General introduction
Full access

This introduction provides an overview of the substance of the book – together with its aims – and, in particular, the approaches that will be adopted. It discusses in some depth a diachronic approach as contrasted with a synchronic approach and then goes on to discuss briefly the various frameworks within which legal reasoning is to be examined.

In an earlier work, by way of its introduction, a range of questions about judging and legal reasoning were posed.1 These questions are just as relevant for this present work, but the emphasis here will be different. The overriding question will be the ways in which legal reasoning might be rethought in the context of legal knowledge (epistemology) in general. What are the frameworks that offer possibilities? Are there institutions, concepts or (and) notions that might provide a different perspective than the ones normally offered? In truth it would be idle to think that there is much to be said that has not, in some way or other, been said before given the very long history of legal thought in Europe. And so it is important to state at the outset that the present project is actually quite a modest one. It is to provide suggestions rather than assertions or answers.


What is legal reasoning? From a practitioner point of view, it is at one level relatively easy to describe and to define: it is the reasoning used by lawyers to solve a legal problem, to advise a client, to justify a legal decision and (or) to comment on a legal text or case. Some might say that it is just ordinary reasoning used in a legal context (see Chapter 10).2 If one sees legal knowledge – as most lawyers seemingly do – as consisting of rules and principles (see 4.2), then legal reasoning is the reasoning attached to the application of these norms to specific problems.3 At another level, however, legal reasoning and argumentation can be seen to be extremely complex. One reason for this complexity is because all method and reasoning in the sciences and social sciences is complex. The knowledge that emerges from the social science reasoning processes depends upon the schemes of intelligibility employed, the inferential reasoning techniques applied, the level of observation adopted,4 the theory brought into play and the paradigm orientations that inform these schemes and reasoning techniques.5 Legal reasoning might seem more straightforward than reasoning in, say, sociology given that it would appear to be just a matter of applying rules to facts, but such a view is simplistic (see for example Chapter 6). Moreover legal reasoning, strictly speaking, needs to be distinguished from legal argumentation: one has the aim of both arriving at and justifying a decision while the other aims to convince.6

In fact one of the problems with legal reasoning is that it is capable of being approached from a variety of disciplinary and theory perspectives, each of which gives a rather different account. This may, of course, be true of reasoning in other areas as well; yet with respect to law it is more acute for what might be called a paradigm reason (cf 10.9). Law, like theology, is subject to a particular paradigm orientation which might be labelled the authority paradigm.7 By this is meant that lawyers work with texts that can be criticised but never questioned in terms of their authority. A law is a law, and so the scope of reasoning about laws is always going to be limited in comparison to work in other, what might be called enquiry paradigm, disciplines. Consequently legal reasoning viewed from inside the authority paradigm – what might be called the internal paradigm view – could well be different from such reasoning viewed from outside the paradigm, that is to say an external paradigm view, in that the epistemological models might well be different.

Perhaps the main difficulty, then, is to determine the approach to be adopted. Does one look at what legal reasoning has been (diachronic) or does one look at what it is now, that is to say legal reasoning as a timeless set of techniques situated in the here and now (synchronic)? Does one adopt an approach internal to the law – that is from the viewpoint of a lawyer and (or) judge – or does one adopt an approach external to the law, for example as a social scientist or a philosopher? If one adopts an internal view, does one envisage legal reasoning to be based on a model of rules, of rights, of policy considerations, of remedies, of causes of action or what (see Chapter 4)? If one adopts an external view, does one employ a causal, structural, functional, hermeneutical, individualist actional or a dialectical scheme of intelligibility (see Chapter 10)?8 Moreover how are the standard techniques of inferential reasoning – induction, deduction, abduction, analogy and metaphor – to be accommodated within these various approaches (see for example Chapter 7)?

In addition, and just as complex, underpinning these models and techniques are questions of theoretical approaches to law. Broadly speaking one can, traditionally, talk of three different schools of legal theory, namely natural law, positivism and realism. How might these theory positions affect reasoning and argumentation? Another distinction is between realism and formalism.9 How might this dichotomy impact upon reasoning? Epistemological issues bring their own distinctions and dichotomies.10 Is law, and its reasoning, any different in essence from scientific reasoning? For example, is there a fundamental distinction to be made between explanation (scientific reasoning) and understanding (social science reasoning)?11 Might one wish to distinguish between axiomatic and casuistic reasoning (see 10.5)?12 Has legal reasoning passed through distinct stages? Thinking about legal reasoning and argumentation will, in principle, require an examination of, and reflection upon, all of these different models, techniques, schemes and paradigm orientations.

What approach, therefore, might be profitable? Two broad epistemological approaches seem, as already suggested, possible. They are Robert Blanché’s distinction between the diachronic and the synchronic, the latter being described as a direct approach while the former is regarded as an analyse historico-critique.13 As this epistemologist noted, the normal approach adopted by scientists is one that sees science in its actuality and this is, in some ways, hardly surprising.14 Scientific research and application is about the here and now. The same can be said about law and thus if one looks both at legal education and at legal theory the emphasis is almost exclusively on law that is currently in force and on theories that are fashioned to comprehend this actuality. As Ronald Dworkin implied, one does not need to know the history of mathematics to be a good mathematician.15 Of course the notion of precedent, relevant to the common law systems, implies a certain historical dimension to legal knowledge, but past cases are normally to be understood in terms of their present relevance rather than as historical documents in themselves. It is perfectly possible (indeed it is the norm), therefore, for law graduates never to have studied legal history. As for historical jurisprudence, it has traditionally been regarded as both discredited and largely ignored since the nineteenth century,16 although by definition natural law theory tends to have something of an historical dimension given its association with the expression ius naturale and with the jurisprudential schools of the past. Legal positivism, in contrast, is seen (or was seen) as scientific and thus a-historical. Indeed why should the positivist study old law?


However, in the twentieth century there was a major resurgence of the diachronic approach in the natural sciences thanks to historians of science such as Gaston Bachelard and Thomas Kuhn. The importance of these two scholars was that they presented an historical vision that was at odds with the traditional linear and progressive view of the development of scientific knowledge. Bachelard saw this development in terms of inertia and epistemological obstacles and thus invited the scientific community to see scientific knowledge more as a matter of rupture (or at least jumping over obstacles) as much as continuity.17 Kuhn presented a similar picture arguing that science went through paradigm changes that were so radical that they amounted to scientific revolutions.18 The relevance of the historical approach is clearly evident in these writings in that, as Blanché says, it ‘offers a good means of analysis in separating, by the date and by the circumstances of their appearance, the various elements which have contributed to form little by little the notions and principles of … science’.19

Do these diachronic insights into the sciences hold any epistemological lessons for jurists? Certainly Blanché’s observation about history permitting one to appreciate the step-by-step formation of science would appear most relevant to legal knowledge. There is a long history of legal thought from early Roman law to modern legal thought and this history is fundamental to legal epistemology in several respects (see Chapters 1–3).

First, the institutions, concepts and categories that are in use today have, for the most part, come originally from the Roman law sources (see Chapter 2). The history of law in continental Europe is, then, in one respect simply a history of Roman law (see Chapters 1 and 3).20 Secondly, the methods used by jurists have themselves been fashioned by history. Indeed the whole formation of the civil law tradition is in some ways a history of differing methods; and so Christian Atias talks not of a progression in method, but a sedimentation. Different methods have built up one upon another.21 Thirdly, many of today’s legal theories have their roots in notions that can be traced back to Roman sources. For example the seeds of natural law, positivism, social contract and even the great nominalism versus universalism debate are to be found in the Digest.22 In fact the philosophical thesis asserted by Ronald Dworkin, despite his disavowal of history, belongs within a tradition that has its origin in the approach of the seventeenth-century French jurist Jean Domat.23 Anyone interested in the formation of the principles, concepts, methods and theories associated with legal knowledge would therefore seem almost obliged to adopt at some point a diachronic approach. Even the development of the common law cannot escape from this civilian historical development since, as Peter Stein has pointed out, while ‘English law has remained relatively free of Roman influences, English jurisprudence has traditionally turned for inspiration to the current continental theories, necessarily based on Roman law’.24

Less easy is the identification, within the history of legal thought, of the kind of epistemological obstacles, ruptures and scientific revolutions that were identified in the sciences by Bachelard and Kuhn. Have there been scientific revolutions? This is a question that will perhaps be implicit in the early chapters of this present work but is actually addressed elsewhere.25 One lesson is that one should not jump to conclusions before Kuhn’s book is read with care. Merely because some jurist or jurists come(s) up with a new way of viewing cases – one thinks of Goff and Jones’ The Law of Restitution first published in 1966 – this does not mean that there has been a change of paradigm. Ian Maclean has written quite convincingly that that the so called humanist revolution in the sixteenth century does not qualify as a Kuhnian revolution.26


Yet when methods are studied synchronically Mark van Hoecke argues that ‘there is a somewhat schizophrenic situation in which one discipline, legal doctrine, is basically studying law as a normative system, limiting its “empirical data” to legal texts and court decisions, whereas other disciplines study legal reality, law as it is’.27 He does not advocate abandoning the legal doctrine approach but he does go on to say that all too often ‘it lacks a clear methodology and the methods of legal doctrine seem to be identical to those of legal practice; it is too parochial’. And thus ‘there is not much difference between publications of legal practitioners and of legal scholars’.28

If legal reasoning is too parochial, as Mark van Hoecke puts it, what are the consequences? One is that there has been relatively little interdisciplinary research with regard to the way jurists analyse and view, for want of a better term, social reality. Some may dispute this, but if one turns to comparative law this want of interdisciplinary research becomes particularly evident. There has been much attention focussed on the functional method29 and, more recently, hermeneutics;30 yet there is little scholarship placing these methodological schemes in some kind of epistemological context. What are the alternatives to functionalism and to hermeneutics?31 In social science epistemology there is a body of literature on schemes of intelligibility and on paradigm orientations; indeed there is considerable work on epistemology in the social and human sciences.32 However, not only is much of the work ignored by jurists but social science theorists themselves think that jurists have little to offer social scientists.33

One particular problem with the historical approach is that it is often considered only to be descriptive. History can tell us what legal reasoning has been but it cannot inform us about what it ought to be. ‘Theories that ignore the structure of legal argument for supposedly larger questions of history and society’, wrote the legal philosopher Ronald Dworkin, end up ignoring ‘questions about the internal character of legal argument’ with the result that ‘their explanations are impoverished and defective, like innumerate histories of mathematics’.34 This internal character was for Dworkin essential in that legal practice, ‘unlike many other social phenomena, is argumentative’. Each participant ‘understands that what it permits or requires depends on the truth of certain propositions that are given sense only by and within the practice’.35 One is back at the philosophical dimension of legal argument in turn implicating the very notion of law itself. Legal argument cannot just be described, implies Dworkin, because the empire of the law ‘is defined by attitude’. And this ‘attitude is constructive: it aims, in the interpretive spirit, to lay principle over practice to show the best route to a better future, keeping the right faith with the past’.36

In terms of this ‘best route’, is legal reasoning and argumentation to be considered uniquely as a system of signs or elements interacting together as a structure which can be described independently of what the system evokes?37 Or is it something more – something that can only be grasped from the inside and described, not by any epistemological discourse, but only by some philosophical notion such as ‘attitude’?


Yet, however ‘descriptive’ the exercise of looking at the past might be, it is arguable that such descriptions are the only way in which one can actually appreciate what Dworkin has called ‘attitude’. Such an attitude must come from somewhere and while it may be true that being a good mathematician is not dependent upon a knowledge of the history of mathematics, Dworkin’s analogy is misleading in as much as it is helpful. Understanding the method – the ‘attitude’ – of mathematicians is, at least in one sense, much less complex than trying to understand the methods – or attitude – of social scientists. Mathematics is not continually engaged, as a discipline, in methodological arguments whereas the social sciences are. This is not to suggest that there are not important epistemological debates to be found within the philosophy of mathematics;38 but mathematical analysis is so precise that it can be replicated by machines. In the social sciences not only is it not yet possible to produce artificial intelligence (AI) programmes that can replicate the methods but, equally, methodological debates form part of the very substance of social science disciplines.39 Law, it might be said, is different. But is it? This is one particular issue that deserves investigation (see for example Chapters 7 and 8). However, it has certainly not yet proved possible to produce any AI programme that thinks just like a judge or jurist and trying to describe legal reasoning raises difficult epistemological questions. Can it be divorced from philosophy? Can it be isolated from reasoning in other disciplines? It is these questions that force one back to the diachronic approach. What has legal reasoning been?

In examining legal reasoning from an historical perspective two assumptions (for better or for worse) will be made. The first is that the object of this examination will be professional jurists and lawyers ranging from Roman to modern times. The justification for this assumption is that it is Europe which has provided the intellectual and social context for the development of the professional discipline of ‘law’ if one sees law as consisting of a specific kind of language and set of practices. Of course if one defines, epistemologically, law more widely – as covering say all forms of dispute resolution – then the European orientation becomes much less justifiable. Nevertheless it can certainly be argued from an historical perspective that the language, theories and practices of law, as seen as a university discipline in many of the world’s law faculties, are an aspect of European social and intellectual culture. James Gordley in fact goes further: he sees the European jurists as ‘team players’ and doubts that ‘one could write a similar history of artists or philosophers’.40

The second assumption, also open to question, is that this European-inspired idea of law embraces the two traditions within this European idea, namely the civil law and the common law. It is certainly true that civil and common law reasoning is often seen as being different; logical, or top-down, reasoning is contrasted with casuistic, or bottom-up, argumentation.41 But theorists such as Chaïm Perelman and Mitchel Lasser have indicated that such a dichotomy must be treated with caution.42 Moreover, from a reasoning point of view, it is arguable that Roman methodology is closer to that of the common law than with contemporary civil law which means that one cannot assign one form of reasoning to the civil law and another form to the common law.43 History reveals a more complex picture. Yet, as Gordley notes, ‘it was not until the nineteenth century that English courts arrived at the “elementary conceptions” that we have used to illustrate Roman law: the protection of possession, as distinct from ownership; liability for fault, as distinct from strict liability; consent in contract formation that could be vitiated by mistake’.44 Accordingly it will not be implied that there is no difference of legal mentality between the two European legal traditions. However, differences will be located in types of reasoning and argumentation – and perhaps in a difference of epistemological emphasis – rather than in differences of legal traditions as such, although of course the histories of the two traditions are by no means the same and this has impacted on method (see Chapter 3). One common factor that might be helpful, as we have seen, is that of attitude. Is there a Western attitude that underpins legal reasoning?

A diachronic approach has, then, much to offer. But so does a synchronic approach if one is prepared to look beyond the legal literature. Nevertheless it is probably impossible to comprehend law as a body of knowledge without a comprehension of the history of legal thought and reasoning, including of course the language through which the discipline has expressed itself. One reason for this can be found within the discipline of history itself. ‘In order to understand what history is’, wrote Donald Kelley, ‘we must first ask: What has history been?’45 Similarly one might say that in order to understand what law is we must first ask what law has been. As a discipline law is both old and, seemingly, remarkably stable. Much of the language and many of the fundamental categories used by Western lawyers were, as has been mentioned, developed by the Romans; the reasoning and interpretation issues also have equally long histories.46 The starting point, therefore, for rethinking legal reasoning will be the ‘Kelley question’. But, going beyond this question, it is the challenges of other disciplines that will help perhaps provoke a rethinking (see in particular Chapters 7, 8 and 10).


What is likely to emerge from this rethinking? Certainly not, it must be said at once, a new ‘theory’ of legal reasoning as such. This is not actually an aim of this present work. And indeed the theories of others will not on the whole be the objects of engagement. The aim is, instead, to examine legal reasoning as it is (that is to say largely in the courts) in relation to its knowledge base and to identify a range of what might be called reference frameworks and legal ‘artefacts’47 that can serve as the means for rethinking legal reasoning. The purpose is ontological and epistemological. This said, two general theses (for want of a better term) will be advanced. These are, first, that the legal notion of an ‘interest’ might perhaps be a very suitable artefact for rethinking legal reasoning (Chapters 11–12); and, secondly, that fiction theory might be the most viable ‘epistemological attitude’ for understanding, if not rethinking, reasoning in law (Chapter 9). However, the main thrust of this ‘rethinking’ work is not as such to advance a thesis; it is to provide the means by which others – the readers – might undertake their own rethinking exercise.48

Accordingly legal reasoning will be examined in terms not just of its history (Chapters 1–3). It will be examined equally in terms of its foundation in legal literature (see in particular Chapter 5), of perceived factual situations (see in particular Chapters 6–7), of its comparative discipline settings (see in particular Chapters 7–8), of its institutional focal points (see in particular Chapters 2 and 6) and of its epistemological contexts and foundations (see in particular Chapter 10). Now, while the aim is not to fashion any new theory of legal reasoning as such, this does not mean that there will be no attempt to develop some epistemological frameworks for understanding legal reasoning today. One framework will be ‘ontological’; that is to say the various elements in which legal reasoning might be seen to be grounded (see for example Chapter 4). The other framework will be epistemological. Here it will be argued, as already mentioned, that fiction ‘theory’ (Chapter 9) offers a possibility for some new thinking with regard to law and legal reasoning. In addition, as also mentioned, interest theory could equally make an important contribution (Chapters 11–12).

These frameworks (rather than theories) will not radically change legal reasoning itself, but they might offer some insights into what legal reasoning actually is as a fully intellectual process in a way that might interest those operating in disciplines other than law. Yet if there is one single aim it is this. A French social scientist once dismissed the discipline of law as having little relevance for social science and humanities theorists;49 perhaps a rethinking of legal reasoning in the manner that will be suggested in this work might help change this attitude. It might hopefully suggest that there is more to law as a discipline than description of, and reasoning about, rules.

1   Samuel (2016a), at 1–3.

2   See eg Alexander and Sherwin (2008).

3   Susskind (1987), at 78–79. And see eg Bengoetxea (1993).

4   Desjeux (2004).

5   See generally Berthelot (2001).

6   One claimed difference between reasoning and argumentation is that the latter always involves the participation of another person: see eg Oléron (1996), at 4. However, legal reasoning, as well as legal argumentation, could be said always to involve another person, either a client or, in the case of a judgment, the wider audience of the legal profession if not the public in general.

7   See Samuel (2008a) and (2009a); and Samuel (2016a), at 109–116.

8   Berthelot (1990), at 62–83.

9   Twining and Miers (2010), at 114–116.

10 Atias (1985), (1994) and (2002).

11 Makkreel (2006) 441.

12 Jonsen and Toulmin (1988). See Samuel (2016a), at 115–116.

13 Blanché (1983), at 33–39.

14 Ibid, at 34.

15 Dworkin (1986), at 14.

16 Cf Stein (1986). Brian Tamanaha contests this assertion about historical jurisprudence and argues that a ‘great deal of contemporary legal theory is informed by history’: Tamanaha (2016), at 329.

17 Bachelard (1938).

18 Kuhn (1970).

19 Blanché (1983), at 36.

20 See generally Stein (1999); Maruotti (2011).

21 Atias (1994), at 66–77.

22 See generally Jones (1940); Gordley (2013).

23 Renoux-Zagamé (2007).

24 Stein (1980), at 123.

25 Samuel (2017a).

26 Maclean (1992).

27 Van Hoecke (2011b), at 2.

28 Ibid, at 3.

29 See eg Michaels (2006).

30 See Legrand (2015).

31 See further Samuel (2009c).

32 See in particular Berthelot (2001).

33 See Berthelot (2001), at 12–13.

34 Dworkin (1986), at 14.

35 Ibid, at 13 (emphasis in original).

36 Ibid, at 413.

37 Blanché (1983), at 120.

38 Ibid, at 78–107.

39 See eg Desjeux (2004).

40 Gordley (2013), at x (preface).

41 See eg Gummow J in Roxborough v Rothmans of Pall Mall (Australia) Ltd [2001] HCA 68, (2001) 208 CLR 516, at para 72.

42 C Perelman (1979); Lasser (2004). On Perelman see now Goltzberg (2013).

43 See famously Pringsheim (1935).

44 Gordley (2013), at 23.

45 Kelley (1970), at 2.

46 See eg Maclean (1992).

47 This very useful notion of a legal artefact is one that has been fashioned by Dr Maks Del Mar in his own work on legal epistemology.

48 See also Samuel (2016a).

49 Berthelot (2001), at 12.