Skip to navigation – Site map

The possibility of naturalistic jurisprudence

Legal positivism and natural law theory revisited
Dan Priel


Contemporary legal philosophy is predominantly anti-naturalistic. This is true of natural law theory, but also, more surprisingly, of legal positivism. Several prominent legal philosophers have in fact argued that the kind of questions that legal philosophers are interested in cannot be naturalized, such that a naturalistic legal philosophy is something of a contradiction in terms. Against the dominant view I argue that there are arguable naturalistic versions of both legal positivism and natural law. Much of the essay is dedicated to showing that such views are possible: I identify naturalistic versions of a “natural law” view, a “positivist” view, as well as a “semi positivist” view, all of which are variants of the familiar (anti-naturalistic) views defended under these labels. I also offer a tentative argument in support of a naturalistic positivist view, one that has more in common with the views of Thomas Hobbes and Jeremy Bentham than with the anti-naturalistic positivist views popular these days.

Top of page

Full text

1 Introduction

1Legal positivism and natural law remain much-used tools in the legal scholar’s toolbox. Among other things, they stand for competing ways of thinking about legal reasoning, about the foundations of political authority, and about the existence of necessary connections between law and morality. However, when contemporary legal philosophers, or more precisely when analytic legal philosophers from the English-speaking world, look at the distinction these days, they consider this unregimented use of the terms a big mess. To understand legal positivism, they say, we must identify a “distinctive proposition” (Gardner 2012: 19). Following an intellectual path that cannot be retraced here (but is discussed in Priel 2015) legal philosophers seem to have settled on a very narrow understanding of the distinction between legal positivism and natural law, one primarily concerned with competing views about the conditions of legal validity. So understood, however, it turned out to be a distinction without a difference, since many natural lawyers do not disagree with legal positivists on the question of legal validity. This essay seeks to show one (out of several) possible understandings of the relationship between legal positivism and natural law that has the virtue of being both better in tune with the history of the field and also has the potential to inject new life to a debate has gone stale. It does that not by tinkering at the edges of existing definitions but by suggesting a new way of thinking about this distinction, one that reinterprets the two positions as naturalistic theses. On the basis of this discussion I will then sketch in a very perfunctory way the approach I believe is the right one for explaining the normativity of law. More than a full-blown argument some of the positions I advance here are a promissory note for a view on the foundations of law’s authority that requires more than space here allows. Before getting there, I will need to make the case for such an approach, one that in the context of contemporary debates is not obvious.

2 Two sorts of jurisprudential anti-naturalism

2In one sense the question whether the study of law should be “naturalized” may seem odd, for the right question should be the opposite one, namely why it should not. “When we study law we are not studying a mystery but a well known profession,” Holmes (1897: 457) famously said. On this view, law is ultimately humans doing things in the world; and humans, including their actions and attitudes, are all part of nature. So to say that an account of law should be “naturalistic” should not sound too controversial; one would think it would be almost universally embraced. A naturalistic theory of law should not be more controversial than a naturalistic theory of football.

3Nevertheless, in contemporary jurisprudence naturalism remains a minority view. In part, this may be because “naturalism” has been used to refer to many different things (Flanagan 2006), which I will not attempt to survey. For the most part, however, the cool reception for naturalism is not the product of a careful examination of various naturalistic ideas. Instead, we are often presented with a flat denial of the very possibility of the idea: “For the understanding of [the normative aspect of law] the methodology of the empirical sciences is useless” (Hart 1983: 13). Though highly compressed, these words contain the key to the perceived problem with any attempt at a naturalistic jurisprudence. Law is not just humans with certain attitudes acting in certain ways; it is also a normative practice, one that tells people what they ought to do. And no naturalistic account can provide a satisfactory explanation of normativity.

4Closer examination shows we can distinguish between two different reasons for rejecting naturalism in jurisprudence. One view, which I will call “weak jurisprudential anti-naturalism” (or just “weak anti-naturalism”) sees the task of jurisprudence as one of explaining an existing human practice, but denies that all aspects of this practice are amenable to a naturalistic explanation. This view can be distinguished from a stronger anti-naturalistic stance according to which legal philosophy is not concerned with explaining a human practice, but engaged in a completely non-empirical (“conceptual”) inquiry into the nature of law (Priel unpublished a).

5This distinction can be presented somewhat more formally if we consider the following two propositions:

(i) An account of (the nature of) law is an account of an existing human practice.

(ii) There can be a naturalistic account of all aspects of (the nature of) law.

6A jurisprudential naturalist accepts both (i) and (ii). A weak jurisprudential anti-naturalist accepts (i) but rejects (ii). A strong jurisprudential anti-naturalist rejects both. As I will attempt now to show, this distinction helps clarifying the existing jurisprudential terrain. Hart, I will argue, was a weak anti-naturalist; by contrast, some of those who claim to adopt his approach to jurisprudence in fact embrace the stronger anti-naturalistic stance.

7To say that Hart was an anti-naturalist may not be obvious. He was, after all, working within the tradition of English empiricism, and more specifically reviving a theory of law that he claimed had its origins in the thought of the arch-naturalist Jeremy Bentham (Hart 1982: 28). And he expressed obvious distaste for theories that relied on what he called “obscure metaphysics” (Hart 2012: 188). Naturalism, one might say, should have come naturally to him. It is thus no surprise that some commentators consider Hart a naturalist (Leiter 2009: 198; 2014: 952).

8Nevertheless, I wish to challenge this view. Hart’s methodology was, as he explicitly acknowledged, hermeneutic (in plainer English, “interpretivist”) in nature. And as he quite explicitly stated, he saw this approach as anti-naturalistic. As he put it, for understanding normative practices, “what is needed is a ‘hermeneutic’ method which involves portraying rule-governed behaviour as it appears to its participants” (Hart 1983: 13). He thought that the methods of science cannot deal with such issues, which is why, as already quoted, he stated unequivocally that the methods of “the empirical sciences” are “useless” for explaining law’s normativity. This, of course, was not a marginal point for Hart. He alludes here to the idea of “the internal point of view,” which he saw the key to resolving the most fundamental questions of jurisprudence (Hart 1983: 13–14; Hart 2012: vi, 88).

9Hart’s overall picture is thus rather ambivalent toward naturalism (Priel 2013: 302–04). The distinction between weak and strong versions of anti-naturalism can help explain this ambivalence. Hart wanted to explain law as a human practice. An explanation of the “concept” of law was an explanation in terms of human attitudes, and it is in this sense that he thought that his account was a form of sociology (Hart 2012: vi). At the same time he argues against a psychological reading of his views (Hart 2012: 88, 139–40) and insisted that it was impossible to explain the normativity of law from an “external” perspective. Hart believed that such a perspective “may very nearly reproduce the way in which the rules function in the lives of certain members of [a] group,” those who treat legal rules as potential threats, but it “cannot reproduce…the way in which the rules function as rules in the lives of those who normally are the majority of society” (Hart 2012: 90). For them legal rules are not merely predictions of likely outcomes, but a reason for action. And reasons, Hart thought, could not be explained by the methods of science.

10Though he does not spell out the argument, it seems that his thinking goes like this: There is a categorical distinction between reasons and causes, and between the two kinds of explanation of human action that they are based on. This in itself, however, does not explain why a reason-based explanation cannot be “external.” Hart added to this distinction the further point that the “reflective critical attitude” (Hart 2012: 57) of those who accept a practice as a rule is something that an external perspective cannot explain. Crucially, Hart did not think this was just a temporary problem, reflecting the limitations of psychological science of his day. He thought that this limit was fundamental, as it reflected a logical distinction between causal explanation and reason-based explanation. At one point in The Concept of Law Hart briefly considered causal explanations that sought to identify a link “being fed and nurtured in certain ways” and the establishment of a certain moral code (Hart 2012: 193). Though not without value, Hart insisted that such explanations were of limited significance for his inquiry: “Connections of this sort between natural conditions and systems of rules are not mediated by reasons; for they do not relate the existence of certain rules to the conscious aims or purposes of those whose rules they are” (Hart 2012: 194, emphasis in original). Explanations of this kind are “for sociology or psychology /…/ to establish by the methods of generalization and theory[.] Such connections therefore are of a different kind from those which relate the content of certain legal and moral rules to” facts like human vulnerability and the approximate equality of humans (Hart 2012: 194, emphases added; also Hart 2012: 291). Such views that draw a sharp divide between two kinds of explanation of human behavior, one naturalistic (causal) and the other anti-naturalistic (reason-based), were popular around the time Hart was writing these words, especially at Oxford (Priel 2011a: 79–81).

  • 1 One might try to distinguish between the normativity of law and the normativity of other social pra (...)

11What is the scope of this view? It is sometimes said that “law is a rather unique normative system /…/ in that the norms of law are typically products of human creation” (Marmor 2011: 2; also Ehrenberg 2016: 4), which would suggest that this problem is rather unique to law. But the claim is mistaken. Our lives are saturated with norms and normative systems that are humanly created: There is the vast world of social norms and customs; games are normative; and above all, language is a normative system if there is one (it is wrong to call a tree a “bear,” it is wrong to say “I is going”). Yet all these are the products of human creation. It is actually the non-humanly created normative systems (if such exist) that are “unique.” Hart shared this view. He did not think the limitations of the methods of empirical science were limited to law; he clearly believed it was true for all “rule-governed behaviour” (Hart 1983: 13; Hart 1982: 149–50).1

12It seems that in addition to the distinction between causes and reasons (but possibly related to it) Hart was also impressed by the way norms “appear” from a first-personal perspective, in a manner that cannot be captured by any third-person (“external”) explanation. Hart can be seen as accepting Dennett’s claim that “[t]here is no such thing as first-person science” (Dennett 2001: 230), but drawing from it the opposite conclusion to Dennett’s. Dennett’s conclusion was that in order to give a scientific explanation of the first-person perspective, we must be able to translate it into third-person language (as Dennett put it: “if you want to have a science of consciousness, it will have to be a third-person science”). Hart, like other followers of hermeneutic explanations of human action, seems to have thought that because there is no third-person perspective on normativity that will fully explain it, science is “useless” as a means of explaining normativity. The kind of argument I suggest Hart is groping for has become far more familiar in the philosophy of mind, but one finds is also in contexts closer to the one Hart was interested in (Baker 1998, 2011; Kim 2010: chs. 5–6).

13I think this is the straightforward reading of Hart’s position. His words written in 1983 cannot be said to be a careless remark later repudiated. On the contrary, they are part of a retrospective reflection and an attempt by Hart to identify the central themes in his work and its intellectual commitments. By then, he no longer had the excuse he may have had in the late 1950s that the cognitive revolution was still in its inception. Hart’s remarks fit his lifelong indebtedness to what he called “linguistic philosophy,” which was conceived by its practitioners (especially at Oxford) in opposition to naturalism, especially of the Quinean kind.

14Against all this Leiter has suggested that hermeneutic explanations can be naturalized (2007: 173–75). If true, this can be seen as a way of salvaging Hart’s naturalistic credentials, but whether or not this is a viable position, it is evident that Hart himself resisted it. He saw this as the fundamental error committed by those who have thought they could give an “exhaustive” account of terms like right and duty in naturalistic terms (Hart 1955: 248). At most, Hart could be interpreted as an advocate of the view sometimes called “liberal naturalism.” This view shares with stricter forms of naturalism the notion that an account of normativity is “nonsupernatural,” but at the same time insists that “normative items do not fall under laws of nature” and, more positively, that the “space of reasons and values” is something that sciences cannot explain. Instead, this view suggests that these aspects of human life are the product of “maturation, learning a language, and enculturation” (Macarthur 2010: 127–28). Anyone familiar with Hart’s work will easily recognize it in these words. Whether we call this view “liberal naturalism,” “second-nature naturalism” or “weak anti-naturalism” is a pointless verbal dispute, but it is quite clear that the view is one that is at odds with the kind of naturalism that Leiter wants to see more of in legal philosophy, one that is avowedly continuous with the natural sciences.

  • 2 In Priel (unpublished a) I draw the distinction between internalists and externalists, which is dif (...)

15It is true, and worth noting, that in much contemporary legal philosophy the rejection of naturalism goes further and its proponents adopt an even stronger anti-naturalist stance.2 But even though I readily acknowledge that Hart’s anti-naturalistic tendencies are weaker than those of some of his present-day followers, the dominance of anti-naturalism in contemporary legal philosophy (which Leiter recognizes and laments) owes much to Hart, and not just to developments that came after him. Given Hart’s extraordinary influence on contemporary legal philosophy, it is hard to see how this could not be the case.

  • 3 It is not just that Hart called his book an essay of “descriptive sociology” (Hart 2012: vi); Hart (...)
  • 4 As mentioned in the text, this view owes much more to Kelsen than to Hart, as Kelsen’s so-called “p (...)

16The difference between Hart’s anti-naturalism and its stronger counterpart can be seen when comparing his ideas to those of Joseph Raz, who attributed Hart’s “failure” to his “naturalism and adherence to empiricist epistemology” (Raz 2009a: 52). The difference between the two versions of anti-naturalism can be put as follows: For weak anti-naturalists like Hart an account of law is at bottom a sociological one; what counts as law is fixed by what people think is the case,3 but the proper method for sociology is fundamentally different from those used by natural scientists. The weak anti-naturalist position is thus a general methodological thesis that is part of the philosophy of social explanation. The strong anti-naturalist view rejects the sociological understanding of legal philosophy. This is the contemporary legacy of Kelsen’s deeply anti-naturalistic, anti-sociological understanding of legal philosophy. For strong anti-naturalistic legal philosophers the inquiry into the nature of law is an a priori inquiry with no sociological components: it “largely run[s] in parallel” or even has some “priority” to sociological inquiry (Green 2012: xlvi, xlvii; also Gardner 2012: 277, 297; Raz 2009b: 43–44, 104–05). This view implies that while sociologists may make use of this philosophical account (as it provides the proper limits of their subject-matter), philosophers have little or no need for sociology.4

17It is perhaps a mark of the strength of the anti-naturalistic sentiment that pervades contemporary legal philosophy that much of the debate within legal philosophy today is not between naturalists and anti-naturalists, but between weak (broadly Hartian) anti-naturalists and strong (broadly Kelsenian–Razian) anti-naturalists. These two anti-naturalistic views are indeed quite different and call for very different responses. I have argued at length elsewhere (Priel unpublished a) that the pre-sociological conception of philosophy is indefensible, whereas the sociological conception is a viable enterprise, but is conducted with the wrong methods. Whether I am right or wrong in my critique of the pre-sociological, strong anti-naturalist view, I will say nothing further here about it. Instead, I will assume legal philosophy is correctly conceived as part of the philosophy of social explanation, as an attempt to explain the practice humans call “law,” and will address the prospects of a naturalistic account of it. Within that enterprise, this essay addresses the prospects of a naturalistic account of the normativity of law. What’s more, I attempt to re-present the familiar debate between legal positivism and natural law as pertaining to the problem of the normativity of law. I do so in three ways: first as a debate conducted entirely on the basis of anti-naturalistic assumptions; then, as a debate in which one view (legal positivism) is reconceived as a naturalistic approach; and finally—my main proposal in this essay—as a way of understanding both camps in naturalistic terms.

3 Anti-naturalistic legal philosophy

18The great influence of Hart and Kelsen (via Raz) on contemporary (Anglophone) legal philosophy explains the dominance of an anti-naturalistic stance within legal philosophy (Priel 2012: 296–307, collects the evidence). This is particularly evident in the significance attached to explaining law from the “internal point of view,” which for Hart and many who follow his approach, indicates a clear commitment to a rejection of methodological naturalism. Though many contemporary legal philosophers purport to dismiss Dworkin’s interpretivism, Stephen Perry (1995: 101) was right to say that the seeds of Dworkin’s interpretivism were sown by Hart; and it is fair to say that much of contemporary jurisprudence (and by this I mean many who are unreservedly critical of Dworkin) is in essence a variant of Dworkin’s interpretivism (Priel unpublished b). And Dworkin’s interpretivism is an unabashedly anti-naturalist position.

19That this is the case can clearly be seen in Andrei Marmor’s (2011: 33–34) remark, that

[n]aturalizing jurisprudence…works fine as long as it is not really jurisprudence—understood as the philosophical question about the nature of law—that one attempts to reduce to a natural science. The philosophical question about the nature of law is one about a scheme of interpretation; it is a question about the collective meaning and self-understandings of a complex social reality.

20For our purposes, the most notable aspect about this claim is that it is a general statement about legal theory, not a view derived from a particular substantive view; and since Marmor is a legal positivist, it is plain that he thinks this statement is true of legal positivism just as much as it is to natural law theory. In other words, on this view the debate between legal positivism and natural law is conducted on anti-naturalist premises that both sides share:




Legal positivism

Natural law theory

Figure 1: Legal positivism and natural law in most contemporary scholarship

21So understood the debate over naturalism seems orthogonal to the debate between legal positivism and natural law. However, I believe this conclusion derives from a very narrow understanding of the debate between these two views. The most familiar way, at least until recently, to explain the divide between legal positivism and natural law theory was in terms of conditions of validity. I do not think this is the most helpful way of explaining the difference, not least because understood in this way it is difficult to see a real difference between legal positivism and natural law theory (e.g., Perry 2009; cf. Gardner 2012).

22A more interesting way of understanding the debate is between two competing accounts of the normativity of law, for it is on this question that we can identify a real difference between the two views. And it is here that we can understand the sense in which (some) legal positivists deny, and natural lawyers insist on, a connection between law and morality. The legal-positivist view seeks to explain what it means for the practice of law to make a demand that something be done (or not done) in a particular way. For these legal positivists the analogy between law and games has proven attractive, because just like law, games make certain ways of acting “in some sense non-optional or obligatory” (Hart 2012: 82) without any apparent need to invoke morality as part of the explanation. (Games still differ from law in one crucial respect, which arguably undermines the analogy. People choose to play games; people do not typically choose to be subject to law. I say more on this below.) The natural lawyer, however, asks what it is for an individual to be actually required to do or refrain from doing something because the law says so (Finnis 2003), and here it seems more difficult to offer a compelling account without invoking morality.

23It is still unclear whether this way of understanding the debate between legal positivism and natural law also reveals a difference between the two, although for different reasons than before. When the dividing line is seen in terms of conditions of validity, legal positivism and natural law seem to be in verbal rather than substantive disagreement. In the context of debates over normativity understood in the way just identified, the disagreement is apparent because the two supposedly contrasting views are concerned with different questions. More for our purposes, this debate in no way addresses the challenge of giving a naturalistic account of the normativity of law. The reason is not because the “location” of the challenge to such an account does not touch on the question of the connection between law and morality; rather, it is because, as we have seen, it focuses on the question of whether the explanation of any normative domain can be naturalized. On this question, the prevailing view in contemporary jurisprudence, one that crosses the natural law–legal positivism divide, is that the answer is no. It is on this front that a naturalistic approach to the explanation of human action challenges contemporary legal philosophy.

4 Enter jurisprudential naturalism

  • 5 Many legal positivists (most of whom are, as mentioned, anti-naturalists) strenuously denied any su (...)

24When Brian Leiter (2007) urged legal philosophers to naturalize jurisprudence he presented it as a general methodological argument. He argued that legal philosophers were a throwback to the anti-naturalistic approach of 1950s and 1960s Oxford, and their continued adherence to conceptual analysis was a remnant of a bygone philosophical era. Leiter also drew an explicit link between sociological positivism (a close relative of naturalism) and legal positivism (Coleman & Leiter 1996: 241).5

25As mentioned, Leiter presented Hart as a naturalist, but my discussion in the previous section should explain why I think Leiter is better read as an attempt to turn legal philosophy towards naturalism and interpret existing debates in its light (cf. Leiter 2007: 134–35, 161, 189). By then recasting legal positivism as a naturalistic view Leiter was able to redraw the jurisprudential map in the following way:


Legal positivism


Natural law theory

Figure 2: Leiter’s recharacterization of legal positivism and natural law

26Understood in this way, I believe Leiter’s view has certain important implications that are worth spelling out. First, this view offered a (seemingly) clear divide between legal positivism and natural law. Against the difficulty of many contemporary legal theorists to find any real difference between legal positivism and natural law, Leiter’s position provided us with a way of seeing the two views as premised on a fundamental difference. Another significant implication of characterizing the debate in this way is that it ties jurisprudential debates to broader philosophical debates. Leiter (2007: 2) described, correctly in my view, the world of analytic jurisprudence as “small [and] hermetic,” suggesting that debates inside of it are often conducted in complete isolation from, or even apparent ignorance of, discussions in other branches of philosophy. By tying the legal positivism–natural law debate to the debates over naturalism in other areas of philosophy, Leiter provided a way of opening up a stale and moribund discipline to new ideas, to draw links between debates within jurisprudence and debates in, among others, ethics, epistemology, the philosophy of mind, the philosophy of social science.

27Perhaps most importantly, this way of characterizing Leiter’s view, can be used as part of an argument in favor of legal positivism. When the difference between legal positivism and natural law is presented as an application of the broader debate between naturalism and anti-naturalism, the naturalist has an almost ready-made argument for legal positivism, or (if it is any different) against natural law theory. Cut to its core, the argument looks roughly like this:

  • 6 There is risk of terminological confusion here. It is not uncommon, especially in literature from t (...)
  • 7 (6) is not strictly speaking entailed by (5). (6) is true to the extent that we think that legal po (...)

   (1) Legal positivism presupposes the truth of naturalism.
   (2) Natural law presupposes the falsity of naturalism.6
   (3) There are good (pragmatic) reasons to favor naturalism.
   (4) (From (3)): There are good (pragmatic) reasons to disfavor theories that presuppose the truth of anti-naturalism.
   (5) (From (2) and (4)): There are good (pragmatic) reasons to reject natural law.
   (6) (From (5)): There are good (pragmatic) reasons to favor legal positivism.7

28I have not been able to find this argument stated clearly in Leiter’s work, and whatever flaws found in it should not be attributed to him. I think, however, in spirit the argument is there, and can be fairly reconstructed from what he writes. Regardless of authorship, what are the merits of this argument? When the stakes between legal positivism and natural law are this high, i.e. when they involve the question of the truth of naturalism, it is open for the natural lawyer to question (3). Indeed, if the natural lawyer successfully challenges naturalism, the opposite argument can be made against Leiter’s naturalistic legal positivism. Now, while obviously the plausibility of naturalism is an issue that must be addressed at one point in discussing the merits of naturalistic jurisprudence, in an important sense the real question in such a debate has little to do with jurisprudence. In the argument above, the debate between legal positivism and natural law is merely a sideshow to the larger philosophical debate over naturalism.

29If for no other reason that the truth of (3) has nothing to do with law per se, I will not consider it here. Assuming (3) is true, is there a way out for the natural law theorist? No doubt, if naturalism is true, this argument is effective against some versions of natural law theory, but in the following section I will suggest that we can reconstruct certain natural law views in naturalistic terms (that is to say, we may reject (2)). For a jurisprudential naturalist, these are the most interesting views to consider.

30Before turning to this possibility, I wish to say a few words about presenting the distinction between legal positivism and natural law theory as an instantiation of a divide between naturalism and anti-naturalism. As mentioned, most legal philosophers who call themselves legal positivists today understand it as a claim to the effect that what counts as law (what is “legally valid”) in a particular jurisdiction is fixed by what certain people think counts as law. This is a rather weak claim but one that may still be controversial; it ceases to be so when it is coupled with the idea that this test of legal validity is itself fixed by the attitudes of the people in question. Unfortunately, this is what one sometimes finds legal positivists to argue. Leiter (2015b: 1192) says he has “never met anyone who has not fallen through the Dworkinian looking-glass who actually thinks it is an open question whether ordinary lawyers describe the Nazis as having laws, just bad ones”. Even if the statement on what “ordinary lawyers” think is true (and I would not be surprised if the views of many ordinary lawyers in Germany are different on this matter), it is an empirical statement. As such, it is not surprising that natural lawyers have no difficulty accepting it; but this is not because, as Leiter thinks, legal positivism has proven victorious (Leiter 2007: 2, 163), but because legal positivism understood in this way does not engage with the claims natural lawyers make. Legal positivism understood in this way is the claim that what according to lawyers is valid law is what ordinary lawyers consider to be law. This is uncontroversial. To challenge natural law theory it needs to be shown that what according to lawyers is valid law is actually law. What ordinary lawyers think about Nazi law has no bearing on the latter question.

31If the latter question is not to be dismissed as a verbal dispute (i.e., what natural lawyers mean by “law” is different from what legal positivists mean by “law”), this question must be answered by turning to normative questions, and specifically questions about political authority. So understood, there is an old idea—far older than any Dworkinian looking-glass—that there may be things that have the external appearance of law, but in fact are something else, specifically a form of robbery, the confusion of power with authority. On this view to say that unjust law is always law is to focus on certain “structural” features while ignoring law’s essence. For legal positivism to pose a challenge to this view it must be understood as a political (not conceptual) thesis about authority. Indeed, the view that what counts as law is fixed by what certain people take to be law follows, and is explained by, a “positivist” thesis about authority. In particular, such a thesis about authority can explain who are people to whose views we should look to know what counts as law, and why it is their views and not the views of others that matter. There are naturalistic and anti-naturalistic ways of understanding this question, and I think these will be more illuminating ways of understanding a debate between a naturalistic version of legal positivism and an anti-naturalistic version of natural law theory.

32This point leads to a second one, which is that Leiter’s interests and my own are somewhat different. Leiter seems to accept the identification of what counts as law as an important task for jurisprudence (Leiter 2007: 181, 189–90; although in Leiter 2013 he is more skeptical about this project), although he rejects the dominant conceptual approach to answering this question, favoring a naturalistic answer instead. By contrast, Leiter (2007: 162–63, 170–72) seems relatively uninterested in explaining the normativity of law, suggesting that legal positivists (and Hart more specifically) do not try to offer an account of law’s normativity. In this respect, Leiter’s position differs from my own. I see the question “what is law?” as a sociological-historical question to be answered by empirical methods: To answer the question “what is law?”—if by this question we mean “what counts as law?”—is in principle no different from the question “what is football?”—if by this question we mean “what counts as football?” If the latter two questions do not raise any problems for a naturalistic account, I do not see why the former two should be any different.

33Leiter’s relative uninterest in the question of the normativity of law might be grounded in his belief that legal rules are not what guides the outcomes of (appellate) legal decisions. (Leiter has written on normativity more generally [2015] where his views are more congruent with what I say below.) If this is true, then legal rules are normatively epiphenomenal, and what is needed is not an account of law’s normativity, but an explanation for why legal norms look like it has normative force and why this appearance is mistaken. This too seems to me to be an empirical question that does not raise any particular problems for a naturalistic account of law.

  • 8 Even claims coming from cognitive science should be treated with caution. As was recently argued by (...)

34It is, of course, possible that judges do not decide cases by following legal norms, but such a claim is highly contentious. As Leiter (2007: 190) himself acknowledges, it may be true only of appellate courts, which leaves open the possibility that law has normative force on trial judges as well as other people. Moreover, the empirical studies on which it is based have been challenged by other empirical studies that argue that some legal norms influence outcomes. Perhaps most important in this context, it is notable that Leiter turns to political science, not cognitive science, for empirical support. As such, these studies do not point to any cognitive impediment to legal rule-following (limits that presumably are true of all humans), but are confined to American courts, and especially the United States Supreme Court.8 There are reasons for thinking that that court, and perhaps American law more generally, are relative outliers compared to other courts in other jurisdictions. If legal rules have normative force (and not just the appearance of one), explaining it remains an open question. And as questions of normativity are often seen as posing a serious challenge to naturalistic views, it is worth laying out the different ways in which naturalistic jurisprudence might respond to it. This is the task of the next section.

5 Legal positivism and natural law recast in naturalistic terms

35With previous positions laid out, it should be obvious where the argument is going: placing both legal positivism and natural law on the naturalistic side of the divide.


Legal positivism

Natural law theory



Figure 3: My proposed recharacterization of legal positivism and natural law

36At first sight, this suggestion may seem odd. As shown above legal positivism is today mostly held by anti-naturalists, but with legal positivism the historical links to naturalism are clear. Hobbes and Bentham, both considered seminal legal positivists, were adamant about the continuity between the natural world, human nature, and the social and political world. They explicitly sought to explain the moral and political domain using the same methods used for explaining the natural world. So placing legal positivism (or at least some version of it) on the naturalistic side is not particularly difficult to stomach (although it does suggest that contemporary legal positivism and its purported intellectual ancestors are less similar than is commonly assumed: Priel 2015). Matters do not look quite so simple on the natural law side of the ledger: the origins of natural law theory lie in the misty days of ancient Greece and Rome, with an image of a universe where god(s) not only set the laws that set the physical world in motion, but also the laws for humans to abide by. To this day natural law theory is closely associated with religious doctrine, and religion and naturalism are not obvious bedfellows. No wonder, then, that when Kitcher (2012: 4) recently wrote about the prospects of naturalistic ethics, he said it meant that “[t]here are no spooks….[I]nvocations of the Forms, or of Natural Law, or of Processes of Pure Practical Reason, or even of Moral Properties accessible to ordinary human faculties all have to be shown to accord with standards of reliable inquiry—or they have to go.” Finally, even if we could offer naturalistic versions of both legal positivism and natural law theory, why should we? One advantage of the strategy pursued in the previous section was that it affirmed the clear divide between legal positivism and natural law. Putting legal positivism and natural law theory in the same metaphysical basket risks blurring the distinction between them once again.

37These potential problems point to the two conditions that a naturalistic natural law account will have to meet. It will have to be based on “standards of reliable inquiry,” and it will have to result in a view that is distinctively and importantly different from legal positivism. The rest of this section is an attempt to show that we can have a meaningful distinction between legal positivism and natural law theory in which both are understood in naturalistic terms. As it turns out, we end up with not two, but three distinct positions. In addition to the two familiar players, I introduce a newcomer (of sorts), which I dub “semi positivism.” Roughly speaking, the first sees morality as the product of biological evolutionary processes, which it then tries to build into an account of the normativity of law; the second sees moral norms as a product of social evolutionary processes; the third does away with connections to morality as the basis for the normativity of law, and sees it as the product of human effort.

5.1 Naturalistic natural law

38It has long been recognized that some writers in the natural law tradition have sought to derive moral principles from (purported) facts about human nature. The further step is to ground the foundation of law in similar notions. Adopting this underlying idea, a naturalistic version of natural law will try to explain human nature, including the moral norms that humans accept, as evolutionary adaptations. Both theoretical arguments and empirical evidence have been presented in support of these ideas. First, theoretical models have shown how certain moral norms can prove adaptive and stable (popular presentations are legion, e.g., Ridley 1998). Second, animal studies have shown that they too behave in ways suggesting a rudimentary set of moral norms (e.g., De Waal 1996; Peterson 2011). Anyone who accepts that humans are products of the same evolutionary processes that are responsible for behavioral traits found in another animals will have a hard time explaining why animal morality (to the extent that it exists) is the product of evolution while similar moral traits in humans have no evolutionary basis whatsoever. Third, developmental psychologists have argued that certain behavioral traits that seem to conform to adults’ moral attitudes are found in babies that are only a few months old, suggesting these attitudes are innate (Bloom 2013). Finally, despite considerable variation in moral views, psychologists have argued for the existence of certain universal moral norms found across cultures (Haidt 2007). It goes without saying that the conclusions drawn from these studies remain highly controversial; but taken together these considerations have been used to construct, or begin constructing, naturalistic accounts of morality.

39One strand within these views could be seen as a naturalistic version of natural law ideas. Mikhail (2012: xv, 6–8, 314–16) has shown parallels between his naturalistic approach to the foundations of law and morality and older thinking on these domains. Mikhail argues that underlying the perceived diversity of moral attitudes and discourse, there is a universal moral grammar. The relationship between the two is understood in roughly the same as the relationship between linguistic diversity and the universal (linguistic) grammar; and in both instances, it is the perceived “poverty of stimulus” that cannot account for the complexity and rapidity with which children acquire moral discourse.

40Though the immediate inspiration of this view is Noam Chomsky’s views on universal linguistic grammar, Mikhail (2012: 296) sees links between his ideas and those of “Cicero, Aquinas, Grotius, and other classical writers on natural jurisprudence and the law of nations”. As he points out, all of them have drawn a similar analogy between the human capacity for language and for moral judgment. The fundamental difference between the two, the one crucial for our purposes, is that where traditional (anti-naturalist) natural law saw the source of both in God, the naturalist version sees both as the product of evolutionary processes.

41Mikhail (2014: 786) attempts to show that the results of several studies about human moral attitudes on intentional infliction of harm to others track the doctrinal structure of the tort and crime of battery: “at least some of the components of the legal norm against harmful battery form essential building blocks of human moral cognition”. One of the upshots of this claim, according to Mikhail (2014: 784), is that human psychology contains a fairly complex structure that might be used to explain “how more complex moral and legal offenses are mentally represented and morally evaluated”.

  • 9 An even more radical critique of the argument I consider in the text would deny there is anything t (...)

42There are two central concerns with this view. One, about which I will have relatively little to say, is whether the linguistic analogy is valid.9 Even among those who accept the general view that morality has innate, possibly evolutionary, foundations, the analogy remains contentious. For instance, Kim Sterelny (2012: 167) accepts that the acquisition of moral norms has biological foundations but dissents from the linguistic analogy: “we are biologically prepared to develop moral cognition, not because our minds are prewired to acquire moral concepts and principles, but because moral cognition is a natural development of our existing emotional, intellectual, and social repertoire.” In particular, we are attuned to the emotional reactions of others, which often cause emotional reactions in ourselves. It is these emotional responses that then form the basis for the development of moral norms. Norms that match these emotional reactions are likely to be more easily learned, endorsed, and eventually become entrenched. However, our moral discourse is further complicated by the way principles may lead to emotional reactions. The totality of moral discourse reflects more, Sterelny argues, the expertise of an experienced craftsman, involving a mixture of principles, rules of thumb, prototypes and analogies, than the articulation of a pre-social innate norms.

43In this respect Mikhail’s arguments with respect to battery, even if entirely successful, are not clearly generalizable (as he acknowledges in Mikhail 2014: 754). The tort of battery and the moral rule against the infliction of harm to others may be moral universals, but our moral discourse includes much more, including much that it is difficult to reduce to causing harm. It is less clear that one can reconstruct a similar account to the one provided by linguists about universal linguistic grammar to explain these aspects of moral discourse.

44The other worry about this view is closer to the questions that lie at the heart of this essay. Assuming this view is true, can it explain the normativity of law? Mikhail does not argue for this view, so I should make it clear that the following reconstruction is not Mikhail’s. It is not entailed by what he says, and it is not obvious that he would endorse it. Therefore, what follows is not to be taken as criticism of Mikhail, only one attempt to examine whether his position can be used to address the question of the normativity of law.

45Mikhail’s general discussion of the linguistic analogy and his example regarding battery raise a possible ambiguity. His general discussion relies on the poverty of stimulus to argue for a universal moral grammar. This view is consistent with considerable moral diversity which can be analogized to linguistic diversity. On the other hand, Mikhail’s example in the case of battery, if I understand him, is meant to illustrate the existence of some universal moral norms, which he seeks to show are far more elaborate and nuanced than people typically assume. This ambiguity has significance for the extent to which we can rely on it for explaining the normativity of law.

46On the latter view (universal moral norms) the normative force of positive law derives from its correspondence to this natural law. Natural law provides a kind of blueprint that positive law ought to imitate. Law succeeds in creating obligations to the extent that it matches morality. The discovery of some universal moral norms in the context of the direct causing of physical harm is valuable in providing a naturalistic grounding for this view, but it is very difficult to generalize from it.

47In traditional natural law, certain moral norms were part of God’s plan for the universe; its naturalistic version puts evolution in place of God. This shift may raise all sorts of difficulties in the domain of morality. The main difficulty in the jurisprudential context is that God’s natural law can be as complex as one wishes it to be: God being omniscient, there are no limits to the complexity of his natural law. (Even here, the common view is that natural law is somewhat underspecified, leaving it open for humans to make determinate decisions regarding the content of positive law.) By contrast, evolution is not omniscient, and it operates very slowly. Evolved moral norms are by no means simple, but they are not nearly as complex as modern legislation. In other words, the existence of a universal moral grammar, if true, is of great significance for understanding human morality as a natural phenomenon. It is less clear how it could be used to explain the normativity of law.

48It might be tempting to argue that moral diversity can be correlated to legal diversity, and use this fact to explain the normativity of law. In this way the existence of a universal moral grammar is consistent with several moral blueprints. While I do not think many will doubt that the prevailing moral and social norms in a given community affect the content of the community’s legal norms, it is still difficult to explain the normativity of law on this basis. For one, there are many well-known examples of divergence between the two even in one community. Even taking moral diversity into account, there remains a significant difference between the vagueness and looseness of prevailing moral norms and the complexity of legal norms, which cannot be explained solely in terms of using law to provide greater specification (determination) for moral demands. This difference makes it difficult to see how the former could be used to explain the normativity of much of the law.

49To conclude, while a naturalistic, evolutionary account could provide us with something that bears important resemblance to one aspect of natural law theory, namely an explanation of the existence of some universal norms that are part of human nature in a way analogous to language, I cannot see how such a view could be the basis for a naturalistic equivalent for the natural lawyer’s explanation of law’s normativity.

5.2 Naturalistic semi positivism

50The second view I consider is also evolutionary, albeit in a different way. Here, social norms emerge in a process of social evolution. These models do not assume that certain norms have been naturally selected, but they show how a social convention can emerge and become self-enforcing once it is in place (Skyrms 1996; Sugden 2004). There are obvious continuities between accounts based on biological and social evolution, and arguments made in one context can often be transposed to the other, not least because evolutionary game theory is the main explanatory tool used in both. In the case of biological evolution, certain behavioral traits we call “morality” are selected because they are adaptive; in the latter, certain social norms emerge and remain stable because they confer advantages on the social group in which they exist. Such models may seem uncontroversial for explaining purely conventional norms (the rule of the road is the classic example), but they have also been used to explain the spontaneous emergence of norms of a more “moral” nature, such as norms requiring assistance to others, or norms relating to ownership and property.

51The label “semi positivism” is meant to capture the way in which this view assumes that social norms are socially created (and in this way they are “positivistic”) and explained completely from the “external point of view,” while at the same time not being the product of conscious, purposeful norm-creation action. At the same time, there is a sense in which such norms can be seen as natural: Not natural in the sense that certain moral norms can be derived from an evolutionary account of human nature, but natural in the sense that they emerge spontaneously and sustain themselves without external enforcement. It is for this reason that Sugden (2004: 150, 176) used the phrase “natural law” to describe his view.

  • 10 The example highlights a point that I cannot explore here but which I think is one of the merits of (...)

52Such views may seem obviously inadequate, even irrelevant, for the explanation of law, much of it is unquestionably consciously and intentionally created. Still, a proponent of this view may argue that the foundations of the law we have, however complex and consciously-created, can still be traced to these natural origins. For instance, it is undoubtedly true that law has developed various forms of complex property rights (leases, easements, different kinds of inalienability, complex ownership arrangements such as condominiums, cooperatives, and so on), which go beyond basic ideas of property that (on this view) can emerge spontaneously. Nevertheless, the foundations of the complex legal arrangement can be traced to these semi-positivist property rights.10

53Is this view relevant to explaining the normativity of law? Recall that we are not interested in the explanation of the existence of legal norms or their content, but in accounting for their normativity. This is a question about the basis of political and legal authority. Such an account thus only requires a social-evolutionary explanation of the norms required to sustain the authority of law, not for the consciously-created legal norms themselves. Even with this qualification, I believe it is impossible to fully explain the normativity of law on this basis, for reasons quite similar to those offered in the previous section. That certain notions of property, punishment, or reciprocity, may have emerged spontaneously may be an important part of an explanation of human society. But just as in the case of innate moral norms, this does not help us with explaining the normativity of law. The question of law’s normativity is puzzling precisely in the context of situations in which it does not arise spontaneously.

54Nevertheless, this does not yet imply such ideas are irrelevant for an account of the normativity of law. While such an account cannot explain the normativity of legal phenomena today, it can be part of broader story in which the normativity of law begins with conventions and ends up with consciously-made legal governance. In other words, there may be a broader evolutionary story of law in which an evolutionary account of norms can figure as part of the story about the development of law. In section 6, I present the bare outlines of such an evolutionary account of legal authority.

5.3 Naturalistic legal positivism

55There is a familiar naturalistic account of law, associated with “classical” legal positivism, according to which the normativity of law is to be explained in terms of the prospects of an adverse reaction to a breach of a legal norm. Such views are thought to have been criticized beyond redemption by Hart’s critique of Austin’s command theory. I believe Hart’s demolition job was less successful than is often assumed. Later in this essay I will argue that his own view is actually rather close to the one he supposedly displaced, and will sketch an argument in support of the idea that coercion is essential for a positivist naturalist theory of law.

56But first things first. The view I dub here positivist denies that law is derived from natural law (however conceived). On this view law is a matter of artifice not just in the obvious, wholly uncontroversial, sense that positive law is in some sense the product of human action; it is also artificial in the sense that it is not conceived of as an imitation of natural law or morality: As a matter of fact, law, or at least most of it, need not reflect innate tendencies, nor does it enforce spontaneously-emerging norms. And the authority of law is none the worse for that, because law’s authority does not depend on its correspondence with any other standard.

57This view is not some kind of novelty. In fact, I think there is a good case to be made that this is the sense, or at least the interesting sense, in which Hobbes and Bentham can be seen as “legal positivists.” For Bentham, law is a human invention, to be designed by human ingenuity for the improvement of human welfare. The idea that such laws get their normative force from imitating some pre-existing morality—“natural law”—makes no sense, because no such thing exists. Accordingly, the view that the normativity of law derives from their imitating natural law is false, because there is nothing for positive law to imitate.

58Despite the centrality of “natural laws” to Hobbes’s argument, I believe he exemplifies this position as well. Hobbes’s natural laws, as he makes abundantly clear, are not really laws at all (Hobbes 1996: 111, 185). They are precepts that a purely self-interested individual will rationally follow if he seeks his own survival. Unfortunately, such natural laws, said Hobbes, are insufficient to maintain peace, which is why the creation of a political authority is necessary. The important point is that the laws enacted in civil society are not supposed to be merely a positive enactment of these rational precepts. Unlike Locke, for whom the point (and therefore also the limit) of legislation was the protection of natural law, Hobbes was clear that the sovereign’s commands should be enacted “to ensure that citizens are abundantly provided with all the goods necessary not just for life but for the enjoyment of life” (Hobbes 1998: 144).

59For my purposes the most interesting aspect of this view is how it explains the normativity of law, the sense law creates oughts. And in a way, the answer in terms of the threat of sanction may at first appear almost a default position: law’s coercive force explains law’s normativity because nothing else could. This view denies that the authority of positive law is in any way derived from some correspondence to another standard, be it natural law or conventional social norms. The only remaining means by which the proclamations written on paper we call “laws” can become something one ought to follow is coercion; or more precisely, acceptance and coercion with often murky boundaries between the two. (This, as is explained further below, is not akin to saying that such laws are legitimate. To argue that the normativity of law derives from acceptance and coercion is not akin to saying that might makes right.)

60Does this suffice for explaining the normativity of law? Acceptance suffices for explaining the normativity of many social norms: they are binding on those who (implicitly) accept them. Closer to the legal domain, acceptance is sufficient for the explanation of the normativity of games. Games are (in part) systems of rules, whose normativity is not explained by reference to morality. Hart’s account of law’s normativity in terms of “acceptance” by “officials” is at bottom no different from his view about the normativity of games.

61There is, however, one significant disanalogy between law and games. Games are voluntary, law characteristically is not. The rules of the game are binding only on those who accept them, and they cease to have normative force the moment one rejects them. If one rejects the rules of association football, one is no longer doing anything wrong when one touches the ball with one’s hands; if one thinks chess would be improved by allowing castles to move diagonally, one is free to play this game with a willing partner. Because such a choice is not available in the case of law, an account of its normativity has to include something in addition to normativity in terms of acceptance. How can we explain naturalistically the normativity of law for those who do not “accept” its norms, who do not wish to play the game of law? As far as I can see, the only remaining possibility is in terms of coercion. This was the answer of the “classical” legal positivists Hobbes and Bentham; and this was Hart’s answer as well. A legal system exists, he argued, when officials accept the law; as far as the others go, nothing more than “deplorably sheep-like” rule following is required (Hart 2012: 117).

  • 11 Here Hart was at odds with strong anti-naturalists like Raz (1990: 56–58) who argued that no social (...)

62Hart did not think there is any difficulty in explaining the normativity of law as a threat for those who do not accept it. As he says (Hart 2012: 90), the view of those who sought to explain all of law’s normativity in this way “may very nearly reproduce the way in which the rules function in the lives of…those who reject [a group’s] rules” and are concerned about “unpleasant consequences [that] are likely to follow violation.” Hart saw his main contribution in adding another component to this account, namely an analysis of the normativity of law for those who accept it. Hart sought to give a social account of this aspect of law’s normativity (which is why, as we have seen, some have been tempted to call his view “naturalistic”), but he clearly thought that the methods of the social sciences are incapable of explaining the normativity of acceptance. To address this problem, he insisted on an anti-naturalistic methodology instead (the only one that could explain the “internal aspect” of rules).11 And, as mentioned earlier, the scope of his claim was vast: social-scientific methods could not explain the normativity of any domain. It was not just law that a naturalistic methodology could not explain; it was also language, chess, and football.

63The naturalistic response here is to say that there is a fully naturalistic, social or psychological account of this question. Hart’s meager attempts at explaining the shortcomings of such a psychological approach are unconvincing. He argued that a “hardened swindler” (Hart 2012: 88) may be under an obligation even if he does not feel any compulsion to obey. This may be true, but it is beside the point. After all, even on his view, the normativity of law for those who do not accept it is to be explained in terms of coercion, not in terms of psychological compulsion. For those who do accept the rules, it remains unclear what a fully naturalistic account is missing. The idea that a naturalistic approach could not explain human cognition is based on an impoverished view of science, perhaps defensible in the 1950s but much more difficult to defend today. If the claim is that there is something deeper that such a view leaves out, the naturalistic response is that there is nothing more to normativity (cf. Leiter 2015).

6 From natural to artificial law

64The preceding discussion presented several theoretical possibilities, but it would be presumptuous to suggest that it offered a decisive argument in support of any of the three views. In this section I attempt to show that such an argument, one that purports to show that one of these views is always true and the others always false, may itself be a mistake. This, together with what I have already said, may lead some to counter that however interesting the ideas presented here may be they have nothing to do with legal positivism and natural law theory.

65In one sense this is not a very important matter. “Legal positivism” and “natural law theory” are labels, and one should be free to use them as one pleases. There is no question that I give legal positivism and natural law a somewhat different meaning from the way they are more commonly understood these days (and I make this worse by adding a third position). But such changes in meaning are to be expected. The philosophy of mind today is vastly different to the philosophy of mind in the 1960s, as is the case with political philosophy. The questions asked, the approaches taken to answer them, the concepts and ideas used are all very different. There is thus nothing wrong with the suggestion that legal philosophy today should look very different from the way it did in the 1960s. Even philosophical labels change and acquire new understandings. A present-day empiricist need not accept much of what George Berkeley believed. The same is true in jurisprudence: The prevailing understanding today of the difference between legal positivism and natural law as a debate concerning the conditions of validity is itself not timeless and by no means corresponds to the way the terms have been understood throughout history. This becomes even more evident when these labels are ascribed to thinkers who never used them. If Hobbes and Bentham are legal positivists (though they never used the word “positivist”), then there is no question they would have objected to many of the ideas nowadays defended under that banner, not least its dominant anti-naturalism (Priel 2015).

66By using the labels I wish to draw attention to important continuities with some ideas defended by some legal positivists: naturalism, and a very particular separation between law and morality. Nevertheless, had it not been to the weight of history, there might have been a better label for their position. The opening words of Hobbes’s Leviathan (1996/1651: 9) are instructive:

  • 12 This view of law and the state as an artificial creation and as an instrument is also central to Be (...)

Nature (the art whereby God hath made and governes the world) is by the art of man, as in many other things, so in this also imitated, that it can make an Artificial Animal. For seeing life is but a motion of Limbs, the begining whereof is in some principall part within; why may we not say, that all Automata (Engines that move themselves by springs and wheeles as doth a watch) have an artificiall life? For what is the Heart, but a Spring; and the Nerves, but so many Strings; and the Joynts, but so many Wheeles, giving motion to the whole Body, such as was intended by the Artificer? Art goes yet further, imitating that Rationall and most excellent worke of Nature, Man. For by Art is created that great leviathan called a common-wealth, or state, (in latine civitas) which is but an Artificiall Man.12

67If we are to identify a clear contrast to “natural law theory,” the most obvious label would be “artificial law theory,” for this view takes seriously the idea that law is artificial, and not just in the uncontroversial sense that positive law is a human creation. Artificial law theory is the view that even the question of normativity needs to be explained with no reference to morality, however we understand its origin. This artificial law theory connects the ideas of Hobbes and Bentham to those of some contemporary legal positivists, and it is on this matter that one can speak of some continued “tradition.” (Unfortunately, this link is much obscured today by the dominance of the Kelsen-Raz version of legal positivism, which unlike the Hartian version, rejects this idea.)

68It is true, however, that when the normativity of law is understood in this way, questions of legitimacy become inescapable in a way that they are not in the context of games. Legal positivism, as conceived of here, is a thesis about the separation not of law and morals, but of normativity of law and its legitimacy (cf. Priel 2011b). We have seen in the previous section that in a naturalistic account of law coercion is necessary for this explanation of the normativity of law. When this is the case, one must complement this account with a political theory that explains what distinguishes law from other forms of coercion. To resort to terminology more familiar to legal philosophers, a naturalistic account requires a political account to explain why law is not the gunman situation writ large. Hobbes and Bentham recognized this, which is why their (naturalistic) theories of law are inseparable from their political theories.

69Hart tried to avoid such questions by asserting as a supposedly value-neutral observation that a system of norms that relies on coercion can still be a legal system: “there is little reason for thinking that it could not exist or for denying it the title of a legal system” (Hart 2012: 117). But it is nothing of the sort; the claim that a legal system exists when those in power accept the system and others obey it in a “sheep-like” fashion is value-neutral only if it is meant to capture what people call “law” (in which case, Hart supports it with no evidence), or if it is meant as a linguistic stipulation (cf. Hart 2012: 209–12, where Hart talks of “preferring the wider concept of law”). To the extent that he tried to go beyond that, or is taken to have gone beyond that, he failed. Hart tried distinguishing a legal system from something that just looks like a legal system (but is more similar to the gunman situation writ large), by relying on the distinction between the internal and the external point of view of rules, which in this context can only mean that a legal system exists when self-appointed “officials” think that one exists. More precisely, law is the gunman situation writ large, so long as its self-appointed “officials” think it is legitimate. If this is not taken as an observation about prevailing linguistic usage, it is not a neutral claim. The surprising, and possibly counterintuitive, naturalistic alternative that emerges from the preceding discussion is that a (naturalistic) theory of law must be accompanied by a political argument showing when (if ever) coercion is legitimate.

70I think there is another, perhaps even more radical, challenge to contemporary jurisprudence that such a naturalistic orientation could yield. Law and its authority can be explained in evolutionary terms, as a development from natural to artificial. That law evolves is a familiar feature as far as its content is concerned, perhaps even that its contents tends to change from more “natural” norms (whether those are innate moral norms or evolved social norms) to more “artificial” ones. The point I make here is that law’s evolution from natural to artificial is also a possible explanation of a change in law’s authority. Indeed, the processes of the artificialization of law’s content is parallel to (and perhaps dependent on) the artificialization of its authority. Once the foundation of authority changes, it becomes normatively possible to justify far more extensive laws that are “artificial” in nature.

71The contrast with the prevailing view of jurisprudence is significant. The approach to explaining the normativity of law is static. True, Hart’s account of the emergence of law did include a story of the emergence of law in terms of a transition from a society regulated only by primary rules of obligation to a society that adds various “rules about rules.” Because this account was largely speculative, and as a historical matter highly unconvincing (cf. Cohen 1962: 408–11), it is now taken by allies and critics alike as a fable only intended to illustrate the difference between law and non-law. By contrast, the view considered here treats the question of the emergence of legal phenomena and the evolution of legal normativity as a matter of historical development. On this view one cannot answer the “philosophical” question of the normativity of modern, complex law without understanding its origin.

72Another interesting implication of the preceding discussion is that it shows how the dominant “positivist” approach to going about explaining law is skewed in favor of a particular answer. It begins with an account of what law is (in general) and then moves to the question of identifying “laws” (through some test of “validity”). In this non-evolutionary story law comes into being in a more-or-less discrete moment, which is why it is thought important (and possible) to identify certain markers of legal existence. Hart’s familiar story about the “foundations of a legal system” remains the model for much positivistic thinking about law to this day. Inherent in this account, and largely accepted without question, is the priority of legal authority to laws. Law can only come into being when there is a legal authority that brings it to existence.

73The different naturalistic approaches considered here show that this way of thinking about law (legal authority prior to laws) is not a neutral perspective. Both the naturalistic natural law view and the semi-positivist view reverse this order. If we need a slogan, for these approaches it is “laws before (or, independently of) legal authority.” The view I am adumbrating here is that the origins of law lie in natural laws in the sense considered here. Here laws are explained in terms of innate moral attitudes or as the product of spontaneously-emerging, accepted social conventions. The important point is that in this world, it is the responsibility of the authority not to make these laws but to make sure that they are protected. In such a world, where legal norms hew closely to these natural norms, the authority of law derives from the “natural” normativity of these norms. With time, the order of priority reverses. Social, political, and technological changes result in legal authority coming to be seen as the source of law. Law is then seen as the product of legal authority, and its normative force is derived from this fact. In this process law in its earlier stages is more “natural” and becomes with time more “artificial” both in terms of its content and in terms of its normativity.

74The significance of the last point is that it allows us to redefine the relationship between natural law and legal positivism. Rather than competing theories, each vying to tell us the whole truth about the “nature” of law (wherever and whenever it is found), we can think of three theoretical positions considered above as three elements in a larger evolutionary story of the development of the idea of law. Rather than presenting certain conceptual truths about law and arguing on the basis of them that some societies have law and others do not (an exercise that I think is wholly circular: Priel unpublished a), we can see that the “nature” of law itself can evolve and with it also the underlying account of the normativity of law.

75Such an evolutionary view does not require the complete displacement of some natural law elements (naturalistically understood) from an account of the law. One can accept Mikahil’s claims that innate moral intuitions govern the law of battery (and possibly a few others) while grounding the authority of other parts of the law on a different foundation. Additional interactions between the three positions identified above are possible. Empirical evidence shows how certain relatively arbitrary norms can become to be seen as “fair,” and as such be considered morally required (Young 1996; Sugden 2004). Thus, it might be possible to incorporate such an account of “acceptance” of legal authority and legal norms, one that goes beyond threat of sanction, into our account of modern “artificial” law.

76At the same time, such an account can acknowledge that there was some artificial law even in more ancient times. What differs is the relative volume of each. As can be expected from an evolutionary account such as this, it is neither the case that law changes its nature overnight, nor that its earlier origins are completely lost. We can expect certain legal norms, like the tort of battery, to remain relatively close to “natural” innate moral norms regarding the direct infliction of physical harm; we can expect tax law or environmental regulation to be far more artificial. Thus, an account of a historical development from a more natural to a more artificial law need not deny that remnants of natural law are still with us (cf. Priel 2016: 70–72). This view will also provide a straightforward explanation, or at least part of an explanation, for the dominance of natural law theory in ancient times and the emergence of a more positivist view at the time of the emergence of the modern state. The explanation is not that the world needed a genius of the stature of Hobbes or Bentham to recognize what law had always been but somehow no-one before them had noticed. It is that they were writing in a time of major political, social, and technological changes. These changes made law increasingly artificial.

77Much flesh needs to be added to these bare bones, but there is at least one theoretical problem that this idea helps solve. The dominant positivist view suffers from a well-known problem of circularity: It is difficult to explain the priority of legal authority to laws, when legal authority (“officials” in Hart’s terminology) is itself a product of laws. One advantage of the alternative view just presented is that it breaks this vicious circle. The origins of today’s artificial law are in natural law, whose normative force did not depend on the existence of a legal authority.

7 Conclusion

78Naturalism remains an unwanted idea in legal philosophy. To the extent that it is allowed some place within jurisprudence, it is though to address questions that are entirely independent of the “traditional” questions of legal philosophy. In this essay I have attempted to show that this is a mistake, that a naturalistic perspective has important things to say even on the most familiar debate in contemporary legal philosophy, between legal positivism and natural law theory. Not only can these two views be given interesting, meaningful naturalistic formulations (and ones that have important corollaries to the views of major historical jurisprudential figures), they can also provide an alternative way of thinking about the proper way of answering some of long-standing questions in the field. The proposal suggested here is that legal positivism and natural law are not competing accounts in the sense that the truth of one view necessarily implies the falsity of the other in any and every place. Nor is it that these two perspectives are simply two views of the cathedral, two complementary perspectives on the same phenomena. Instead, I argued that these views are competing, albeit in a more subtle sense in that they compete only in relation to particular times, places, or even particular laws. Over the whole range of complex and diverse phenomena that make up the law, one may well need both.

Acknowledgments.I thank participants in workshops on naturalistic jurisprudence at Jagellonian and York Universities for their comments. I also thank two anonymous referees for Revus for their helpful comments, which forced me to refine and clarify my argument. Brian Leiter’s clarifications of his views (which I discuss in section 4) has led to several changes. I hope the result is an accurate presentation of his views. Leiter also attempted to convince me that H.L.A. Hart was more naturalistic than I present him. Here, I remain largely unconvinced, but Leiter’s remarks have led me to substantially expand my discussion in section 3 in the hope of better supporting my position. Priel (unpublished b) further bolsters these conclusions with a discussion of the intellectual ancestry of Hart’s self-professed “hermeneutic” approach. Despite all these helpful remarks, I must acknowledge that some of the issues discussed in this essay, especially in section 6, call for a more detailed discussion. I hope to provide that discussion in the future.

Top of page


Lynne Rudder BAKER, 1998: The First-Person Perspective: A Test Case for Naturalism. American Philosophical Quarterly 35. 327–48.

Lynne Rudder BAKER, 2011: Does Naturalism Rest on a Mistake? American Philosophical Quarterly 48. 161–73.

Paul BLOOM, 2013: Just Babies: The Origins of Good and Evil. New York: Crown.

L. Jonathan COHEN, 1962: Critical Notice of H.L.A. Hart, “The Concept of Law”. Mind 71. 395–412.

Jules L. COLEMAN & Brian LEITER, 1996: Legal Positivism. The Blackwell Companion to Jurisprudence and Legal Philosophy. Ed. In Dennis Patterson. Oxford: Blackwell. 241–60.

Frans DE WAAL, 1996: Good Natured: The Origins of Right and Wrong in Humans and Other Animals. Cambridge: Harvard University Press.

Daniel DENNETT, 2001: Are We Explaining Consciousness Yet? Cognition 79. 221–37.

John R. DINWIDDY, 1989: Bentham on Invention in Legislation. Enlightenment and Dissent 8. 25–42.

Kenneth EHRENBERG, 2016: The Functions of Law. Oxford: Oxford University Press.

Nicholas EVANS & Stephen C. LEVINSON, 2009: The Myth of Language Universals: Language Diversity and Its Importance for Cognitive Science. Behavioral and Brain Sciences 32. 429–92.

John FINNIS, 2003: Law and What I Truly Should Decide. American Journal of Jurisprudence 48. 107–29.

Owen FLANAGAN, 2006: Varieties of Naturalism. The Oxford Handbook of Religion and Science. Ed. Philip Clayton. Oxford: Oxford University Press. 430–52.

John GARDNER, 2012: Law as a Leap of Faith: Essays on Law in General. Oxford: Oxford University Press.

Leslie GREEN, 2012: Introduction. In Hart 2012.

Jonathan HAIDT, 2007: The New Synthesis in Moral Psychology. Science 316. 998–1002.

H. L. A. HART, 1955: Theory and Definition in Jurisprudence. Proceedings of the Aristotelian Society, Supplementary 29. 239–64.

H. L. A. HART, 1982: Essays on Bentham: Studies in Jurisprudence and Political Theory. Oxford: Clarendon Press.

H. L. A. HART, 1983: Essays in Jurisprudence and Philosophy. Oxford: Clarendon Press.

H. L. A. HART, 2012: The Concept of Law. 3rd ed. Oxford: Oxford University Press.

Joseph HENRICH et al., 2010. The Weirdest People in the World? Behavioral and Brain Sciences 33. 61–135.

Thomas HOBBES, 1996: Leviathan. Cambridge: Cambridge University Press.

Thomas HOBBES, 1998: On the Citizen. Cambridge: Cambridge University Press.

Oliver W. HOLMES, 1897: The Path of the Law. Harvard Law Review 10. 457–78.

Jaegwon KIM, 2010: Essays in the Metaphysics of Mind. Oxford: Oxford University Press.

Philip KITCHER, 2012: Naturalistic Ethics without Fallacies. Nomos 52. 3–30.

Brian LEITER, 2007: Naturalizing Jurisprudence: Essays on American Legal Realism and Naturalism in Legal Philosophy. Oxford: Oxford University Press.

Brian LEITER, 2009: Naturalizing Jurisprudence: Three Approaches. The Future of Naturalism. Eds. John R. Shook & Paul Kurtz. Amherst: Humanity Books. 197–207.

Brian LEITER, 2013: The Demarcation Problem in Jurisprudence: A New Case for Skepticism. Neutrality and Theory of Law. Eds. Jordi Ferrer Bertrán. José Juan Moreso and Diego Papayannis. Dodrecht: Springer. 161–75.

Brian LEITER, 2014: Legal Realism, Old and New. Valparaiso University Law Review 47. 949–63.

Brian LEITER, 2015a: Normativity of Naturalists. Philosophical Issues 25. 64–79.

Brian LEITER, 2015b: Marx, Law, Ideology, Legal Positivism. Virginia Law Review 101. 1179–96.

David MACARTHUR, 2010: Taking the Human Sciences Seriously. Naturalism and Normativity. Eds. Mario de Caro & David Macarthur. New York: Columbia University Press. 123–41.

Andrei MARMOR. 2011. Philosophy of Law. Princeton: Princeton University Press.

John MIKHAIL, 2012: Elements of Moral Cognition: Rawls’ Linguistic Analogy and the Cognitive Science of Moral and Legal Judgment. New York: Cambridge University Press.

John MIKHAIL, 2014: Any Animal Whatever? Harmful Battery and Its Elements as Building Blocks of Moral Cognition. Ethics 124. 750–86.

Stephen R. PERRY, 1995: Interpretation and Methodology in Legal Theory. Law and Interpretation: Essays in Legal Philosophy. Ed. Andrei Marmor. Oxford: Clarendon Press.

Stephen PERRY, 2009: Beyond the Distinction Between Positivism and Non-Positivism. Ratio Juris 22. 311–25.

Dale PETERSEN, 2011: The Moral Lives of Animals. New York: Bloomsbury.

Dan PRIEL, 2011a: Jurisprudence and Psychology. New Waves in Philosophy of Law. Ed. Maksymilian Del Mar. Basingstoke: Palgrave Macmillan. 77–99.

Dan PRIEL, 2011b: The Place of Legitimacy in Legal Theory. McGill Law Journal 57. 1–35.

Dan PRIEL, 2012: Jurisprudence Between Science and the Humanities. Washington University Jurisprudence Review 4. 269–323.

Dan PRIEL, 2013: The Scientific Model of Jurisprudence. Neutrality and Theory of Law. Eds. Jordi Ferrer Bertrán, José Juan Moreso and Diego Papayannis. Dodrecht: Springer. 239–54.

Dan PRIEL, 2015: Toward Classical Legal Positivism. Virginia Law Review 101. 987–1022.

Dan PRIEL, 2016: Holmes’s ‘Path of the Law’ as Non-Analytic Jurisprudence. University of Queensland Law Journal 35. 57–73.

Dan PRIEL, unpublished a: The Misguided Search for the Nature of Law. Available at

Dan PRIEL, unpublished b: The Prevalence of Interpretivism.

Joseph RAZ, 1990: Practical Reason and Norms. 2nd ed. Oxford: Oxford University Press.

Joseph RAZ, 2009a: Between Authority and Interpretation: On the Theory of Law and Practical Reason. Oxford: Oxford University Press.

Joseph RAZ, 2009b: The Authority of Law: Essays on Law and Morality. 2nd ed. Oxford: Oxford University Press.

Matt RIDLEY, 1998: The Origins of Virtue. London: Penguin.

Brian SKYRMS, 1996: Evolution of the Social Contract. Cambridge: Cambridge University Press.

Kim STERELNY, 2012: The Evolved Apprentice: How Human Evolution Made Humans Unique. Cambridge: MIT Press.

Robert SUGDEN, 2004: The Economics of Rights, Co-operation, and Welfare. 2nd ed. London: Routledge.

H. Peyton YOUNG, 1996: The Economics of Convention. Journal of Economic Perspectives 10. 105–22.

Top of page


1 One might try to distinguish between the normativity of law and the normativity of other social practices and argue that while the latter can be explained naturalistically, the former cannot. This, however, would not be a good idea for a legal positivist. It seems to me that the only way of reaching such a conclusion requires holding, first, that the normativity of morality (unlike the normativity of other human practices) cannot be explained naturalistically; and second, that the explanation of law’s normativity (unlike the normativity of other human practices) cannot be detached from an explanation of the normativity of morality. Such a view sounds very much like natural law theory.

2 In Priel (unpublished a) I draw the distinction between internalists and externalists, which is different but often corresponds to the distinction drawn here between weak and strong anti-naturalists.

3 It is not just that Hart called his book an essay of “descriptive sociology” (Hart 2012: vi); Hart also stated that the rule of recognition, perhaps his central idea and what determines what belongs to a legal system is an “empirical, though complex, question of fact” (Hart 2012: 292).

4 As mentioned in the text, this view owes much more to Kelsen than to Hart, as Kelsen’s so-called “pure theory of law” was meant to be devoid of any sociological component. As part of the trend toward strong anti-naturalism in contemporary legal philosophy, there is a marked tendency in some recent interpretations of Hart to “Kelsenify” him. Green (2012: xlv–xlvi) and Gardner (2012: 277), both strong anti-naturalists, interpret Hart as accepting this view as well. Regardless of the merits of this view, I argue elsewhere (using a somewhat different terminology) that this is an implausible reading of Hart’s work (Priel unpublished a: 31–35). It eliminates from Hart’s work precisely those aspects that he saw as the main grounds for distinguishing his work from Kelsen’s. See also note 3.

5 Many legal positivists (most of whom are, as mentioned, anti-naturalists) strenuously denied any such connection (see sources cited in Priel 2013: 275 note 13).

6 There is risk of terminological confusion here. It is not uncommon, especially in literature from three of four decades ago, to find natural law theory being called “naturalism.” In this sense, of course, legal positivism is the anti-naturalistic thesis and it is a tautology that natural law theory is naturalistic. To avoid this confusion, I do not use the term “naturalism” as a short hand for natural law theory.

7 (6) is not strictly speaking entailed by (5). (6) is true to the extent that we think that legal positivism and natural law exhaust the domain of possible (or plausible) general theories of law, such that one of them must be true. Even if not true, the argument in the text clearly paves the way in the direction of a certain version of legal positivism.

8 Even claims coming from cognitive science should be treated with caution. As was recently argued by Henrich et al. (2010) many purportedly universal findings are of limited general applicability because they are largely based on experiments with American college students who may be very unrepresentative of global population.

9 An even more radical critique of the argument I consider in the text would deny there is anything to analogize from, because it rejects the idea of a universal linguistic grammar. Evans and Levinson’s (2009) target article and responses present the competing views on this matter. Strictly speaking, the fact that universal linguistic grammar is false does not refute the existence of a universal moral grammar. However, much of the attraction of Mikhail’s position rests on the analogy between the two. If there is no universal linguistic grammar, universal moral grammar is likely to go as well.

10 The example highlights a point that I cannot explore here but which I think is one of the merits of this naturalistic reinterpretation of the legal positivist/natural law debate. As currently understood, the debate is “conceptual” and as such has little connection with debates on substantive areas of law, which are seen as normative (Marmor 2011: 10). By contrast, the view presented here shows the continuity between discussions on law in general and on (say) the foundations of property law.

11 Here Hart was at odds with strong anti-naturalists like Raz (1990: 56–58) who argued that no social account of normativity is possible. As Raz (1990: 53, 56) put it, a social account of normativity “deprives rules of their normative character”.

12 This view of law and the state as an artificial creation and as an instrument is also central to Bentham’s thinking (Dinwiddy 1989).

Top of page


Electronic reference

Dan Priel, « The possibility of naturalistic jurisprudence  », Revus [Online], 32 | 2017, Online since 18 June 2017, connection on 17 January 2018. URL : ; DOI : 10.4000/revus.3832

Top of page

About the author

Dan Priel

Associate professor at Osgoode Hall Law School, York University (Canada).

Address: Osgoode Hall Law School – Ignat Kaneff Building – 4700 Keele Street – Toronto, Ontario  M3J 1P3 – Canada


Top of page


All rights reserved

Top of page
  • OpenEdition Journals