Fuller’s Clock: A Case for Legal Non-Positivism in Artefactual Theory of Law¹

Abstract

This article covers the traditional problems of the philosophy of law: the controversies between realism and normativism, on the one hand, and positivism and non-positivism, on the other. The author, adopting the ontological perspective of the artefactual theory of law, attempts to attain two research aims. First, he argues that artefactual theory of law paves the way towards a moderate position overcoming realism-normativism duality. Second, he advocates the thesis that the supposedly ontological difference between positivist and non-positivist views might be dispelled and described as merely conceptual. Therefore, the opposite points of view could be considered as metaphysically equivalent ways of grasping the same social reality. According to the crucial premise of the argument law is a meliorative thick concept. The essay was inspired by an analogy between the philosophy of law and the philosophy of science first discussed by Lon L. Fuller.

In his classic oeuvre, The Morality of Law, Lon L. Fuller suggests that ‘the parallels between legal and scientific philosophies are indeed striking’¹. Following Fuller’s reasoning carefully, it can be noticed — although he does not explicitly claim it — that in fact the analogy turns out to be threefold, concerning not only (1) the philosophy of law and the philosophy of science, but also (2) law and science themselves and (3) their corresponding theoretical (and arguably folk) concepts. What could a philosopher of law learn from a philosopher of science? In what respects is the aforementioned accordance accurate? How deep does this analogy go? And how profound or insightful is it? Fuller’s observation — admittedly laconic and therefore enigmatic — nevertheless seems penetrating. Its consequences may reach ontology, semantics, and some methodological aspects of the theory and the philosophy of law. The American thinker took the first step towards answering these questions. My aim is to follow the path that he showed a little further. It is said — and these words come from Peter F. Strawson, who referred to Aristotle and Kant — that ‘the greatest philosophers, more than any others, repay this effort of rethinking’². It is my belief that the same could be said of Herbert L.A. Hart, Ronald Dworkin, or Fuller.

The article has two central objectives, that can be referred to as Analogy Project and Advocacy Project. As for the Analogy Project, my goal is to show some unexpected parallels between the philosophy of science and the philosophy of law. I try to shed some light on Fuller’s observation on the analogy of (among others) his own non-positivism and elitism of Michael Polanyi. Taking up the theme of The Morality of Law, I elaborate on it and point to the further (and until today overlooked) similarities between two seemingly distant fields of philosophy. In particular it is essential to highlight the analogy between Fuller’s theory of inner morality of law and one of the currents of the rational reconstruction methodology initiated by Imre Lakatos and Larry Laudan. As for the Advocacy Project, I present an argument in favour of a moderate non-positivism in the artefactual theory of law. The key premise of the argument is that law is a meliorative thick concept. The conclusion is that there is a distinction between law and bad law that is not law, independent of any individual legal system, although there is no demarcation criterion being a benchmark in each and every case. Moreover, artefactual theory of law makes it possible (1) to overcome the dualism of (legal) realism and normativism, and (2) to reconceive the ontological difference between positivism and non-positivism as a conceptual difference. As it turns out, the methodological and ontological consequences of advocated view on law are in line with corresponding theses concerning science (its nature and concept) on the basis of Lakatos’ and Laudan’s theory. This conclusion additionally reinforces the analogy which is to be outlined.

1 Institutional Artefacts

Distinguish three categories of entities: artefacts (e.g., watches, scientific theories, or legal systems), artefactual kinds or types of artefacts (to graphically highlight this difference, one might use capital letters: WATCH, SCIENCE, LAW), and the correlative kind or generic concepts (e.g., clock, science, or law)³. Now, for the purpose of further considerations, assume the following pair of equalities:

x is law = (1) x is an institutional artefact, (2) x belongs to the artefactual kind LAW, (i)

x is science = (1) x is an institutional artefact, (2) x belongs to the artefactual kind SCIENCE. (ii)

To clearly explain what artefactual kinds and, consequently, artefacts actually are, it is useful to contrast them with natural kinds. Natural kinds have their essence through brute physical facts, as John Searle labelled them⁴. Notice that with regard to the two-component scheme of the genus-differentia definition, an essence corresponds to the latter. It correlates with the bunch of properties that any x of a given kind must exemplify. Natural essences are internal and mind-independent. Contrarily, in the case of artefactual kinds, each differentia is constituted by the intentional attitudes of those who created an artefact⁵. Therefore, types of artefacts are mind-dependent entities, although in a weak sense referring to collective, not individual intentionality. Furthermore, artefacts of certain genres thus (indirectly) existentially depend on the content of some collective intentions.

Herein I shall accept the characteristics of this ontological category provided by Amie Thomasson⁶. Let an item x be a member of an artefactual kind K_a if and only if:

(1a) x’s authors have a substantive artefactual concept of K_a,

(1b) this concept is generally adequate and defines the essential properties of K_a-type objects,

(1c) the authors intend to impose these K-relevant features on the object,

(2) and their intention has been largely successfully realised.

Points (1a)–(1c) jointly constitute the intention condition. Point (2) defines, according to the terminology proposed by Jonathan Crowe⁷, the success condition. Every once in a while, a product does not conform to the standard of what was to be produced. Subsequently, it is either an imperfect token of a given type K_a (like a broken or stopped clock that shows the wrong time), or it does not fall under that kind at all (in this case, it is not a clock, but just a piece of iron).

Institutional artefacts are creations of mind. They are based on a collective recognition of a set of constitutive rules. According to John Searle, constitutive rules share a common logical form: in the circumstances of z the object x counts as y (e.g., as K_a(x), i.e., x being an exemplar of K_a). By means of these rules, an artefact — classified as an instance of certain institutional kinds (LAW and SCIENCE seem to be exemplary here) — acquires its deontic powers⁸. This privileged and normative status is one of the K_a-relevant features: x that lacks this feature is not K(x). In the following lines, I would like to investigate the sine qua non conditions for being a token of artefactual types such as LAW or SCIENCE.

Let us safely assume that K-relevant features of K(x) might be reduced to some functions x performs⁹. Artefacts that actually do not perform their intended functions nevertheless remain tokens of a certain type (e.g., a broken clock, a disproved scientific theory, or a historical, thus non-binding, legal system). However, it seems justified to accept that items that could not perform (i.e., did not, do not, and never will do) a K_a-essential function simply do not belong to the category of K_a (e.g., an object imitating a clock, but lacking a proper watch mechanism). Note that this thesis is laid out in stronger modal terms: we speak of what an item could not be, not what by accident it is not. The function of science is to explain and predict natural phenomena. On the other hand, the function of law is to provide its subjects with non-redundant reasons for action¹⁰. Legal systems accomplish this, as Crowe¹¹ notes, by being a deontic marker, indicating the boundaries of permissible social behaviour. The set of constitutive rules that simply does not satisfy the above-mentioned condition would be a product of some kind, or even an institutional artefact, but arguably not a law. Artefactual theory of law is a thesis that law is an institutional artifact that satisfies the success condition for falling under the kind LAW.

One might agree that the existence of institutional artefacts depends on collective intentional attitudes. This observation, however, does not entail that they are merely useful fictions. Consider a putative difference between real and fictional institutions. The criterion for existence of an institution can be expressed in terms of reasons or, to be more precise, reason-giving. Recall the so-called Eleatic Principle famously introduced by David M. Armstrong.¹² The principle states that only those objects that are causally active are real. How do we understand the causal activity of law, and perhaps institutional artefacts in general? It is to notice that institutions yield objective — i.e., sufficient (good or bad) — reasons for action. Yet the concept of objective reasons cannot be explicated in causal terms, and thus it is rather related to justification (which is not a casual notion). For an agent who successfully refers to such reasons, they constitute a justification — or an excuse — suspending negative reactions towards him, such as anger, resentment, or blame. The function of the law, then, is to be seen as creating reasons that, under certain circumstances, bring about social approval or disapproval and manifest reactive attitudes in the well-known sense set forth by Strawson¹³. In virtue of one’s knowledge of the laws as they are, objective reasons turn into motivating ones, which encourage the members of the communication community to act. The latter concept is psychological and therefore causal as well. Hereby, it is possible to grasp the distinction between a real legal system, which creates reasons or merely triggers them¹⁴, and a fictitious one. The reality of an institutional artefact is in a way synonymous with rationality (however understood in a specific manner). Such narrowly defined rationality belongs to the substantive concept of law — i.e., the one that belongs to the content of its author’s intentions.

Science enjoys epistemological authority. The practical authority of law consists in providing reasons for our everyday actions. This is (at least) what we mean by saying, following Mark Murphy¹⁵, that ‘necessarily, law is a rational standard for conduct’. Its rationality, thereafter, might be understood in two ways: in a weak sense (when legal norms constitute only motivating reasons) and in a strong one (when they constitute motivating as well as justificatory reasons¹⁶). I am tempted to claim that weak rationality is a necessary but insufficient condition for the success of producing an institutional artefact that falls under the kind of LAW.

2 Law as a Meliorative Thick Concept

Bernard Williams famously distinguished two genres of ethical concepts: thin (e.g., good, evil, or right, wrong) and thick¹⁷. By way of example of the latter category, Williams pointed out ‘treachery and promise and brutality and courage’. According to the author of Ethics and the Limits of Philosophy, the peculiarity of thick concepts is that they contain both a descriptive and a normative component. Consequently, the following dependence obtains:

if proposition p contains a thick concept, then p semantically implies some description,

and p semantically or pragmatically implies some (positive or negative) evaluation. (iii)

Consider a couple of speech acts. Under certain circumstances (e.g., pointing with a finger to a piece of tombac some jewellery is made of), subject S says: ‘x is gold, but false’. Without any doubts, we might say that the sentence is false. False or fake gold is not gold whatsoever. When one labels an object as gold, one semantically implies a positive evaluation or assessment that it is a noble metal of high value and not its (presumably cheap) counterfeit. Suppose by way of contrast that in some other circumstances S says (e.g., jokingly): ‘x is gold, but gold for fools’. It appears that the meaning of this statement might be explained more accurately in terms of pragmatic, not semantic, implication, as a presupposition or a conversational implicature. The latter sentence cannot be rejected simply as false. It would be more accurate to say that the speaker intends to cancel the presupposition that he was referring to a piece of gold at all. On the contrary, he completes the former speech act by providing additional information. Having learned that, one understands that the factor ‘for fools’ has a modifying and non-determining character (so it cannot be a contradiction in itself, a contradictio in adiecto). Fool’s gold is not gold, just as a king of cards is not a king. The imaginary statement, however, does not contradict itself, but, after a comma, explains that the speaker’s intention refers to pyrite, which is an iron mineral that can be easily found in almost all types of rocks.

It is my belief that the concepts of law and science are meliorative thick concepts (the term meliorative comes from the Latin word melior, which means better)¹⁸. With reference to science, this observation seems to be apparent and goes without saying. When one characterises a theory as ‘scientific’, one does not merely describe it. At the same time, one expresses a positive opinion about it. In widespread disagreements about what is science and what is not, interlocutors discuss both facts and values. Such disputes seem to be not purely descriptive but also normative by their very nature. Let us clearly distinguish three well-known concepts. By science we refer to a cognitive activity that meets the highest standard of reliability. We speak of pseudoscience, meaning all cognitive activity that is at once competitive with science and unreliable. Pseudoscience is not science any more than pyrite is gold. On the other hand, bad science can be defined as unreliable research proceedings nonetheless within a scientific paradigm (e.g., carelessly gathered evidence or results of experiments that are falsified). Bad science is just like fake gold — and in no case is it science.

It is worth considering whether there is an analogy between science and law in this regard. We can certainly assume that pseudo-law (e.g., created as a result of the self-proclaimed act of a person or institution) is not law. The demarcation criterion between these two categories is usually provided by a legal system itself. Nonetheless, it would be interesting to look into an opposition between law and bad law (legal system or legal norm). Are there any generic success conditions, independent of any particular legal system, that allow us to recognise that, at least sometimes, a bad law is not a law, just as a piece of iron is not a clock (even a broken one)? In advance of addressing this question, I would like to make a few remarks on the philosophy of science.

3 The Methodology of Rational Reconstruction in the Philosophy of Science

The conviction that the scientific method constitutes the highest standard of any reliable cognition becomes an impulse for the methodologies of rational reconstruction of episodes in the history of science. In this section, I will briefly explain what makes this philosophical standpoint distinctive.

The notion of rational reconstruction comes from Hans Reichenbach¹⁹, but the very idea could be traced as far back as Gottlob Frege. According to Reichenbach, such reconstruction serves as a demarcation criterion between proper and improper reasoning. Premises and a conclusion of each reasoning are usually connected by both logical and psychological links. Rational reconstruction — restricted exclusively to the context of justification (according to Reichenbach, the context of discovery was hopelessly irrational) — consists in replacing unreliable psychological links with reliable and unbreakable nodes of logical inference. Only when such a procedure turns out to be successful can we be assured of the rationality of a reasoning, and therefore its scientific character²⁰.

The idea of ​​rational reconstruction as the logic of scientific discovery was developed by Karl R. Popper in the 1930s²¹. ‘Logic’ in the title of Popper’s famous book refers to methodology of science that studies rationality (correctness, or effectiveness) of methods of empirical research. Following Reichenbach Popper strictly separated the context of discovery from the context of justification, recognizing the latter as the only domain of the philosophy of science. The context of discovery, placed out the scope of rational reconstruction, has been left to sociology or history of science as distinct fields of inquiry.

Popper believed that the demarcation problem was the most fundamental challenge to the philosophy of science. The problem was to point both necessary and sufficient conditions that enable us to distinguish science from metaphysics, on the one hand, and pre-science or pseudo-science, on the other. The solution to Kant’s problem (as Popper otherwise called the aforementioned challenge) was to define science and to provide the criterion that a theory ought to meet in order to deserve the name of science (or ‘Science’, written with the capital letter). Unlike Reichenbach (who adopted a verificationist position), Popper came up with falsificationism. He claimed that a scientific hypothesis must have had empirical conditions for falsifiability, and thus must have been a general sentence with testable implications the falsehood of which would falsify the hypothesis itself. The opposition of the context of discovery and justification is a form of facts–values (or descriptive–normative) dualism.

According to Popper, the philosophy of science is, therefore, inherently normative: it indicates what the method of empirical research ought to be, regardless of how scientific theories were actually conducted and justified. The history of science can only be instructive to its philosophy in the context of discovery, not justification. Popper’s methodology was, in consequence, presented as a priori, lacking any in-depth analysis of historical material. The normative precepts of falsificationism cannot be falsified by the actual history of science. What is of particular significance to our considerations is the suggestion that there are normative criteria for scientificity, independent of any facts.

The methodology of rational reconstructions was challenged by the sociology of scientific knowledge (or shortly: sociologism) of Thomas Kuhn — a historical and hence a posteriori perspective intentionally emptied of normative elements. Kuhn believed that an adequate explanation of scientific discoveries should be described in terms of their psychological or social causes. Therefore, in the final analysis, the history of science — the conduct of the scientific community — was not governed by reasons, and thus irrational. Unlike Popper, Kuhn sees science not as a theory in the logical space of reason, but as one of the social institutions that changes along with historical, political and ideological conditions (which are external to science, and often irrational). For the author of The Structure of Scientific Revolutions, the philosophy of science is descriptive, not normative, and – as any other empirical research – a posteriori, not a priori. A philosopher of science explores the actual history of science, and only by noticing regularities can abstract or generalize rules that govern the practice of a scientific community. These rules, often adopted implicitly, are a component of a paradigm accepted by a scientific community at some point in history. The criteria for scientificity are always relative to the paradigm. Theories that have been rejected over the years (e.g. caloric theory or phlogiston theory), therefore, are no less scientific than those currently accepted. There are no trans- or supra-paradigmatic criteria for scientificity. The concept of science is purely descriptive: a theory is scientific when it is widely accepted by the scientific community. Thus, there are no universal standards of scientific rationality that could possibly serve as a demarcation criterion, or as a basis for the evaluation of correctness of a method agreed by scientists.

The mature methodology of rational reconstructions, along with its meta-methodology, was presented by Lakatos²² and Laudan²³. Lakatos’s crucial idea was to reconcile the normative approach (represented by Popper) with the descriptive one (represented by Kuhn).²⁴ He believed that in the philosophy of science there was room for a normative category of reasons. Notwithstanding, it also must take into account the history of canonical exemplars of scientific discoveries (as the Copernican, Newtonian, or Darwinian revolution). Pursuant to the famous utterance of Lakatos²⁵, ‘[h]istory of science without philosophy of science is blind, philosophy of science without history of science is empty’. His own methodology of scientific research programmes should therefore be regarded as an attempt to take a ‘third way’ between the unilateral theories of Popper and Kuhn (this dictum applies equally to the methodology of Laudan’s research traditions).

We may attempt to sketchily list basic assumptions of the variety of methodologies of rational reconstruction:

(R_s1) it is adequate to talk about scientificity only in the context of a series of theories that follow one another (and never in the context of an isolated theory),

(R_s2) there are criteria for selecting the best theories available in a given research situation that are constitutive for science,

(R_s3) these necessary conditions are objective, universal, and normative, and

(R_s4) serve as a demarcation criterion between SCIENCE and NOT-SCIENCE, hence including BAD SCIENCE.

To explain the idea behind the above postulates, let us present the differentiation made by Lakatos. Namely, he distinguished the internal, external, and real history of science. Internal history takes into consideration solely the reasons that the scientific community should follow when accepting (resp. refuting) a theory or a hypothesis in a given research situation. Thus, in relation to the actual history, it is to be regarded as a prescriptive standard, which cannot be falsified by any facts whatsoever. This is precisely the normative component of Lakatos’s methodology. External history, on the other hand, is the totality of non-rational circumstances that caused decisions made by a scientific community. We shall underline the key meta-methodological directive laid out by the author of Proofs and Refutations. As he maintained, the methodology that allows to reconstruct the most extensive part of the real history of science as its internal history is to be considered as the most suitable as a rational reconstruction. This stipulation constitutes, in turn, the descriptive component of Lakatos’s view. Additionally, one might rightly highlight that any description of the real history of science includes both its internal and external history²⁶.

In this light, Popper’s and Kuhn’s positions seem to be one-sided. The former took into account only the internal history of science, and the latter – its entire real history. But how are these considerations related to the philosophy of law?

4 Philosophy of Science and Philosophy of law

The link between two philosophies — as I remarked at the preface of the text — was firstly observed by Fuller. For historical reasons, however, he could not fully understand its consequences. Fuller’s best-known books, The Morality of Law and The Anatomy of Law, were published in 1964 and 1968, respectively. Nevertheless, the crucial works of Lakatos and Laudan were not brought up until the late 1970s. Therefore, my aim is to emphasise this parallelism and to grasp it from the contemporary perspective. Before I move on, I would like to highlight a pair of rudimentary facts.

First, similarly to Lakatos — who considered a sequence of theories, not an isolated theory, as a fundamental notion — Fuller put forward a diachronic rather than synchronic definition of law. This is to say, he understood it as a ‘complex human undertaking’²⁷ stretched over time. For this reason, emphasizing the importance of law in action — contrasted with law in books — he was mainly interested in the legal system not as a set of static norms, but the variety of dynamic acts of making and applying laws. The transition from law in books to law in action can be juxtaposed with the shift from the logic of justification to the logic of discovery in the philosophy of science. Fuller asks about the necessary conditions for creating good law, and therefore in his considerations the dualism of facts-values (normative-descriptive) becomes blurred.

Second, Fuller defined law more precisely as ‘the enterprise of subjecting human conduct to the governance of rules’²⁸. Fuller’s position, one of the 20th -century versions of non-positivism, is in line with the contemporary artefactual theory of law. Moreover, his contribution in this respect may be considered pioneering. Subsequently, he portrayed a legal system as a product of a persistent and purposeful effort of collective agency. The purpose of law is to provide its subjects with reasons for action (and at least in this weak sense is law rational). As one might say in terms of artefactual theory, providing reasons for action is to be one of the relevant features or intended function of institutional artefacts that fall under the type of LAW. That being the case, it becomes a constituent of the success condition for creating instances of that artefactual kind. A set of constitutive rules that does not meet this condition would presumably count as an institutional artefact, and yet it would fail as an exemplification of LAW.

4.1 Between Realism and Normativism

In the passages of The Morality of Law, Fuller literally drew only a superficial analogy. It is supposed to compare, on the one hand, the positions he argues with, and, on the other hand, his own theory of internal morality of law and the notion of ‘internal morality of science’ ascribed by him to Polanyi. The predictive theory of law by Oliver W. Holmes and the normativism of Hans Kelsen are the doctrines Fuller succinctly criticised. On the basis of the philosophy of science, they correspond, respectively, to Percy W. Bridgman’s operationalism and to Reichenbach’s ‘scientific empiricism’²⁹ (as Fuller dubbed it), i.e. logical positivism. In the following lines, I attempt to elucidate the putative similarity (depicted in the first figure).

Fig. 1

Figure one

In line with the characteristics proposed by Bridgman, ‘science is what working scientists do, not what other people or even they themselves may say about it. No working scientist, when he plans an experiment in the laboratory, asks himself whether he is being properly scientific, nor is he interested in whatever method he may be using as method’³⁰. What seems to be of particular significance is the operational nature of the cited definition. As claimed by adherents of operationalism, the meaning of theoretical concepts should be grasped through the operations we perform. In this spirit, Einstein famously defined time as what clocks measure (cf. the definition of the second, i.e., the basic unit of time measurement, adopted in the SI metric system). In the same manner, Holmes defined the law as ‘[t]he prophecies of what the courts will do in fact, and nothing more pretentious’³¹. To paraphrase the quotation from Bridgman, one could say in conformity with Holmes that the law is what judges do when issuing (final) judgements. This is one of the elementary approaches to legal realism.

Operationalism is indeed a kind of neo-positivism. However, it is emphatically opposed to scientific empiricism. As noted above, Reichenbach believed that all problems belonging to the context of a scientific discovery could not be rationally reconstructed, and therefore were beyond concern of the neo-positivist philosophy of science. Correspondingly, Kelsen has put all problems of the making and interpreting of laws outside his pure legal theory and relegated them to the realm of so-called ‘meta-juristic’³², afterwards deliberately ignoring the latter. Legal positivism, as Kelsen states it, limits the science of law to abstract norms composing a hierarchically arranged legal system.

For the purpose of these considerations normativism may be understand as a conjunction of a pair of theses: ontological and methodological. The ontological thesis claims that the law is a system of norms that are abstract entities (en passant, Kelsen’s theory reveals the influence of Frege’s platonism and his understanding of a sentence as an ideal object existing in so-called third realm). The validity of a norm is independent of whatever subjects of law or legal officials do. The methodological thesis postulates narrowing down the science of law to norms, in contrast with facts. It is a normative science that indicates what a legal norm ought and ought not to be. Jurisprudence is an autonomous field, detached from both metaphysics and empirical sciences. Kelsen, therefore, viewed jurisprudence – as Popper philosophy of science – as normative and a priori undertaking.

Realism, on the other hand, is a methodological rather than an ontological position. According to realists, the law is just a part of a social and psychological realm. However, they do not seek an explanation of its of law. Neither Holmes nor other important representatives of this standpoint (such as Wesley N. Hohfled or Karl N. Llewellyn) were interested in ontological quandaries. The methodological thesis of realism enhances the need to apply methods of empirical sciences (e.g. psychology and sociology) in jurisprudence, emphasizing the non-autonomy of the latter. The adherents of realism, therefore, view jurisprudence – as Kuhn philosophy of science qua history of science – as both descriptive and a posteriori.

Note that the nature of law was the very subject of the realism–normativism controversy. According to realists, the law is just a part of a social and psychological realm. In other words, the answer to the ontological question ‘What is law?’ is to be investigated within ordinary psychosocial facts. According to normativists, on the other hand, law is a set or system of norms, existentially independent of what officials or laymen actually do. The common assumption of both positions — or, as Robin Collingwood put it, the absolute presupposition of the realism–normativism controversy — is the necessity of a choice between dichotomous options: law belongs to the domain of either Sein or Sollen (to be exact, like Collingwood’s other presuppositions, this is a presupposition of a question rather than an answer). This is precisely the axiom Fuller strongly wishes to reject. He is convinced that a proper explanation of the phenomenon of law should take into account both facts and norms. Within a framework of an adequate philosophy of law, there is no room for the Sein and Sollen dualism³³.

Therefore, he sets his own theory together with Polanyi’s elitism (as Lakatos labelled it³⁴). According to the Hungarian-British polymath, science ‘is a team activity consisting in searching for institutional forms and practices appropriate to its specific goals and problems’³⁵. As an institutional fact, it has a two-faced nature, both factual and normative. Institutional facts consist in sets of constitutive rules (as noted in Sect. 2), which, however, cannot exist, as Searle correctly admits, apart from any material substrate³⁶. As Fuller believes, the aforementioned characteristics could be equally applied to law. It may be presumed that the artefactual theory of law is one of attainable ways to clarify Fuller’s very general comments.

Briefly, Polanyi’s account can be characterised by the following four theses:

(E₁) in accordance with objective criteria, scientific theories are divided into better and worse,

(E₂) the scientific elite can make right choices in this regard,

(E₃) these criteria are elitist in the epistemological sense and cognitively inaccessible to laymen,

(E₄) science becomes better and better with the passage of time.

Throughout his classic book, Fuller simply contented himself with noting this appealing parallelism. One of his motives — among other possible reasons — was the obvious disagreement with the thesis (E₃). If the demarcation criteria contrasting science and bad science, law and bad law (or even a broken clock and a piece of iron), were to be intersubjectively unavailable for cognition, they could not be the matter of the philosophy of law or science. In particular, any rational reconstruction of the context of justification or discovery would be ruled out by the very fact. And how else — we are to question rhetorically — is one to understand the eight requirements of the internal morality of law set forth by Fuller? These specify criteria for the activities of lawmakers to count as law-making. Unlike Polanyi, the American philosopher of law was persuaded that the internal morality of law (or science) was not merely a fragment of the ‘tacit knowledge of lawyers’ (or scholars). On the contrary, the epistemology of science, and thus law, seems to be within the scope of our intellectual activity. However, in a methodologically mature form, this enterprise was not taken up and developed until the 1970s by Lakatos and Laudan. Let us elaborate the line of thought sketched by Fuller.

4.2 Between Positivism and non-positivism

At this point, I would like to consider the hypothesis that there is a partial analogy between Fuller’s methodology and that of rational reconstructions. It consists in — as similarity described in the previous section — the rejection of the common assumption that underlies the dispute between pairs of radical positions: on the one hand, two oppositional views of neo-positivism (in the philosophy of science), and, on the other, realism and normativism (in the philosophy of law). The first of these symmetries was outlined against the background of the dispute over the nature of law. The second, however, concerns the question of its validity, and hence the positivism and non-positivism controversy. The second figure illustrates the putative parallelism, which is specified in more detail in the ensuing paragraphs.

Fig. 2

Figure Two

As pointed out before, sociologism explains the history of science in terms of social and psychological causes. In unfolding his falsificationism, Popper meaningly ignored the analysis of any historical material. Its normative postulates were justified in a purely a priori way. The examples that Popper used to illustrate his reasoning turned out to misrepresent the true history of science. Subsequently, Lakatos’s intention was to synthesise these prima facie incompatible approaches. He achieved it through rejecting the common assumption made by Kuhn and Popper that the proper explication of a scientific discovery exclusively focused either on its non-rational causes or the reasons that guide a scientific community in a given research situation. The solution he submitted was already mentioned. Any adequate rational reconstruction has to acknowledge both factors: causes and reasons. The regrettable dualism is thus dispelled.

The following question points out the bedrock of the positivism and non-positivism controversy: are there any normative — but independent of any legal system — demarcation criteria delimiting law and bad law that is not a law at all (not a clock, but just a piece of iron)? Adherents of positivism may agree that law is a meliorative thick term (though, they do not always do so, and John Austin’s account seems to be a representative example here). On that assumption, they must recognise that the proposition ‘x is a law (i.e., a legal system or a legal norm)’ implies a positive appraisal only in some pragmatic sense. It entails a certain presupposition or implicature that might be cancelled in a regular conversation. Regardless of moral evaluation, bad law invariably falls under the artefactual kind LAW.

The advocates of non-positivism, on the other hand, admit that in some cases bad law does not constitute any law at all. The famous maxims of St. Thomas Aquinas or Gustav Radbruch — lex iniusta non lex est or lex iniustissima non lex est — assume that the judgement ‘x is law’ implies semantically that x is not unjust (as in St. Thomas’s dictum) or that it is not unjust or wrongful in the highest degree (as in Radbruch’s formula). Radbruch opposed the thesis that the sufficient condition for a set of (constitutive) rules to be law is to establish it in a formally legal way. The author of Five Minutes of Legal Philosophy gave priority to moral legitimacy over formal legality, distinguishing law from statute. In his opinion, every legal system should correspond to the a priori idea of law (including universal, ahistorical norms of natural law). The category of ‘statutory lawlessness’ in Radbruch is somewhat analogous to Popper’s pseudo-science (such as astrology or phrenology³⁷). The common assumption of these positions is that morality marks the external standard that law must satisfy by the very definition. Fuller, on the contrary, speaks of the internal morality of law.

Immoral law in the sense of St. Thomas or Radbruch resembles rather a behaviour of scientists who make discoveries by breaking some ethical norms (e.g., by carrying out cruel experiments on animals). However, this type of conduct — although immoral and worth condemning — actually constitutes science as long as it meets the standards of its internal rationality. The same stipulation applies to law.

Within the scope of the artefactual theory of law, the position of non-positivism can be supported by the following reasoning. The argument leads to the conclusion that — contrary to one of two fundamental theses of the metaphysics of positivism (next to the social theses) — between law and morality there is a conceptual link. First, the artefactual type LAW is existentially dependent on the concept of law. Second, the concept of law defines the necessary conditions for falling under the artefactual kind LAW. Third, the concept of law is a meliorative thick concept. Thus, there are normative conditions for falling under the kind of LAW that give us a demarcation criterion contrasting law and non-law, particularly law and bad law. The question of whether a set of constitutive rules or a systematic and rule-governed practice of officials constitutes law concerns not only facts but also — to put it mildly — values.

As a result, it seems that Fuller was rightly speaking of the morality of law and ascribing it — unlike the theorists of natural law — an internal status. This may be interpreted as a ‘compromise’ feature of his position. It is true that Fuller did not analyse the historical material, contenting himself with imagining fairy-tale scenarios (recall, e.g., the story about King Rex from The Morality of Law). The eight requirements of good law (or eunomy, as he referred to it) were justified only by a priori thought experiments. His methodology was a prominent example of so-called armchair philosophy. A posteriori justification would have to be relativised to a given legal system (e.g., American or Polish) and aim at a historical analysis of the law in action.

5 Ontology of law

In this light, Fuller appears primarily as a theorist of the positivism and non-positivism controversy. He wondered what common assumptions, shared by the adherents of extreme doctrines, led the 20th -century philosophy of law into a deadlock. The importance of his thoughts lies, among many others, in that he successfully identified not one but two absolute presuppositions of that dualism. He decided to deny both of them, creating his own original and fruitful doctrine. The first dogma has already been briefly discussed in the section above. In line with positivism, Fuller recognised that there are no normative and external criteria for the validity of norms or legal systems. Against positivism, however, he attached importance to the internal morality of law. Furthermore, in The Morality of Law Fuller revealed the second of Collingwood’s presuppositions of the aforementioned controversy. As he observes acutely: ‘[i]t is truly outstanding to what an extent there runs through modern thinking in legal philosophy the assumption that law is like a piece of inert matter – it is there or not there’³⁸. Let us rethink this thought in terms of the artefactual theory of law. (In the margin, I would like to emphasise that the tenet revealed by Fuller dominated not only the philosophy of law, but the mainstream of metaphysics in general, starting with Aristotle and ending with Tadeusz Kotarbiński. In reverse, Alfred N. Whitehead was one of the thinkers who made explicit and rejected this presupposition³⁹).

We distinguish two pairs of conflicting statements in the positivism and non-positivism controversy, respectively radical and moderate ones. The common assumption of radical positions is the thesis that the answer to the question of whether a given set of constitutive norms is in fact a law must always be either ‘yes’ or ‘no’. In other words, they reckon that the artefactual kind LAW corresponds to the notion of set in the classic sense of the set theory. Hence, the function of belonging to a given collection takes value in the set with two elements {0, 1}. Therefore, a claim ‘x is law’ is either true or false, with no third option: tertium non datur.

Yet, as Fuller teaches us, this logical axiom does not necessarily apply here⁴⁰. Consider some moderate answers to this essential question. One was coined by Fuller himself, and the other comes from John Finnis. According to Fuller, sometimes institutional artefacts may resemble pieces of iron. This is precisely the case when they do not provide any reasons for any subject of law. As a matter of fact, it seems to be true only of some of the ‘desiderata’ put out by Fuller, namely of promulgation, prospectivity (non-retroactivity), consistency, and possibility of obedience. That in sum makes four postulates, that is, half of their total number (the others are generality, clarity, constancy through time, and congruence between official action and declared rules). In this context, Fuller points to emphatic examples taken from Albert V. Dicey (recall that Dicey considered them as ‘the highest exertion and crowning proof of the sovereign power’):

• (i) an act ‘the object of which is to make legal transactions which when they took place were illegal, or to free individuals to whom the statute applies from liability for having broken the law’ (‘the legislation of illegality’),

• (ii) an act of a parliament that dissolves itself and ‘leaves no means whereby a subsequent parliament could be legally summoned’ (self-dissolution of a parliament which Fullers comperes to the act of suicide).⁴¹

Such laws do not constitute a law, for they do not provide any agents with reasons to act. They are not deontic markers in the sense defined by Crowe. Although they might be institutional artefacts in some sense, they are not tokens of the artefactual type LAW.

On the other side of the coin, Finnis, widely regarded as one of the influential theorists of the law of nature, argued that laws are ‘presumptively morally binding’⁴². However, according to the terminology introduced in this paper, this statement allows to classify him as a representative of moderate positivism. Undoubtedly, Finnis would agree that law is a meliorative thick term. However, he would argue that propositions about law imply a positive evaluation only pragmatically, not semantically. In some contexts, therefore, such a presupposition or conversational implication can be cancelled. A bad law, however imperfect it might be, still remains law: a clock, even if broken, is not a piece of iron. Contrastingly, Herbert L.A. Hart, a proponent of radical positivism (although sophisticated), upholds that law is ‘defeasibly morally binding’⁴³. Finnis’s view may be put into words following Kenneth Ehrenberg: although bad law is a defective member of the class of laws, it is legally binding⁴⁴.

Fuller adopted a slightly different approach to this matter. He assumed that an institutional artefact that does not meet the ‘desiderata’ of internal morality is not law in a full-blooded sense. Therefore, Fuller challenged the second absolute presupposition of the positivism–non-positivism controversy. An artefactual type LAW is not a set for which the membership function is defined as zero-one. Rather, it corresponds to a fuzzy set whose elements have degrees of membership. In this case, the membership functions are valued in the real unit interval [0, 1]. Fuller’s moderate non-positivism amounts to the statement that an institutional artefact that does not satisfy the requirements of internal morality is not completely a law. ‘Of course’, as he maintains, ‘both rules of law and legal systems can and do half exist’⁴⁵.

On what basis one can determine whether Finnis’s or Fuller’s standpoint is accurate? It is impossible to refer to the mind-independent ‘essence’ of the law. After all, we have said that in the case of artefactual kinds, an essence is constituted by the collective intentions of creators. The substantive concept of law belongs to the content of these intentions. The corresponding artefactual kind depends existentially on the notion of law. The Finnis–Fuller controversy could only be resolved on a conceptual basis. Therefore, whatever position is apt, it must be true analytically, simply by virtue of the meaning of the concepts. The allegedly contradictory theories are metaphysically equivalent descriptions of the same artefactual world.

There is no ‘real’ or ontological difference between the theories laid out by Finnis and Fuller. At most, the difference is conceptual. However, it is very unlikely that the concept of law provides us with rigid demarcation criteria. Rather, the notion is vague, and hence the speculative distinction between law and bad law might be only arbitrarily established. Fuzziness of LAW as artefactual kind implies that corresponding membership function does not satisfy a bivalent condition. Notice that philosophers of science abandoned former attempts to seek rigorous demarcation criteria for scientific activity (e.g., provided by experimental physics). This observation suggests that we should not expect to find an analogous touchstone in the philosophy of law.

However, the lack of such necessary conditions does not exclude rational reconstructions as a philosophy of law. By reconstructing the internal history of a given legal system, we could gain a useful instrument to criticise some practice of making and applying the law. Remember that internal history is prescriptive towards the real history. To someone acquainted with continental philosophy, the methodology outlined above might seem congenial with the so-called normative reconstruction of social institutions and practices. Such methodology — revealing the extent to which our institutions and practices implement and strengthen socially legitimate values ​​— has been worked out and developed by one of the representatives of the fourth generation of the Frankfurt School, Axel Honneth⁴⁶.

***

Finally, for the sake of clarity, let me summarise the above considerations in the following points:

• (1.1) Law belongs to the ontological category of institutional artefacts, and thus it is concept-dependent.

• (1.2) The concept of law is a meliorative thick concept.

• (2.1) A creation that falls under the artefactual kind LAW must satisfy a success condition, being a deontic marker for legal subjects⁴⁷.

• (2.2) Therefore, there is a distinction between law and bad law that is not law, and the demarcation criteria are independent of any particular legal system.

• (2.3) Bad law that does not serve as a deontic marker and hence provides no one reasons for action is fictional, not real (a piece of iron, not even a broken clock); imaginary norms of Dicey fit as an example here.

• 6. (3.1) Between the moderate positions of positivism and non-positivism (examples of which are the theories of Finnis and Fuller), there is no ontological but a conceptual difference.

• (3.2) However, the concept of law does not provide exact criteria that would allow to determine in each and every case whether a given institutional artefact is a bad law, but a law, or a bad law and therefore not a law.

• (3.3) Consequently, the function of membership to the LAW as artefactual kind takes the value in the interval [0, 1] rather than in the set {0, 1}; artefactual kind LAW resembles a fuzzy set.

• (4) In practice, rational reconstruction as a philosophy of law, recognising the difference between its internal and actual history, may result in insightful critique of making or applying the law.