Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 17 September 2019

Philosophy of the Common Law

Abstract and Keywords

This article presents current practice and historical evidence about the common law. Common law and its widely shared conceptualizations are considerably very complex. It discusses the concept of artificial reason in common law that differs in two respects from natural law. The aim of this artificial reason is as a convergence of judgment on common solutions, thereby securing effective practical guidance. Its characteristic mode of functioning may raise important issues for the philosophical understanding of the nature of law and legal reasoning. The article expresses material conventionalism and the congruence thesis and raises various arguments. This article further outlines key themes and broad notions of common law jurisprudence and translates them into a contemporary philosophical expression and explores arguments that might be given in support of them.

Keywords: common law, artificial reason, jurisprudence, philosophical understanding, legal reasoning

1 Introduction

Common law is judge-made law. Every student of Anglo-American law knows that. By virtue of the doctrine of stare decisis, judicial decisions are held to be binding not only on the parties in the case before the court, but also on future courts in the jurisdiction deciding similar cases and thus binding on all those to whom the judge-made rules in question are addressed. In novel cases, where law arguably is silent, judges fill the silence with new binding precedents. Of course, stare decisis allows courts to ‘distinguish’ the cases they face from what might first appear to be relevant and binding precedents, and allows courts to ‘extend’ precedents beyond their explicit four corners by analogy. The doctrine, reluctantly and within narrow limits, even allows judges to overrule precedents when judges find them to be especially troubling. In this way, common law courts massage precedents and in the process make new law. There is nothing puzzling in this familiar tale. It poses no especially interesting challenge to existing theories of law and fits very comfortably into a variety of positivist accounts. (We might also be able without too much difficulty to square it with rival natural theories.) Law is essentially the product of lawmaking—the only major difference between common law and other forms of law lies in the institution doing the legislating.

Well, it may be just a little puzzling. After all, it appears that judicial legislating is done in the course of adjudicating. Legal rules are made in the course of applying the (p. 589) rules to the very case that called for the rulemaking. This is enough to convince Fred Schauer at least that common law is ‘uncommonly puzzling’ (Schauer 1989: 455). And Bentham attacked it relentlessly for confusing these two very different functions, resulting in radical uncertainty and a mask for judicial abuse of power (Bentham 1970: 184–95). We expect judges to follow rules, but it appears that in common law practice rules follow the judges. Common law becomes even a little more puzzling when we look into the history of English law. In a Yearbook report of a fifteenth-century case we find it argued, ‘precedents and usages do not rule the law, but the law rules them’ (Tubbs 2000: 45). Legal historians widely agree that before the eighteenth century there was no firm doctrine of stare decisis in English common law (Simpson 1973: 77; Gray 1992: 157–8). Indeed, it may not have been established until sometime in the nineteenth century. It is not entirely a coincidence that the positivist understanding of common law as strictly judge-made law emerged at the time of the hard-ening of the doctrine of precedent in English common law. Indeed, it was Bentham who first used the term ‘judge-made law’, hurling it against English law as a term of contempt and abuse. The fact that it seems so innocent, and even descriptively appropriate, to us in the contemporary common law world is testimony to the distance we have travelled from the jurisprudential world of classical common lawyers—a journey launched by Bentham's withering critique of eighteenth-century common law and John Austin's recasting of common law understanding to make it fit classical positivist jurisprudence.1

But if historians are correct, English common law functioned well enough for over 500 years without the one thing that, according to current orthodoxy, held the practice together as a form of law. Moreover, it was this common law practice in its seventeenth-century form, understood as seventeenth-century common lawyers understood it, that spread with English colonialism to the New World and beyond. Thus, current practice and historical evidence pose a challenge to familiar views about the common law. Common law, and widely shared conceptualizations of it, may have been (and may continue to be) considerably more complex than our common knowledge admits. What we have taken for common knowledge about common law may not be knowledge at all. We should take a closer look. Its characteristic mode of functioning may raise important issues for our philosophical understanding of the nature of law and legal reasoning.

How, then, are we to understand common law? This question appears to be unwieldy, especially if one expects a survey of common law jurisprudence from its birth in the twelfth century to its modern manifestations around the world. To make the task a bit more manageable I propose to look at common law jurisprudence in the seventeenth century, a very critical point in its development. By the seventeenth century common law practice had matured and there had emerged a group of reflective (p. 590) common lawyers who sought to articulate, albeit in a piecemeal, occasional, and sometimes partisan political fashion, a coherent understanding of the law they practised. They were engaged participants in the legal practice of their day: lawyers, judges, royal counsellors and parliamentarians, not philosophers. Hence, they never articulated a full-fledged philosophical theory of law. Yet, they shaped a distinctive perspective on questions about the nature of law and legal reasoning, and the normative authority of law, questions that are still at the center of philosophical reflection on law. In Section 2, I outline key themes and broad notions of common law jurisprudence; in Section 3 I translate them into a contemporary philosophical idiom and explore arguments that might be given in support of them.

2 Classical Common Law Jurisprudence

To classical common lawyers, law was not something laid down either by will or nature; rather, it was something taken up,2 used in deliberation and argument, and followed in practice: ‘the only method of proving, that this or that maxim is a rule of the common law’, Blackstone wrote in the mid-eighteenth century, ‘is by shewing that it hath been always the custom to observe it’ (Blackstone 1765: i. 68). Law was regarded not as a structured set of authoritatively posited, explicit norms, but as rules and ways implicit in a body of practices and patterns of practical thinking all ‘handed down by tradition, use, [and] experience’ (Blackstone 1765: i. 17). These rules were the product of a process of a common practice of deliberative reasoning, and constituted the basic raw materials used in it. Common law was ‘reasonable usage’ (Hedley 1610: 175), observed and confirmed in a public process of reasoning in which practical problems of daily social life were addressed. ‘Custom’ and ‘reason’ were the twin foci of this conception of law. These two notions were complementary, mutually enhancing and supporting, and mutually qualifying.

2.1 Common Law as Custom of the Realm

All general discussions of the common law started with the claim that common law was common custom of the whole realm (Hedley 1610: 175; Hale 1971: 17, 30; Blackstone 1765: I, 67). This was ‘general custom’ (as opposed to local custom of manor or shire) that was ‘immemorial’, existing from ‘time out of mind’. For some (p. 591) common lawyers this was sufficient proof of its wisdom. Hedley argued, for example, that common law is common reason ‘tried by time’, the ‘trier of truth [and] author of all human wisdom’ (Hedley 1610: 175; see Coke 1628: 97b). Other seventeenth-century writers more modestly held that long usage fitted the law to the nature of the English people (Hale 1971: 30). There was also some disagreement over what it meant to claim that common law was immemorial. Coke claimed that most of the key doctrines and rules of common law remained essentially unchanged since Roman times (Coke 1793, Second Reports, preface), but in this he was no doubt eccentric. Matthew Hale expressed a more moderate and common view when he wrote in the last third of the seventeenth century, ‘the strength and obligation, and the formal nature of a law, is not upon account that the Danes, or the Saxons, or the Normans, brought it in with them, but [rather that] they became laws, and binding in this kingdom by virtue of their being received and approved here’ (Hale 1971: 43).3 Not to be taken literally, the image of ancient origin, according to Hale, stood for three other key features of common law.

First, the common law was characterized by historical continuity. It may have gone through vast changes over its history, but through these changes it maintained its integrity as a single, coherent body of law. Despite the variations over the centuries, we can say it is the same law, just as the Argonauts' ship was the same when it returned home as when it departed, even though during its long voyage it had been repaired so often that it was made up of scarcely any of the original materials (Hale 1971: 40). And nothing of this ship of law was immune to change.

Secondly, continuity depended on integration of each part into the whole. To claim that common law existed ‘time out of mind’ meant that the validity and binding force of any rule of law depended not on who made it or when, but on its being ‘received and approved’ in the kingdom. This ‘reception’ was manifested in their integration into the body of doctrines and practices that makes up the common law. Integration, not origin, was the key (Hale 1971: 3, 6, 8). This integration of custom, statute, or judicial decision is not simply a matter of logical consistency or coherence. It is a practical and historical matter: practical, because it is a matter of whether the rule is ‘taken up’, practised, and used (by its subjects and by officials who must assess their actions in light of the law); and hence, historical because only time can tell whether a rule, however it happened to be introduced, is thus integrated and becomes part of the common law.

Finally, for Hale, and many of his contemporaries, integration involved accommodating the rule or maxim to the nature of the nation, ‘such as by a long experience and use is as it were incorporated into their very temperament, and, in a manner, become the complexion and constitution of the English Commonwealth’ (Hale 1971: 30; see Davies 1615: ii. 252, 255). The common law was said to be the ‘constitution’ of the nation—its basic normative structure and the root of its collective health. The (p. 592) constitution Hale had in mind was not limited to the constitution of government; it was the constitution of the people. Also, the process of accommodation worked two ways. The rules of law, at first rough and clumsy, are refined over time, softened to fit the contours of the community's daily life. Simultaneously, following the rules and practices shapes the dispositions, beliefs, and expectations of the people. Thus, what they took to be reasonable and practicable solutions to the problems of social interaction depended on a sense of continuity of present practice with the past; but also, what counted as continuous with the past depended heavily on what were regarded by participants as reasonable projections from the arrangements and practices of the past to present conditions and problems. This also explains Hale's claim that common law was ‘received and approved’ in the kingdom. This was not a matter of mere expression of consent, but rather acceptance of it in virtue of and manifested in the law's integration into their lives.

Yet, for Hale, like all common lawyers, custom always had its status as law in and through the activities of the courts. But this raises the important question about whose custom is the custom of the common law, that of the people or that of the courts? There is no doubt that already by the sixteenth century, common law had become highly technical and no longer, if it had ever been, merely a reflection of customs of the land. So, surely, common law was, as Bentham later called it, custom in foro rather than custom in pays (Bentham 1977: 182–4, 217–18). Classical common lawyers did not deny this, but they insisted nevertheless on a fundamental link between the two. The groundwork for articulating this link was laid by St German in the sixteenth century. He distinguished clearly between general customs, which were diffused throughout the realm and known to lawyers and lay people alike, and maxims, which were the specialized rules of law known only in the king's courts (St German 1974: 59). Maxims had their ‘strength and warrant’ in, and ‘take their effect by’, the general customs of the realm. At the same time, they were rooted in a shared sense of their reasonableness, that is, of their suitability in, and their contribution to the coherence of, the rest of the common law and the practices of which it consists (ibid.). Thus, although lawyer's law might have seemed arcane to the ordinary Englishman, nevertheless, its claim to validity was thought to rest on the fact that it was congruent with the customs that were second nature to the people, and with the body of the common law as a whole refined by its distinctive discipline of reasoning. Thus, common law was not to be equated with custom of the realm, but the latter was the radical source of its validity, not literally by derivation, but by source and congruence.

2.2 Common Law as Common Reason

Common reason and natural law. The mantra of classical common lawyers was: ‘the common law is no other than common reason’ (Coke 1628: 97b, 183b; Hedley 1610: (p. 593) 175; Finch 1759: 75; Doddridge 1631: 242). However, despite the echoes of ancient natural law doctrine,4 common law writers had something quite different in mind. They piously granted that natural law was the ultimate ground of all law, but rarely sought to bring this ground into their workaday world. Natural law was too abstract and theoretical for their pragmatic, concretely focused minds; it was too often silent, or contested, or simply out of touch with concrete human affairs (Hale 1956: 502–4). ‘[R]eason is the life of the law, nay the common law itself is nothing else but reason’, Coke famously wrote, but he rushed to add that by this he meant not ‘natural reason’ but the ‘artificial reason’ of the trained common lawyer, ‘an artificial perfection of reason gotten by long study, observation, and experience, and not every man's natural reason …’ (Coke 1628: 97b). Similarly, in typical common law fashion, Hale insisted that the reason of the common law was the embodied prudence and deliberative judgment of the judge who, through his emersion in the concrete details of common law is fluent in the common language of human affairs, and thus best able to articulate notions of the ‘just and fit … common to all men of reason’—better than philosophers or theologians who seek to do so ‘transported from the ordinary measures of right and wrong’ and cut off from ‘the common staple of humane conversations’ (Hale 1956: 502, 503).

This ‘artificial reason’ of common law differs in at least two respects from natural law as commonly understood. First, it was not thought of in terms of broad general principles, and by itself provided no contentful tests by which to assess the legitimacy of a given legal rule or doctrine. It was regarded, rather, as disciplined practice of reasoning. If ‘reason’ legitimated some doctrine, this was only because that doctrine survives critical scrutiny in a process of reasoning and disputation. Secondly, when the clear and uncontested law (what we now call ‘black-letter law’) yielded no unambiguous solution to a legal problem, the tendency of the common lawyer was not to consult extra-legal moral sources, as a natural lawyer might do, but rather to look deeper and longer into the accumulated fund of experience and example provided by the common law. Common lawyers put their faith in the ability of a trained reasoning capacity, immersed in the vast resources of experience supplied by law to yield reasonable and sound solutions in even the most difficult or apparently novel cases.

Artificial reason. Several features characterize ‘artificial reason’ as conceived by classical common law jurisprudence. First, it was pragmatic, focused on practical problem solving. It addressed concrete situations and problems and sought to forge solutions from the materials that were ready to hand. It measured success in terms of whether the proposed solution works; and the measure of its ‘working’ was whether it was ‘taken up’ in further cases.

(p. 594) Secondly, artificial reason was a contextual competence. ‘[M]en are not born common lawyers’, Hale remarked, ‘neither can the bare exercise of the faculty of reason give a man a sufficient knowledge of it, but it must be gained by the habituating and accustoming and exercising that faculty by reading, study, and observation to give a man a complete knowledge thereof’ (Hale 1956: 505). The problem-solving typical of the common law judge was seldom merely a matter of looking up a relevant rule and applying it to the facts of the case before him. Typical common law reasoning was neither deductive nor inductive, but analogical, arguing from one case to the next on the basis of perceived likenesses and differences and the location of the instant case in the landscape of common experience painted by the judge or lawyer in command of the full resources of the common law.

Thirdly, artificial reason was self-consciously nonsystematic. Not hostile to theoretical reflection—it was far too practical for that, willing to use whatever tools lay ready to hand—but it was decidedly not a theoretical or systematic turn of mind. At the same time, it was not strictly particularistic in the current philosophical sense of that term. Common lawyers were not opposed to thinking in terms of universalizable reasons, and they were keen to secure coherence of their judgments with other solutions and parts of the law. Still, they typically sought local coherence. Overall coherence of moral or practical vision was less important to common lawyers than concrete workability. Thus, classical common law resisted reduction to a system of axioms or first principles from which its constituent maxims, rules, and decisions could (at least in principle) be inferred. Several prominent common lawyers sought to identify especially important general principles running through much of the law, but these collections were manifestly unsystematic. Bacon, for example, clearly meant his collection of maxims to be an aid for the student of the common law, but it was important, he thought, not to give the student the mistaken idea that the common law was to be found in these general principles or maxims (Bacon 1630: B3).

Fourthly, the artificial reason of common law was essentially discoursive,5 that is, a matter of deliberative reasoning and argument between interlocutors. The unwritten common law was deposited in the experience and memory of practitioners, Doddridge wrote, ‘thence to be deduced by deceptation and discourse of reason: and that when occasion should be offered and not before’. Common law, he added, is reason ‘tried and sifted upon disputation and argument’.6 Coke made explicit a point that is clear but still implicit in Doddridge's thought: the disputation they have in mind is specifically forensic. In difficult cases, Coke argued, no individual alone and outside a court of justice, could ever discover the right reason of a rule of common (p. 595) law. For it is only in the process of argument, regarding concrete cases, in open court subject to reasoned challenge, that law is to be found and forged (Coke 1793, Ninth Reports, preface).

Finally, then, the reason of common law is itself common or shared. It is not ‘natural’ in the sense of being a merely individual capacity. It is an intellectual competence, a discoursive faculty that is learned through participation in a practice of public forensic argument, situated in and moving about in a world of recorded experience of ‘human affairs and conversation’. The philosopher and theologian are not suited for this task, Hale argued, for it is not an enterprise of discovery of general practical principles through the exercise of abstract reason, but rather an enterprise of judging particular cases through grasp of concrete relations and arrangements woven into the fabric of common life (Hale 1956: 502–3). In his History, Hale traces the process by which a judge seeks a rule of decision in a particular case. He goes first, says Hale, to the settled common law and custom of the realm, then to authorities and decisions in past cases, and finally to ‘the common reason of the thing’ (Hale 1971: 46). This, of course, is not the Hobbesian idea that once the sources of law run out the judge must appeal to his natural reason, or the civilian view (adopted by English equity practice) that the judge must appeal to conscience. Rather, Hale's judge goes back again to the cases and the ‘human conversations’ in which they are rooted and by sensitive judgment aided by analogy to other relevantly similar cases finds a solution. It is the common reason of the thing that the judge seeks.

This explains in part the willingness of a common law judge to seek only local coherence, rather than broad theoretical coherence of a single moral vision or systematic rationality. The aim of this artificial reason is convergence of judgment on common solutions, thereby securing effective practical guidance. Larger theoretical coherence, when it does not serve the end of convergence of judgment, was regarded as, at best, a luxury, and more typically an obstacle to achieving the end. But law, common lawyers maintained, is not concerned with the moral vision of any individual, however soundly argued it may seem to be, but rather with the convergence of the views and judgments of the larger community, and forging and maintaining a common sense of reasonableness. Salience, not vision, and pragmatic convergence, not theoretical coherence, were its fundamental aims.

2.3 Precedent and Statute in Classical Common Law Jurisprudence

Precedent. Already in the fifteenth century orthodox common law judges could write, ‘precedents and usages do not rule the law, but the law rules them. … Precedents are not in all cases binding upon the courts’ (Long Quinto, M.f. 110, quoted in Tubbs 2000: 45). Of course, this was not meant to deny absolutely the legal relevance of (p. 596) judicial decisions, but rather to undermine the claim of particular judicial decisions to binding authority. The general legal significance of a case, in the eyes of a common lawyer, lay in the nature and quality of the argument for the decision. The gradual appearance of more sophisticated recording techniques in the seventeenth century did not fundamentally alter this underlying doctrine. Vaughn, CJ in Bole v Horton wrote, ‘If a court give judgment judicially, another court is not bound to give like judgment, unless it think that judgment first given was according to law’ (Tubbs 2000: 182). Similarly, Hale maintained that judicial decisions, while they bind ‘as a law between the parties thereto, as to the particular case in question, ‘till revers'd by error or attaint, yet they do not make a law properly so called’ (Hale 1971: 45). A century later Blackstone echoed Hale and Vaughn: ‘the law and the opinion of the judge, are not convertible terms, or one and the same thing; since it may happen that the judge may mistake the law’ (Blackstone 1765: i. 71; emphasis in the original). Hedley added,

if a judgment once given should be peremptory and trench in succession to bind and conclude all future judges from examining the law in that point or to vary from it, then the common law could never have been said to be tried reason … for it should then be grounded merely upon the reason or opinion of 3 or 4 judges … [Therefore no] judgment should be so sacred or firm that it may not be touched or changed. (Hedley 1610: 178–9)

Thus, no single judicial ruling has the authority of law (beyond res judicata for the parties) just in virtue of the judge's having decided it, and future judges are free to test a prior court's formulation of a rule or doctrine of common law in light of the legal community's shared sense of reasonableness. Only in so far as the decision can be integrated into the body of the law—made consistent and reasonably coherent (at least within its local context)—and is taken up in the deliberation and argument of the legal community, is it to be given legal credit. If a judicial ruling is entrenched and regarded as peremptory, thereby blocking all subsequent assessment of its reasonableness and coherence with the whole (in the forensic context of ‘deceptation’), just by virtue of a judge's decision alone, the common law in general could no longer claim authority.

Still, Hale hastened to add that, although judicial decisions fail to make law properly so called, ‘they have a great weight and authority in expounding, declaring, and publishing what the law of this kingdom is’, they are the best evidence of the law (Hale 1971: 45). A century later Mansfield wrote in the same vein, ‘precedent, though it be evidence of law, is not law itself, much less the whole of the law’.7 Classical common law jurisprudence resolutely resisted the theoretical pressure to identify law with canonically formulated, discrete rules of law. Law, on this view, is not a set of rules or laws, but a practised framework of practical reasoning, and this practised framework provides a form of social ordering. Its rules and norms can be formulated, perhaps, but no such formulation is conclusively authoritative. In De Augmentis, Bacon (p. 597) counselled ‘not to take the law from the rules, but to make the rule from the existing law [that is, the body of argued opinions and decisions]. For the point is not to be sought from the words of the rule, as if it were the text of the law. The rule, like the magnetic needle, points at the law, but does not settle it’ (Bacon 2000, Book 8, aphorism, 85, Bacon 1858: v, 106). Moreover, each formulation of a rule is in principle vulnerable to challenge and revision in the course of reasoned argument and dispute in the public context of litigation.

Is it possible, then, to say that common law jurisprudence recognized binding precedent? The answer is yes, although it differs rather sharply from the more familiar view inspired by classical positivism. The view has three salient features. First, past judicial decisions claim judicial respect and attention not in virtue of merely having been decided—laid down or posited—but in virtue of having been taken up by subsequent courts and thereby having found a place within that body of common experience. They have this place because they were products of a process of discoursive reasoning and contextually situated reflective judgment. Secondly, while individual cases are not regarded as establishing authoritative rules, they are taken to illustrate the operation of proper legal reasoning, to exemplify the process of reasoning within the body of experience. Thirdly, past cases do not preclude deliberation and reasoning in subsequent cases, but rather they invite and focus that reasoning. The prior court's formulation of the issues, and the reasons for resolving them as it did, and the rule on which it rested its decision, are not regarded as final. It is always open to judges in future to test any precedent court's formulation of the rule of its decision. Hence, subsequent courts participate with the precedent court in reasoning about issues raised by the case and extend that reasoning to the case before them. Yet, judicial formulations of the issues and the rules are due respect, because the prior court has claim to expert authority.

Statutes. The attitude of common lawyers towards statutes was complex and conflicted, especially in the heyday of classical common law jurisprudence. By the seventeenth century, it was no longer possible to deny the legislative power of Parliament. This forced common law jurisprudence to address directly the place of the products of parliamentary legislation in the common law of England. Hale articulated a subtle account of the relationship between enacted and common law that remained influential until Blackstone and still deserves attention.8

Recall that Hale argued that debates over the antiquity of common law were wrongheaded. It was impossible to deny that the common law over its long history had been subject to a great deal of change (some of the most dramatic changes resulting from royal legislation). No part of the law, including laws authorizing lawmaking, had been immune to change; yet, the common law, like the Argonauts' ship, maintained its identity. Continuity, he insisted, not antiquity, was the key. The ‘formal and (p. 598) obliging force’ of common law lay not in its origin but the reasonable conviction that the laws fitted well together and fitted the common life of the English people.

Hale distinguished between written and unwritten law. As he understood them, these terms did not refer to two kinds of laws, but rather to two modes of existence of law (or, as we might say, forms of legal validity). Some laws were valid in virtue of having been explicitly made by an authorized lawmaker; other laws were valid in virtue of incorporation into the common law. The class to which a given law was assigned was not determined solely by the way it came into being, but by its present mode of validity. The validity of written law was a function of its having been enacted according to established formal, constitutional rules. Unwritten law, regardless of its origin, drew its authority from its present incorporation into the use and practice of common law. The process of incorporation or integration worked on precedent and statute alike. Only through continual use, exposition, interpretation, and exten-sion—through being taken up and appropriated by practitioners of the common law—was a novel rule or doctrine made part of the common law. Through ‘contrary usage’ the opposite effect was also possible: the doctrine could be narrowly limited, for example, precedent may be ‘distinguished’ to the point of extinction of its general legal effect. The same might happen to laws initially introduced by legislative act. The statute might not be taken up and incorporated into common law practice, but rather it might be narrowly interpreted, limiting its scope and legal significance. While the laws authorizing its creation exist the statutes remain valid law, but they do not enjoy deep or wide impact on the law without incorporation. Moreover, statutes that are not eventually incorporated are vulnerable to changes of constitutional or lawmaking authority. This, Hale speculates, is probably what happened to many ancient statutes that did not survive in memory. The continuity of law, on this view, is guaranteed not by some posited norm according to which a validly enacted rule remains valid until repealed,9 but rather on the more strenuous test of incorporation into the use and practice of common law judges and lawyers, and more fundamentally (and more indirectly) into the use and practice of the people. The common law and its custom, Hale insisted, was ‘the great substratum’ of the law (Hale 1971: 46).

Thus, on Hale's view, the status of statutory law, which may initially have greater claim to treatment as valid law than precedent, can change—indeed, he suggests that if it is ultimately to survive it must change—from resting on formal rules of authorized lawmaking to incorporation into the normative family of common law. This represents a shift from formal to substantive validity, from dependence on what are (p. 599) now called ‘content independent’ criteria of validity to substantive integration into the law and the life of community to which it gives structure.

2.4 Common Law Jurisprudence as a Theory of Law

This, in brief outline, is the classical common law conception of law. We might even call it a theory of law, but we would have to add immediately that qua theory it was relatively modest. It sought to capture general and fundamental structural features of law, but its focus was largely local. Its account was not meant to apply to law anywhere, at any time, but in the first instance at least only to the common law they practised. It did not strive to contribute to universal jurisprudence, the enterprise that seeks to articulate conceptually necessary features of law wherever it is alleged to exist. This is not to say that the conception, if coherent and plausible as an account of the defining features of common law, would have no implications for universal jurisprudence, but only that those implications would be largely limiting and negative—in the way that a counter-example forces revision of a universal thesis.

Thus, the theoretical ambitions of common law jurists were very different from those typical in the natural law tradition. This may explain in part why common law thinkers were willing to rely rather uncritically on broad natural law concepts when they, albeit rarely, did venture to make universal pronouncements about the nature of law. They did so, confident that their own local conception of law was broadly consistent with natural law orthodoxy, but they could not be bothered to set out the grounds for their confidence. Their lack of theoretical persistence was partly due to the fact that they were, in almost every case, not philosophers and theoreticians but active practitioners, and there was not any immediate practical need to work out the theoretical details. It may also have been due to the fact, perhaps only vaguely evident to them, that were they to find that they could not borrow freely from the natural law tradition, they would have no publicly recognized framework for presenting their conception and would have to work one out systematically on their own. This was, perhaps, an issue best left unaddressed, since they lacked the resources to battle on the broader theoretical front.

Thus, we are left the task of considering how to fit the common law conception of law into the long tradition, or rather overlapping traditions, of philosophical reflection about the nature of law. The common law conception, I think, represents a view of law that is in important respects incompatible with both orthodox natural law thought and with orthodox legal positivism. It represents a distinctive approach to understanding the nature of law and legal reasoning, a third way of conceptualizing the phenomena of modern law. It is too early to rule out rapprochement with one or the other of the dominant jurisprudential traditions, but we benefit from an appreciation of insights we might gain from approaching the theoretical explanation of legal phenomena from the distinctive starting-point that classical common law theory (p. 600) provides. Keeping in mind the modesty of its theoretical pretensions, I propose to develop some themes of classical common law jurisprudence in the idiom of contemporary jurisprudence and to try to make a case for the plausibility of its most important theses.

3 Common Law Conventionalism

3.1 Positivist vs. Common Law Conventionalism

The emphasis of common law jurisprudence on the customary nature and foundations of common law suggests a possible liaison with mainstream positivism. Like contemporary positivism, common law jurisprudence conceives of law as a kind of institutionalized convention, or complex set of conventions. So, in what follows I will refer to the contemporary articulation of this theory as ‘common law conventionalism’. Yet, it differs in important ways from Hart's positivist conventionalism (Hart 1994) and alternative accounts of law inspired by Hart's theory (e.g. Coleman 1998, 2001; Marmor 1998; Waluchow 1994). It departs from positivist conventionalism at two crucial points.

First, it rejects the model of laws that many positivist theories adopt.10 On that model, law is understood as a set or systems of rule-like directives. The existence, validity, and content of these directives is said to be determinable by appeal to content-independent criteria. The set as a whole, also, has its unity largely in terms of relations among the directives external to their content. Moreover, according to this model, these directives guide action by providing law subjects with reasons for action that pre-empt their action-directing deliberation on potentially competing reasons they may have.

However, as we have seen, the common law conception of precedent, at least one recognized source of law, is strictly inconsistent with this model. Lon Fuller, perhaps the most important contemporary common law theorist, wrote in his unjustly neglected work, The Anatomy of Law, ‘a judicial decision is always an explained thing’ (Fuller 1968: 90). Perhaps to be more accurate, he should have said, ‘a judicial decision (p. 601) is reasoned thing’, but his point is clear. In common law the normative force and authority of a judicial decision extends beyond the facts and the parties litigating in the particular case, yet that precedential force does not depend solely on the authoritative utterance of a general rule in the body of the judge's opinion. The judge in a prior case does not unilaterally and finally fix the scope or meaning of a rule through his or her decision, regardless of how carefully crafted the language of the opinion is. In the end, it is the quality and force of the reasoning, not the public utterance of it, that lends authority to a court's rationale (Chapman 1994: 43). Common law conventionalism shifts theoretical attention from laws—the authoritative directives produced by lawmaking institutions—to the process of practical reasoning with and within law. Law, on this view, is a matter of convention, but it is a convention of a special sort, namely a practised discipline of practical reasoning. This departs decisively from the model familiar in positivist jurisprudence.

Secondly, common law conventionalism offers a distinct alternative to Hart's account of the conventional foundations of law. According to Hart's theory legal rules exist not as socially practised conventions but as systemically valid rules, and systemic validity is a matter of being a member of a set of rules identified by a common rule of recognition. The rule of recognition is not itself a valid legal rule, but an entirely different kind of rule, a convention constituted by the practice of law-applying officials. The rule of recognition is a social fact, a fact about the regular practice of officials. Yet, it is a rule; its normativity depends on its being treated as such by those who practise it. The rule of recognition guarantees the unity and continuity of the legal system. (Actually, Hart's main argument for the rule of recognition is that it alone can explain the more immediate and theoretically fundamental facts of the persistence and unity of law.)

Thus, necessarily, at the foundation of any legal system is this recognition convention consisting in the convergent behaviour and appropriate attitudes of law-applying officials and their professional associates. The existence of such a rule is a conceptually necessary condition of the existence of law in a given time and place. As a matter of conceptually contingent but ‘natural’ necessity, and a condition of the efficacy of law, law subjects must generally behave in ways consistent with the laws identified by the rule of recognition. It is not at all necessary, however, and may even be rare, that citizens themselves will take an ‘internal attitude’ towards the law, and it is very unlikely that they will grasp, let alone endorse, the governing rule of recognition. Hence, on this view, the scope of the convention on which law in any society is founded will, by a combination of conceptual and natural necessity, be limited uniformly to the practice of law-applying officials and (some) lawyers.

I will call this ‘formal conventionalism’ to contrast it with the ‘material conventionalism’ regarding the foundations of law endorsed by common law conventionalism. By ‘formal’ I do not mean to imply that it is committed to the view that the criteria of recognition practised by officials in any given legal system necessarily concern only matters of non-evaluative social facts, facts about the sources of the laws. (p. 602) Some conventionalists following Hart—so-called ‘inclusivists’ (Coleman 1998; Waluchow 1994; and even Hart 1994: 248–9)—argue that it is an open question, to be determined by contingent facts about the practice of law-applying officials in actual legal systems, whether their rule of recognition includes only source-based criteria, or also includes moral tests of some sort. Hart-inspired conventionalism is ‘formal’, rather, because its locates the conventional foundations of law in a narrow, structural, proto-constitutional feature of a legal system, namely, the criteria practised largely just by law-applying officials by which valid rules of law are distinguished from invalid ones. In contrast, common law conventionalism insists that, as a conceptual matter, a structure of control could not operate as law unless it were built on a broader conventional base, that is, unless there is a substantial degree of congruence in substance, and continuity of modes of practical reasoning, between formal, institutional elements of law and wider social practice.

Thus, the contemporary version of common law jurisprudence insists that law is conventional in two respects, in both of which it decisively departs from key positivist doctrines: (1) law is a special kind of convention, a practised discipline of practical reasoning, and (2) it depends for its existence on substantial congruence and continuity with broader practices in the community. In what follows, I shall attempt to explain more fully and make plausible these two key theses.

3.2 Convention of Common Reasoning

The deliberative discipline of common law. Classical common lawyers held that common law was common yet ‘artificial’ reason. Although Coke was inclined at times to dress this notion in mystery and legal mysticism, more sober common lawyers used this phrase to capture the pragmatic, nonsystematic, contextual, and essentially discoursive nature of common law. Common law conventionalism follows the lead of Coke's sober colleagues. It maintains that law, in the jurisdictions in which common law is dominant, offers ordinary practical reasoning a multi-layered, practised discipline of deliberating and reasoning together regarding public matters. Common law conventionalism reorients thinking about the nature of law dominated by positivist and natural law conceptions. Its theoretical point of departure is not a set of norms, prescriptions, or propositions of law, but rather a practice of common practical reasoning. Rather than a metaphysical thesis, it urges a methodological thesis, a point about order of explanation and understanding, not an ontological point about the ultimate order of being.

Common law conventionalism calls attention to a number of key ways in which law disciplines practical reasoning. First, the pragmatic spirit of common law forces deliberation and argument about practical matters—matters that can touch large issues of political morality and pervasive aspects of social life—to focus on concrete situations, and the relatively specific problems that arise in them. It focuses on the (p. 603) matters at hand, mindful of implications that go well beyond the case, but allowing the specific circumstances and problems to orient deliberations. Secondly, legal practical reasoning is historical: it anchors deliberation and what classical common lawyers call ‘deceptation’ to past decisions and actions taken by or on behalf of the community—decisions which, for that reason, are understood to be prima-facie normative for the community and its members. Explicit lawmaking activities (and the constitutions, statutes, and regulations they produce), as well as precedent-setting decisions and actions (primarily of officials, but also in some cases of ordinary citizens) direct their deliberation and provide the resources with which participants in this practice build their arguments. Thirdly, while the discipline makes use of all appropriate and valid forms of reasoning and argument, its central and distinctive technique is analogical thinking. Fourthly, this discoursive process is essentially collaborative: it is a practice of thinking, arguing, deliberating, and deciding in common. Finally, this essentially collaborative enterprise is formally institutionalized in a public forum. This provides a public focus, forum, and exemplar for a practice with the potential for wide participation in society.

This sketch of the discipline of common law reasoning needs to be developed at each of its key points, but within the limits of this chapter I will focus only on the third and fourth, which I think most need elaboration and explanation.

Analogical thinking. The distinctive technique of the common law discipline is analogical thinking,11 which consists of two analytically distinct intellectual processes: analogical reasoning and reflective assessment. They typically work in tandem in particular instances of common law reasoning, but in many cases the second may be tacit.12 Consider first analogical reasoning proper.

There is no formal logic of analogical argument. Deductive logic governs arguments—ordered structures of premises leading to conclusions—and so it is appropriate to speak of deductive arguments, but there is no deductive reasoning. Reasoning often uses, but cannot be restricted to deductive argument. Deductive logic regiments our thinking as it moves from one proposition to another, but it is (p. 604) powerless to do more. It cannot compel, even in a normative sense, a reasoning person to accept a proposition as true. For that we need reasoning, for it is an exercise of reasoning, with the help of rules of deductive logic of course, that brings us to accept a conclusion, feeling the force of the argument for it, rather than, for example, abandoning one of its premises. Analogical thinking involves a form of reasoning in this sense. There is no analogical argument in the strict sense in which there is deductive argument—there are no formal rules of inference, either sui generis or derived from deductive logic. Nevertheless, it is a form of reasoning, not mere feeling or particularistic intuition. While it boasts no rules akin to those of deductive logic, analogical reasoning has a general structure and is subject to important constraints—this is especially true of its use in law—and in virtue of them it displays a recognizable, indeed indispensable, form of reasonableness.

Analogical reasoning is reasoning by or from example, a similibus ad similia, from like to like. It moves from particular to particular without relying in any fundamental way on articulated prior rules (Aristotle, Nicomachean Ethics, bk. II, ch. 24). The process is familiar. First, one or more past cases (decisions and the factual circumstances they addressed) are brought into a frame with a more or less detailed narrative of the facts of the case to be decided, and relevant similarities and differences are noticed. A tentative comparison class is constructed on the basis of rough assessments of relevant similarities. At work already here is a sense of relevance—call it threshold relevance. In typical instances of analogical reasoning in law, these two movements are mutually referential and mutually dependent. While the facts of the case lead one to retrieve a collection of cases to serve as the rough comparison class, the narrative of the facts is influenced by familiarity with the cases and the categories they exemplify. As Llewellyn once said, judges ‘have been law-conditioned. … [T]hey see significances … through law-spectacles’ (Llewellyn 1960: 19).

Next, a more robust sense of relevance groups together some of the cases falling in the comparison class into a class of ‘like cases’, while it distinguishes others. A pattern takes shape that makes sense of treating some of the cases in the same way; an intelligible guide for action is identified, although it may not be possible to articulate it completely. Rational pressure giving shape to the emerging sense of the likeness of these cases comes from all parts of the initial rough comparison class: cases judged similar to the instant case, those judged dissimilar and ‘distinguished’, and the instant case itself. The construction of the class of like cases is the result of a kind of triangulation among these focal points.

Note that even the process of distinguishing presupposes some degree of (threshold) relevant similarity. Distinguished cases are members of the initial rough comparison class; otherwise they would not present an intellectual challenge to the deliberator. It is in virtue of this initial relevance that distinguished cases exert pressure on the shape of the class of ‘like cases’. The present case also exerts its share of pressure. It can urge reconsideration of received views of the salience or importance of facts of previous cases and thereby of the reasons on which their decisions rest. (p. 605) This is one reason why rules of law may seem to change as they are applied (Levi 1949: 3 f.) and why common lawyers regard each formulation of a ‘rule’ of past cases as corrigible, vulnerable to revision.

This process can yield a determinate result, an understanding sufficient to ground a decision in the present case that is consonant with precedent and fit to guide future decisions and actions. Success is not guaranteed, however. In some cases analogical reasoning will substantially narrow the range of alternatives for dealing with the problem facing the court without uniquely determining a solution. At other times, the guidance will be largely unhelpful. In these cases, analogical thinking is forced to move to the stage of reflective evaluation to which we will turn presently.

The process of constructing the rough comparison class (based on a sense of threshold relevance) raises interesting theoretical questions, but assessments of robust relevance are the most puzzling to students of analogical reasoning in law. It is the deliberators' sense of robust relevance and absence of it that leads them to bring the present case into a family of like cases including only some members of the initial comparison class, and on the basis of this classification decide the present case. The details of this process deserve more extensive treatment than can be given them here, but we need to comment on a few important features of the process.

First, while judgments of similarity and dissimilarity in analogical reasoning presuppose, or rather manifest, a sense of relevance, it would be a mistake to claim that such judgments presuppose a rule in terms of which relevance is defined. It is the existence of some facts in common that brings the general rule into play (Levi 1949: 3). The point, however, is not that the judgment of relevance is logically particularistic. Repeatable properties of the analogues and disanalogues are selected from an uncountable host of other facts and features of the compared cases, and the family that is constructed must make sense as a pattern for future action, in ways I will mention presently. The force of the analogical reasoning is a function of how normatively compelling the pattern is. So, it is possible for judges to express their conclusions of analogical reasoning in terms of a rule, or at least a rule-like pattern. However, the rule is not an input into the reasoning process, but part of its output (Chapman 1994: 67–8), and not the most important part at that. In analogical reasoning, no rules of fixed and determinate scope are in play as inputs.

Explicitly formulated rules purport to offer complete and mature normative categories, but examples offer something more open ended and partial: normative categories that are incomplete, not fully articulated, and still capable of growing (Levenbook 2000: 202–5). It may be useful to try to formulate the rule implicit in a range of cases, but common law judges are always sensitive to the tentative and vulnerable nature of that enterprise. Understanding of the topography and termination point of the path of a legal doctrine is often imperfect and always subject to reconsideration in light of further cases. This rather cautious and conservative point is what underlies Levi's otherwise radical-sounding claim that it is not the intention of the prior judge that fixes the meaning of a precedent decision but what the present judge, (p. 606) attempting to see the law as a fairly consistent whole, thinks it should be (Levi 1949: 3).

Secondly, the assessments of robust relevance on which analogical reasoning depend are not the results of more general theoretical reflection, moral or otherwise. It is a mistake to treat these assessments as the outcome of attempts to locate the examples in the comparison class under some moral (or other practical) principles that purport to justify them. Locating and assessing such principles is part of analogical thinking, but not of analogical reasoning proper. As we shall see, this further intellectual process presupposes results of analogical reasoning. The two must be distinguished because the ‘meaning’ or content of a precedent example is not strictly determined by the justification of complying with it (Levenbook 2000: 192–6).

Thirdly, analogical reasoning in law is in aid of practical deliberation; it serves the normative purpose of guiding actions. Thus, the class of like cases must not only be interesting, suggestive, or evocative, as an apt metaphor might be, but it must also be practically intelligible. That is, it must project an intelligible pattern that can be followed, not only by the decision-maker, but also by the community governed and directed by it. Thus, the pattern must make normative sense to those who are expected to follow it and to those expected to accept its application as vindication of a decision based on it.

In this respect, common law reasoning stands like Janus at the gate between the past and the future, looking to decide the case at hand while looking both backward to the field of examples and commitments from which we have come and forward to the field of the future into which deliberators and those who depend on them are passing. It looks to the past not only for help and guidance, but for normative direction, for commitments undertaken that have implications for decisions for cases at hand and for actions and decisions in the future. This is not merely an aspiration; it is also a significant constraint on common law reasoning.

Fourthly, because it seeks to provide normative guidance, leads to and purports to ground a decision of the community on a matter of public importance, and claims to be binding, analogical reasoning in law proceeds in the service, and within the constraints, of a concept of (a species of) justice, sometimes misleadingly called ‘formal justice’ (Chapman 1994: 67). Levi puts the structuring question of analogical deliberation this way: ‘When will it be just to treat different cases as though they were the same?’ (Levi 1949: 3). This notion is indispensable (albeit not sufficient, see Postema 1997) to an explanation of whatever normative force precedent and analogues extending from them purport to have, but it is not an easy notion to understand. It is perhaps easier to say what it is not than to capture its positive content. On the one hand, it cannot be a generic or conclusive notion of justice. While it may have some prima-facie and defeasible moral force, it is likely to fall well short of full-dress moral justification, or even such justification in the name of justice. On the other hand, we cannot equate it with a purely formal notion of consistency, or treating like cases alike according to some rule (i.e. according to some rule or other). Understood in this way, (p. 607) the notion is entirely empty (Westen 1990, ch. 9). Any case is ‘like’ any other case in some respect and the likeness can be expressed in the form of a rule. The problem with this empty notion is not that its guidance is utterly indeterminate; but rather that it is utterly indiscriminate: it fails to capture a notion of justice relevant to analogical reasoning. Since anything goes as far as this notion is concerned, it follows that it cannot be a notion of justice; for if justice does anything morally important, it discriminates, it leads us away from some courses and leads us to others. Thus, the normative notion involved must impose a substantive constraint. It cannot, in the name of justice, simply require that like cases be treated alike however ‘like’ is determined, for it helps us assess likeness and unlikeness. How are we to characterize the substantive idea of justice involved in our assessments of robust relevance in analogical reasoning in law? I doubt whether anyone has been able to say with any confidence. But the fact that we do not yet have an adequate account of this notion should not lead us to conclude that it is not at work at an inarticulate level in these assessments. Any attempt to theorize the operation of justice in this context will be answerable to our intuitive capacity to make such assessments. The important point to keep in mind here is that a sense of justice is indispensable to the assessments of robust relevance in analogical reasoning in law.

Finally, the collaborative nature of common law reasoning (Fuller 1968: 93) exerts additional pressure on our assessments of robust relevance in analogical reasoning. While it is always individuals who participate in analogical reasoning in law, they proceed with a keen sense that they deliberate, as Hart put it, not each for his own part only (Hart 1994: 116), but as members of a larger whole. They regiment their perceptions and judgments to a common point of view, just as we adjust our judgments of size of objects in our visual field for distortions of light and perspective. This capacity for reflective judgment is a social capacity, the ability to reason from a body of supposed shared experiences to solutions to new practical problems, to judge what one has good reason to believe others in the community would also regard as reasonable and fitting. These judgments can be made with confidence, not because one is a good predictor of others' behaviour, but because one understands at a concrete level the common life in which they all participate. To become fluent in the language of ‘human affairs and conversation’ is to acquire the social capacity to make judgments that even in novel cases one can be confident will elicit recognition and acceptance as appropriate in one's community. This capacity is rooted not in shared general beliefs or values, but rather in living, working, and especially talking together, in the concrete activities of ordinary life, making the adjustments of action and perspective necessary to achieve understanding of our common world and acting intelligently and purposefully in it.

The reflective component of analogical thinking. Often analogical reasoning of the form we have just sketched will be sufficient practically speaking to ground a decision of law. Decision-makers will legitimately feel no need to raise further questions (p. 608) regarding the results of the process. The decisions will appear to the decision-makers and those depending on their decisions as reasonably coherent with the existing law and intelligible enough to give relatively determinate guidance for the future. However, in other cases the results may not be as happy. The process may yield troublingly indeterminate results (practically, not merely logically, indeterminate), and decision-makers may be forced to make a principled choice among the alternatives produced. Or the practically determinate result may be compelling when considering the instant case with its nearest analogical neighbours, but out of phase with law viewed more broadly. This apparent lack of more global coherence may have a merely technical legal focus, or it may cast its eye even more broadly on important matters of moral concern. In these cases, those involved in analogical thinking may feel pressure to reflect more broadly on the results of the prior process.

The methodology of this reflective process has been described in various ways, but Dworkin's ‘interpretive’ account of legal reasoning (Dworkin 1986: chs. 2, 3, 7), while the target of criticism from many quarters, may be the best available characterization of the intellectual process at this level. Briefly stated, the methodology requires that the deliberator seek that set of general principles (theory) that makes the best overall sense of law, such that the principles not only imply the more specific rules or decisions under review but also show them to be justified (for this the principles must approximate true or rationally warranted principles of morality). Alternative principles are ranked according to the extent to which they ‘fit’ the legal data and ‘appeal’ from a moral point of view.

There is much to be said for Dworkin's characterization of this process as an account of an important part of common law reasoning. Even severe critics of Dworkin's theory of law admit that the process he described is ‘the dominant methodology in both the practice of law and legal scholarship’ (Alexander and Kress 1995: 288). However, common law conventionalism insists that this process of reflective evaluation of the results of analogical reasoning is subject to constraints Dworkin himself did not recognize. First, the pragmatism of analogical thinking significantly reins in the urge to achieve global systematic coherence. As we have seen, common law is more concerned with workability on the ground, than with coherence of broad moral vision. While it cannot accept blatant incoherence of a result with other fundamental parts of the law that tend to undermine the legitimacy of large parts of the law, it is tolerant of incompleteness of vision and a degree of lack of overall systematic coherence. It is more concerned with coherence of legal doctrine with the activities, practices, and lives of the citizens whose interactions it seeks to guide than with more abstract theoretical coherence of legal doctrine. It is willing to sacrifice some theoretical coherence for substantial resonance of the law in the community it serves. Hence, while there is pressure at the reflective stage to achieve systemic and moral coherence, even this part of analogical thinking tends towards relatively local rather than more global coherence, coherence of practice rather than of theory.

(p. 609) Secondly, the essentially collaborative, public dimension of common law reasoning reinforces this tendency. Sound analogical thinking will tend to seek out doctrines arising from past decisions that other players, officials and citizens alike, can anticipate or at least recognize and find intelligible. Thus, Dworkin's mythical superhuman judge ‘Hercules’ is not a hero of common law reasoning. His theoretical successes, if they fail to take fully into account their dependency on intelligibility to others in a context of public justification, fail as grounds for assessing rules or patterns arising from past decisions.13

3.3 Material Conventionalism

Classical common law jurisprudence insisted on the conventional (customary) foundations of law. Critics from Bentham 1977 to the present (Waldron 1998) have challenged this claim. Yet, classical common lawyers would have regarded these criticisms as misplaced. As we have seen, they were the first to deny that common law is nothing more than social custom. They were keenly aware of the difference, indeed the distance, between practice of the courts and social mores and practices. Nevertheless, taking a cue from St German, they held the subtler view that the force and validity of law were rooted in broad social practice. But it was integration of the law into the life of the community, congruence of practice and continuity of modes of practical reasoning, that they stressed, not identity of law and social norms. It is this subtler, but not uncontroversial view, ‘material conventionalism’, that I will now try to articulate and defend.

Hart's formal conventionalism theory reduced the necessary role of conventions in law to an absolute minimum: law is fundamentally conventional, but only because, like an inverted pyramid, it balances the entire normative weight of the legal system on the single point of official practice. It is formal, because the conventions on which law rests are concerned with structural features of law, not with the internal coherence of its doctrines and their congruence with the lives of citizens. Material conventionalism maintains that law can exist and function properly only when law institutionally identified and applied sinks its roots into the soil of broader social practice. Continuity of reasoning practice and congruence of substantive standards is necessary for law. This is not a claim about the role that extra-legal ‘social norms’ typically play in legal reasoning (Eisenberg 1988), neither is it a claim about necessary conditions of good or effective law; rather, it is a claim about the conditions that must obtain if a set of rules and institutions, and the coercive machinery it directs, is to (p. 610) operate as law, whether good or bad law. These contrasts suggest a point from which an argument for material conventionalism can begin.

The argument for the congruence thesis at the heart of material conventionalism rests on two key premises.14 The first (‘normative guidance’) premise is, in brief, that law by its nature seeks to provide wholesale normative guidance to rational, self-directing agents. The second (‘interdependency’) premise concerns the social environment in which the agents to which law's norms are addressed find themselves. This social environment is characterized by complex interdependence of actions; it is a vast network of webs of interactions. Before we proceed we need to clarify these two premises.

Normative guidance and interdependency. We should notice, first, that the normative guidance premise takes no position on the issue of the fundamental tasks or functions of law. It merely holds that law by nature and design seeks to guide action. Normative guidance is not a function of law; it is, rather, a defining technique or instrumentality that may be put to many different uses. Normative guidance is a defining feature of law, not its function or purpose. Secondly, the normative guidance premise maintains that law as such seeks to guide action, or purports to guide action, not that it always or necessarily does so successfully.

Thirdly, we need to explore briefly the notion of normative guidance. Following Shapiro, we should distinguish normative guidance from normative governance (Shapiro 1998: 472–3). Let us call the agent whose action falls within the scope of a norm the norm agent with respect to it. Norms govern behaviour when they are valid and actions of their norm agents fall within their scope. They provide a basis for evaluating the actions, and may encourage the evaluator to act in certain ways in response to them. Thus, they may provide secondary guidance even when they do not guide their primary norm agents. Norms guide the actions of their norm agents when they function as rules or norms in the practical reasoning of their norm agents leading to the formulation of intentions and actions on them. Norms that govern action typically are designed also to guide those actions, but this is not always the case.

We can distinguish further two kinds of norm governance. For mere norm governance it is sufficient that the norm is valid and applies to the norm agent's action. Norm governance is robust when, in addition, the norm shapes the way in which the norm agents publicly present and justify their actions to themselves and others. Thus, it is possible for agents to be governed by a norm, and to acknowledge that governance implicitly by using it in good faith to justify or vindicate their actions taken (or seek to justify or excuse violations of the norm), even though they did not use the norm in deliberations that led to their actions. They may be, but need not be, insincere in such cases. Insincerity does not disqualify it as robust norm guidance in the eyes of the law, since law is inclined only to look on behaviour not on motivation.

(p. 611) Following Shapiro again, we can also distinguish two forms of normative guidance (Shapiro 1998: 489–92). A norm offers epistemic guidance if the norm agent learns what she must do—what her duties and responsibilities, or her rights, are, how she is to proceed in order to achieve a desired legal result—from the norm. The norm can guide epistemically, even if the norm agent is motivated to comply by considerations independent of the norm. It offers motivational guidance if the norm agent is motivated, at least in part, to comply with the norm by the fact that the norm governs the agent's conduct in the circumstances. Typically, agents that are motivationally guided by a norm will also be epistemically guided by it. To be motivated by a norm is to regard it as legitimate and to act at least in part on that assessment. In both forms of normative guidance, the norms play a role in the norm agent's practical reasoning, but the roles are different. The most robust form of normative guidance is motivational (or motivational-cum-epistemic) guidance.

Fourthly, it is distinctive of law to seek to provide normative guidance wholesale, as it were. That is, it seeks to guide not by issuing directives to individual norm agents in the specific circumstances in which they find themselves case by case, but rather by promulgating norms that are relatively general both with respect to norm agents and to circumstances. Moreover, the normative guidance premise makes a claim about a legal system in general, not about every component (every rule or norm) of the system. Law seeks to guide action through the instrumentality of its component laws, but not through them alone, for the way the laws are linked together can also play a role in practical guidance. This takes some of the burden off any specific rule or norm. Law as a whole may succeed in guiding without every single component doing so or even making pretence of doing so. These two features imply that normative guidance is a matter of degree. It is a matter of degree how general its norms are, how widespread in a community the law's normative guidance is, and how much of the legal system succeeds in offering normative guidance. Thus, the claim that by its nature law seeks to guide action involves a scalar measure. If some attempt to exercise control of social behaviour fails as law, presumably this is because its normative guidance potential falls below some minimal threshold.

Before we proceed we should note two important implications of the normative guidance premise. First, it presupposes that norm agents are intelligent, rational, self-directing agents. That is, they are able to understand themselves, their natural and practical environment, and the actions available to them in their environment, and they have the capacity to direct their actions in accord with the norms or reasons that apply to them in that environment. If officials are to undertake to guide or robustly govern action in a legal manner, they must do so in ways that acknowledge and engage these capacities of rational self-direction. It must address rational, self-directing agents. Secondly, the normative guidance premise places limits on the kinds of behaviour and modes of communication and of control that law can enlist in the project of normative guidance. Interventions that trigger action without addressing norms or rules or reasons or normative examples to them may achieve (p. 612) compliance but fail to do so as law. Thus, bulk compliance with its norms (pace Hart) is not enough for the existence of law in a community, for it may be entirely epiphenomenal, or it may achieve compliance exclusively through means that do not even attempt to address its subjects as rational, self-directing agents.

The second, ‘interdependency’, premise holds that the social environment in which rational, self-directing agents typically act is characterized by complex interdependence, that is, it is a vast network of webs of interactions. This is a pervasive, fundamental fact about the environment in which rational self-directing agents find themselves. Unlike the first premise, this premise does not express a conceptual truth about law. But it is in certain respects like a conceptual truth. It has the same status as the circumstances of justice have with regard to the nature and shape of justice according to Hume (Hume 1975: 183–92). That is to say, not only is it generally true that the agents whom law seeks to guide find themselves in this environment, but further, that outside of this environment, law and its distinctive technique lack intelligible point or purpose. This is not to say that we can derive from this premise a thesis about the necessary task or function of law, but only that whatever that task may be, if there is one, it must take this environment into account. Law cannot ignore this fact without utterly silencing its normative voice.

So, we have arrived at the following point. We know that law as such seeks to provide wholesale normative guidance and the resources for robust normative governance, that this presupposes that citizens (and officials), the norm agents of law, are self-directing rational agents, engaged in webs of interdependent actions. And thus to achieve normative guidance laws must be addressed to citizens and must be of such a nature that self-directing agents can understand them and appreciate their practical force, and on this basis apply the norms to their own actions. The conclusion material conventionalism needs to establish is the following: wholesale normative guidance and robust normative governance of rationally self-directing and complexly interacting agents is possible in this environment only if law is congruent with background social practices and widespread public understandings of them. Congruence, like normative guidance, is a matter of degree. What needs to be shown is not that law must be maximally congruent, but only that if it falls below some threshold it ceases to function as law. This argument does not define the threshold. It seeks to establish a foothold for material conventionalism against its rival, formal conventionalism, which denies the relevance of congruence and settles for a ‘bulk compliance’ condition. So, it is sufficient if it can show that a substantial degree of congruence is necessary for law to function without defining the threshold.

Congruence is not merely a condition of the existence of law. It is also an aspiration of law. Addressing norms to rational self-directing agents is not merely something law as such seeks to do, but it is also something we demand as a condition of respect for those individuals and as a condition of their liberty. It is no surprise, then, that lovers of liberty have often thought law to be an ally. The rule of law ideally extends the demands for congruence well beyond the minimum necessary for the bare (p. 613) existence and minimal functioning of law as such. However, it is a mistake to conclude from this natural link to the ideal of the rule of law that the congruence thesis is merely a thesis about an ideal for law (or ideal of law). It is first of all a condition of something's existing and functioning as law. On the other hand, it is also a mistake to conclude that ideally there should be no discontinuity between law and social practices. For part of what makes it possible for law to do useful things in society is its formality and institutionality, its resolute focus on details, its commitment to keeping present to mind the normative past of the community, and its introduction of finality into discussion of matters that see no present prospect for closure. These matters are not always on the minds, nor do they always deeply inform the practices, of people in the community. Part of law's value lies also in its distance from practice and from the values and systems of belief of the people it seeks to guide.

Argument for the congruence thesis. The argument proceeds by counter-position. Suppose legal norms as a system were largely incongruent with ordinary social practice. To provide normative guidance for those it purports to govern it must promulgate its norms in such a way that norm agents can both understand their meaning—that is, understand what is involved in complying with them—and grasp their practical point or force—that is, see why someone might think she has at least a prima-facie reason to comply with them. Moreover, because the norms operate wholesale and are addressed to rational self-directing agents whose actions are woven into thick networks of interdependence, the meaning and practical force must be publicly accessible. That is, it must be possible for people to have a rough idea how others with whom they interact read the norms. Without substantial congruence this is simply not possible. Let us see why.

First, understanding the meaning of a rule or example (knowing what one is to do) cannot be separated from grasping its practical point (knowing why, i.e., what reasons someone might have to do it). The what and the why are interdependent. It is easy to see how the why depends on the what: one has a reasonably determinate, intelligible reason to do something only when there is something sufficiently determinate to do. More importantly, in contexts of social interaction, the practical force (the why) of rules often depends heavily on reasonable expectations of how others on whom one's actions depend are likely to act, and that, in turn, depends on how they understand what the rule requires. Equally, the what depends on the why: norms, whether they are explicit rules or informal examples, lay out patterns of actions over time and different circumstances. To understand the meaning of a norm is to grasp what is involved in following it. This understanding depends on its practical point. Intelligibility is not merely a function of understanding words in a language, but understanding how those words connect with the range of actions available to one, and this latter understanding requires a capacity to give the words practical significance. That, in turn, requires that one be able to see some point in carving up the world of action in some ways rather than others. Grasping the practical point of a (p. 614) norm is essential to understanding what is involved in following it. And this requires that the norm agent be able to set the language of the norm, or the example of an action or decision, into a practical context that makes intelligible, practical sense to her.

Secondly, cut off from ordinary daily life and social practices, law cannot hope to make its meaning accessible to norm agents. If the law as a system is largely incongruent with ordinary social practice, then lawmakers cannot rely on understandings that already inform their daily lives to help norm agents understand the meaning or grasp the practical force of the law's directives. For example, precedents would not have the rich environment of similar examples and patterns of interacting necessary for them to have meaning. It would be hard for individual agents not only to determine how to generalize beyond specific decisions, but also even to determine what a case ‘on all fours’ with the precedent might be. The same is true for explicitly made, carefully articulated, and publicly announced rules. They, too, have meaning only when norm agents know how to apply the general categories they define to specific cases, only when they can identify with confidence a specific case as an instance of the rule. This is never merely a function of the linguistic meaning of the rules. For it is as normative guides for behaviour, with a certain texture, in contexts of action and interaction, that the words of the explicitly made rules must be grasped. As Kant recognized, this requires a special exercise of judgment—he called it ‘Mutterwitz’ (common sense)—and that in turn requires a grasp of a context of application. Cut off from that context, the rule is a fish out of water.

Note that a regime of sanction-backed commands or authoritative directives (content-independent and peremptory rules) cannot succeed. The problems they both face are rooted in the fact that they seek practical force for the commands or directives entirely external to their content. Sanctions and authority provide reasons that are designed precisely to operate entirely independently of the norms they underwrite. They are all-purpose motivational means. They are no respecters of content. Moreover the command or exercise of authority is intended to have a peremptory effect in the practical reasoning of norm agents. Thus, no guidance regarding the meaning of the directives can be expected from the sources of their practical force. This is thought to be the point of relying on commands and authoritative directives. But, then, all the weight in determining the content of the norms must fall on the linguistic resources of fixed meaning as elaborated by consideration of the lawmakers' intentions. Without sinking their roots deeply into the soil of ordinary daily life, however, the efforts of the lawmakers will be utterly ineffectual. Their intentions for the shape, scope, and meaning of the rules cannot be determined apart from considering how law subjects will take up the promulgated rules. Unless they are willing to tailor norms to the situations of each law subject, they will have to depend on the same resources as we have considered above.

The importance of a sustaining environment of social understandings is amplified by the need for individual agents to anticipate the understandings of those with (p. 615) whom they regularly interact. Legal norms must have publicly accessible meaning and practical force. Even if it were feasible for governing officials to communicate clearly to citizens individually the norms by which they were expected to guide their actions, the officials would fail—or at least fail to govern by law—because the individuals addressed in this way could not count on other agents, on whose actions they depend, to understand the norms in the same way. The routines, customs, conventions, and practices of ordinary daily life—the affairs and conversations of common social life, as Hale would put it—provide a common context in which to locate public understandings of the examples and rules of law. Cut off from this context, law's attempt to provide normative guidance could not hope to succeed. Robust normative governance would suffer the same fate, since the legal norms could not provide resources with which individuals might seek publicly to justify their actions to others.

This argument does not assume that citizens must accept or endorse or commit themselves to the norms. Neither does it depend on background consensus in society on basic values or general principles, let alone consensus on what Rawls calls comprehensive moral, religious, or philosophical doctrines. What it requires, rather, are understandings on the ground, i.e. common or overlapping activities and practices that have practical meaning and force for those who participate in them. Material conventionalism maintains that it is necessary that law be incorporated into the ordinary social life of the community it seeks to govern, while it will always be different and to a degree distant from it. For this, it is not necessary that legal norms be incorporated into any general theory or comprehensive doctrine about that social life or its underlying principles. Providing the soil into which law must sink its roots are ‘conversations’ not creeds, practices not principles, ordinary affairs and activities not theories and doctrines. These resources give practical life to law's norms. Entirely without them, law's normative guidance and robust normative governance would be rootless.

Of course, this is not to say that it is unfeasible to seek to control social behaviour without deference to ordinary social life. It is to say, rather, that if minimal conditions of congruity of substance and continuity of reasoning are not met, any attempts at social control will have to be massively intrusive, dependent on direct coercion and on situation-by-situation, narrowly formulated commands. If a legal system were to lose its balance and topple into this condition, people subject to its increasingly arbitrary twists and turns (as they would inevitably perceive it), would at some point find themselves in a large-scale version of scorer's discretion (Hart 1994: 141–7). No longer would it be possible to regard the administrators of control as playing by the same rules as the people they sought to control, but rather by some other rules or no rules at all. The game of law would have been transformed into a game of administrators' discretion. There would be no point in trying to understand the rules in order to use them as guides, even epistemic guides, to action. All attention would turn rather on the decisions, instance-by-instance of the administrators. Subjects would become entirely dependent on the wills of the administrators. The problem with that is not (p. 616) that it is costly, or unfeasible, or morally abhorrent, but, rather, that it is unrecognizable as functioning (even badly functioning) law. Whatever is at work, it is not law, for it does not manifest the defining feature of seeking to provide wholesale normative guidance. Thus, law as such can function only if it is congruent to a substantial degree with the social life of the community it seeks to govern.

We can now draw implications of this argument for the contest between Hart's formal conventionalism and material conventionalism. Formal conventionalism holds that at its foundations, law rests on a narrow convention, the practice of law-applying officials with respect to identifying valid legal norms, requiring in addition only the bare fact of bulk compliance of law subjects with the norms they identify and seek to apply. This, we now see, is insufficient. On the contrary, what is necessary rather is incorporation of the formal, institutionalized system of law into the life of the community it seeks to govern at least to the extent required by the congruence thesis defended above. Moreover, the validity of legal norms depends at least as much on their substantive incorporation into the body of the law as on their conformity to external criteria whether purely factual or moral. This twofold incorporation explains at least as well the persistence of law over time, and offers a more compelling explanation of the unity of the legal system than Hart's rule of recognition hypothesis.

3.4 Authority and Authorities in Common Law Conventionalism

‘It is not Wisdom, but Authority that makes a Law’, Hobbes argued in his Dialogue, adding even more pointedly against Coke, ‘Statutes [i.e. proper law] are not Philosophy as is the Common-Law, and other disputable Arts, but are Commands, or Prohibitions’ (Hobbes 1971: 55, 69). Bentham carried this challenge further in the next century, charging that common law is nothing but a matter of ‘unauthoritative jurisprudence’ (Bentham 1970: 153). Necessarily, law is a matter of general rules, he argued, yet on orthodox common law assumptions these rules were not set out publicly in judicial decisions, but were to be constructed from them, and no judicial formulation was authoritative. But, then, he charged, this treats legal rules as nothing more than ‘inferential entities’, fictitious constructions. ‘From a set of data like these [namely, a set of judicial decisions] a law is to be extracted by every man who can fancy that he is able; by each man perhaps a different law: and these are the monades which meeting together constitute the rules … of common or customary law’ (Bentham 1970: 192). The alleged rules of common law exist, not as publicly accessible general prescriptions, but as private conjectures, personal inferences—not as law, Hobbes would say, but as ‘philosophy’. Thus, common law jurisprudence fails as an intelligible coherent account of law. Whatever the alleged rules of common law are, they cannot be rules of law. Common law is ‘a thing merely imaginary’ (Bentham 1977: 119).

(p. 617) This critique suggests an important challenge to common law conventionalism, not to its endorsement of the convergence thesis we have just considered, but to its model of law and its role in practical reasoning. The challenge might go like this. Law seeks to provide public normative guidance, but it does so in a distinctive way, namely, by addressing authoritative directives to citizens whom it seeks to govern and guide. These directives are able to make a difference in the practical reasoning of norm agents because their status as rules of law is determined by content-independent criteria and they function in a pre-emptive fashion in the practical reasoning of citizens. That is, the content and scope of the directives can be determined by strictly non-evaluative forms of investigation and reasoning. If they are indeed authoritative, they offer reasons to act and reasons for not considering potentially competing moral or practical considerations. In this way, legal authorities secure finality with respect to those issues law is deployed to address. Common law conventionalism cannot account for the normative guidance that law as such seeks to provide, because identification of its prescriptions depends on substantive, content-related deliberations, the results of which are said to be constantly open to reassessment. Thus, it cannot account for the distinctive authority of laws or the role of institutionalized authorities in the legal system. It cannot hope to provide normative guidance in precisely those contexts in which, by its own admission, law is most concerned to address, namely, complex social interaction calling for public standards that enable citizens to coordinate their interaction. Authoritative directives, in contrast, eliminate the need for agents to engage in evaluative deliberation about the appropriate courses of conduct open to them. They can appeal, rather, to publicly accessible rules, the validity and content of which can be determined by non-evaluative marks of their authoritative sources.

This challenge can be refined and developed in various respects, but it may suffice for present purposes, since my aim here is merely to indicate in broad strokes how common law conventionalism might try to defend itself against a challenge coming from this quarter. Note, first, that common law conventionalism, like its seventeenth-century ancestor, was as unhappy as Hobbes was with philosophy as a model for law, but it likewise resists the model of command (or its refined cousin, authoritative directives). It seeks to shift the weight of attention from the directive aspects of law to its deliberative aspects. Common law conventionalism and its critics start from the same premise: law as such seeks to provide public normative guidance. But common law conventionalism refuses to equate law's normative guidance exclusively with the kind of difference that authoritative directives are said to make in practical reasoning. Its conception of normative guidance is broader, including within it a role for authorities and their authoritative directives.

Common law conventionalism denies that authorities and authoritative directives are theoretically central to law. It is useful in this regard to recall the familiar debate over the role of coercion in the law. Legal theorists have long been tempted to explain the distinctive manner of law's functioning in terms of coercive sanctions. Critics of (p. 618) sanction theories correctly charge that this strategy promotes an undeniably salient and important aspect of law to an undeserved position of theoretical prominence, treating what is of secondary importance as core with the result that we are blinded to important features of law. For example, reasoning about legal obligations is reduced to predictions of official coercion, nullity of an attempted exercise of a legal power is treated as punitive, rules of law are treated as external impositions from above, and the complex way in which laws seek to guide actions is obscured. The solution, critics argue, is not to deny that coercion has a role in law, but rather to find an account of law and normative guidance that makes room for, but does not reduce to, the operation of coercive sanctions. We run the same risk, common law conventionalism argues, if we equate law with discrete legal rules regarded as authoritative directives. Again undeniably salient and important but secondary features of law that are treated as core, with the result that important aspects of law and its distinctive mode of operating are obscured or mischaracterized. Making law's authoritative directives central to our understanding of law highlights its role in giving finality to issues that threaten otherwise to disrupt or confuse social interaction. But to focus on it exclusively blinds us to the fact that law not only institutionalizes the execution of deliberation, but also institutionalizes deliberation itself. It also distorts our understanding of the normative guidance that law seeks to provide. It encourages us, for example, to think of precedent in terms of rules, as ill-drafted statutes, rather than as examples, and to think of analogical reasoning as a form of discovery of official rules, the content and scope of which is already fixed by prior decisions. It also forces us to think of judicial decision-making as a kind of unauthorized and shadowy lawmaking and the law arising from common law decisions as judicial legislation. At the same time, it obscures the collaborative aspects of the common law deliberation and the discipline to which law subjects this deliberation.

Common law conventionalism's notion of normative guidance is wider than that captured by the notion of authority. Rather than regarding looking (exclusively) to discrete rules or norms of the legal system as the primary guiders of action, common law conventionalism looks to resources in the system of law as a whole, to its process and not just its product, to provide normative guidance. As a result, the kind of guidance offered is not limited to supplying to a class of agents specific reasons for the actions represented in a general rule, reasons for acting in a certain way and reasons precluding deliberation on other reasons the agent might consider. Rather, law is seen to provide a framework for common deliberation about courses of action and reasons for them. Law situates the practical reasoning and deliberation of individual rational, self-directing agents in a three-part framework: (1) a disciplined practice of practical deliberation exemplified in, but not restricted to, the activities of legal officials, (2) a body of examples that enables agents to define the practical problems they face and work towards a solution of them, and (3) an institutionalized public forum in which the practice is carried on, one that models a form of public, collaborative reasoning and which is open to active and passive participation by citizens. As part of (p. 619) this institutionalized process, the decisional conclusions of this deliberation are in certain respects and for certain purposes regarded as final and congeal into authoritative directives. Thus, this broad strategy for normative guidance gives authority an important role to play, but it does not take the lead. The aim of this wider kind of normative guidance is not to provide a surrogate for individual deliberation leading to action in public social contexts, but rather to redirect and provide a discipline and a body of resources for such deliberation.

Is law construed in this way utterly unable to provide guidance to rational self-directing agents where it is needed most, namely, in negotiating the complex webs of social interaction they face day-to-day? There are reasons to think it does not. First, we must keep in mind that the question is not whether law as modelled by common law conventionalism does the best job possible in providing normative guidance for social interaction. Common law conventionalism only claims that the normative guidance law offers takes this form; whether it is greatly successful in providing such guidance or even whether its claim to provide such guidance is generally warranted in any particular case is not at issue (any more than the alternative view holds that law always has the authority it necessarily claims). The issue is whether its claim to do so is practically intelligible. It would not be practically intelligible if it were obvious on the surface that it could not in any likely social circumstances perform as it promises, or if certain structural features systematically undermined its ability minimally to so do. It is true that the practical deliberation of isolated individuals, viewing parametrically the social world they face, and without any anchors of common experience or common judgment, would probably fail to meet this minimum standard. But that is not the kind of disciplined practice of practical deliberation institutionalized in law as conceived by common law conventionalism.

Secondly, to the extent that this broader form of normative guidance makes judicious use of authority, it shares whatever success it can hope to achieve. Of course, it is conceivable that by locating authoritative directives in this larger context of institutionalized deliberation all the advantages of the former are lost, but that surely is not clear at this point. Moreover, the most plausible versions of the view that identifies law with authoritative directives do not deny that judges and courts regularly and legitimately engage in deliberation with wide latitude (using, but not restricted to, reasoning with fixed legal rules), but they regard such activities as falling outside the purview of a theory of law (Raz 1995, ch. 13). Whatever the general merits of such a view might be, it is even more vulnerable to this worry about undermining the public coordination potential of law, should the worry prove to be serious. Indeed, common law conventionalism may have an advantage since it has an account of how law seeks to discipline deliberation in a way that arguably enhances its public normative guidance.

Finally, a more fundamental question is what should count as successful public normative guidance as viewed not from a moral-political view, but from the point of view of the aims of law in any particular community. Suppose it turns out that a more (p. 620) Hobbesian conception, precluding as far as feasible individual practical deliberation, promises more tranquil social interaction, a more orderly public life. It does not follow, obviously, either that this is what law as such must seek. We may demand with Hume that law ‘cut off all occasion of discord and contention’, but what reason have we to think that this demand is built into the nature of law itself. Indeed, we may ask more of law and of ourselves as participants in the public life governed by law. We may ask, for example, that it help us as a community to seek greater justice for our life together, even though we disagree in deep and important ways about what justice requires.15 An institutionalized practice of public argument and deliberation that enables an enterprise of this sort does not lose its status as law just because it risks being somewhat less orderly than the Hobbesian ideal. Neither is its claim to provide normative guidance thereby proved practically unintelligible. According to the Iliad, the god Hephaestus made a shield for Achilles depicting a city of war and a city of peace. The city of peace was represented by a wedding and a trial. Orderly and effective normative guidance may be evident not in the absence of tension, conflict, or dissension, but in the presence of a framework in which the conflicts can be articulated and resolution of them can be sought deliberatively in public. Normative guidance so construed has a certain moral appeal, but for present purposes that is not the important point. It is not the appeal but the intelligibility of the conception that was challenged. So far, we do not seem to have reason to doubt its intelligibility.

Conclusion. Taking a cue from seventeenth-century common law jurisprudence, we have sketched a theory of law that departs in important ways from familiar natural law theories in stressing the conventional foundations of law and from familiar positivist accounts in taking a convention of a special sort, a practised discipline of public practical reasoning, as a defining feature of law. Bentham took common law jurisprudence seriously enough to spend a large part of his life trying to refute it and undermine its grip on lawyers in England and abroad. While his refutation failed, he was right to spend his philosophical energies on this conception of law. It is time again to enter it—or its descendant, common law conventionalism—in the jurisprudential sweepstakes alongside its more familiar rivals, positivism and natural law theory, and give it a run for its money.


Alexander, Larry, and Kress, Kenneth (1995), ‘Against Legal Principles’, in Andrei Marmor (ed.), Law and Interpretation (Oxford: Clarendon Press), 279–327.Find this resource:

    (p. 621) Austin, John (1885), Lectures on Jurisprudence, 2 vols. (London: John Murray).Find this resource:

      Bacon, Francis (1630), Elements of the Common Laws of England (London; (repr. New York: Garland Publishing Company, 1978).Find this resource:

        ——(1858), On the Dignity and Advancement of Learning (1605), in Works, 7 vols., eds. James Spedding, Robert Leslie Ellis, and Douglas Denon Heath (London: Longman).Find this resource:

          Bentham, Jeremy (1970), Of Laws in General, ed. H. L. A. Hart (London: Athlone Press).Find this resource:

            ——(1977), A Comment on the Commentaries and A Fragment on Government, ed. J. H. Burns and H. L. A. Hart (London: Athlone Press).Find this resource:

              ——(1998), Legislator of the World: Writings on Codification, Law, and Education, ed. Philip Schofield and Jonathan Harris (Oxford: Clarendon Press).Find this resource:

                Blackstone, William (1765), Commentaries on the Laws of England, 4 vols. (Oxford: Clarendon Press, 1765–9).Find this resource:

                  Brewer, Scott (1996), ‘Exemplary Reasoning: Semantics, Pragmatics, and the Rational Force of Legal Argument by Analogy’, Harvard Law Review, 109: 923–1028.Find this resource:

                    Chapman, Bruce (1994), ‘The Rational and the Reasonable: Social Choice Theory and Adjudication’, University of Chicago Law Review, 61: 41–122.Find this resource:

                      Cicero (1928), De Legibus, tr. Clinton Walter Keyes (Cambridge, Mass., Harvard University Press, Loeb Classical Library).Find this resource:

                        Coke, Edward (1628), First Institute of the Laws of England (rep. New York: Garland Publishing Company, 1979).Find this resource:

                          ——(1793), The Reports of Sir Edward Coke, in Thirteen Parts, 7 vols. (Dublin: J. Moore).Find this resource:

                            Coleman, Jules (1998), ‘Incorporationism, Conventionality, and the Practical Difference Thesis’, Legal Theory, 4: 381–425.Find this resource:

                              ——(2001), The Practice of Principle (Oxford: Oxford University Press).Find this resource:

                                Davies, John (1615), Irish Reports, in Alexander Grosart (ed.), The Works … of John Davies, 2 vols. (Edinburgh, 1876).Find this resource:

                                  Doddridge, John (1631), The English Lawyer (London: Assignees of I. More).Find this resource:

                                    Dworkin, Ronald (1986), Law's Empire (Cambridge, Mass.: Harvard University Press).Find this resource:

                                      Eisenberg, Melvin A. (1988), The Nature of the Common Law (Cambridge, Mass.: Harvard University Press).Find this resource:

                                        Finch, Henry (1759), Law, or a Discourse Thereof (repr. New York: Augustus M. Kelly Publishers, 1969; 1st pub. London 1627).Find this resource:

                                          Finnis, John (1973), ‘Revolutions and the Continuity of Law’, in A. W. B. Simpson (ed.), Oxford Essays in Jurisprudence (Oxford: Clarendon Press), 44–76.Find this resource:

                                            Fuller, Lon L. (1968), The Anatomy of Law (New York: F. A. Praeger).Find this resource:

                                              ——(1969), The Morality of Law, 2nd edn. (New Haven: Yale University Press).Find this resource:

                                                ——(1981), ‘Human Interaction and the Law’, in Lon L. Fuller, The Principles of Social Order, ed. Kenneth Winston (Durham, NC: Duke University Press), 211–46.Find this resource:

                                                  Gray, Charles M. (1992), ‘Parliament, Liberty and the Law’, in J. H. Hexter (ed.), Parliament and Liberty from the Reign of Elizabeth to the English Civil War (Stanford, Calif.: Stanford University Press, 1992), 155–200.Find this resource:

                                                    Hale, Matthew (1956), ‘Reflections by the Lrd. Chiefe Justice Hale on Mr. Hobbes His Dialogue of the Lawe’, in William Holdsworth, A History of English Law, 7th edn., 17 vols. (London: Methuen, 1956; 1st pub. 1670s?).Find this resource:

                                                      ——(1971), A History of the Common Law of England, ed. Charles M. Gray (Chicago: University of Chicago Press).Find this resource:

                                                        Hart, H. L. A. (1983), Essays in Jurisprudence and Philosophy (Oxford: Clarendon Press).Find this resource:

                                                          (p. 622) Hart, H. L. A. (1994), The Concept of Law, 2nd edn. (Oxford: Clarendon Press).Find this resource:

                                                            Hedley, Thomas (1610), Speech in Parliament on Royal impositions, in Elizabeth Read Foster (ed.), Proceedings in Parliament 1610 (New Haven: Yale University Press, 1968).Find this resource:

                                                              Hobbes, Thomas (1971), Dialogue Between a Philosopher and a Student of the Common Laws of England, ed. Joseph Cropsey (Chicago: University of Chicago Press).Find this resource:

                                                                Hume, David (1975), An Enquiry Concerning the Principles of Morals, ed. L. A. Selby-Bigge, 3rd ed., rev. by P. H. Nidditch (Oxford: Clarendon Press).Find this resource:

                                                                  Kornhauser, Lewis A., and Sager, Lawrence G. (1993), ‘The One and the Many: Adjudication in Collegial Courts’, California Law Review, 81: 1–59.Find this resource:

                                                                    Levenbook, Barbara Baum (2000), ‘The Meaning of Precedent’, Legal Theory, 6: 185–240.Find this resource:

                                                                      Levi, Edward (1949), An Introduction to Legal Reasoning (Chicago: University of Chicago Press).Find this resource:

                                                                        ——(1965), ‘The Nature of Judicial Reasoning’, University of Chicago Law Review, 32: 395–409.Find this resource:

                                                                          Lieberman, David (1989), The Province of Legislation Determined (Cambridge: Cambridge University Press).Find this resource:

                                                                            Llewellyn, Karl N. (1960), The Common Law Tradition: Deciding Appeals (Boston: Little, Brown Publishers).Find this resource:

                                                                              Marmor, Andrei (1998), ‘Legal Conventionalism’, Legal Theory, 4, 509–31.Find this resource:

                                                                                Pattaro, Enrico, Gerald J. Postema, and Peter Stein (eds.) (forthcoming), A Treatise of Legal Philosophy and General Jurisprudence, 12 vols. (Dordrecht: Kluwer Academic Publishers).Find this resource:

                                                                                  Postema, Gerald J. (1986), Bentham and the Common Law Tradition (Oxford: Clarendon Press).Find this resource:

                                                                                    ——(1994), ‘Implicit Law’, Law and Philosophy, 13: 361–87.Find this resource:

                                                                                      ——(1997), ‘Integrity: Justice in Workclothes’, Iowa Law Review, 82: 821–55.Find this resource:

                                                                                        ——(2001), ‘Law as Command: The Model of Command in Modern Jurisprudence’, Nous, Philosophical Issues, 11: 470–501.Find this resource:

                                                                                          ——(unpublished), ‘If this keeps up …: Analogical Thinking in Law’.Find this resource:

                                                                                            Raz, Joseph (1995), Ethics in the Public Domain (Oxford: Clarendon Press).Find this resource:

                                                                                              St German, Christopher (1974), Doctor and Student, ed. T. F. T. Plucknett and J. L. Barton (London: Selden Society; 1st pub. 1523, 1530).Find this resource:

                                                                                                Schauer, Fred (1989), ‘Is the Common Law Law?’ California Law Review, 77: 455–71.Find this resource:

                                                                                                  Shapiro, Scott (1998), ‘On Hart's Way Out’, Legal Theory, 4: 469–507.Find this resource:

                                                                                                    ——(2000), ‘Law, Morality, and the Guidance of Conduct’, Legal Theory, 6: 127–70Find this resource:

                                                                                                      Simpson, A. W. B. (1973), ‘The Common Law and Legal Theory’, in A. W. B. Simpson (ed.), Oxford Essays in Jurisprudence (Oxford: Clarendon Press), 77–99.Find this resource:

                                                                                                        Sunstein, Cass (1993), ‘On Analogical Reasoning’, Harvard Law Review, 106: 741–91.Find this resource:

                                                                                                          Tubbs, J. W. (2000), The Common Law Mind: Medieval and Early Modern Conceptions (Baltimore: The Johns Hopkins University Press).Find this resource:

                                                                                                            Waldron, Jeremy (1998), ‘Custom Redeemed by Statute’, in M. D. A. Freeman (ed.), Current Legal Problems 1998: Legal Theory at the End of the Millennium (Oxford: Oxford University Press), 93–114.Find this resource:

                                                                                                              Waluchow, W. J. (1994), Inclusive Legal Positivism (Oxford: Clarendon Press).Find this resource:

                                                                                                                Westen, Peter (1990), Speaking of Equality (Princeton: Princeton University Press).Find this resource:


                                                                                                                  (1) Bentham (1970: 152–5, 184–95, 1977: 188–273, 1998: 123–40); Austin 1885: ii. 525–33, 620–60); Postema 1986, ch. 8).

                                                                                                                  (2) I am indebted to James Murphy for this apt phrase.

                                                                                                                  (3) Note: throughout this chapter I have silently modernized spelling, capitalization, and punctuation.

                                                                                                                  (4) Cicero captured the Stoic doctrine, later elaborated in Thomist natural law theory, when he wrote, ‘law is the highest reason … when firmly established and completed in the human mind’ (De Legibus, i. 6.18).

                                                                                                                  (5) This word is now obsolete, but it was current in the seventeenth century, and it nicely captures the interlocutory as well as the strictly discursive aspect of common law reason (OED).

                                                                                                                  (6) Doddridge (1631: 241, 242). ‘Deceptation’, or ‘disceptation’, is an archaic term meaning ‘disputation, debate, or discussion’; ‘to discept’ is to dispute, debate, express disagreement or difference of opinion (OED). I am grateful to Michael Lobban for this reference and generally for deepening my understanding of classical common law jurisprudence. His work on seventeenth-century common law will appear in vol. 10 of Pattaro, Postema, and Stein (forthcoming).

                                                                                                                  (7) Jones v Randall (1774) Lofft 383, 385, quoted in Lieberman 1989: 126); see also Coke (1628: 254a) and Blackstone (1765: i. 69).

                                                                                                                  (8) I have defended this reading of Hale's view of the validity of statutes in Postema 1986: 19–27). I summarize conclusions of that argument here.

                                                                                                                  (9) Hart 1983: 16) endorses this explanation of the continuity of law, relying on Finnis (1973: 61–5). Note that for both Hart and Finnis this norm is an explanatory postulate. They attribute it to courts (judges ‘must tacitly accept’ it, Hart says) in order to explain why courts freely accord validity to statutes even after the ‘parent laws’ authorizing their enactment go out of existence. Hale offers an alternative explanation. It has the virtue, apparently lacking in Hart's proposal, of being falsifiable. To test his hypothesis, one would merely have to determine whether courts tend to endorse as valid orphaned statutes even though the courts regard the statute as failing the ‘incorporation’ standard.

                                                                                                                  (10) Some positivists may not be wedded to this model, for example, those who adopt an ‘inclusivist’ view of the conventional practice at the foundations of law (Coleman 1998; Waluchow 1994). They might find common law conventionalism's account of laws and legal reasoning (to some degree) congenial, but they are committed to a version of what I will call Hart's ‘formal conventionalist’ account of the foundations of law. I should add that positivists are not the only theorists attracted to the model of laws that common law conventionalism decisively rejects. There is a long and important line of theorists within the natural law tradition, especially active in the seventeenth century, who also embraced it (see Postema 2001).

                                                                                                                  (11) For more extensive discussions see Chapman 1994: 64–106), Sunstein 1993, Brewer 1996, Levenbook 2000, and Postema (unpublished).

                                                                                                                  (12) Brewer 1996 distinguishes three stages of analogical reasoning, the first two of which bear some similarity to the two components I identify. This is not the place to work out in detail the differences between my account of analogical thinking and Brewer's analysis of ‘exemplary reasoning’. However, it may help understanding of the account sketched in the text to note a few of the more important differences. (1) Although I agree that the intellectual processes are components or stages of analogical thinking, unlike Brewer, I think distinctively analogical reasoning goes on only at the first stage. The intellectual processes at work in the stage of reflective assessment (Brewer: ‘confirmation stage’) are of a different nature. (2) Brewer regards the process I call analogical reasoning proper as a matter of abductive reasoning. I do not think this captures adequately the nature of the reasoning or the kind of intellectual capacities involved in analogical reasoning. Chapman 1994: 64–106) sketches a more promising alternative account consonant with classical common law jurisprudence. (3) Brewer gives more prominence to construction of rules in analogical reasoning than in my view is warranted. I say a few words about this issue in the text.

                                                                                                                  (13) For a striking example of the pressures on courts to achieve collectively coherent decision even when individual members of the court are drawn to very different individual theories of the relevant law, see the discussion of the so-called ‘doctrinal paradox’ in Kornhauser and Sager 1993.

                                                                                                                  (14) The argument developed here elaborates and extends an argument familiar to readers of Lon Fuller's work (Fuller 1969, 1981). Fuller's argument is sketched in Postema (1994: 368–80).

                                                                                                                  (15) Developing an idea familiar from Dworkin's work, albeit in a direction he may not endorse, I have called this enterprise the search for ‘integrity’, which I argue is the pursuit of the public virtue of justice in circumstances in which justice is broadly in dispute (Postema 1997).