Skip to navigation – Site map

HomeNuméros48Symposium on the internal legal p...The concept of law as a functiona...

Symposium on the internal legal positivism

The concept of law as a functional concept

Rodrigo E. Sánchez Brigido
This article is a translation of:
El concepto de derecho como concepto funcional [es]

Abstracts

In her book Positivismo jurídico "interno", Cristina Redondo claims that legal theory can be neutral from a moral point of view. An essential part of her argument is based on the criticism of an anti-positivist conception of law developed by Fernando Atria according to which, since law is a functional concept, no descriptive or morally neutral theory of law can be provided. The essay claims that, while Atria's conception of law encounters some difficulties, Redondo's criticisms miss the mark leaving their impact inconclusive. More importantly, there are two kinds of natural law theory that also claim that law is a functional concept but that do not face Atria's difficulties. One such theory seems particularly suitable for avoiding not only Redondo's criticism but also standard objections from the positivist camp. If this type of theory is correct, a theory of law cannot be morally neutral.

Top of page

Full text

1 Introduction

  • 1 Redondo 2018.
  • 2 A substantial part of the book is devoted to discussing skepticism. I shall not consider it here.

1Legal scholars have been debating the aspirations and limits of legal theory for quite a long time. The book Positivismo Jurídico “Interno” is an important contribution to that discussion.1 In her book, Cristina Redondo defends a positivist account of legal theory according to which it is possible to develop an empirical description of a normative object such as law. She also argues that anti-positivist accounts, which deny this possibility, are inconsistent or face serious difficulties.2

2Redondo’s book cogently presents the pros and cons of each kind of approach. It further explains with unusual clarity how the literature argues about the different ways in which legal theory can analyze its object of inquiry, and develops an argument in favor of the idea that such an approach can be morally neutral. Although the book has all these virtues, I believe that it also has an important weakness. Redondo bases much of her view on a criticism of an anti-positivist conception of law developed by Fernando Atria, according to which law is a functional concept such that it cannot be philosophically explained in a descriptive or morally neutral fashion. The problem is that, even though Atria’s account indeed faces several problems, Redondo’s criticisms miss the mark leaving their impact inconclusive. More importantly, there are two kinds of accounts of natural law that Redondo strikingly does not discuss, which on the idea of a functional concept, but that do not face the difficulties faced by Atria’s account. One of them relies on the idea of law as a functional kind, while the other uses the idea of law as a normative-functional concept. I do not intend to show that these views are correct. But I do believe that they should be seriously considered. In particular, the account that relies on the idea of the concept of law as a normative-functional concept looks particularly apt not only to meet Redondo’s criticisms but also to avoid standard objections from the positivist camp. If this kind of approach is sound, a theory of law cannot be neutral.

3 The paper is structured as follows. First, I will describe the theses that, according to Redondo, are shared both by positivism and anti-positivist theories of law (section 1). I will then consider their points of disagreement and I will focus my attention on the issue of whether an account can be descriptive or morally neutral (section 2). Redondo criticizes Atria’s anti-positivism and, although her general criticism is well oriented, I will claim that it does not really identify the main problem in Atria’s theory and that, in any event, it does not allow us to reject the view that law is not a functional concept (section 3). I will then consider the two other main types of accounts available in the literature that rely on the idea that law is a functional concept. The first has been defended by Michael Moore and claims that law is a functional kind. The second claims that the concept of law refers to an institution constituted by a normative standard of success and has been defended by, among others, Mark Murphy. Although my intention here is not to determine whether these accounts are correct, I will attempt to show that neither of them face the difficulties of Atria’s theory and that the second kind of theory is the most attractive. Considering this kind of approach is important because, if correct, a theory of law is necessarily incomplete and superficial unless it holds a view on the moral value of law (section 4).

2 Positivism and anti-positivism according to Redondo

  • 3 Redondo 2018:159.
  • 4 Redondo 2018: 164.
  • 5 Redondo 2018: 165.
  • 6 Redondo 2018: 163.

4To determine the way in which Redondo characterizes the dispute between positivism and anti-positivism it will be useful to start by describing her disagreement with Atria. According to Redondo, Atria’s anti-positivist account shares three ideas with positivism. On the one hand, a concept refers to a set of properties necessarily present in any instance of law (or in its paradigmatic instances), which allows one to distinguish between what is and what is not law.3 On the other hand, one of these properties is that law is an institutional entity that exists only if a social group believes that it exists and accepts it. That is, there is law only if there are acceptants.4 Finally, both Atria positivist accounts hold that a theory of law is not necessarily committed to the values accepted in a given legal system.5 Thus, the theory does not have to commit itself to the view that law is valuable or must be obeyed in virtue of its very existence. Put otherwise, legal theory must adopt a point of view that is different from the acceptants’ point of view.6

  • 7 The idea that legal theory should identify the necessary features of law is widely shared. For inst (...)
  • 8 Redondo 2018: 165.
  • 9 Redondo 2018: 166.

5Redondo then distinguishes positivism and Atria’s anti-positivism on the bases of the theses that they do not share. Although both accounts attempt to identify those features that law necessarily exhibits,7 positivism claims that the theory can do so in neutral terms (that is, without assigning moral value to its object of inquiry), while, for anti-positivism, the theory defines its object of inquiry in non-neutral terms.8 Call this the “problem of neutrality”. On the other hand, according to Redondo, there is a difference in relation to the theory’s direction of fit. According to positivism, the theory’s aim is to describe or inform about law’s essential features. While acceptants’ claims about the law have a world-to-mind direction of fit (or, in other words, they have an internal or constitutive relationship with law), the theory’s claims have a mind-to-world direction of fit.9 On the contrary, for an anti-positivist account such as Atria’s, the relation of fit of both acceptants and legal theorists is one and the same. A legal theorist cannot capture, analyze, or explain its object without, at the same time, demarcating it. Call this point of disagreement between positivism and Atria’s account “the problem of the direction of fit.”

  • 10 For a persuasive argument to the effect that natural law theorists such as Aquinas, Finnis, Moore o (...)

6This way of analyzing the dispute is important not only because it is the starting point for Redondo’s criticism of Atria, but also because it allows us to characterize the debate with anti-positivism in general. Indeed, no anti-positivist account has to deny that the concept of law refers to a set of properties, which are necessarily present in any instance of law, or that law is an institution that exists only if there are acceptants, or that jurisprudence is not necessarily committed to the values accepted in a real legal system.10

  • 11 Redondo claims correctly that it is implausible to think that legal theories are part of the set of (...)
  • 12 Later on I shall clarify why.

7With respect with the theses that are not shared, the non-neutrality thesis also characterizes anti-positivism as such. I believe, nonetheless, that it is extremely doubtful that that same thing occurs with the thesis related to the direction of fit. For instance, as far as I can see, well-known natural law theorists such as Finnis, Moore, or Murphy do not claim that jurisprudence cannot analyze the concept of law without simultaneously demarcating it. In fact, there are strong reasons to think that this idea is not only wrong (in this respect, Redondo’s arguments look sound)11 but that anti-positivism does not need to endorse it. The thesis does not really help ground anti-positivism’s basic commitments.12

8 Since I am interested here not only in analyzing Redondo’s criticisms of Atria’s view but also in the plausibility of anti-positivism as such, I will put the direction of fit thesis to the side and I will focus my attention on the neutrality thesis only.

3 The problem of neutrality

  • 13 Atria 2016: 112.
  • 14 Atria 2016: 115.
  • 15 Atria 2016: 119.
  • 16 Atria 2016: 126.
  • 17 Atria 2016: 133.

9It might be helpful to start with Redondo’s criticisms of Atria. Atria puts forward an argument in favor of the non-neutrality thesis, grounded on a distinction between kinds of concepts. The nature of any object X is that in virtue of which X is what it is, in such a way that the property in question is not alien to X but is in some sense in X.13 Now, according to Atria, a class of objects is a nominal class if the properties that define the objects do not share any characteristic on top of its denomination. On the other hand, a class of objects is a natural class if what is internal to the class (the class’s objects, one is inclined to think) is its structure or shape (structural class) or function (functional class). Nominal classes lack this nature. The possibility of something being called X but not being in fact an X is what distinguishes natural from nominal classes. In this framework, Atria claims that although there are structural concepts (such as the concept of water), legal concepts are functional concepts.14 For instance, something is a sanction if it fulfils the function of expressing blame.15 Legal concepts are, nevertheless, functional in a special sense: they refer to an object whose task can only be fulfilled if the object is identified by its structural properties and not by its function. For instance, the concept of a law (a general and abstract norm that is in everyone’s interest) is manifested in a structure (the legislative procedure) that allows identifying a law without referring to its function.16 And the theory should make the structure intelligible by referring to such function.17

10As should be clear, Atria’s argument attempts to defend the non-neutrality thesis. If legal concepts are functional and not structural concepts, the theory cannot adopt a neutral point of view when giving an account of law.

  • 18 Redondo also claims that Atria does not need the notion of a structural concept to develop a theory (...)
  • 19 Redondo 2018: 185.

11Redondo puts forward three criticisms against this theory.18 The first criticism is that Atria’s distinction between concepts is inadequate because Atria himself claims that the notion of a structural concept is “a necessary notion we cannot do away with if we wish to give an account of the institutional nature of law”.19 Even though he attempts to do the opposite, Atria in fact grants, Redondo says, that we cannot give up the notion of a structural concept and, in fact, recognizes that both structure and function are necessary.

  • 20 When claiming that there can be mixed concepts, I am only saying that there are concepts that are n (...)

12The scope of this criticism is not very clear but, it seems to me that Redondo is saying is that Atria is inconsistent when claiming that legal concepts are structural concepts, while at the same time also claiming that legal concepts are functional concepts. If this is an accurate account of Redondo's position, then I believe that this criticism is plausible but that it does not get us very far. Atria’s distinction between functional and structural concepts is in effect problematic. It claims that the two relevant categories are mutually exclusive, when it is in fact possible that there are concepts that are both functional and structural at the same time (e.g., a concept according to which an object is an X if it has certain properties that allow it to fulfil a certain function). The very reconstruction of legal concepts put forward by Atria, according to which they are functional in a particular way, shows that there are in fact “mixed” concepts, that is, concepts that are both functional and structural in nature. In other words, it may the case that for Atria legal concepts are not functional but mixed concepts.20 If this is what Redondo is trying to say, then I believe that she is right. The problem is, nevertheless, that she has offered no argument to show that legal concepts are not mixed concepts.

  • 21 Redondo 2018: 186, 189 fwd.
  • 22 Redondo 2018: 187.

13Redondo’s second criticism is that Atria’s account is committed to the wrong idea, that legal concepts cannot serve as criteria for individuating cases of legal institutions because, according to Atria, the institution’s content must be identified without taking its function into consideration.21 This consequence is paradoxical, says Redondo. Normally, having a concept implies having a criterion of individuation. In contrast, for Atria, to have a concept one needs a theory that makes it intelligible and explains it, although surprisingly, it does not serve as an individuation criterion. In fact, Redondo concludes, if we put this issue to one side, both positivism and Atria’s anti-positivism accept an institutional identification criterion based on the structure, even though it is fitting to rely on the function they play.22

14I have serious doubts about the scope of this criticism and tend to believe that it is wrong. It is useful to recall that Atria’s functional concepts are, in fact, mixed concepts in the sense explained above. The structure plays the role of an individuation criterion in Atria’s account because (as he understands it) the constitutive function of the object referred to by the concept demands that the structure plays that role. For instance, the concept of a law refers to an institution constituted by a function (a general and abstract norm that is in everyone’s interest) that manifests itself in a structure (the legislative procedure) that allows identifying something as a law without referring to its function. And this is because, in Atria’s terminology, if that were not the case, the structure would not make the realization of the function probable. Put otherwise, that concept refers to an institution constituted by a moral function (a law) that demands individuating the concept’s instances of application (the concrete dispositions that count as laws) by reference to a certain structure (the legislative procedure). The function, then, does serve as an identification criterion, though only indirectly. It is then not true, contrary to Redondo’s suggestion, that an account such as Atria’s presupposes that the concept cannot individuate its instances of application.

15Redondo’s last criticism is that it is not clear how Atria could avoid the threat of skepticism. In this sense, Redondo says that Atria’s theory faces cannot settle a disagreement about the true function that makes an institution intelligible if we have two or more interpretations (accounts) with equal or almost equal rational support. Redondo (2018: 192) asks:

How can one claim that not just any interpretation captures the true function of an institution (the function to which it is internally related) without admitting the contrast between propositions that describe or report the function that the institution in question in effect performs, regardless of the theory, and propositions that identify the functions that the institution should perform, according to the theory?

16This criticism is a bit misguided. If we are not provided with an example of two interpretations and an argument that shows that they both have equal theoretical credentials, then the view that there can be two or more interpretations of an institution with equal rational support is merely speculative. Presumably, Atria and other philosophers who belong to his school of thought believe that there are criteria to evaluate theories such that they all do not have the same rational support and, possibly, that only one theory (the correct one) has it. On the other hand, the criticism that is put forward as a question presupposes that Atria cannot distinguish between the function that a specific instance plays and the function that, according to the theory, law as such and that particular instance should play. But Atria does not have to deny that distinction. If the concept of law is a mixed concept, that does not deny that there can be legal institutions within which their morally valuable function is performed in a suboptimal, incomplete, or degraded fashion.

17At the end of the day, Redondo’s three criticisms are inconclusive. It does not look right that Atria’s functional concepts cannot have individuation criteria, nor are there strong reasons to believe that the threat of skepticism could not be avoided. And, even though it is true that Atria’s classification of concepts is a little bit too idiosyncratic, Redondo has not discarded the possibility that law is not a mixed concept.

  • 23 An anonymous referee suggested that Atria may not be interested in giving an account of the general (...)

18Of course, I am not saying that Atria’s thesis is correct. The key problem with Atria’s account, in my view, is that, even if the legal concepts he has in mind were mixed concepts (Atria focuses on concepts such as will, sanction, and law), the very concept of law is not mixed. The concept of law is an ordinary notion, not a legal notion. To be a mixed concept, that concept would have refer to an object (and nobody denies that it is an institution) whose function can be achieved only if the institution itself is identified by reference to certain structural properties and not by its function. But this does not look right at all. To make the case as favorable to Atria’s view as possible, assume that law is an institution that, as he says, attempts to make probable something naturally improbable (that we are not alienated). It is not true that, to achieve this, the institution itself must be identified by structural properties without referring to its function.23

  • 24 This book will be identified here as Atria 2016.

19In a recent paper Atria has replied to this objection. There (Atria 2020: 857-860), he argues that even though this is a point that his book La forma del Derecho (LFD) has dealt with,24

it may be useful to consider that it is the same type of argument that Raz employs in his theory of authority to move from his normal justification thesis to his preemption thesis: if the normal way of justifying the authority of law is that following its directives makes it more likely that their subjects conform to the reasons that already apply to them, then those directives must be identifiable without relying on those reasons because, if that were not so, the authority of law would be redundant. The same applies, as LFD claims (p. 154), to contractualist justifications of the state (Hobbes, Locke). To take Sánchez Brigido’s idea of my own view, the law makes a non-alienated life a probable prospect. Then one asks why that would imply that the structure is opaque in relation to its function. A possible answer, discussed in LFD, (pp. 402-406), is that the law makes possible that, when our interests are in mutual conflict, something that we can recognize as a common interest emerge. This requires procedures for creating a common will, the outputs of which will be recognized as such not because we are in agreement from a substantive point of view but because the procedural requirements have been satisfied.

20In my opinion, Atria here misidentifies the scope of the objection. The objection is that there is a crucial difference between legal concepts, such as the concept of a law (such as a legal disposition), and the concept of law (the institution). For instance, even if it were true that I can only identify a certain norm by focusing on its structure (e.g., that it was enacted by a legislative procedure), because if that were not the case then its function (being a general norm in everyone’s interest) could not be achieved, it does not look right that I can only identify a certain legal system (e.g., the Italian one) focusing on its structure (e.g., there is a general rule of recognition accepted by certain officials, and the Italian people follows the norm identified by such a rule) without focusing on its function (e.g., to achieve that the Italians are not alienated). This is because for the system to achieve its function, this cannot be the case. I can perfectly identify that in Italy there is a legal system by focusing both on its structure and its function, without that making it improbable that its function be achieved. Relying on Atria’s wording in his reply, I can identify that there is a legal system in Italy that has procedures of creation of a common will that allows the Italians to recognize its results as their own. When I do that, I have identified a legal system (that is, the object referred to by the concept of law) focusing both on its structure and function without that conspiring against the function achieving its aim. Put otherwise, I see no way of saying that the very concept of law is a mixed concept.

21Now, a criticism such as the one I have just put forward might be decisive against Atria’s theory but it is still inconclusive in relation to anti-positivism as such. This is because it can be the case that law is a functional concept in a sense less idiosyncratic than the one defends. One may argue, for instance, that the concept refers to an institution that plays a moral role without saying that it must be identified by its structure without referring to its function. And it is plausible to think that, if the concept of law is functional in that sense and that the relevant function is a moral one, the non-neutrality thesis is correct.

22It is true that Redondo puts forward theses that exceed what Atria says about functional concepts. In this sense, she claims that positivism does not deny that legal institutions necessarily play one or several social functions, such as regulating behavior, and neither does she deny that one can assign them moral value. What she says is that:

  • 25 For a similar argument, see Ehrenberg 2009: 101.

these general functions…are also performed by other normative institutions such as customs, morality or religion; they are not distinctive of law, i.e. they do not allow us to tell what is law and what it not. At the same time, more specific functions change from one instance of law to the other and, accordingly, these functions cannot be considered a common feature of all instances. In other words, what is essential of law is not the functions that it performs but the specific way in which it does. (Redondo 2018: 184)25

  • 26 For a good description of the different functions that natural law theorists such as Aquinas, Finni (...)

23As far as I can see, this argument is too hasty. On the one hand, the view that the law shares general functions with other institutions is underdeveloped. There are well-known naturalist accounts that defend well-articulated ideas about what the relevant and distinctive function of law is.26 Yet Redondo has not said a word about them. On the other hand, even if it were true that the law shares certain general functions with other institutions and that what is distinctive about it —as Redondo claims— is the way in which it does that, it is not clear why the theory could not refer to this feature in a morally neutral fashion. One might claim that one would not be able to fully understand the way in which law plays its function without understanding that very function, which in turn makes plausible the non-neutrality thesis if, as anti-positivists argue, that function is morally relevant.

24In short, Redondo’s criticisms of anti-positivism are also inconclusive. It is indeed necessary to further develop the arguments instead of discarding the possibility that the concept of law is functional. In fact, as I will argue now, there are other theories that rely on the idea of law as a functional concept that Redondo has not examined and that do not face the problems Atria’s account faces.

4 Two alternative accounts

  • 27 Duke 2016: 486.

25Although there are several ways of characterizing naturalism, perhaps the best summary of this view says that there is a necessary link between law and the fundamental requirements of practical reason.27 In the recent literature, there are two kinds of theories that rely on this thesis, which are grounded in the notion of function to develop the specific characterization of the kind of necessity involved. The first one uses the notion of functional kind while the second uses the notion of normative-functional concepts.

4.1 “Law” as a functional kind

  • 28 Moore 1992: 206-208.

26The theory of law that uses the notion of functional kind is the one first put forward by Michael Moore. It says that “law” is a term that refers to an object directly, without any third element (certain properties designated by the concept) serving as an intermediary. The meaning of “law,” according to this view, is based on the nature of the thing referred to rather than by the concept. This kind of metaphysical necessity depends only on what the law is, in the same vein that happens with natural kind terms. That water is H2O is a metaphysical truth because something cannot be water if it were not H2O. But “law” is different from natural kind terms because, says Moore, it refers to artifacts whose nature is given only by its structure. Moore calls them “functional kind terms.” They are terms that refer to artifacts whose nature is given by the function they play and in which the structure is in the service of that function.28

  • 29 Moore 1992: 208-213.

27Moore is aware that the identification of the function of an object is a controversial issue, yet he believes that we are familiarized with this kind of approach. For instance, terms that refer to organs that are members of the human body are of this kind. Consider the human heart. In order to identify the heart’s function, according to Moore, we should involve ourselves in a procedure subject to review on the basis of the available evidence that identifies the good the body produces and the way in which the organ at stake causally contributes to the promotion of such good. According to the best theory available, says Moore, the heart’s function is to pump blood because physical health is the goal of the human body (i.e., that for which the body is good) and pumping blood causally contributes to achieving that goal.29

  • 30 Moore 1992: 213-216.
  • 31 Moore 1992: 218.

28Moore says that the term “law” refers to an institution that also has a function related to the human good and its ways of achieving it. The best account available, which according to Moore is Finnis’s, shows that law’s function is achieving the common good understood as a set of conditions that allow a given community’s members to achieve for themselves goals that are reasonable, or to reasonably realize certain values that can only be achieved if they act together.30 Moore then concludes that, “given certain metaphysical truths about human nature, and given certain necessary truths about morality that make certain goals of law good goals, then the law must have certain structural metaphysical features”.31 And such structural features are those that serve to achieve its function.

29As should be clear, if this theory is correct, then the non-neutrality thesis cannot be correct. This is because to explain law, in a sense that fits both positivism and anti-positivism, one should explain that that makes it be what it is. That what makes the law what it is, according to Moore, is something that is metaphysically determined by the function that law plays, which is something that implies conceiving of it as an instrument for the realization of the human good. Determining what that value is necessarily requires, in turn, relying on the best moral theory available. Thus, the theory of law cannot explain its object (law as an institution) unless it adopts a moral point of view. In other words, it cannot explain law in a neutral fashion.

30This kind of approach does not face the difficulty faced by Atria’s account because it does not say that the institution must be identified only by its structure, without reference to its function. On the other hand, Finnis’ view, on which Moore relies, has a reasonably concrete and articulated elaboration of the idea of the common good, and Moore puts forward a reasonably well-constructed argument about why its function is specific and distinctive of law. That function would allow us to distinguish law from other institutions. For this reason, Moore’s view, if correct, could also deal with Redondo’s criticisms. In short, it is a functional approach that does not face the problems faced by Atria’s account and which, if correct, would show that Redondo’s defense of the non-neutrality thesis is inadequate.

31This does not mean, of course, that Moore’s view is not without problems of its own. On the one hand, saying that law is a functional kind implies adopting a particularly strong metaphysical commitment. It assumes that law has a nature in some sense, in a similar vein to other natural kind terms (H20, etc.), which is controversial to say the least. On the other hand, the accounts of the good and the right that Moore’s approach relies on, as well as his explanation of the function that law plays in light of such accounts, is controversial. Be that as it may, if one wishes to defend the neutrality thesis, a view such as Moore’s should be seriously considered and eventually discarded.

4.2 “Law” as normative-functional notion

32The second kind of functionalist argument does not say, as Moore does, that “law” is a functional kind. On the contrary, the concept of law refers to certain properties, and these properties are those that allow identifying its object. In an elaboration that I find particularly attractive, the argument is based on two premises.

  • 32 Semantic externalism is elaborated in Burge 1979: 73–121.
  • 33 About the relevance of the variety of concepts for a theory of law, see Stoljar 2013: 230.
  • 34 Murphy 2005: 21. A similar argument can be found in moral theory by the so called “constitutivist” (...)
  • 35 Duke 2016: 485-509.

33First, it is possible that there are different types of concepts (of natural kinds, of kinds identified according to the so-called “semantic externalism,”32 interpretative concepts, prototypical concepts, amongst others).33 The concept of law, together with others, is of a special class. It is a functional concept in the sense that it designates a property that refers to an item constituted by a normative standard of success. Call these concepts, lacking a better name, “normative-functional concepts.” For instance, to rely on a particular version of this premise, Mark Murphy argues that in the natural law tradition, law necessarily designates a rational pattern of conduct; that standard (rationality) is internal, it is a constitutive part of the nature of the object. The kind of necessity at stake, then, is not that of a “a triangle necessarily has three sides”. Rather, it is of the kind purporting claims like “necessarily cheetahs are fast.” An object that does not have three sides is not a triangle but, according to this way of understanding things, a legal system that does not meet the standard is still a legal system, only it is one that is defective in relation to the standard. On this way of understanding the question, according to Murphy, necessity is predicated of the kind, not of the individuals that belong to the kind.34 Other philosophers defend a similar view.35

  • 36 Murphy 2006: 23. Moreover, Murphy claims that, if one accepts that being constitutively capable of (...)
  • 37 This is one of the ways in which the law is connected, according to this view, to morality. This do (...)

34Second, the standard in play is genuinely normative. This means that the concept designates an entity (in the case of law, an institution) such that satisfying the standard is relevant for us from a normative or, more generally, a practical point of view. That the standard is satisfied is relevant for the question of how we should act. This entails that any acceptable theory about an institution should be grounded in the dominion of practical philosophy. And given that analyzing a concept involves identifying necessary properties that distinguish law from other institutions, adopting a moral point of view to identify law is indispensable. For instance, Murphy says that a theory of law is not adequate “without a full understanding of the requirements of practical reasonableness”.36 In other words, if the concept of law is functional in the sense emphasized, we must determine the relevant normative standard that distinguishes law from other notions, and determine why it is genuinely normative. Any other approach is incomplete and superficial.37 Thus, if this view is correct, the neutrality thesis cannot be correct.

  • 38 The idea can be elaborated by employing Aristotle’s or Aquinas’s metaphysics, as Duke suggests (Duk (...)

35This kind of approach has several advantages. On the one hand, it avoids the difficulties that marred Atria’s view because it does not say, as Atria did, that legal systems must be identified without reference to the function because, if not identified in that way, they could not achieve their function. Contrary to what happens with Moore’s view, it does not claim that the concept of law is a natural kind term, and as a consequence, it does not have a dubious or strong metaphysical commitment. The theory only says that the concept of law belongs to the dominion of practical reasoning, and there are several ways of understanding that idea.38 Ex ante, there is no reason to doubt its conceptual or metaphysical credentials. On the other hand, this kind of theory can plausibly avoid several positivist objections.

  • 39 Ehrenberg thinks that a functional explanation need not adopt a view as to whether the function is (...)

36Thus, the positivist cannot argue that it is possible to explain law as one could explain any other artifact, that is, by explaining what people believe about is function and what people believe about the alleged value of that function.39 This possibility is blocked because, as I said, the theory must necessarily explain its normative standard and what makes it genuinely binding.

  • 40 Leiter claims that, understood charitably, Finnis’ project consists of trying to explain the featur (...)
  • 41 Murphy 2006: 23.

37The positivist cannot say that it is merely a theory about how ideal law should be, and that this is not problematic because no one really denies that there can be morally ideal instances of law.40 This is because the theory claims that, unless one explains the morally ideal instances of law, one cannot explain its real instances. As Murphy says, one cannot have a full account without a complete theory of the way in which law can be defective, and one cannot have such a complete theory without a full understanding of the requirements of practical reasonableness.41

  • 42 Redondo 2018: 203-208.
  • 43 Redondo 2018: 208-210.
  • 44 Redondo 2018: 244.

38 Further, because of the previous point, the theory correctly deals with the habitual positivist objection according to which the view would be committed to the implausible idea that any real instance of law is morally valuable. Redondo also seems to rely on this line of criticism in the last part of her book. She argues that there are different points of view from which the philosopher may explain the law. The first distinction between different points of view is semantic. Thus, the point of view she calls “external1” is the point of view that attempts to describe and explain only the empirical or behavioral aspects of an institution. The point of view referred to as “internal1” explains the way in which those that participate in the institution refer to it.42 In turn, one can classify points of view in terms of the different pragmatic assumptions that one can assume, that is, depending on whether one morally accepts the institution or not. Thus, the point of view called “internal2” of those who believe that the institution is justified, while those that adopt “external2” do not assume that.43 In my view, Redondo believes that her classification is exhaustive and implies that an anti-positivist must adopt the point of view according to which, if one explains social institutions, one accepts that they are justified (“internal2”). For the anti-positivist account that I just sketched, this is false. The classification is not exhaustive because there is another possible normative point of view, and therefore, the anti-positivist must not accept the idea that any real institution is morally justified. This, incidentally, shows why it is not true, contrary to what Redondo says,44 that the anti-positivist must admit that the point of view of the one that describes the law is that of an acceptant. Put otherwise, there is no reason to accept that the direction of fit of theory is the same as the direction of fit (constitutive) of the acceptant. This is why, as I said above, an anti-positivist may not accept this idea.

39 Finally, the positivist cannot say that one can choose philosophically describing law from different points of view depending on the objective one pursues. I tend to think that Redondo believes that the correct point of view is optional because when she introduces the different points of view at the end of her book, she does not say that only one of them is correct. This suggests that this issue is merely optional. But according to the argument I just sketched, this is false. The only possible point of view is the moral point of view. Any other approach is necessarily superficial and incomplete.

40 To sum up, the view I just sketch does not carry any of the problems Atria’s faces and looks perfectly able, if correct, to deal with the usual positivist objections. More importantly, if correct, the non-neutrality thesis is true.

5 By way of conclusion

41The theories that employ the idea of function are, as I see it, the most recent and interesting attempt to deploy the non-neutrality thesis. Redondo criticizes this kind of approach and she does so with her characteristic deepness and philosophical clarity. I have tried to show, however, that her argument against functionalism faces two crucial problems.

42On the one hand, Redondo criticizes Atria’s version of functionalism but, even though the criticism is well oriented, it does not identify the main difficulty of that theory and, in any case, it does not show that the concept of law is not a functional concept.

43On the other hand, Redondo’s criticism is almost exclusively centered on Atria and ignores other functionalist theories. One of them employs the idea of law as a functional kind and the other considers law as a normative-functional concept. The latter kind of theory looks attractive for many reasons: it does not face the difficulties of Atria’s theory, it does not have unnecessary metaphysical commitments, and it seems fit to avoid usual positivist objections.

44These considerations show, in my view, the main weakness of Redondo’s excellent book. It does not propose a direct argument to show that the concept of law cannot be functional, nor does it consider the version of functionalist approaches that seems more promising. The point is important because this sort of account belongs to an old tradition —the natural law theory— and there is no straightforward discussion of the main thesis of that tradition: law is a concept that belongs to the domain of practical reasonableness and, consequently, a philosophical account of the item referred by the concept cannot be provided unless one adopts a practical point of view, that is, unless one determines how the law bears on our reasons for action.

—Acknowledgment.— A previous version of this book was discussed in a workshop organized by Paula Gaido at the Asociación Argentina de Filosofía del Derecho (University of Buenos Aires, Law Faculty, September 2019). I thank Juan Iosa and two anonymous referees for their valuable comments on an earlier draft.

Top of page

Bibliography

Atria, F. (2016). La forma del derecho. Marcial Pons.

Atria, F. (2020). Una réplica para seguir conversando. En J. Gallego Saade (ed.), El derecho y sus construcciones. Diálogos con Fernando Atria (pp. 803-1015). Palestra.

Burge, T. (1979), Individualism and the Mental. Studies in Epistemology: Midwest Studies in Philosophy, 4(1), 73-121.

Dickson, J. (2001). Evaluation and Legal Theory. Hart Publishing.

Duke, G. (2016). The Weak Natural Law Thesis and the Common Good. Law and Philosophy 35, 485-509.

Ehrenberg, K. (2009). Defending the Possibility of a Neutral Functional Theory of Law. Oxford Journal of Legal Studies, 29 (1), 91-113.

Gardner, J. (2001). Legal Positivism: 5 ½ Myths. American Journal of Jurisprudence, 46(1), 199-227.

Katsafanas, P. (2018). Constitutivism about Practical Reason. En D. Star (ed.), Oxford Handbook of Reasons and Normativity (pp. 367-391). Oxford University Press.

Leiter, B. (2018). Legal Positivism about Artifact Law: a Retrospective Assessment. En L. Burazin, K. Himma & C. Roversi (Eds.), Law as an Artifact (pp. 13-28). Oxford University Press.

Moore, M. (1992). Law as a Functional Kind. En R. George (Ed.), Natural Law Theories (pp. 188-244). Oxford University Press.

Murphy, M. (2004). Natural Law Theory. En M. Golding & W. Edmunson (Eds.), The Blackwell Guide of Law and Legal Theory (pp. 15-28). Blackwell Publishing.

Murphy, M. (2006). Natural Law in Jurisprudence and Politics. Cambridge University Press.

Murphy, M. (2013). The Explanatory Role of the Weak Natural Law Thesis. En W. Waluchow & S. Sciaraffa (Eds.), Philosophical Foundations of the Nature of Law (pp. 3-21). Oxford University Press.

Raz, J. (2004). Can There Be a Theory of Law. En E. Martin & W. Edmundson (Eds.), The Blackwell Guide to the Philosophy of Law and Legal Theory (pp. 324-342). Blackwell Publishing.

Redondo, C. (2018). Positivismo jurídico “interno”. Klub Revus.

Sánchez Brigido, R. (2020). Atria sobre análisis conceptual. Hacia una teoría del derecho no descriptiva. En J. Gallego Saade (Ed.), El derecho y sus construcciones. Diálogos con Fernando Atria (pp. 114-138). Palestra Editores.

Shapiro, S. (2011). Legality. Harvard University Press.

Stoljar, N. (2013). What Do We Want Law to Be: Philosophical Analysis and the Concept of Law. En W. Waluchow & S. Sciaraffa (Eds.), Philosophical Foundations of the Nature of Law (pp. 230-256). Oxford University Press.

Top of page

Notes

1 Redondo 2018.

2 A substantial part of the book is devoted to discussing skepticism. I shall not consider it here.

3 Redondo 2018:159.

4 Redondo 2018: 164.

5 Redondo 2018: 165.

6 Redondo 2018: 163.

7 The idea that legal theory should identify the necessary features of law is widely shared. For instance, for many positivists the theory of law is a set of propositions about the law that are necessarily true and adequately explain the nature of law (cfr. Raz 2004: 324; Dickson 2001: 17). There are many ways of conceiving what “explaining the nature” means, but under any plausible conception, it involves the task of identifying the features that distinguish the law from other phenomena. On the other hand, as the argument proposed in the text shows, natural law theories also accept this way of understanding legal theory’s task.

8 Redondo 2018: 165.

9 Redondo 2018: 166.

10 For a persuasive argument to the effect that natural law theorists such as Aquinas, Finnis, Moore o Murphy share this view, see Duke 2016: 485-509.

11 Redondo claims correctly that it is implausible to think that legal theories are part of the set of beliefs constitutive of law. Thinking so would be incompatible with claiming that the law is a social entity, not a theoretical entity, and, besides, denying that a theory can be descriptive is self-defeating (Redondo 2018: 172). I have put forward arguments to the same effect elsewhere (Sánchez Brigido 2020: 114-122).

12 Later on I shall clarify why.

13 Atria 2016: 112.

14 Atria 2016: 115.

15 Atria 2016: 119.

16 Atria 2016: 126.

17 Atria 2016: 133.

18 Redondo also claims that Atria does not need the notion of a structural concept to develop a theory of adjudication (Redondo 2018: 187) but, in my view, her argument is too schematic and brief. Besides, its scope is very limited because it relates to the concept of adjudication only. In any case, it seems clear that the thesis of non-neutrality would survive even if this argument were successful.

19 Redondo 2018: 185.

20 When claiming that there can be mixed concepts, I am only saying that there are concepts that are not functional in the sense in which Atria understands this notion, since part of what is internal to them is their structure and not only their function. I am not claiming that structure and function are independent of each other. In fact, Atria criticizes such a view because it ignores that, as he claims, the structure makes the function possible (Atria 2016: 123). On the other hand, when claiming that Atria’s concepts are mixed concepts I am not ignoring Atria’s main idea, namely, that these concepts refer to legal forms that make what is improbable probable. A mixed concept, in the sense I am referring to in the text, still refers to an institution in which structure and function work as Atria supposes.

21 Redondo 2018: 186, 189 fwd.

22 Redondo 2018: 187.

23 An anonymous referee suggested that Atria may not be interested in giving an account of the general idea of law when developing his theory of concepts. Although Atria’s book may be read in that way, I do not think that it is Atria’s view. When making the distinction between functional and nominal concepts, Atria is explicit in claiming that the very concept of law is a functional concept: “It is in virtue of the structure of water that our answer to the question about the identity of water is correct or incorrect. And if the question goes up one level, why H2O?, the answer may be: simply because this is how the world is. Our explanation of the nature of water leaves water just as it was before explaining it. This means that the analysis of the concept of water may take the concept (its nature) for granted, and it may suppose it is a datum that exists previous to the analysis (in the sense that our conceptual explanations may be refuted, for instance, by experimentation, that is, by looking at what water simply is). And it is precisely here where the law is different from water. The law —and the same happens with legal concepts such as contract, felony, testament—does not have a nature before the analysis that may work as a criterion of correction by way of experimentation" (Redondo 2018: 64, emphasis added). Besides, as I claim in the text below, when answering my objection Atria concedes that the concept of law is a functional concept. Finally, in any case, if the concept of law were not a functional concept (or a mixed concept), the considerations that I mention in the text below would apply.

24 This book will be identified here as Atria 2016.

25 For a similar argument, see Ehrenberg 2009: 101.

26 For a good description of the different functions that natural law theorists such as Aquinas, Finnis, Moore, and Murphy attribute to the law, and of how they are distinctive functions, see Duke 2016: 495-502.

27 Duke 2016: 486.

28 Moore 1992: 206-208.

29 Moore 1992: 208-213.

30 Moore 1992: 213-216.

31 Moore 1992: 218.

32 Semantic externalism is elaborated in Burge 1979: 73–121.

33 About the relevance of the variety of concepts for a theory of law, see Stoljar 2013: 230.

34 Murphy 2005: 21. A similar argument can be found in moral theory by the so called “constitutivist” approaches. See Katsafanas 2018.

35 Duke 2016: 485-509.

36 Murphy 2006: 23. Moreover, Murphy claims that, if one accepts that being constitutively capable of being a rational standard is not only a merit but also part of the existence conditions of law, then a proposition about these conditions includes reference to the merits. Murphy 2013: 19-20.

37 This is one of the ways in which the law is connected, according to this view, to morality. This does not, however, mean that this is the sense in which positivism is different from natural law theory from a historical point of view. For instance, one way in which this difference is traditionally expressed claims that it is not necessary for a norm to be legally valid, that it satisfy a moral criterion, or that it is necessary that it does not satisfy a moral criterion (Leiter 2018: 7). The first sense may be shared by a natural law theorist. The second sense, depending on its precise content, possibly not. Cfr. Gardner 2001: 201.

38 The idea can be elaborated by employing Aristotle’s or Aquinas’s metaphysics, as Duke suggests (Duke 2016: 491). Or, one can appeal to the notion of constitutive concepts (cfr. Katsafanas 2018).

39 Ehrenberg thinks that a functional explanation need not adopt a view as to whether the function is really valuable. He claims that there is a difference between a reasonable and prudential point of view, one that claims that “the law is a good thing because it does good things”, from a theoretical point of view, which claims that “the law is an important thing to study and understand because people believe that it does good things” (Ehrenberg 2009: 91-113, 111).

40 Leiter claims that, understood charitably, Finnis’ project consists of trying to explain the features of morally ideal legal systems and that this need not be disputed by a positivist (Leiter 2018: 21). For a similar argument, see Shapiro 2011: 408-409.

41 Murphy 2006: 23.

42 Redondo 2018: 203-208.

43 Redondo 2018: 208-210.

44 Redondo 2018: 244.

Top of page

References

Electronic reference

Rodrigo E. Sánchez Brigido, “The concept of law as a functional concept”Revus [Online], 48 | 2022, Online since 31 August 2022, connection on 01 October 2023. URL: http://journals.openedition.org/revus/8369; DOI: https://doi.org/10.4000/revus.8369

Top of page

About the author

Rodrigo E. Sánchez Brigido

DPhil (Oxford). Professor of Philosophy of Law at the University of Cordoba (Argentina). Professor of Fundamental Rights at the University San Andrés (Argentina).

Address: M. Acha 1908 - CP 1437 - Ciudad Autónoma de Buenos Aires - Argentina

E-mail: rodrigosanchezb74 [at] gmail.com

Top of page

Copyright

CC-BY-SA-4.0

The text only may be used under licence CC-BY-SA-4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search