alone what determines that content. Whether legal outcomes are “rule-governed” is more relevant to understanding and predicting what will happen to them or their loved ones as they navigate a legal system that can protect or threaten their interests.

With this in mind, consider Oliver Wendell Holmes's views in “The Path of the Law” that Hart famously criticizes in The Concept of Law.Footnote ⁷⁰ Hart criticizes the view (which he attributes to Holmes) on which claims about what the law is are identified with claims predicting the behavior of legal officials. Hart gives us strong reasons to think that this view can't provide a good account of what legal content is; if this view were true, then when judges deliberate about what the law entails in a given case they are engaged in a complicated prediction about their own behavior (along with the behavior of other legal officials). That fits poorly with all sorts of data about the activity of legal officials. As Hart discusses, the view also fits poorly with how many ordinary people orient themselves to the law when they take the content of the law to have a kind of normative significance in guiding their behavior.Footnote ⁷¹ These points suggest that Holmes's view does not capture all important parts of legal reality. But, it might be a promising account of some aspect of legal reality, and one that ordinary people often care a great deal about. To take a mundane example, someone trying to figure out how to navigate real estate law when building a home might consult a lawyer because she cares a lot more about what is likely to happen with her home—given how judges are likely to rule, what legal consequences she is likely to face, and so on. Knowing what the law actually is might (at most) be instrumentally important in such contexts, insofar as the facts about the behavior of legal officials might well in part be determined by facts about what the law actually is (and their forming correct beliefs about that, and then acting based on those beliefs). But whether it is instrumentally important in this way can't be settled from the armchair. If that is right, then it might well be that Holmes's theory is helping illuminate one aspect of legal reality, and an important one at that, even if it fails as an account of other parts of legal reality for exactly the reasons Hart articulated.Footnote ⁷²

So far, we've been suggesting that there can be ways in which positivist or antipositivist views about “the law” and views about “the law” from legal realists and critical legal theorists might well be less in conflict than one might initially think, since they might well be best understood as about different parts of legal reality. But we want to close by noting a few ways in which these views might still come into conflict, even if they are indeed about different parts of legal reality.

First, as we noted above, there might well be reasons for some philosophers defending a form of positivism or antipositivism to take on board certain commitments about the core issues that animate discussions about legal indeterminacy or other issues that matter to legal realists and critical legal theorists. For example, this could be because the best defense of the form of positivism someone favors relies on a premise about the extent of legal indeterminacy, or has implications for that issue. The same kind of point is also possible in reverse: a certain kind of legal realism may turn out to rely on a commitment to positivism.Footnote ⁷³

Second, it's possible that, as a matter of sociological practice, an extensive focus on one issue might be a way of “crowding out” reflection on a second issue. Thus, for example, it might well be that extensive focus on the nature of legal content may, in certain contexts, crowd out reflection on facts about the way legal norms operate, which in turn might (in an ideologically objectionable way) occlude people grasping their actual operation.Footnote ⁷⁴ The same possibility could be true in reverse, of course, where an extensive focus on the operation of legal norms crowds out consideration of the nature of legal norms. Such “crowding out” might well play a certain kind of “ideological” function in certain contexts by helping to stabilize an image of legal reality that obscures important aspects of legal reality.

Third, although different theorists might be focused on different parts of legal reality, they might have those focuses because of background disagreements. To illustrate, suppose Anand focuses on theorizing about topic A, and Beatrice focuses on theorizing about B, and each one does so because each one thinks that their preferred topic is the one that is more explanatorily important for answering such-and-such question. They then have a background disagreement: namely, what's more explanatorily important for the question at hand (which, we can note, might be either a descriptive or normative question). Such background disagreements about the relative explanatory or normative importance of different parts of legal reality might drive more disagreements in legal theory than legal theorists standardly appreciate.

Furthermore, this kind of background disagreement about relative normative or explanatory importance we introduced above might be implicitly part of many disputes in legal theory, even if it is not part of the literal content of what legal theorists say. Consider people making claims about what X “really is” or what “real Xs” are, even if both sides agree that each one uses ‘X’ to refer to something that exists (e.g., each uses ‘law’ to refer to some part of legal reality). That kind of dispute might, in certain cases, be best understood as a form of “metalinguistic negotiation,” wherein speakers appear to use (rather than mention) a given term to put forward a view in “conceptual ethics” about how that term should be used in a given context.Footnote ⁷⁵ These conflicting normative views about the word in question might well be tied to different views about which properties we should focus on in a given context (each picked out by a conflicting use of that term), which in turn might be based on different views about the relative explanatory or normative importance of one property vs. others. In this way, claims about what “law really is” might (in certain contexts) be best understood as indirect ways of expressing disagreements about which aspects of legal reality have a given salient feature, such as which aspect is more explanatorily or normatively important.Footnote ⁷⁶ If we focus explicitly on those disagreements, we might better locate the disagreements that underwrite conflicts in theorizing about the law, and then work to make progress on them.