Karl Olivecrona s Legal Philosophy: a Critical Appraisal

Transcription

1 1 Karl Olivecrona s Legal Philosophy: a Critical Appraisal TORBEN SPAAK * * Torben Spaak, Professor in Jurisprudence, Department of Law, Uppsala University (Sweden). This article reports research carried out under the auspices of the Bank of Sweden Tercentenary Foundation. I would like to thank Åke Frändberg and Uta Bindreiter for helpful comments on the article. I would also like to thank Jes Bjarup, Michael Steven Green, and Thomas Mautner as well as the participants in the advanced seminar in practical philosophy at the Department of Philosophy, Uppsala University, and the participants in the advanced seminar in jurisprudence, at the Department of Law, Uppsala University, for helpful comments on an earlier version of the article. In addition, I would like to thank the participants in a seminar on Karl Olivecrona s legal philosophy at the Department of Law, Stockholm University 20 November 2007, in a seminar on Karl Olivecrona s legal philosophy at the Department of Law, Lund University, 27 February 2008, and in a workshop on Hans Kelsen s legal philosophy at the Department of Law, University of Oslo, 25 November 2007, for helpful comments on my presentation of the ideas put forward in the article. Last but not least, I would like to thank Robert Carroll for checking my English. The usual caveat applies, however: The author alone is responsible for any remaining mistakes and imperfections.

2 2 1. Introduction Karl Olivecrona and Alf Ross were the most prominent of the Scandinavian Realists, who were active from the late 1920 s into the 1970 s. While Ross was better known on the international arena, and also more appreciated as a legal philosopher, 1 Olivecrona was in my view a more interesting thinker. Hence I offer in this article an overview and a critical appraisal of Olivecrona s legal philosophy, and argue (i) that Olivecrona s legal philosophy, especially the critique of the view that the law has binding force, the analysis of the concept and function of a legal rule, and the idea that the law is a matter of organized force, is a significant contribution to twentieth century legal philosophy. I also argue, on a more critical note, (ii) that Olivecrona s thoughts on judicial law-making are somewhat confused, (iiii) that Olivecrona fails to substantiate some of his most important empirical claims, such as the claim that imperatives have a suggestive character, and (iv) that the distinction espoused by Olivecrona between the truth and the correctness of legal statements is problematic but not needed in Olivecrona s legal philosophy. 1 Ross s most important publications are all available in English: Ross (1946; 1957; 1959; 1968).

3 I begin with a few words about Olivecrona s main publications (Section 2), continue by 3 introducing an important theme in Olivecrona s legal philosophy, viz. naturalism (Section 3), and say a few words about Olivecrona s meta-ethics (Section 4). I proceed to discuss four distinct but related topics in Olivecrona s legal philosophy, viz. the critique of the view that the law has binding force (Section 5), the view that legal rules are as so-called independent imperatives and that their function is to cause human behavior (Section 6), the analysis of the concept of a right (Section 7), and the view that the law is a matter of organized force (Section 8). This way of dividing up the subject matter tracks the presentation in the First Edition of Law as Fact and allows us to see how Olivecrona s legal philosophy developed over the years. I add a few words about Olivecrona s thoughts on judicial law-making, which are to be found in the Second Edition of Law as Fact (Section 9). Having done that, I discuss the distinction between the truth and the correctness of legal statements (Section 10). I conclude with an assessment of Olivecrona s legal philosophy, conceived of as a contribution to legal philosophy in the twentieth century (Section 11).

4 4 2. OLIVECRONA S MAIN PUBLICATIONS Olivecrona was born in Uppsala, Sweden, in 1897, and died in Lund, also Sweden, in Although he had already published a few short articles in legal philosophy, notably a piece on legal concepts (Olivecrona 1928), his first major publication in the field of legal philosophy was the First Edition of Law as Fact (Olivecrona 1939), which saw the light of day in 1939 the year in which World War II broke out and in which Olivecrona s mentor and philosophical hero, Axel Hägerström, passed away. After the publication of the First Edition of Law as Fact, Olivecrona began devoting himself to legal philosophy in earnest. His most important publications include Om lagen och staten [On Law and the State], 2 which is a slightly expanded Swedish version of the First Edition of Law as Fact; Lagens imperativ [The Imperative of the Law]; 3 Realism and Idealism: Some Reflections on the Cardinal Point in Legal Philosophy ; 4 The Problem of the 2 Olivecrona (1940). 3 Olivecrona (1942). 4 Olivecrona (1951).

5 Monetary Unit; 5 Rätt och dom [Law and Judgment]; 6 Legal Language and Legal Reality ; 7 5 The Imperative Element in the Law ; 8 Grundtankar hos Hägerström och Lundstedt [Fundamental Ideas of Hägerström and Lundstedt]; 9 the Second Edition of Law as Fact; 10 and Rättsordningen [The Legal Order], 11 which is a Swedish version of the Second Edition of Law as Fact, where the latter is a slightly expanded version of the former. His last legalphilosophical publication was an article on Axel Hägerström and natural law theory, which was published in a collection of essays on the legacy of the Uppsala School of philosophy (Olivecrona 1978). I shall base my overview of Olivecrona s legal philosophy mainly on Olivecrona s two monographs, viz. Olivecrona (1939; 1940) and Olivecrona (1971; 1976). 5 Olivecrona (1957). 6 Olivecrona (1960). 7 Olivecrona (1962). 8 Olivecrona ( ). 9 Olivecrona (1964). 10 Olivecrona (1971). 11 Olivecrona (1976).

6 6 3. THE INFLUENCE OF THE UPPSALA SCHOOL: NATURALISM Olivecrona s legal philosophy reflects the influence of the Uppsala school of philosophy, especially as put forward by Axel Hägerström, and Olivecrona was more than willing to acknowledge in a general way the influence of Hägerström s ideas on his legal philosophy. Here is how he expressed his feelings of indebtedness to Hägerström in the preface to the First Edition of Law as Fact. When this book was going to the press news arrived of the death of Professor Axel Hägerström. He was my revered and beloved master. I cannot make any attempt here to describe the nature and extent of his philosophical research, which will certainly in time be more widely known and appreciated than it is now. For my personal part it is enough to say that my endeavour to treat law as fact could not have been made without the basis supplied by his work. (1939, Preface). The Uppsala philosophers, led by Hägerström and Adolf Phalén, maintained (i) that conceptual analysis is a central philosophical task, (ii) that subjectivism (conceived of as the view that any perceived or imagined entity exists, at least partly, in someone s mind) is

7 false, (iii) that metaphysics (conceived of as the view that there is a real reality beyond the 7 world of time and space) is false, even pernicious, and (iv) that there are no objective values. 12 Although Olivecrona refrained from abstract discussions of these tenets, it is clear from his writings that he shared the view that conceptual analysis is a central legal-philosophical task, that metaphysics is false, even pernicious, and that there are no objective values. That he embraced (i) is clear from the fact that he devoted his writings to analysis of fundamental legal concepts, such as law, legal rule, right, duty, and court judgment. That he embraced (iii) and (iv) is also clear from his writings. Thus the rejection of metaphysics is explicit in his programmatic statement of his naturalism (see below) and in his critique of the view that the law has binding force, and his belief that there are no objective values is explicit, inter alia, in his critique of the view that the law has binding force, in his analysis of the function of so-called independent imperatives, and in his discussion of judicial lawmaking. The second tenet, the critique of subjectivism, does not seem to have played much of a role in Olivecrona s legal philosophy. Let us now take a closer look at the philosophical position that is commonly referred to as naturalism, and which appears above all in thesis (iii) above. Although the term naturalism 12 For more on the Uppsala School of philosophy, see Oxenstierna (1938).

8 appears to lack a definite meaning in contemporary philosophy (Papineau 2007, 1; Bedau ), writers on naturalism make a fundamental distinction between (i) ontological (or metaphysical) and (ii) methodological (or epistemological) naturalism. John Post (1999, 596-7), for example, explains that metaphysical (or ontological) naturalism is the view that everything is composed of natural entities... whose properties determine all the properties of whatever it is that exists, and that methodological naturalism is the view that acceptable methods of justification and explanation are continuous, in some sense, with those in science. 13 Ontological naturalism is thus a thesis about the nature of what exists: there are only natural entities. But what is a natural entity? I shall assume that it is an entity of the type that is studied by the social or the natural sciences, 14 though I recognize that it is difficult to find a fully acceptable characterization of natural entities. 15 On a more fundamental level, we might perhaps say that a natural entity is an entity that can be found in (what I shall refer to as) the 13 See also Wagner & Warner (1993, 12). 14 This seems to be the view taken in Brink (1989, 22-3) and in Lenman (2008). 15 Discussing moral non-naturalism, Ridge (2008, 4) calls the attempt to choose between the many available ways of making the distinction between natural and non-natural properties a fool s errand.

9 all-encompassing spatio-temporal framework. 16 On this analysis, if a contemplated entity, 9 such as God, a natural number, a scientific theory, or a legal norm, cannot find a place in this framework, it isn t a natural entity. 17 Methodological naturalism, on the other hand, requires that philosophical theorizing be continuous with the sciences. But what, exactly, does continuity with the sciences mean? Brian Leiter makes a distinction between methodological naturalism that requires results continuity with the sciences and methodological naturalism that requires methods continuity, and explains that whereas the former requires that philosophical theories be supported by scientific results, the latter requires that philosophical theories emulate the methods of inquiry and styles of explanation employed in the sciences. He states the following about methods continuity : Historically, this has been the most important type of naturalism in philosophy, evidenced in writers from Hume to Nietzsche. Hume and Nietzsche, for example, both construct speculative 16 Armstrong (1978, 261) takes (ontological) naturalism to be the doctrine that reality consists of nothing but a single all-embracing spatio-temporal system. 17 For a spirited rejection of ontological (and methodological) naturalism, see Popper (1978).

10 10 theories of human nature modeled on the most influential scientific paradigms of the day (Newtonian mechanics, in the case of Hume; 19 th century physiology, in the case of Nietzsche) in order to solve various philosophical problems. Their speculative theories are modeled on the sciences most importantly in that they take over from science the idea that we can understand all phenomena in terms of deterministic causes. Just as we understand the inanimate world by identifying the natural causes that determine them, so too we understand human beliefs, values, and actions by locating their causal determinants in various features of human nature. (2007, Footnotes omitted) But one may well wonder whether talk about continuity with the sciences is not too abstract a formulation to be helpful. The question, of course, is: Which sciences do the naturalists advocating such continuity have in mind? Although Leiter does not go into this, it is clear that Olivecrona had in mind the social sciences, especially psychology. 18 One may also wonder about the logical relation between ontological and methodological naturalism. Leiter does not touch on this issue, since he is concerned solely with methodological naturalism. It is, however, tempting to assume that methodological naturalism 18 Olivecrona never addressed the question of whether there might be kinds of psychological research that are not acceptable from the standpoint of (methodological) naturalism.

11 implies ontological naturalism. 19 For one might argue that it wouldn t make sense to aim at 11 emulating the methods of inquiry and styles of explanation employed in the sciences, unless one also believed that the world is such that this approach is likely to be successful, that is, that everything that exists is composed of natural entities, and that these entities determine all the properties of that which exists. But having thought about it, I am inclined to think that a believer in methodological naturalism may be agnostic about the ontological question, in the sense that he may allow that there may or may not be non-natural entities, such as a God, provided that these entities, if they exist, are unable to causally interact with the natural world if there were a God, who could causally interact with the natural world, we couldn t really know that metal expands when heated, say, since the God might then choose to stop a heated piece of metal from expanding This appears to be the view of Wagner & Warner (1993, 12). I shall leave it an open question whether the reverse also holds, that is, whether ontological naturalism implies methodological naturalism. 20 I want to thank Folke Tersman as well as Brian Bix and Michael Green for having emphasized in conversation and in correspondence the possibility of a believer in methodological naturalism who is agnostic about the ontological question.

12 At any rate, Leiter also distinguishes a third main type of naturalism, which I shall refer to 12 as semantic naturalism, according to which a concept must be analyzable in terms that admit of empirical inquiry, if the analysis is to be philosophically suitable. Leiter calls it semantic S-naturalism, because he conceives of it as a special kind of substantive naturalism. Here is Leiter: S-naturalism in philosophy is either the (ontological) view that the only things that exist are natural or physical things; or the (semantic) view that a suitable philosophical analysis of any concept must show it to be amenable to empirical inquiry. / / In the semantic sense, S-naturalism is just the view that predicates must be analyzable in terms that admit of empirical inquiry: so, e.g., a semantic S-naturalist might claim that morally good can be analyzed in terms of characteristics like maximizing human well-being that admit of empirical inquiry by psychology and physiology (assuming that well-being is a complex psycho-physical state). (2002, 3). I believe, however, that we should make a distinction between a narrow and a broad conception of semantic naturalism. On the narrow conception (NCSN), which Leiter appears to accept, a philosophically acceptable analysis of a concept entails that the concept strictly speaking, the term that expresses the concept refers to natural entities. On the broad

13 conception (BCSN), on the other hand, a philosophically acceptable analysis of a concept 13 entails that it does not refer to non-natural entities. This distinction is of some interest in this context, because the non-cognitivist analysis embraced by Olivecrona according to which moral terms like right, good, or duty have no cognitive (or descriptive) meaning, and do not refer at all 21 is in keeping with the broad, but not the narrow, conception. For, on this type of analysis, while such terms do not refer to non-natural entities, they do not refer to natural entities either. However, the broad conception of semantic naturalism is difficult to square with what we might call the classical conception of philosophical analysis, according to which such analysis aims to establish an analytically true equivalence between the analysandum (what is analyzed) and the analysans (what does the analyzing). 22 Since on the non-cognitivist analysis, moral terms have no cognitive meaning and do not refer at all, one cannot specify the analysans by saying A has a right to X if, and only if,... or A ought to do X if, and 21 Instead of cognitive meaning, they may have emotive meaning. On this, see Stevenson (1937). 22 On the classical conception of philosophical analysis, see, e.g., Langford (1942); Urmson (1956, 116-8); Sosa (1983); Strawson (1992, ch. 2); Anderson (1993).

14 only if,.... Accordingly, someone who embraces the classical conception of philosophical 14 analysis will almost certainly prefer the narrow conception of semantic naturalism. Although Leiter does not touch on this issue either, it seems to me that neither type of semantic naturalism implies ontological naturalism. 23 Like the methodological naturalist, the semantic naturalist may allow that there may or may not be non-natural entities, provided that these entities, if they exist, are unable to causally interact with the natural world. For both the belief that a philosophically acceptable analysis of a concept entails that the concept refers to natural entities, and the belief that a philosophically acceptable analysis of the concept entails that the concept does not refer to non-natural entities, is compatible with the belief that there may be non-natural entities that cannot influence the natural entities. Olivecrona was clearly an ontological naturalist, and he appears to have been a semantic naturalist in the narrow sense as well. For he clearly believed that everything that exists is composed of natural entities whose properties determine all the properties of that which exists; and he appears to have believed that a philosophically acceptable analysis of a concept entails that the concept refers to natural entities. But whereas Ross (1959, chs 3-4) endorsed a predictive analysis of the concepts of valid law and legal statement, Olivecrona (1962, I shall leave it an open question whether ontological naturalism implies semantic naturalism.

15 60; 1975) rejected this type of analysis on the ground that it does not capture our common 15 sense understanding of these concepts. And since he would probably have accepted the predictive analysis if he had accepted methodological naturalism, it seems that he could not rationally have accepted methodological naturalism. Olivecrona s adherence to ontological naturalism is clear from his claim in the First Edition of Law as Fact that any adequate theory of law must eschew metaphysics and treat the law as a matter of social facts. The aim, Olivecrona explained, is to reduce our picture of the law to make it correspond with objective reality: I want to go straight to this question [of law as fact] and treat directly the facts of social life. If in this way we get a coherent explanation, without contradictions, of those facts which are covered by the expression law, our task is fulfilled. Anyone who asserts that there is something more in the law, something of another order of things than mere facts, will have to take on himself the burden of proof. / / The facts which will be treated here are plain to everybody s eyes. What I want to do is chiefly to treat the facts as facts. My purpose is to reduce our picture of the law in order to make it tally with existing objective reality, rather than to introduce new material about the law. It is of the first importance to place the most elementary and well-known facts about the

16 16 law in their proper context without letting the metaphysical conceptions creep in time and again. (1939, 25-7). That Olivecrona s commitment to and understanding of naturalism remained the same in all essentials throughout his long carrier is clear from his treatment of the various legalphilosophical problems that he engaged with, but also from what he said on the few occasions when he explicitly considered his methodological stance. For example, he explained in the preface to the Second Edition of Law as Fact, that even though it is not a second edition in the usual sense, but rather a new book, the fundamental ideas are the same, viz. to fit the complex phenomena covered by the word law into the spatio-temporal world. (1971, vii) 4. META-ETHICS Olivecrona accepts, as we have seen, the main tenets of the Uppsala school of philosophy, including the tenet that there are no objective values. But the leading meta-ethicist among the Uppsala philosophers, Axel Hägerström, was not content simply to assert that there are no objective values. Instead, he defended a fairly radical version of emotivism, arguing in his

17 inaugural lecture of 1911 called Om moraliska föreställningars sanning ( On the Truth of 17 Moral Ideas ) that moral judgments express the speaker s feelings or attitudes, and that therefore they cannot be true or false (Hägerström 1939). 24 Olivecrona never spoke of moral values or moral rights or obligations, as distinguished from other types of value, right, or obligation, but preferred to speak more generally of values, rights, or obligations, etc. Nevertheless, it is clear from the context that he usually had in mind precisely moral values, rights or obligations. However, he rarely went further than to assert that there are no objective values and no objective ought. But this claim, or these claims, could be accepted not only by non-cognitivists, but by moral constructivists, including meta-ethical relativists (see, e.g., Harman 1996), and by error-theorists (see, e.g., Tegen 1944, 50; Mackie 1977, ch. 1; Joyce 2001). As a result, the precise nature of his meta-ethical position is somewhat unclear. I suggest, however, that in his early writings Olivecrona vacillated between an error-theory and a non-cognitivist theory in regard to rights statements 24 For an in-depth analysis of Hägerström s inaugural lecture, see Bjarup (2000).

18 and judgments about duty, while accepting non-cognitivism in regard to value judgments 18 proper, and that in his later writings he embraced a non-cognitivist theory across the board. 25 Olivecrona s analysis of the concept of a legal rule in the First Edition of Law as Fact suggests a non-cognitivist stance. Olivecrona maintains, as we shall see in Section 6, that an independent imperative can also be expressed by a sentence in the indicative mood, such as It is the case that you shall not steal. And he points out that this is the reason why we believe in objective values and an objective ought. But, he objects, [w]e do not impart knowledge by such utterances, we create suggestion in order to influence the mentality and the actions of other people. There is no real judgment behind the sentences. The ob- 25 Konrad Marc-Wogau argued already in 1940 that Olivecrona vacillates between two different ways of understanding the existence of rights, duties, and the binding force of law. On the first interpretation, these entities exist only as ideas or conceptions in human minds. As Marc-Wogau puts it, on this interpretation they have subjective, but not objective, existence. On the second interpretation, the entities exist neither in reality nor as ideas or conceptions in human minds. On this interpretation, they have neither objective nor subjective existence. Marc-Wogau suggests that Olivecrona really wanted to defend the second interpretation, although he frequently spoke as if he were concerned with the first. Marc-Wogau (1940). For more on this topic, Swedishspeaking readers may wish to consult Mac Leod (1973, 19-20); Petersson (1973, 162-5).

19 19 jective nature of an action is not determined by saying that it should, or should not, be undertaken. What lies behind the sentences is something other than a judgment. It is that, in our mind, an imperative expression is coupled to the idea of an action. This is a psychological connection only, though of the utmost importance in social life. But for certain reasons the connexion appears to us as existing objectively. Thus we get an illusion of a reality outside the natural world, a reality expressed by this shall. That is the basis of the idea of the binding force of the law. (1939, 46). Olivecrona s analysis of the concept of a right in the First Edition of Law as Fact, on the other hand, suggests an error-theoretical analysis. Olivecrona argues that since we have seen that the idea of the binding force of the law is an illusion, we must conclude that the idea of duties is subjective: Duty, he explains, has no place in the actual world, but only in the imagination of men. (Ibid., 75) He then maintains that the situation is essentially the same with regard to the concept of a right: It is generally supposed that the so-called rights are objective entities. We talk about them almost as if they were objects in the outer world. On reflection we do not, of course, maintain that this is the case. But we firmly believe that the rights exist outside our imagination as objective realities, though they are necessarily something intangible. We certainly do not confine their existence to the world of imagination. Suggestions to that effect are commonly rejected with scorn and

20 20 indignation. Yet on close examination it is revealed that the rights just as well as their counterpart the duties exist only as conceptions in human minds. (Ibid., Emphasis added) I take Olivecrona to be saying that while statements about rights or duties assert, or perhaps imply, that rights and duties exist outside our imagination as objective realities, the truth of the matter is that they exist only as conceptions in the human mind, and that therefore all statements about rights or duties are false. 26 Olivecrona returns to the topic of legal rules in an article published Having reiterated the claim made in the First Edition of Law as Fact that the grammatical form of value judgments, which here appears to include rights statements as well as judgments about duty, but not value judgments proper, deceives us into believing in objective values and an objective ought, he proceeds to clarify the real nature of value judgments: These statements have the verbal form of judgments; that is to say, they are verbal propositions concerning reality. When we, for instance, qualify actions as good or bad, we apparently ascribe 26 Strictly speaking, only statements that assert the existence of rights or duties are false on this interpretation, whereas statements that there are no rights or duties are true.

21 21 the property of goodness or badness to them. Yet, it is obvious that no such property can be detected in the actions among their natural properties. The qualification represents our own emotional attitude; it would be senseless to describe an action as either good or bad if it were to leave us completely unmoved. The statements on goodness or badness are supplied with meaning by the corresponding feelings. But our feelings are entirely subjective; it is senseless to ask whether they are true or not. They exist, or do not exist: that is all. (1951, Footnote omitted) As I see it, the reference to what we apparently ascribe to actions and to our emotional attitude, and the claim that it is senseless to ask whether our feelings are true or not, indicate that Olivecrona now embraces non-cognitivism. Olivecrona returns to the concept of a right in the Second Edition of Law as Fact, where he makes a distinction between two different ways of rejecting the reality of rights. He explains that we may say that there is no facultas moralis of natural law theory or no Willensmacht of the imperative theory of law, or we may instead say that the noun right as commonly used does not signify anything at all, not even something that exists in imagination only (1971, 183). He is explicit that he now prefers the second, non-cognitivist

22 22 analysis, though he does not comment on the fact that he used to prefer the first, errortheoretical analysis. We see that although Olivecrona was, on the whole, a non-cognitivist of the emotivist type, he appears to have been somewhat unclear not only about the differences between value judgments, judgments about duty, and rights statements, but also, and more importantly, about the difference between non-cognitivism and error-theory. 5. THE BINDING FORCE OF THE LAW Olivecrona begins the First Edition of Law as Fact with a consideration and rejection of the view that the law has binding force. He introduces the topic to be discussed in the following way: The most general definition of law seems to be that law is a body of rules, binding on the members of the community. Vague as it is, we may take this as our starting point for our investigation into the true nature of the law. It contains at least one element which, beyond doubt, is common to practically all those who have treated the subject. This is the assumption that the law is binding. Leaving aside for the time being the question how a rule is to be defined, we will first ask what is

23 23 meant by the binding force of the law and try to decide whether the binding force is a reality or not. (1939, 9). While Olivecrona does not explain what, exactly, he takes the binding force of the law to be, the core idea must surely be that a binding legal rule binds the subjects of the law in the sense that it obligates them, and that in this way it differs from a draft rule that has not yet been enacted into law. One way to account for the binding force thus conceived would be to maintain that law and morality are conceptually connected, so that legal rights and duties turn out to be a special case of moral rights and duties. But, as we shall see, Olivecrona focuses on Hans Kelsen s analysis of the concept of binding force, and Kelsen accepts the conceptual separation of law and morality. Hence Olivecrona must have conceived of the binding force of legal rules as being sui generis. Having rejected several attempts to explain the nature of the binding force by reference to social facts, such as feelings of being bound, or inability to break the law with impunity, Olivecrona concludes that the binding force has no place in the world of time and space, but must be located in some sort of supernatural realm (Ibid., 14). But, he objects, this is absurd.

24 The law could not be located in a supernatural world beyond the world of time and space, 24 because there could be no connection between such a world and the world of time and space: There is one very simple reason why a law outside the natural world is inconceivable. The law must necessarily be put in some relation to phenomena in this world. But nothing can be put in any relation to phenomena in the world of time and space without itself belonging to time and space. Therefore all the talk of a law, which in some mysterious way stands above the facts of life, is selfcontradictory. It makes no sense at all. (Ibid., 15-6) As Olivecrona sees it, we have here the dividing-line between realism and metaphysics, between scientific method and mysticism in the explanation of the law. To believe that the law has binding force and that therefore the law belongs in a supernatural world is to give up any attempt at a scientific explanation of the law and legal phenomena and to indulge in metaphysics: The binding force of the law is a reality merely as an idea in human minds. There is nothing in the outside world which corresponds to this idea. (Ibid., 17) Olivecrona does not, however, explain why there can be no connection between the world of the ought and the world of time and space; he just asserts that there can be no such connection. But, even though he does not say so, his critique owes a lot to Hägerström s

25 critique of Hans Kelsen s theory of law, put forward in a 1928 review of Kelsen s 25 Hauptprobleme der Staatsrechtslehre (Hägerström 1953). Hägerström argued that the very idea of the world of the ought is absurd, because this world cannot be thought of as even existing alongside the world of time and space. For, he reasoned, no knowledge of any reality is possible, except through relating its object to a systematically interconnected whole, and the fact that the two worlds the world of the ought and the world of time and space are different in kind means that they cannot be coordinated in a systematically interconnected whole. As he puts it, so far as I contemplate the one [world], the other [world] does not exist for me. (Ibid., 267). Note, however, that whereas Olivecrona appears to be concerned with the existence of the world of the ought, Hägerström is clearly concerned with knowledge about the world of the ought. Olivecrona then turns to consider Kelsen s theory of law, because he believes that this theory illustrates the necessity for believers in the binding force of the law, to make a distinction between the world of the law and the world of time and space (1939, 17-8). He seizes on the fact that on Kelsen s analysis, there is a connection between operative facts and legal consequence in legal norms that is as unshakeable as the connection between cause and

26 effect in nature. And this connection, he continues, is such that the legal consequence ought to 26 ensue when the operative facts are at hand. But, he objects, it is simply impossible to explain in a rational way how facts in the world of time and space, such as the activity of the legislature, can produce effects in the world of the ought. As he puts it, [a]t one time Kelsen bluntly declared that this is the Great Mystery. That is to state the matter plainly. A mystery it is and a mystery it will remain forever. (Ibid., 21) We see, then, that the binding force of legal rules, as Olivecrona understands it, is sui generis, and amounts to the idea that valid legal rules (or norms) obligate the subjects of the law. We also see that Kelsen s analysis amounts to a non-naturalist realist understanding of legal norms (or rules). As Kelsen puts it, [t]o speak... of the validity of a norm is to express first of all simply the specific existence of the norm, the particular way in which the norm is given, in contradistinction to natural reality, existing in space and time. The norm as such, not to be confused with the act by means of which the norm is issued, does not exist in space and time, for it is not a fact of nature. (1992, 12). And again: One will not be able to

27 deny... that the law qua norm is an ideal reality, not a natural reality. (Ibid., 15) 27 And this 27 is precisely what Olivecrona has in mind when he says that, on Kelsen s analysis, the ought establishes an objective connection between operative facts and legal consequence. (Ibid., 18) Although Olivecrona does not say so in the First Edition of Law as Fact, it is clear from his analysis in the Second Edition of Law as Fact that he takes the absence of binding force to imply the absence of legal relations: Since there is no binding force, there can be no legal relations. As we shall see in Section 6, Olivecrona introduces in the Second Edition of Law as Fact the concept of a performatory imperative, in order to account for those legal rules that do not immediately concern human behavior (1971, chs 5, 8). The introduction of this concept is of interest in this context, because Olivecrona adds to it a consideration of the nature of the legal effect that is supposed to follow from the utterance of a performatory imperative (Ibid., 221-6). Such legal effects, he points out, are clearly supersensible: 27 See also Kelsen 1945, Kelsen speaks about the specific existence of norms in the Second Edition of Reine Rechtslehre, too, though he is more cautious here and is never, as far as I can see, explicit that norms do not exist in time and space. Kelsen (1960, 5-6, 9-10).

28 28 Relatively uniform ideas of ownership, monetary claims, many other kinds of right, corresponding duties, and legal qualities are disseminated among the general public. These rights, duties, and legal qualities are supposed to be created, modified, transferred, and extinguished through operative facts by virtue of the law. They form a supersensible world: in the sensible, natural world there are no rights and duties, or legal qualities. (Ibid., 223) He adds that in the world of time and space, there is only the psychological fact that people tend to believe that there is a legal effect, and, of course, the (sociological) fact that they tend to act accordingly (Ibid., 224-6). It is not absolutely clear from his account whether Olivecrona (i) takes the claim that legal rules have no binding force to imply the claim that there are no legal relations, and, perhaps, vice versa, or whether he (ii) takes these two claims to be synonymous, to be two sides of the same coin, as it were. Although I do not think much depends on whether (i) or (ii) is the better interpretation, I shall assume that (i) is closer to the truth. In any case, it is clear that Olivecrona s critique of the view that the law has binding force is premised on a commitment to ontological and semantic naturalism: Since Olivecrona is an ontological naturalist, he cannot accept the existence of a world, viz. the world of the ought, located beyond the world of time and space; and since he is a semantic naturalist in the

29 narrow sense, he cannot accept an analysis of the concept of a binding legal rule that does not 29 entail that the concept refers to natural entities. I cannot, however, accept Olivecrona s critique of the view that the law has binding force, because I believe we can account for the existence of binding rules in a naturalistically acceptable way, viz. along the lines of the conventionalist analysis proposed by the Finnish philosopher Eerik Lagerspetz (1995, ch. 1). On this analysis, a binding social rule, R, exists if, and only if, the members of the relevant group of people (1a) believe that R exists and (1b) believe that the others in the group believe that R exists, and (2) act accordingly, that is, speak of R as existing and, if occasion arises, treat R as existing, at least partly because they have the beliefs (1a) and (1b). 28 As we shall see in Section 6, this type of analysis is in keeping 28 Note that, strictly speaking, Lagerspetz s analysis lays down sufficient, but not necessary, conditions for the existence of conventional facts. As Lagerspetz points out, there may be some rules (or other conventional entities) that are not even known by large parts of the population, but which nevertheless may be said to exist in the sense that they are part of a system of rules that satisfies the conditions laid down in the analysis (1995, 19). However, I choose to render Lagerspetz s analysis in terms of necessary and sufficient conditions, because I believe the rules (or other conventional entities) just mentioned play only a minor role in any system of rules.

30 with Olivecrona s stance that R does not really ( objectively speaking ) exist, though we tend 30 to think that R exists, and to act accordingly. One might, however, object that Lagerspetz s analysis is not relevant to my concerns, because it shows only is what it means for a group of people to accept (or believe) that R is binding, not that R really is binding. 29 I cannot accept this objection, however, because it assumes that there really is in this context a difference between (i) being binding and (ii) being considered to be binding, and I cannot see that there is any such difference in the case of legal entities and qualities. The only sense in which, say, A s offer to B to buy A s bicycle is legally binding is that courts and other legal actors consider (or recognize) A s offer to be binding. But, the critic might continue, if (i) and (ii) were really identical, we couldn t explain how courts and other legal actors can be uncertain whether a certain rule or a certain offer is really binding, as in fact they sometimes are. For in contemplating the question whether A s offer is binding, judges and other legal officials will not be attempting to answer the question Do I or my colleagues consider this offer to be binding, but the question whether the purported offer satisfies certain conditions laid down in the relevant legal norms. 29 I would like to thank Jan Österberg for suggesting this and the next objection to my analysis.

31 This objection fails, however, because it ignores both (i) the distinction between 31 individual and collective beliefs, and (ii) the fact that the collective beliefs that constitute the binding force of the relevant entity need not be explicit and simple beliefs of the type R has binding force, but may be in various ways indirect. Let me explain what I mean. First, while it is clear that one cannot without contradiction maintain that a person who believes that R is binding can at the same time be uncertain as to whether R is binding, one may well maintain that a person, who is part of a group of people the collective beliefs of whom constitute the binding force of R, is uncertain as to whether R is binding. Moreover, we shall see in Section 5 below that it may be a fairly complicated matter to characterize and identify the relevant beliefs, and this means that any given member of the relevant community may be uncertain about the precise content and incidence of the relevant beliefs and, therefore, uncertain whether R is binding. Similarly, a meta-ethical relativist, such as Gilbert Harman (1975), who maintains that what is morally right or wrong in any given community depends on the (unrefined) beliefs of the people in that community, may without contradiction accept that two or more members of the community can be uncertain whether a certain a certain rule is binding or whether a certain action is morally permissible. The relevant conventions in the community can easily survive minor uncertainties or disagreements.

32 Nevertheless, Lagerspetz s analysis is not without its problems, and I shall discuss those 32 problems when I return to the analysis in Section LEGAL RULES AS INDEPENDENT IMPERATIVES But if, as Olivecrona maintains, legal rules do not establish legal relations, what do they do? Olivecrona clearly needs to conceive of the function of legal rules in some other way than Kelsen and others do. And his view is that legal rules are psychologically effective, and that in this way they are part of (what he calls) the chain of cause and effect. Olivecrona makes a distinction between the form and the content of a legal rule. The content of a legal rule, he explains, is an idea of an imaginary action by a judge in an imaginary situation (1939, 28-9). He points out, however, that the technique of law-giving obscures the actual purport of legal rules to a considerable extent, because on the face of it many legal rules do not concern human conduct at all, but rather the existence of rights and duties. But, he insists, ultimately all legal rules concern human conduct (Ibid., 30). The form of legal rules, he continues, is imperative, because the lawmakers do not aim to inform us about the existence of certain ideas in their minds, but to impress a certain behavior

33 on us (Ibid., 31). He is, however, careful to point out that he does not have the grammatical 33 imperative form in mind when he maintains that legal rules have imperative form. Statutory provisions are often phrased in the indicative or the subjunctive mood, but they always express an imperative (1942, 9). On Olivecrona s analysis, what is important is that an utterance (or a gesture) functions as an imperative, and although he is not explicit about it, Olivecrona appears to believe that an utterance functions as an imperative if it is intended to be an imperative. But, he continues, to determine whether a particular utterance is intended to be an imperative, one needs to consider the whole situation in which the utterance takes place (Ibid., 16-7). Pointing out that the command is the prototype of the imperative, Olivecrona explains that a command works directly on the will of the recipient of the command, and that this means that it must have a suggestive character. He states the following: A command is an act through which one person seeks to influence the will of another. This may be done through words or signs or perhaps by a determined look only. It is characteristic of the command that the influence on the will is not attained through any appeal to things that constitute values for the receiver of the command. The command may be supported and strengthened by a threat or by a promise. But this is something secondary. The command as such does not contain

34 34 any reference to values. It works directly on the will. In order to do this the act must have a suggestive character. Whether words or other means are used, the purpose is obviously suggestion. (1939, 33-4.) Olivecrona maintains, more specifically, that if a command takes effect there arises in most cases in the addressee s mind a value-neutral intention to perform the commanded action, that is, an intention that is not motivated by the addressee s own wishes, and adds that in some cases a command may actually trigger an action without the addressee s having had any intervening value-neutral intention (1942, 7, 10-1). Olivecrona proceeds to explain that legal rules are not commands (1939, 35-40). Pointing out that the imperative theory, or, as I would say, the command theory, presupposes that there is a commander, he objects that there simply is no one person or group of persons who could be the commander(s). He notes that it is often suggested that the state, or someone representing the state, is the commander, but points out that this cannot be so. For one thing, there is simply no one who commands anything in the process of law making. For, on closer inspection, we see that neither the members of the parliament or of the government, nor the head of state commands anything: What they do is push certain buttons (when voting) or sign

35 certain documents (when promulgating the law). Moreover, the lawmakers have their position 35 as lawmakers on the basis of certain rules, and it would amount to circular reasoning to maintain that those rules are commands issued by the lawmakers themselves. Olivecrona concludes that the imperative theory of law could work only if there were some kind of superhuman entity who functioned as commander, but points out that there can be no such entity. But if legal rules are not commands, although they have imperative form, what are they? Olivecrona explains that in addition to commands, there is a class of imperatives that we may refer to as independent imperatives. And he maintains that legal rules are best conceived of as such independent imperatives (Ibid., 42-9) On Olivecrona s analysis, there are three important differences between commands and independent imperatives. First, whereas a command is always issued by a certain person, an independent imperative is not issued by anyone in particular (Ibid., 32-41). Second, whereas a command is always addressed to a certain person or persons and concerns a certain action or actions, an independent imperative concerns a kind of action and is not addressed to anyone in particular. Olivecrona s idea, then, is that an independent imperative concerns a class of persons (the norm-subjects) and a class of actions (the action-theme), not particular persons

36 and actions. 30 Third, whereas a command is in no way equivalent to a judgment, we have seen 36 (in Section 4) that an independent imperative can sometimes be replaced by a sentence that expresses a judgment; and Olivecrona believes, as we have also seen (in Section 4), that this explains why people (wrongly) believe that they can have knowledge of what we ought to do (Ibid., 45-6). Olivecrona objects, however, to the view that an independent imperative can be replaced by a sentence that expresses a judgment, that there are no real judgments behind the sentences that (appear to) express such judgments, but only a psychological connection, viz. a connection in a person s mind between the imperative expression and the idea of an action. What is really going on in the process of legislation, he explains, is that the legislature attempts to influence human behavior by making use of the imperative form: The word ought and the like are imperative expressions which are used in order to impress a certain behaviour on people. It is sheer nonsense to say that they signify a reality. Their sole function is to work on the minds of people, directing them to do this or that or to refrain from something else not to communicate knowledge about the state of things. By means of such 30 Of course, this is also a characteristic property of rules, including legal rules.

37 37 expressions the lawgivers are able to influence the conduct of state officials and of the public in general. The laws are therefore links in the chain of cause and effect. (Ibid., Emphasis added) We should note here that the claim that imperatives are psychologically effective is of central importance to Olivecrona s naturalistic theory of law if they weren t psychologically effective, the theory would be seriously incomplete, since, on Olivecrona s analysis, there are no legal relations the knowledge of which could somehow motivate the citizens to act accordingly. Olivecrona points out, however, that the way the individual mind works is a matter for the science of psychology, and that for the purposes of his investigation into the nature of law, he need only point to the general conditions that must be satisfied for legislation to be effective in society (Ibid., 52). And he identifies, in keeping with this, two general conditions for the efficacy of legislation in society. First and most important, the citizens must display an attitude of reverence toward the constitution: Everywhere there exists a set of ideas concerning the government of the country, ideas which are conceived as binding and implicitly obeyed. According to them certain persons are appointed to wield supreme power as kings, ministers, or members of parliament etc. From this their actual power obtains.