5 - Law As Religion, Religion As Law

Summary

This paper examines the methodological problems presented by the relationship between law and religion, and the tensions between internal and external approaches. It argues for a (neutral) semiotic approach: the basics of sense construction, as understood by the Greimasian model of semiotics, are the same for both law and religion. At the same time, the model allows for the identification of differences. But it also problematizes the value of the concepts themselves: who, we may ask, needs to talk about either “law” or “religion” as such a very different question from that of the characterization of particular acts or norms as “legal” or “religious?”. “Law as Religion: Religion as Law” is thus a secondary (or meta-) question addressing the relationship between institutional concepts rather than human behaviour in either its factual or normative dimensions. It may, however, figure large in the rhetoric of religious politics, whose full understanding requires us to narrativise the pragmatics (speech behaviour) of its various participants.

I Introduction

The title of this project, “Law As Religion, Religion As Law,” no doubt deliberately begs several vitally important conceptual and methodological questions. Are the terms “law” and “religion” to be understood as conceptual absolutes, universally applicable to any culture, or are they to be regarded as themselves cultural constructs? My starting point will assume the latter. But then we are faced with further questions: From whose viewpoint may we debate whether, in any particular culture, “law” is to be regarded as a form, or an essential part, of “religion,” and from whose viewpoint may we debate whether “religion” is to be regarded as a form, or an essential part, of “law”? And what is the relevant data? In the Jewish context, we may seek to explore what particular conceptions of “law” and “religion,” and what type of interrelationships between them, we find in the whole range of traditional data of the halakhah.

But the issue turns out to be more complicated. The cultural world which we inhabit is not exclusively Jewish, nor should it be. Ḥareidim may give a different answer, and may even interpret differently the famous dictum of Ben Zoma in Pirke Avot 4:1: “Who is wise? He learns from all men” – stressing the proof-text which follows: “As it is written (Psalm 119:99) I have gained an understanding from all my teachers.” But can we be sure that, even within Jewish discourse, the very concepts of law and religion are exclusively Jewish? I would argue not. It may, therefore, be necessary to distinguish between internal and external viewpoints on the matter, not only as an historical exercise, but also in order to clarify some of the conceptual issues involved. Yet even this, I would argue, will suffice only if we are not able to identify some neutral viewpoint from which to address the issues.

In this paper I argue for the use of a particular version of semiotics as providing such a neutral viewpoint. In Sections II–IV, I outline much of the work I have done over the years addressing these issues from more traditional internal and external (particularly jurisprudential) viewpoints, and the present chapter draws heavily on those earlier publications. Thereafter, I sketch both the general theoretical claims of the version of semiotics which I have used, and its applications to secular law, with a view to its use both in the context of halakhah, and more generally in discussing the relationship between law and religion.

This is an issue with which I have had to grapple for much of my career. My undergraduate degree was in secular (English) law; my doctorate was on early Jewish law, supervised by David Daube. Between 1968 and 1997, I taught in law faculties but with increasing opportunities to teach Jewish law, especially biblical law, for comparative purposes. From the early 1980s I began to explore interdisciplinary approaches to legal study, and quite quickly adopted the semiotic approach which I describe in this chapter. At that stage, I made little attempt to apply this to Jewish law, though I continued to research in Jewish law. I published four books on legal semiotics, the last being Making Sense in Jurisprudence of 1996.¹ At that stage, I felt that I had exhausted what I had to say in relation to secular law, and in 1997 I took up a chair in Jewish studies at the University of Manchester. I explained this to myself on the grounds that Jewish law was more central to teaching and research in Jewish studies (even in a theology department primarily oriented towards Christianity) than it was to any law faculty outside Israel. Increasingly, I applied semiotic approaches to biblical law.² I later became interested in the agunah problem (the “chained wife,” whose divorce is blocked by her husband), and in 2004 established a research unit in Manchester in which I collaborated with four colleagues (including two PhD students). We worked together until 2009, and I have since published six books and over twenty working papers.³ These publications adopted in part an analytical approach sufficiently “external” to disinterest the halakhic authorities, but they remained immune from semiotic analysis. Increasingly, however, I began to realise the relevance of semiotic analysis to some of the fundamentals of that research, and to Jewish law research more generally, and this is reflected in the present chapter.

II Law

Despite the title of Hart’s classical work, The Concept of Law (including the definite article), there is no such universal concept (unless, perhaps, we are Platonists), nor, indeed, is it clear that we need one. In fact, even Hart’s book deals very little with such a concept; rather, its concern is with legal rules and the legal system. Much the same could be said in the Jewish context. Amongst the plethora of terms used in the Bible and later, it is hard to identify one which uniquely denotes the concept of law. And when we speak of “the halakhah,” we too have in mind a particular system of rules rather than an abstract concept.

In the secular context, the positivist paradigm⁴ dominates our understanding of the legal system – arguably, for the citizen as much as the academic. This paradigm is often described as a “sources” theory: law is defined (or recognized) in terms of its sources, and these sources are human and political. While Austin and Bentham in the nineteenth century understood this primarily in terms of power (sovereignty and the imposition of sanctions), both Hart and Kelsen introduced the necessity of legitimacy, understood as sources of law defined by the legal system itself.⁵ This movement on the part of Hart and Kelsen had a twofold purpose: on the one hand, to distinguish the rule of law from any illegitimate power source (a tyrant or gangster); on the other, to guard against abuse of power by legitimate political authorities and to ensure that citizens always had the capacity to choose their course of action in the light of its known legal consequences. This generated (for Hart, in particular) the “demonstrability” thesis: the (“secondary”) rules of the system should be such as would enable “demonstrable” judgments to be made regarding the legitimacy (and, wherever possible, also the meaning) of any “primary” rule of the system.⁶

Such emphasis on the value of certainty is reflected in various ways in the methodology of reading legal texts in modern, positivist legal systems. Law is expressed in language and that language ought to be accessible to the citizen, albeit often advised by a professional. The starting point for interpretation is thus literalism: the law “covers” every situation within the syntax and semantics of the particular legal rule. Moreover, if the statute lays down conditions for any particular legal consequence, those conditions are regarded as not only sufficient but also necessary for that consequence. To put it differently, “If … ” means “If and only if … .”⁷

The political ideology underlying the positivist paradigm entailed a negative view regarding the relationship between law and morality.⁸ On the one hand, morality (whether from religious or other sources) did not count as an independent source of law; on the other, it did not count either as a criterion of the validity of law: an unjust law could not, on those grounds alone, be regarded as invalid, unless such criteria of validity were explicitly adopted by the system of positive law itself. But this prompted attention to a more fundamental question: Does the law consist only of the rules themselves, or also the reasons for the rules? Kelsen took a very strong line on this: the law consisted only of the norms of the system, not the arguments for or against those norms.⁹ Attention thus turned to the status and role of legal argument, and in particular “legal principles.” Dworkin, in particular, argued that law consisted in both rules and principles; that the latter could be implicit in the system as a whole; and that they reflected the political morality implicit in that system.¹⁰

Important aspects of the positivist paradigm have been adopted by some Jewish law scholars, particularly those in the mishpat ivri movement, often for the pragmatic motive of advancing the adoption of Jewish law in the State of Israel.¹¹ Menachem Elon, in particular, adopted a version of the “sources” theory, and Lamm and Kirschenbaum followed the lead of Dworkin in stressing the role of principles in the halakhah.¹² Neither, of course, accepted the basic assumption of legal positivism that sources had to emanate from human, political institutions. We may note, however, that both Austin and Kelsen made some concessions to the idea of a system of religious law.¹³

There are, however, major conceptual problems involved in such a strategy of conceptualizing Jewish law. Radical change (from rules regarded as having “biblical” status) is virtually excluded in Jewish law: there can be no revolution against divine law; indeed in recent centuries (partly in reaction against progressive forms of Judaism), Orthodoxy has proved increasingly reluctant to make any changes in halakhah, as may be seen from a particular problem in the Jewish law of divorce.¹⁴ Elon himself had to accept a significant qualification: in a religious system based on revelation, the Constitution is unamendable.¹⁵ Moreover, there is an acute ideological problem at the very foundation of the application of Jewish religious law within the State of Israel. The State, whose own legal system is secular, in that ultimate authority belongs to a democratically elected legislature (the Knesset), may delegate certain powers to religious courts (Jewish, Muslim, and Christian), but that itself delegitimizes those courts in the eyes of some religious circles, who claim that the ultimate authority is divine, and therefore should operate solely through religious institutions rather than through the State. It is the religion, they would claim, which should define the powers of the State, rather than vice versa. This is a fundamental issue for “law and religion” issues everywhere, and may be regarded as defining the distinction between theocracy and democracy.

But this is not the only conceptual problem. The history of Jewish law shows that charismatic rather than rational sources have played an important role from biblical times. It is striking that a series of instructions to judges¹⁶ fail to mention written sources of law: rather, they tell the judges to avoid partiality and corruption and apply their intuitive sense of justice (in the context, no doubt, of orally transmitted custom). Their authority was charismatic. Thus, the ninth century BCE King Jehoshaphat of Judah charges his judges to avoid partiality and corruption and assures them that “(God) is with you in giving judgement” (2 Chronicles 19:6), rather than their referring to written sources.¹⁷ This survived in a residual power (explicit in talmudic sources but increasingly submerged since then) for the judge to depart from the strict law when intuitions of justice demanded it. Ben-Menahem has studied some thirty cases recorded in the Babylonian Talmud where it is said that the rabbinic judge decided the case “not in accordance with the Halakhah.”¹⁸ Indeed, the halakhah is regarded as a minimum standard, and the judges are expected to go beyond it, wherever possible, in pursuit of justice. Thus, the Talmud quotes R. Yoxanan as saying (Baba Mezia 30b): “‘Jerusalem was destroyed only because they gave judgements therein accordance with Torah law …’ Where they then to have judged in accordance with untrained arbitrators? – But say thus: because they based their judgements [strictly] upon Torah law, and did not go beyond the requirements of the law.”

III Religion

Much the same critique may be addressed to the concept of religion as to that of law. There is no single, universal concept and every reason to doubt whether we really need one.¹⁹ Just as it may suffice to judge whether particular rules and courses of action are “legal,” so too it may suffice to judge whether particular rules and courses of action are “religious.” The sense of both the “legal” and the “religious” may be constructed without presupposing any metaphysical concepts of “law” or “religion.”²⁰ And in both cases, the meanings of “legal” and “religious” are culturally contingent.

Equally variable is the role which any system of religious thought may attribute to law. In the Judeo-Christian tradition, this is discussed in the context of the divine-human relationship (covenant), including, but not restricted to, reward, punishment, and salvation both in this world and in the eschatological context.²¹ Such matters are, indeed, much discussed in the context of Jewish theology, far less in the study of Jewish law (here too implicitly reflecting the positivist paradigm).

There has been much debate in academic departments of religion and theology as to the conceptualization and methodology of religion. “Religious Studies” are nowadays often presented as the study of “World Religions.”²² But does such a comparative approach imply a single, external criterion of religion, or admit the primacy of internal conceptions within any particular system?²³ The latter may appear the more authentic, but a neutral standpoint is required for comparison. Recourse is often made to social science approaches, or more philosophical approaches informed by phenomenology,²⁴ where much attention is paid to the experiential dimension.²⁵ But can religious experience be isolated from the broader religious culture concerned?²⁶ In fact, some aspects of the phenomenological critique of the study of religions share points in common with the semiotic approach I elaborate below, while other aspects radically diverge.

The most explicit Jewish application of the phenomenological approach known to me is that of Rabbi Joseph Soloveitchik in his Halakhic Man, where his concern was to explore the experience of Halakhah, and this not from the (“public”) vantage point of the judge but rather from the (“private”) vantage point of the ordinary Orthodox Jew who is observant of the Halakhah.²⁷ Soloveitchik sought “to penetrate deep into the structure of halakhic man’s consciousness” (p. 4). Halakhic man seeks to fashion mundane reality in accordance with the halakhic model, and this creative activity has an emotional (or spiritual) as well as a cognitive dimension: “Both the halakhist and the mathematician live in an ideal realm and enjoy the radiance of their own creations” (p. 25). Indeed, the mundane world of halakhic man is, from this vantage-point, even more desirable than is the world-to-come:

The Halakhah is not at all concerned with a transcendent world. The world to come is a tranquil, quiet world that is wholly good, wholly everlasting, and wholly eternal, wherein a man will receive the reward for the commandments which he performed in this world. However, the receiving of a reward is not a religious act; therefore, halakhic man prefers the real world to a transcendent existence because here, in this world, man is given the opportunity to create, act, accomplish, while there, in the world to come, he is powerless to change anything at all. … When the righteous sit in the world to come, where there is neither eating nor drinking, with their crowns on their heads, and enjoy the radiance of the divine presence …, they [can only] occupy themselves with the study of the Torah, which treats of bodily life in our lowly world.²⁸

Thus, the reward for observance of the halakhah is conceived by Soloveitchik as the capacity (merely) to contemplate the application (in the empirical world) of halakhah from the (nonempirical) domain of the world-to-come. The reward is thus less satisfying than the act by which the reward is merited, for it is the latter which transforms, which is creative.

The ideal of halakhic man is the redemption of the world not via a higher world but via the world itself, via the adaptation of empirical reality to the ideal patterns of Halakhah. If a Jew lives in accordance with the Halakhah … then he shall find redemption. A lowly world is elevated through the Halakhah to the level of a divine world.²⁹

IV Jewish Law

In various respects, the practice of justice in Jewish law differs from that typical of secular law, in that (1) Jewish law functions through a system in which rationality is a function of revelation;³⁰ (2) its system of rule-making institutions exhibits a far less clear hierarchy, as indeed do its courts;³¹ (3) adjudication functions in the private, rather than the public, domain, its operation not being, in principle, subject to public accountability;³² (4) it frequently lacks enforcement powers in order to give effect to its decisions.³³

In approaching these issues, it is important to address an underlying theoretical issue: In what sense is Jewish law to be regarded as a “religious system of law”?³⁴ For this purpose, I suggest, one may distinguish between “dualistic” and “monistic” models of Jewish law.³⁵ Is it conceived as a separate system from the direct operation of divine justice, operating under delegated authority from God in some semi-autonomous manner, and sharing significant elements in common with secular models of human justice (the “dualistic” model), or is it to be regarded as an integral part of a single system of divine justice (the “monistic” model)? I have argued that there is much biblical support for the monistic model, although rabbinic sources (increasingly) tend to follow the “dualistic” model, as does most modern scholarship on Jewish law. The mishpat ivri movement, which seeks the incorporation of Jewish law within the law of the State of Israel, strongly advocates it for ideological reasons. The two models, however, may not be mutually exclusive.

The Torah contains many rules, but it is far from clear that they were intended for judicial application in the manner of a modern statute in secular courts. Indeed, most of the pentateuchal sources which describe the judicial role make no reference to the application of written sources, but rather commit the resolution of disputes to the sense of justice of the judges, sometimes with a direct hint that their intuitive sense of justice will be divinely guided.³⁶ It is noticeable that two sources relating to the reign of the ninth-century Judaean King Jehoshaphat have him, on the one hand, appointing local judges to whom he says: “Consider what you do, for you judge not for men, but for the Lord; and he is with you in giving judgment” (ve’imakhem bidvar mishpat):³⁷ no mention here of recourse to any written law, nor is it clear that the Superior Court in Jerusalem (whether of Jehoshaphat or as conceived in other sources)³⁸ has such recourse either; on the other hand, the same king sends out a commission including princes, Levites, and priests to teach the people from the “book of the law of the Lord” (sefer torat adonai, 2 Chron. 17:7–9). The implication would appear to be that the first step in any dispute would be for the parties to try to resolve it themselves,³⁹ guided by torah laws, failing which they should have recourse to local judges applying a divinely guided sense of justice, and only thereafter should they refer to the central court in Jerusalem. In a study of the mishpatim of Exodus 21:1–22:16 which I published in 2006, I argued that there was considerable evidence from the substance of the rules that they were designed for “self-execution” rather than third-party adjudication.⁴⁰ I called them “wisdom-laws.”

The situation becomes even more complicated in rabbinic law. On the one hand, the talmudic cases of adjudication “not in accordance with the halakhah”⁴¹ may represent a survival of the divinely guided intuition of which we read in the bible. More generally, the “secondary rules” (Hart) allowing one to identify binding halakhah have many areas of doubt and controversy, as we found in the work of the Agunah Research Unit. Even such a basic rule as that of majority decision generated controversy as to whether it applies only to opinions expressed within the contemporary generation, or trans-generationally.⁴² Indeed, the doctrines of safeq and sfeq sfeiqa (“doubt” and “double doubt”) both attest to the significance of doubt within the halakhic system, and provide considerable discretion in dealing with such cases.⁴³

The orthodox understanding of revelation goes beyond the conferment of divine authority on that which is revealed. The medium itself is understood to have divine features. Thus, the bible is written in Hebrew, the lashon hakodesh,⁴⁴ the language of creation, which has special properties of multiple signification. Moreover, the torah is normally regarded as a single discursive whole, complete, consistent, and without redundancy – any apparent deviations from these standards being resolved in the oral law.⁴⁵ I have argued that the forms of analogy in Jewish law are distinctively religious rather and secular, insofar as they depend very often upon literary rather than substantive features of the biblical text.⁴⁶

Does all this lead to the conclusion that we are able to reach (objective) truth via halakhic argumentation? Some halakhic authorities argue that we can.⁴⁷ I fear that they are unduly influenced by the modern secular model. There is much in the tradition which points in a different direction:⁴⁸ thus, famously, the Talmud records that a divine voice intervenes in a long-standing dispute between the Schools of Hillel and Shammai with the words: “These and these [contradictory opinions] are [both] the words of the living God” (Erubin 13b), but with an instruction to follow Beth Hillel on pragmatic grounds: pragmatic, both in the sense that we need such a decision for communal consistency, and in the sense that the grounds for preferring Beth Hillel are pragmatic in the linguistic sense: their speech behavior was respectful to the opinions of their opponents, and they took the views of the latter into account.⁴⁹ Indeed, the very concept of truth (and its possible application to law)⁵⁰ is hotly debated in philosophy, and there is no reason to assume that its meaning is identical cross-culturally. As for early rabbinic sources, discussion must now commence with the remarkable chapter 5 of Christine Hayes’ What’s Divine About Divine Law,⁵¹ in which she concludes that “divine law does not always align with formal logical proof”; that “the value placed on peace and mercy rather than truth … shows that divine law did not always promote judicial truth”; and that “divine law does not always align with mind-independent ontological reality” (p. 243). In short, the rabbis “did not embrace a broader conception of divine law that assumes the latter’s verisimilitude and correspondence to some kind of objective truth” (pp. 243 f.). In Jewish philosophy, the relationship between emet and emunah, truth and trust (the latter understood in interpersonal terms) is often discussed.⁵² One possible conclusion of our agunah study is that much depends on which contemporary authorities one chooses to trust (and what, in turn, are the sources of trust of those contemporary authorities themselves). I shall argue in the next section that there are strong semiotic reasons for such a conclusion (not only in religious, but also in secular law).

Finally, we may pose the question whether all rules of Jewish law are equally “religious.” It is not difficult to make the case for the religious nature of the rules in the areas of ritual, life and death, and sexual relations, and the language of the biblical sources often reinforces this. But what of those “secular” rules which have direct counterparts in the civil law of other jurisdictions? I have addressed this question in relation to the mishpatim of Exodus 21–22.⁵³ Whatever the origins of the collection, we now find them incorporated into the Sinaitic covenant. But why, I asked, should God be interested in goring oxen (and many other aspects of the mishpatim)? I made some suggestions, but left the matter without any comprehensive substantive answer. Perhaps it represents an endorsement not merely of the necessity but also the value of negotiated settlement of civil disputes, on the basis of divinely endorsed wisdom-laws.⁵⁴ When we proceed beyond the biblical period, to a period when institutional adjudication has become available also for monetary matters (dine mamonot), we do find that they are treated, in significant respects, differently from the “more religious” categories of “prohibition and permission” (issur veheter). There is much more freedom to contract out of rules of dine mamonot,⁵⁵ to modify them by custom (minhag), and even to override them in favor of the law of the State (dina demalkhuta). Yet all these freedoms may be taken to manifest the same value, that of human autonomy and responsibility, as was represented by the underlying values of the biblical wisdom-laws themselves.

A great deal of this section is in line with the argument of an article of Hanina Ben-Menahem, on Talmudic Law as a Religious Legal system⁵⁶ – in particular, his preliminary remarks under the headings: “The Notion of a ‘Religious Legal System’ Is Not Given in Nature, but Is a Product of Reflection”;⁵⁷ “Religiosity is a Relative Concept: Whether a System is Religious is a Matter of Degree, as Opposed to an Either-Or Determination^”;⁵⁸ “A Legal System May Use Specialized Tools to Advance an Ideology”;⁵⁹ and “Self-Reported Accounts of Legal Practice Are Not Necessarily Veridical,”⁶⁰ as well as his account of many of the features of a legal system which justify its designation as a religious legal system.⁶¹ He concludes: “In my view, governance by judges rather than by rules is the primary feature of the talmudic legal system. Of the various elements of the judicial process that, as we have seen, may attest to the religious nature of a legal system, it is by far the strongest.”⁶² I shall argue, however, that adoption of a semiotic methodology in discussing these matters leads to an even more radical conclusion: it treats “the legal system” as a set of separate discursive practices rather than as a unified system in which one particular practice is dominant.⁶³

V An Introduction to Greimasian Semiotics⁶⁴

“Semiotics” is the name given to the research field which asks whether we can construct any general theory of sense construction: one which will not be confined to language or behavior, but which will both provide a common theoretical foundation for the different disciplines of sense-construction, and contribute to their hypotheses and methodologies. If so, it provides a neutral model for the comparison of sense construction in different areas and disciplines.⁶⁵

Contemporary semiotics is dominated by the thought of two “schools,” one inspired by the work of the American philosopher Charles Sanders Peirce (1839–1914), the other based upon the structural linguistics of Saussure, as most systematically developed in the work of Algirdas J. Greimas, the founder of the “Paris School” of semiotics. While the former remains philosophical in its orientation (Peirce was a leading logician, and this is reflected in his work on signs) and is concerned with the history, structure, and interrelations of concepts, the latter leans towards linguistics and to the analysis of empirical data. It is to this latter school of semiotics that my work in this area belongs.

The Saussurean basis of Greimasian semiotics, and particular its understanding of sense and reference,⁶⁶ has profound implications for ontology. Despite the claim that there exist universal processes of sense construction, Greimasian semiotics is fiercely empirical, in that it insists on starting with something accessible to our senses, such as a text (whether oral or written). While legal positivism also commences from an empirical claim – that there exists something that has been laid down (“posited”) by a (normally human) authority – it goes on to assert the ontological existence (independent from any text) of laws and the legal system. The legal norm is claimed therefore to have an existence distinct from its linguistic expression.⁶⁷ This form of legal positivism thus makes ontological claims; semiotic positivism, on the other hand, seeks merely to explain how such ontological claims are constructed and make sense to those who accept them. It does not, and cannot, validate those claims; it can only show how, within legal discourse, such claims are validated through processes of recognition (an essential part of the “narrative syntagm”).⁶⁸ This follows from the basics of Saussurean understanding of the relationship between sense and reference. The sense of terms used in language (its semantics) is not a function of its relationship (correspondence) to the empirical world outside, but rather its relationship to other terms within that linguistic system.⁶⁹ But when such language is used in the context of an assertion, a claim is thereby made that the language does indeed point to something in the outside world. This, however, is a speech act, and can itself be analyzed in terms of the narrative syntagm. It is, no more and no less, the performance of a speech act communicated by a sender to a receiver, in which the sender makes a claim about the outside world, to which the receiver accords some form of recognition. This referential skepticism entails the understanding that truth cannot be established merely within language; the recognition of truth claims, rather, most commonly depends upon interpersonal trust.⁷⁰

My most recent work in this area has reinforced my view of the importance of a number of basic epistemological positions of Greimasian semiotics: its referential skepticism (better: its understanding of reference within pragmatics rather than semantics) and its methodology of approaching texts – these leading to what I would take to be its (universally relevant, if implicit) view of the relationship between truth and trust. All of this, in different ways, involves the relationships between syntactics, semantics and pragmatics.

For Greimasian semiotics, “is” and “ought” are simply different modalities applied to accounts of behaviour patterns (whose sense is constructed in narrative terms).⁷¹ When we affirm the modality of “is,” we are making a truth claim; when we affirm the modality of “ought” we are making a validity claim (just as when we go to an art gallery and affirm the beauty of a painting, we are making an aesthetic claim). When we affirm that something is “divine” (or, for that matter, “legal”), we are similarly attributing a linguistically constructed “modality” to it. We may therefore question whether the secular/religious distinction (or, for that matter, the “legal/moral” distinction) is ontological or itself a social construction of sense. But this is neither to deny nor to affirm the reality of either legal or religious experience. Greimasian semiotics is purely descriptive; its position on sense and reference (the latter being the concern of pragmatics rather than semantics) means that it cannot validate (or invalidate) truth claims within linguistic resources.

The Greimasian school looks for “basic structures of signification,” and finds them in a “deep” level of sense construction based upon a particular model of narrative. The concept of narrative has become popular in the social sciences in recent years, albeit in the form of a range of different particular conceptions of what narrative is. In legal studies, too, narrative models of different kinds have been used for a number of different purposes. The semiotic approach of Greimas and his followers proceeds from first principles. It stresses the role of narrative in the deep structure of signification of any form of discourse. Thus, it is equally applicable to both legal and religious discourses; it also provides a methodology for explaining how we make sense of particular discourses as “legal” or “religious.”⁷²

Greimas sought to extend the ideas of Saussure regarding the semantic structure of individual sentences by applying them also to “discourse.” We have to make sense not merely of individual sentences, nor even of the relations of contiguous sentences, but also of texts as a whole. The folklorist Vladimir Propp had analyzed one hundred Russian folktales, and identified some thirty-one narrative themes in them, recurring in different forms. Through a reanalysis of this material, Greimas derived a much more abstract, general (and, as he claimed, universal) model in which his “basic structures of signification” (“structures élémentaires de la signification”) consist of two axes:

1. (a) the syntagmatic axis of Saussure generated an “actantial” model, in which discourse makes sense in terms of underlying patterns of intelligible action. This is worth stressing: our sense-making capacities are primarily oriented towards making sense of human action, and in particular making sense of elements of human action which appear in a diachronic sequence.

2. (b) the associative (or “paradigmatic”) axis generates choices (often structured as oppositions) of story elements which are used in the story-sequence (or “syntagm”).

The combination of these two levels, Greimas claimed, represents the “deep level” of all discourse. We make such sense by understanding the data in terms of meaningful sequences of action. What is a “meaningful sequence of action” can be specified (the “syntagmatic axis,”). Thus the Greimasian theory claims to provide an account of the construction of the sense of behaviour (including speech behaviour), and not merely of the language we use to tell stories. In subsequent work, members of the Parisian school have applied the Greimasian scheme to many different forms of discourse and behaviour. And psychologists of perception endorse this approach, as when they note the phenomenon of “confabulation” – “fake perception” (my phrase) to make an otherwise non-meaningful sequence meaningful.⁷³

The syntagmatic axis (the discursive version of the syntax of a sentence) is “semio-narrative”: we can thus speak of a “narrative syntagm.” Every human action, for Greimas, begins with the establishment of a goal, which thereby institutes someone as the subject of that action, with the goal of performing it. In realizing the action, the subject will be helped or obstructed by other actions of other social actors. This help or obstruction will affect the “competence” of the subject to perform the action. The desired action itself will be achieved, or not achieved. But it is a characteristic of human action that the sequence does not finish there. Human beings reflect upon past actions. As a consequence, the syntagmatic axis of Greimas concludes with the concept of recognition (otherwise, “sanction”). Human action (whether real or fictional) thus appears meaningful in terms of a basic three-part “narrative” sequence, for which Greimasians have adopted the following technical vocabulary:

1. (1) “Contract”: the institution of the subject through the establishment of goals and competences. The goals of action may be of any kind: they may include communicative goals.

2. (2) “Performance” (or nonperformance) of those goals.

3. (3) “Recognition” of that contract and performance (or nonperformance).

Any narrative involves a set of interactions, in the course of setting goals, performing them, and assisting or obstructing their accomplishment. At the “deep level,” the theory sees these interactions as involving a set of abstract actors, termed actants, which appear in pairs: Sender-Receiver, Subject-Object, Helper-Opponent. A Sender invests a Receiver as Subject of the story, by communicating a goal to him/her (though the actual Sender and Receiver may be the same; in other words, the Subject may be self-motivated). In achieving this goal (normally involving action upon an “Object”), or performing this task, the subject may be assisted by a “helper,” or obstructed by an “opponent.” The figures of “helper” and “opponent” were derived from the plots of Propp’s Russian folktales. Later, the “helper” and “opponent” disappeared from the model as independent actants, and were replaced by a more abstract notion of the presence or absence of the competences – the savoir-faire and pouvoir-faire – required to perform the action. In “recognition,” too, a Sender sends a message to a Receiver, which produces the sense that the task is recognized as having been performed, not performed, well performed, badly performed, etc. The scheme may be represented diagrammatically as seen in Figure 5.1.

Figure 5.1 The Narrative Syntagm according to Greimasian Semiotics

The paradigmatic axis is based on Saussurian and Lévi-Straussian foundations. At every point on the syntagmatic axis (as in the single sentence), there are choices to be made. But such choices are limited to things which may be substitutable, one for the other. At each point in the narrative syntagm, there exist conventionally defined semiotic constraints, as to what elements are substitutable for each other without altering the meaning of other elements in the syntagm. These constraints may reflect binary oppositions (analyzed in terms of the “semiotic square”) or larger groups of substitutable elements, as in relations of “hyponymy.” Discourse elements, just as individual words, have similar relations, and may similarly be substituted (or not substituted). Greimas, however, offers a more formal description of how this works. Binary oppositional relationships, in particular, reflect a structure called the “semiotic square,” in which “contradictories” and “contraries” are distinguished. Where two terms are contradictory, not only does the assertion of the one entail the negation of the other; the negation of the one entails the assertion of the other. In a conventional opposition (contrariety), on the other hand, the assertion of one term entails the negation of the other, but the negation of one term does not entail the assertion of the other. Black v. White is such a relationship: if something is black, it cannot be white, but if it is not black, it is not necessarily white, as seen in Figure 5.2.

Figure 5.2 The Semiotic Square

“Opposites” like black and white, cat and dog, guilty and innocent may be analyzed in this way. Thus, in the conventional understanding of black and white as “opposites,” we go up the sides of the square, in treating “white” as the privileged manifestation of “not black” (the contradictory of “black”) and “black” as the privileged manifestation of “not white” (the contradictory of “white”). This may be illogical, but it aids considerably the construction of sense within social interaction (where we are not all logicians). It is this same process which prompts the lay assumption that a person found “not guilty” (the contradictory of guilty) is in fact “innocent,” choosing to privilege that particular choice from amongst the possible senses of “not guilty.” This also illustrates the proposition that the meanings of terms depends upon their internal sense relations rather than their referents in the outside world. “Not guilty” has different meanings in the English and Scottish legal systems, since the latter also provides the possibility of a “not proven” verdict.⁷⁴

In the sending and receipt of messages regarding performance, which constitutes the “Recognition” part of the narrative syntagm, the evaluations thus communicated are often termed “modalities.” They are not, however, confined to “evaluations” in the normal (judgmental) sense. They may include aesthetic and emotive evaluations. The behavior of a lawyer, for example, may be accompanied by such modalities as helpful, formalistic, polite, and efficient. Modalities may also be “deontic,” expressive of different types of normativity: in Western legal systems, behavior may be permitted, required, or prohibited, while in Islamic jurisprudence it may also be recommended or disapproved.⁷⁵

“Above” the “deep level,” there is a level of construction of meaning contributing more directly to the meaning of the surface data (for Greimas, the “level of manifestation”), which may be termed “thematic.”⁷⁶ Early in my career (and before my exposure to semiotics) I reviewed George Fletcher’s Rethinking Criminal Law⁷⁷ and was particularly struck by his use of the notion of “collective images.”⁷⁸ Fletcher argued that the technical intricacies of the common law of larceny, particularly in the seventeenth and eighteenth centuries, were best understood in terms of the relationship of different fact situations to the “collective image of acting like a thief.” That collective image served as a kind of paradigm: if one asked people what their typical image of the activity of a thief was, the response was likely to be the nocturnal burglar. (Of course, collective images are temporally and culturally contingent.) From this it does not follow that the daytime pickpocket would not be regarded as within the collective image of acting like a thief (though it might generate its own independent collective image – thus burglar versus pickpocket). Clearly, there will be cases closer to or more distant from the paradigm: the daytime burglar, the nighttime thief who puts his hand in the window but not his body, etc. But the collective image was not merely a description of typical action; it also included a tacit social evaluation. This collective image of acting like a thief involved a particular frisson, a fear of the intruder, a horror of the crossing of territorial boundaries, with all the attendant danger which, socially, such boundary-crossing is thought to entail. We may translate such tacit social evaluations into our semiotic model, as the transfer of modalities: the activity of the thief is recognized as (though here, perhaps, not necessarily performed as) dangerous.

In my version of semiotic theory, I call such paradigms “narrative typifications of action.”⁷⁹ From this example, three important elements of the notion of a “narrative typification” emerge. First, a narrative typification is not a definition in terms of necessary and sufficient conditions; as a model which informs our perception, it does not generate (as the positivist might wish) “demonstrable” judgments as to what is “within” or “outwith” the image. But it is capable of generating judgments of relative similarity. Second, such a typification is not a neutral description; it comes, always, laden with a form of evaluation (for even “indifference” is a form of evaluation). Third, some typifications belong to particular social and/or professional groups – “semiotic groups,” distinguished one from another by the (often overlapping but still distinct) systems of signification operating within them.

However, there is no complete divide between semiotic groups, even those of the laity and professionals. Even though (institutional) Law and Religion are forms of secondary symbolization (using the primary, natural language in a particular, specialized way), they remain (being human constructs) parasitic upon social understandings, despite any ideological claims to autonomy, such as the notion of the “artificial reason of the law.” It is a primary claim of the work I have done in legal semiotics that even lawyers cannot escape their own humanity and social settings. Underneath every legal formulation, the primary social understanding may be submerged, but not eliminated.

Greimasian semiotics stresses that both the narrative syntagm and narrative typifications of action are applicable not only to the content of a narrative (semantics), but also to the act of enunciation (or communication) of that narrative (pragmatics). Indeed, no complete understanding of the sense of the message itself is complete without making sense of the purpose of communicating it. I have called this the narrativization of pragmatics.⁸⁰ The enunciation of messages is as much an action as anything else which human beings do (as speech act theory has recognized). It is generally assumed to be meaningful action. We therefore have to ask whence derives its meaning. A Greimasian would claim that the very act of communication of a message is equally analyzable according to the model of the Greimasian and narrative syntax, and thus reflects (at the “deep level”) the semio-narrative syntagm. But within any particular society or social group, what counts as a successful enunciation, and what type of enunciative meaning is attributed to it (assertion, threat, play, irony, etc.) are matters of internalized social knowledge (at the “thematic level”),⁸¹ as indeed are the signs (at the “level of manifestation”) of successful performance.

Application of the Greimasian model to issues of law and religion thus requires us to pose the following questions:

Who is the Subject of the act (whether linguistic or not) in question?

How is that Subject constructed, which entails:

1. (1) Who sends to (or invests with) the subject the appropriate modalities?

2. (2) What are the appropriate modalities in this particular case?

What is the act which the subject performs?

Who performs the recognition of that act?

How is that recognition constructed, which entails:

1. (1) Who sends to whom the appropriate modalities?

2. (2) Who receives them?

3. (3) What are the appropriate modalities in this particular case?

Crucial to the application of this model is the understanding of the existence of different semiotic groups, defined by their distinctive ways of constructing and communicating sense.⁸² Thus, the same act may be understood (recognized) in quite different ways within different semiotic communities, not least because of their adoption of different semantic relations within the same natural language.

Space does not permit the giving here of a systematic account of the application of all aspects of the Greimasian model to “law,” “religion,” or “religious law.” But it is not difficult to locate within the model the nature of the differences between secular law and religious law. Two issues may here be highlighted. First, the identity of the sender who invests the subject with the desire/obligation/competence/authority to act. In secular legal theory this would normally be the state or constitution; generally, in religious law it would be God.⁸³ But our very use of the terminology of State, Constitution or God raises the very same (Saussurian) questions of sense and reference, to which there can be no objective answer within language.⁸⁴ A second issue is the forms of recognition which may be accorded, whether the act be one of a legal institution or a private individual. As noted above,⁸⁵ in western legal systems, behavior may be permitted, required, or prohibited, while in Islamic jurisprudence it may also be recommended or disapproved. The Halakhah does not have a formal typology of such modalities, but it is clear that there are forms of recognition which reflect the recommended⁸⁶ or disapproved, rather than the permitted, required, or prohibited.

VI Greimasian Semiotics Applied to Law (Secular and Religious)

In what follows, I attempt to summarize some of my earlier work on the application of semiotics to secular law and apply the same type of analysis to the halakhah, in order to clarify the areas of commonality, and those of difference.⁸⁷ It also provides an opportunity to show the interaction of various aspects of the semiotic model described in Section V, and even to add one or two refinements to it.

A The Discursive Basis

A fundamental starting point is the realization that “law,” whether secular or religious, is in fact a collection of quite distinct discursive practices, linked by a set of inter-discursive claims. There is legislation, adjudication (which includes both decision-making and justification), and doctrine. Legislation puts forward rules applicable in general; adjudication seeks to resolve the disputes of individuals in the light of both legislation and whatever other sources of general rules are accepted by the particular system; doctrine encompasses both commentary on the reasons for particular areas of legislation and adjudication, and theorizing about the nature of the particular legal system or systems in general. When assessing the inter-discursive claims made in both adjudication and doctrine, we must start by examining how the “other” discourse is constructed in the discourse under consideration (typically, how the sense of a statute is constructed in the course of adjudication). And this involves not only an examination of the syntagmatic and paradigmatic structures of the “home” discourse, but also (so far as may be possible) its pragmatics: Who is communicating this message to whom, and for what purpose? Indeed, this emphasis on the individual discourse may be applied, reflexively, to the present exercise, and the particular senses of the “legal,” “secular,” and “religious” used in it, even if that discourse is thought to have incorporated a sense originating elsewhere (whether in literary or other sources).

There is no single, “natural,” relationship between these different discursive practices. Different “legal systems” may construct them differently. For example, “doctrine” plays a far more important normative role in civil law systems than in the common law. For the halakhah, “doctrine” may take the form of aggadah, including, in biblical literature, narratives. Thus, for example, in studying the Book of Ruth, we must start by examining how the levirate and women’s inheritance rights are understood in that narrative, before considering any possible relationship with the relevant pentateuchal laws.⁸⁸

Of course, the Torah bears little resemblance to any modern legal document. The laws, even those found in the Pentateuch, are embedded in a wider narrative; they do not claim to be the enforceable law of a state, but are presented as divine revelations (or teachings)⁸⁹ of what such a law ought to be. They consist largely in concrete, individual rules rather than legal concepts and institutions. Biblical scholars have highlighted the “casuistic form” as typical of Biblical law. This form comprises two elements: a protasis (a conditional clause: “If … ”) and an apodosis, stating the expected (“normative”) consequences of such an action. The protasis contains the first two elements of the narrative syntagm: the “contract” (instituting the subject) and the “performance”; it does not indicate the third element (the “recognition”) explicitly; instead, the apodosis provides a normative form of recognition. In the Jewish tradition, the latter may be either religious or sanctions of the type expected by nineteenth-century positivism (though in the religious tradition we cannot ignore the religious connotations of the death penalty). The religious forms of recognition are not confined to the ritual sphere, as may be seen from the responsum discussed below.⁹⁰

There is a tendency among some commentators, both academic and rabbinic, to downplay the role of the biblical Torah text in their understanding of Jewish law: at most, the text of the Torah is to be understood through rabbinic eyes. But this in itself has major consequences. The Rabbis recognize the foundational character of the biblical text for the later tradition. In so doing, they have to grapple with the problems posed by the literary character of the biblical text (including its frequent linguistic ambiguity regarding modalities, particularly between permission and obligation). They construct the text as presumptively perfect in terms of human rationality, and thus adopt a hermeneutics which most commonly assumes comprehensiveness, consistency and lack of redundancy, reserving the residual safety net of superior divine cognition for difficulties which appear to defy human resolution.

The distinctiveness of the discourses of legislation, adjudication and doctrine has occasionally been recognized in both secular jurisprudence and the literature of the halakhah. For the former, we may cite in particular the work of Hans Kelsen. From the beginning, his basic conception of the norm had a distinctly semiotic tone: the norm, he argued, was to be understood as the meaning of an act of will.⁹¹ It was thus a form of sense attributed to an act of human behaviour. In his later work, he stressed that the justification of the norm (even in terms of reference to other legal sources) was not part of the norm itself, and should not be treated as part of the science of jurisprudence.⁹² He even abandoned the use of logical subsumption (of the particular case within a general norm) as a form of justification, on the grounds that both the general norm and the particular norm were the meanings of acts of human beings (legislators, judges) but there could be no logical connection between one such act of will and another. I have debated logical subsumption in linguistic terms with Neil MacCormick, stressing that the judge always has to make a decision as to the meaning of the general norm s/he is applying, even if that is endorsement of its conventional meaning.⁹³ On less philosophical grounds, both Scandinavian and American legal realism are also skeptical about the relationship between legislation and adjudication.⁹⁴

In the literature of the halakhah, Elon cites a story of regarding a poseq (R. Ḥayyim of Brisk) who wished not to hear the reasoning of a colleague charged with deciding a case (R. Isaac Elḥanan of Kovno), on the grounds that he respected the authority of the latter to make the decision, but did not wish to hear his reasoning, in case he disagreed with it.⁹⁵

I have analyzed the Hart/Dworkin debate in terms of the discursive structures presupposed by each, using the semiotic square and Blanché’s development of the square of logic into a hexagon⁹⁶ and concluded that the focus of Hart was on legislation (which could have gaps, reflecting Blanché’s hexagon) while that of Dworkin was on adjudication (which could not have gaps, reflecting the semiotic square, with its conventional closure rule): for legislators, gaps could always be filled in the future; for judges, a decision had to be made in each particular case reflecting the best possible construction of the respective rights of the parties.⁹⁷ In the light of all this, I would argue that the conclusion of Ben-Menahem that “governance by judges rather than by rules is the primary feature of the talmudic legal system”⁹⁸ is no mere oddity of a particular religious system, but rather reflects the nature of the relationship between of decision-making and justification in general. I have argued, moreover, that decision-making is a private mental activity often involving subconscious invocation of particular forms of social knowledge,⁹⁹ while justification is a public communication to a particular audience, using the conventions and closure rules of that audience.¹⁰⁰

B The Modalities of Recognition

A full analysis of the “not guilty” verdict¹⁰¹ indicates the following:

1. (1) People are not restricted to attributing a single modality to the action they are evaluating. “Guilty” and “not guilty” may be understood in terms of either “fact” or “proof” – or both: actually not guilty (= innocent) or not proved guilty (whether or not actually innocent).

2. (2) While the structure of particular modalities may be “binary,” as in “Guilty” versus “Innocent,” reflecting the most common form of sense, other, more sophisticated sense constructions are also possible, as in the Scottish triadic structure of verdicts, which is structured as a hyponomy.¹⁰²

3. (3) Both the structure (binary or hyponomic) and the substance of modalities may vary between different semiotic groups (reflecting their particular discursive structures), this in itself reflecting the different narrative syntagms in which they are involved.¹⁰³ In general, binary structures are a feature of orally communicated “common sense” (the sense we share in common), while hyponomic are more characteristic of and derive from literacy-based discourse.¹⁰⁴ But even this, as other aspects of semantics, is subject to socio-environmental factors, as the great variety of words for “snow” in Eskimo languages shows.¹⁰⁵

4. (4) In some cases even the semantic choices made at the paradigmatic level may themselves evoke narrative images.

I have stressed the link of narrative typifications of action (at the “thematic” level) with “tacit social evaluations.”¹⁰⁶ The latter also function as manifestations of the “recognition” element of the narrative syntagm,¹⁰⁷ specifically by the sending of nonlegal modalities in relation to aspects of the “Contract” and/or “Performance.”¹⁰⁸

The nature and variety of institutional recognitions available within the halakhah, despite the absence of the typical sanctions available in a national legal system, is illustrated by a remarkable fifteenth-century responsum by R. Israel b. Ḥayyim of Brunn, which shows the weakness of the distinction between the legal and the religious in this context. A murder, of one Nissan, had occurred in the German town of Posen, committed by two people: “Simḥah” (in a state of drunkenness) and Naḥman (described by a witness as “ignorant and illiterate”). The latter had shown no remorse or penitence, and R. Israel indicates that there is nothing he can do about him. Nevertheless the responsum opens with a strong religious condemnation: “I name them both murderers, utterly guilty before the divine tribunal, but, (alas), beyond the jurisdiction of the human court!”¹⁰⁹ On the other hand, “Simḥah” was reported to have been “filled with remorse and seeks to repent.” For him, R. Israel recommends as follows:

He shall journey about as an exile for a full year. Every day he shall appear at a synagogue – or at least on every Monday and Thursday. He shall make for himself three iron bands, one to be worn on each of his two hands, which were the instruments of his transgressions, and one to be worn about his body. When he enters the synagogue, he shall put them on and pray with them on. In the evening he shall go barefoot to the synagogue. The ḥazan shall seat him (publicly) prior to the Vehu Raḥum prayer. He shall then receive a flogging and make the following declaration: “Know ye, my masters, that I am a murderer. I wantonly killed Nissan. This is my atonement. Pray for me.” When he leaves the synagogue he is to prostrate himself across the doorsill; the worshipers are to step over him, not on him. Afterwards he is to remove the iron bands … After one year he shall continue his fasts on Mondays and Thursdays. He shall, for the rest of his days, carefully observe the anniversary month and the anniversary date of the killing. He shall fast at that time (the date) three consecutive days if he is healthy or only two days, the day of the wounding and the next day, the day of Nissan’s death, if he is infirm. He shall, for the rest of his days, be active in all enterprises to free imprisoned Jews (i.e., hostages held by gentiles), charity, and the saving of lives. He shall work out an arrangement with his (Nissan’s) heirs to support them properly. He shall ask their pardon and the widow’s pardon. He shall return to God, and He shall have mercy on him.¹¹⁰

Here, it is the “sentencing” which constitutes the form of communal (or institutional) recognition which the respondent proposes. The regime includes a physical mirroring punishment, elements of public humiliation, and the public confession and acceptance of the punishment as an atonement. But that is not the only form of recognition provided within the responsum: The question and answer¹¹¹ both also use moral condemnation: both perpetrators are “wicked.” This would be the primary form of recognition for the wider community: a social modality which evokes an image of a typical murder.¹¹² Indeed, R. Israel commences: “You have called Naḥman by his name Naḥman; you have called Simḥah a murderer. I name them both murderers.”¹¹³ Moreover, R. Israel concludes with reintegration within the community: “And since Simḥah has expressed remorse and seeks repentance and atonement, immediately upon his submission to the program of public degradations, he becomes our brother once again for every religious purpose.”

Generally, we may think that the classification of the institutional sanctions, by the attribution of the modalities of the legal (in the secular sense) or the religious, represent a quite separate recognition process, of purely academic interest.¹¹⁴ If so, this case provides an exception: the whole point of this regime, applied to Simḥah but not Naḥman, is that the offender must understand the religious significance of the regime, and indeed the community as a whole is being taught an important religious lesson in terms of how to treat offenders. Moreover, “legal” and “religious,” as applied to the sanctions, are not here constructed in a relationship of “contrariety”:¹¹⁵ there are clear communal sanctions, partly of the physical nature, but they have a religious purpose, and depend upon the religious point of view of the recipient of those sanctions himself.

C Drafting, Decision-Making, and Narrative Typifications

The distinction between decision-making and justification is well illustrated in the legal context. Narrative structures underlie judicial decision-making, and frequently explain why a decision has been made in an area of legal difficulty and where the purely legal justification appears problematic, or eminently contestable. I have applied such an analysis, inter alia,¹¹⁶ to Riggs v. Palmer,¹¹⁷ the American case which Dworkin used in Taking Rights Seriously to demonstrate the need to use legal principles (and not merely legal rules) in deciding hard cases,¹¹⁸ even though such principles might have to be inferred from the legal tradition, and would not pass Hart’s “demonstrability” test.¹¹⁹ Here, the courts agonized, in doctrinal terms, over whether a grandson who – knowing that he had been named his grandfather’s heir in his will – had murdered him in order to secure and accelerate his inheritance, should be allowed to inherit, in the absence of any exclusion of such cases from the New York statute on wills. Ultimately, the New York Court of Appeals decided (by a majority of 2–1) that there was a principle of law, to be applied in this case, that “A person should not profit from his wrong.” Dworkin rejected the view of Hart that the legal system consisted of rules only, so that if there was a gap in the law, it could only be filled by creation of a new rule by the exercise of judicial discretion. Against this, Dworkin argued that there was always a correct, or at least a best available, answer in the existing law, if the judges looked hard enough, even though only a “Hercules,” a “lawyer of superhuman skill, learning, patience and acumen,”¹²⁰ might be able to persuade his colleagues of such. But once one looks at the facts in “common sense” social rather than purely legal doctrinal terms, one realizes that this narrative is so distant from the typical situation of testamentary succession, which presupposes peaceable if not loving (a “tacit social evaluation”) family relationships, that it would “stink” (a nonlegal modality) to give the inheritance to the murderer. That means that we privilege the peaceable versus violent opposition in the narrative, and regard its deviation from the typical as sufficiently important to justify not applying the legal consequence normally associated with (peaceable) testamentary succession. In short, and despite the naive legal assumption that reasons stated by judges in their judgments represent fully and accurately the very bases of their decisions, we have to make a distinction between the private, mental processes of decision-making on the one hand, and the public, discursive processes of justification on the other. The “artificial reason of the law” may dominate in the latter, but the (no doubt culturally contingent) sense we have in common cannot be excluded at the psychological level.¹²¹

Both the syntagmatic axis and the paradigmatic (through narrative typifications) were prominent in the famous “Brother Daniel” case¹²² decided under Israeli (not Jewish) law on who counted as a Jew for the purposes of Israel’s Law of Return 1950.¹²³ The petitioner, Oswald Rufeisen, was born in Poland of Jewish parents, reared as a Jew, and was active as a youth in a Zionist Youth movement. When the Germans occupied Poland he managed to infiltrate a German police station and was able to pass information to the local Jewish population, thus saving them from deportation (and worse). He then fled to the forest and joined a group of Russian partisans. There came a stage when he had to take refuge in a small Catholic nunnery. There, he converted to Catholicism, while still regarding himself as a Jew and adopting the name of “Brother Daniel.” After the war he maintained his earlier commitment to emigrate to Israel and ultimately arrived there in 1958, claiming his right to Israeli citizenship as a Jew under the Law of Return. By a majority of 4–1, the Israel Supreme Court denied his petition, while allowing him to enter and ultimately obtain citizenship by naturalization. Rufeisin’s story combines elements from different Jewish narrative stereotypes. We commence with an unremarkable narrative of a Jewish boy seeking to fulfil a Zionist ideal, proceed to the quite remarkable story of a war hero, and then to the story of a Jew who has converted (deserted) to Christianity. To make sense of all this, we need either to select and privilege some aspects of the story and suppress others, or to create a new synthesis and with it a new evaluation. We do not normally find the hero and the deserter united in a single person. Justice Silberg, for the majority, described his own “psychological difficulty” in the case and speaks in terms of conflicting evaluations: “But this sense of profound sympathy and obligation [to Rufeisen] must not be permitted to mislead us and to justify our profaning the concept of ‘Jew’ both in name and in meaning.” The differences between the majority and minority on the court as to whether a Jew who had adopted Christianity could still be regarded as a Jew, I argued, depended on whether “Jew” and “Christian” were contraries within a semiotic square or whether they admitted of further possibilities (in accordance with the logic of Blanché). For the majority, they were contraries: a Jew could not be a Christian (and this, in the light of the malign history of Jewish-Christian relations, rather than on theological grounds).

But the differences between the majority and minority went beyond their conceptions of “a Jew” and extended to the Jewish nature of the State itself – whether it should be viewed simply as a continuation of the Diaspora experience (which maintained tradition in a defensive way) or as representing a rebirth, a fresh start. This latter issue, I suggested, reflects differences in constructing the relation between Past and Future, which could also be analyzed in terms either of the square or the hexagon. Justice Cohn, the one judge finding in favor of Rufeisen, contrasted the petitioner’s openness on arrival at the Port of Haifa (observing: “Had he folded his gown, hidden his cross and concealed his creed, the gates would have been opened wide without protest”) with the historic position of Diaspora Jews who, “loyal to their ancestral faith, donned the outward garb of the Christian religion so that they might continue to dwell in the lands beloved to them and harvest the fruit of their toils.” Citing Isaiah xxvi, 2, “Open ye the gates that the righteous gentile which keepeth the truth may enter in,” Justice Cohn argued that it was such openness which the State of Israel should now display, showing its own liberation from the constraints of the past.

There is much scope for examining the role of narrative typifications in both general norms and adjudication in Jewish law, but the issue presents itself differently according to the period, literature, and semiotic groups concerned. I confine myself here to the early period. In any legal system, the earlier we go back the more their general norms reflect ordinary speech rather than technical formulations. I have argued that many of the laws of the mishpatim of Exodus 21–22 reflect socially understood narrative typifications, even when those narratives are not fully stated. A simple but clear example is Exodus 22:2–3 (MT: 1–2),¹²⁴

1. 2 If a thief is found breaking in, and is struck so that he dies, there shall be no bloodguilt for him;

2. 3 but if the sun has risen upon him, there shall be bloodguilt for him.

The biblical provision has appeared to many as badly drafted: the first verse, which allows self-help, makes no explicit mention of the time of the incident; the only explicit reference to the time occurs in the second verse, which denies the legitimacy of self-help during the day. It is in the light of that qualification, apparently, that the permission of self-help has to be restricted to the nocturnal incident. This would appear at first sight to be a very strange type of drafting. Apparently, the audience is first given the impression that self-help is always available, then this is qualified by denying its availability when the incident occurs during the day. Indeed, some scholars have wondered whether the second verse may be a later addition: originally the householder was entitled to kill the intruding thief at any time of day or night; later this form of self-help was restricted to the daytime intruder.¹²⁵ But if we pose the question of meaning of v.2 not in the (semantic) form of: “what situations do the words of this rule cover?” but rather in the (narrative) form of: “what typical situations do the words of this rule evoke?” then we are entitled to take into account the image of typical thieving presented in the book of Job:

Job 24:14, 16

The murderer rises in the dark;

                                                that he may kill the poor and needy

                                                            and in the night he is as a thief.

In the dark they dig through houses;

                                                       by day they shut themselves up;

                                                               they do not know the light.

This indicates that nocturnal activity was the primary image of acting like a thief. Exodus 22:2 did not have to make that explicit: it was part of the narrative image evoked by the words.

As regards decision-making, it may be tempting to suggest that many of the thirty talmudic instances of decision-making “not in accordance with the Halakhah” collected and analyzed by Ben-Menahem,¹²⁶ may repay analysis in this light. And indeed, some of them bear substantive comparison with Riggs v. Palmer, insofar as a party is penalized for immoral conduct which is technically legally valid.¹²⁷ More generally, it is not easy to elicit the social understandings underlying the talmudic sources. Later responsa appear to be a more fruitful ground, as in the murder case on which R. Israel b. Ḥayyim of Brunn was asked to advise,¹²⁸ and in which a host of socio-religious understandings are explicit. But here we are dealing with the equivalent of a modern “sentencing” decision, in which considerable discretion, within legal parameters, may be given to the judge. Jewish family law, however, continues to use narrative typifications of female behaviour, with social evaluations influencing legal outcomes: “it is better (for a woman) to dwell in grief than to dwell alone” (literally: in widowhood);¹²⁹ and the fear that a woman seeking divorce may in fact have “cast her eyes upon another man.”¹³⁰

D The Narrativization of Pragmatics

It is not only in the spheres of the content of adjudicatory decisions that the notion of narrative typifications comes into play. I have applied it to the secular trial process, distinguishing within it the “story in the trial” from the “story of the trial,” the latter manifesting the narrativization of pragmatics.¹³¹ We can ask about the Subject, Performance, and Recognition of each of the plethora of different communicative acts within the trial process. In principle, the same model is applicable to religious institutions and discourse, and to the components of “religious law.” Nor is recognition of the speech behavior of participants in any formal legal process restricted to legal modalities. In 1992, a case was reported of an American judge deciding to convict for speeding on the toss of a coin.¹³² The case was not appealed, and so the conviction stood as legally valid. However, the judge was subsequently censured by the Judicial Conduct Commissioner of the State of Washington, on a complaint charging him with violating judicial canons governing the integrity of the court, the avoidance of impropriety and the impartial performance of duties.¹³³ Similarly, in an English case, R. v. Langham and Langham,¹³⁴ the defense appealed on the grounds that the judge “was, or appeared to be, asleep during part of the trial and thus justice was not seen to be done.” The Court of Appeal refused the application for leave to appeal on the grounds of lack of evidence that the judge was actually asleep. Nevertheless it commented that if it was true that the judge appeared to be asleep, that “was a matter which the court would certainly deplore but was not a sufficient grounds for saying that justice was not seen to be done.” Given the essentially private nature of proceedings of rabbinical law courts, official parallels to such adverse social recognitions are unlikely to be found. However, there is much in unofficial media sources deploring judicial behaviour in relation to women, particularly in divorce cases.

E Authority, Truth, Trust, and Objectivity

We may turn to semiotics for assistance also in understanding the notion of authority in the legal system. It is part of the competence – pouvoir-faire – with which the Subject (whether of the legislative or adjudicatory act) is invested. But what is the appropriate legal modality of recognition of that Subject’s act? Normally, we would use the language of validity (the legal appropriation of true). But here as elsewhere, the argument advanced above about the relationship between truth and trust¹³⁵ becomes relevant, and is justified by the understanding that semantics are necessarily mediated via pragmatics. In short, the legal system cannot objectify itself so as to eliminate the dominance of persons over ideas. Of course, notions of political and legal authority are inculcated through education from an early age. But why do we trust in education? Because we trust our teachers, the delegates of our parents. Is the authority of the gedolim, regarded as the supreme halakhic scholars of the age, constructed through analysis of their works, or through trusting the judgements of others? Inevitably, for the most part it must be the latter, since intellectual access to such scholarship is restricted.

Modern secular law prides itself on its objectivity. The classic account of how this works is that of the legal philosopher H. L. A. Hart, for whom legal systems were characterized by the “union of primary and secondary rules.” “Primary rules” were the rules of substantive law; “secondary rules” were the rules which determined how the legal system worked, and included in particular “rules of recognition” formulated so as to provide a “conclusive affirmative indication” as to whether a purported primary rule was (recognized as) valid or not. This model thus generated the “demonstrability thesis”: with little exception, one could always know, objectively, what the law was.¹³⁶ One might, perhaps, expect a system of religious law to be even stronger (“more objective,” if that were possible). But this turns out, in the case of Jewish law, not to be the case. The end result, in my view, is that the system depends on trust rather than truth, and what it “demonstrates” is the mediation of semantics via the narrativization of pragmatics.

In the course of a five-year research project which I directed at Manchester¹³⁷ on the vexed, practical problem of the “chained wife” (agunah) whose husband refuses to cooperate with the court in granting his wife a divorce (a get), thereby preventing her remarriage and rendering adulterous any subsequent relationship into which she may enter and the children thereof “illegitimate” (mamzerim), we discovered that what modern legal philosophers call the “rules of recognition” of the legal system remain, in Jewish law, subject to significant controversy. This, we found,¹³⁸ applied to (1) the rule of majority decision itself (often regarded as the most basic “secondary rule” of Jewish law); (2) the rule that, as amongst post-talmudic authorities, later opinions are followed in preference to earlier ones (hilkheta kebatra’ei), but leaving a discretion to the contemporary judge not to follow the later authority when that the latter’s decision might have been different had he been aware of a (later discovered) earlier authority; (3) uncertainties also exist in such areas as the status of newly discovered MS sources; (4) the status and identification of the “leading authorities of the day” (gedoley hador).

Indeed, doubts as to what is authoritative in the system are so extensive that a (creative) doctrine of doubts has developed, granting discretion to be lenient in various types of case given the existence of different levels of doubt (safeq): a single doubt relating to an issue of rabbinic status is sufficient to justify exercising leniency; a single doubt relating to an issue of biblical status is not sufficient to justify exercising leniency, but a double doubt is sufficient.¹³⁹ Yet reliance on these rules regarding doubt is discretionary. In the absence of any universally accepted central rabbinic authority, different rabbinic courts will exercise that discretion in different ways. Some may even deny that the discretion exists. Can there then be any objectively correct answer in such a situation? Despite the continuing endorsement by some influential rabbinic voices of a positivist-inspired objectivity, the better answer – and one, I would maintain, itself supported by Jewish tradition – is that the halakhah is based at least as much on the concept of trust¹⁴⁰ as on that of truth.¹⁴¹

We should not, moreover, assume a universal conception of “truth.” The Rabbi, philosopher and theologian Steven Schwarzschild wrote: “In Judaism truth is primarily an ethical notion: it describes not what is but what ought to be,”¹⁴² citing the association of truth with ethical notions in the Bible¹⁴³ and rabbinic literature.¹⁴⁴ Hermann Cohen designated the normative unity of cognition and ethics as “the fundamental law of truth.”¹⁴⁵ And Martin Buber is said to have identified faith (emunah¹⁴⁶) with truth, here conceived as interpersonal trust.¹⁴⁷ And such conceptions derive support from classical rabbinic sources, as is shown by the already cited¹⁴⁸ talmudic passage where a heavenly voice (bat qol) affirms apparently contradictory opinions of the rival rabbinic schools of Hillel and Shammai (Erubin 13b) in the words “these and these are the words of the living God,” but concludes that in practice one should follow the views of the School of Hillel. The Talmud then asks: “what was it that entitled Beth Hillel to have the halakhah fixed in agreement with their rulings? – Because they were kindly and modest, they studied their own rulings and those of Beth Shammai, and were even so [humble] as to mention the action of Beth Shammai before theirs.”

Though my conclusion regarding the relationship between truth or trust was arrived at largely through the above research project on contemporary Jewish divorce law, there is reason to suspect its pertinence also to secular law. Critical theorists have long maintained that the objectivity of the legal system is an ideological construction, designed to preserve while concealing the play of power, through invoking logic and in other ways.¹⁴⁹ Dworkin’s invocation of Hercules as a “lawyer of superhuman skill, learning, patience and acumen”¹⁵⁰ may appear to be a giveaway, but even Hercules seeks to base trust in himself on rationality rather than intuition or other personal characteristics. But there are good semiotic reasons to prioritize trust: trust is an interpersonal relationship, thus integrally involved in pragmatics; truth (on the Saussurean approach to reference) is merely a claim, which must take account also of the narrativization of pragmatics.

F The Medium: Orality and Literacy

No semiotic account may properly ignore the media through which the message is communicated. Even where this is language, there remains an important distinction between orality and literacy. There are (associated) diachronic and cognitive aspects to the relationship, as seen in the important work of Walter Ong and Basil Bernstein.¹⁵¹ We can trace the process by which oral norms of behaviour increasingly assume a literary form. In the Bible itself, Daube pointed to what is there an unusual feature of drafting. Num. 35:16–19 commences: “But if he struck him down with an instrument of iron, so that he died, he is a murderer; the murderer shall be put to death.” As far as the (punitive) consequences (“recognition”) of the act of striking are concerned, “he is a murderer” (rotseax hu) is logically redundant,¹⁵² although verse 19 does add a further consequence of being “a murderer,” allowing the avenger of blood to kill him when he meets him. So we have here a classification of the act, or more accurately the actor, to which a further normative consequence is attached. Daube called it the “diagnosis form,” on the analogy of medicine.¹⁵³ Similarly, in commenting¹⁵⁴ on the development from the use of verbal forms (“if a man steals”) to nominal forms, particularly the use of “action nouns,”¹⁵⁵ Daube, observed: “there has been some reflection on the activity in question, there is some trend towards instruction, systematisation, classification, perhaps, and the thing is becoming more of an institution” while stressing that such a development was by no means restricted to law but rather characterizes “much of the intellectual history of civilization.”¹⁵⁶ But then, we must ask, cui bono? In whose discourse (in which “semiotic group”) is such a development useful? We very quickly arrive at the (surely, obvious) conclusion that there is no single “legal” discourse: we need to distinguish between laity and professionals, and indeed subgroups within each.

One consequence of this is that there is no single, or privileged, form of “legal” interpretation. The closer a written text may be to its oral (social) origins, the more important the “oral residue” which remains submerged, in the form of “narrative typifications of action.” It is not that the written form necessarily “defines” such typifications; rather, such typifications may be evoked by example.¹⁵⁷ This requires a different form of reading/understanding from that of modern law. In particular, there is no warranty whatsoever for applying the logic of “necessary and sufficient conditions” (“If” = “if and only if”). Rather, it becomes a matter of assessing the relative similarity of other cases to that offered in the text and to judge the utility of deciding similarly.¹⁵⁸ To take an example from my recent work on the Book of Ruth: the institution of levirate may be typified by Deuteronomy’s “when brothers live together … ” (Deut. 25:5) but should certainly not be restricted to that situation.

VII Conclusions¹⁵⁹

In one important respect, our semiotic analysis makes a difference to the way in which we understand the relationship between law and religion, at least as regards religions which include a doctrine of revelation. The starting point of the narrative syntagm, the “Contract,” involves the construction of a (human) subject, who in the legal context may be legislator or judge. In each case, it presupposes the transfer, through communicative means, of a “competence.” For orthodox Judaism, the Sender of that competence is God. Revelation (in its various forms)¹⁶⁰ thus cannot be confined to theology;¹⁶¹ it plays an integral part in the way we make sense of actual legal practices. Of course, human understanding of God as the source of revelation requires, from a semiotic point of view, the human attribution of particular modalities of competence and reliability to God, as Subject of the communicative act of revelation. “God” in human discourse is necessarily a human discursive construction, even though in so using it the claim is made that it refers to a real God. The existence of this God, not to mention its nature, is beyond human proof, notwithstanding the (largely anthropomorphic) heuristic devises which we use to get close to it. And essentially the same argument may be applied where “the Law” is personified as the Sender.

A sceptic may wish to go no further, and say that much of the “religious” discourse is an exercise in fictions. But even if we accept that human language cannot “refer” (in the strict sense) to matters which are not sensory and about which deictic (in the strict sense) statements cannot be made, this cannot prove the non-existence of such phenomena, any more than positive statements about them can prove their existence. On this argument, truth cannot be established by discursive means, even though the credibility or force of such statements can be explained in terms of interpersonal trust.

But there are issues where the tension is more acute. Take the example of a “cultural Jew,” brought up in a household of moderate observance, who continues to observe some of the rituals, such as the Pesach seder, but without any sense of religious obligation. How do we characterize such a practice? Do we recognize it through the modality of “religious” or not? And can we make that judgment independent of the religious politics of any particular community (or semiotic group)? I am reminded, in this context, of the famous observation of Maimonides (in what appears to be its most authentic version) in relation to the eschatological fate of Noahides. They do merit the world to come, says Maimonides, but only if their observance of the Noahide commands is because they believe that they were given by God; otherwise, such Noahides (who observe those commands because of their rationality) may be regarded only as wise men.¹⁶² All the more so, it may be argued, in relation to cultural Jews? But they too have an internal point of view, that of their own semiotic group, and if that group qualifies such behaviour as “Jewish,” even if not as “religious,” who are we to quarrel with the sense of Jewish identity which they construct? That, I may observe, was certainly the approach adopted by the late Justice Haim Cohn in the Shalit case and in his (wonderful) dissenting judgment in the case of Oswald Rufeisin (“Brother Daniel!”).¹⁶³

The overall effect of the argument of this paper may be a simple affirmation, at least in the following sense: the basics of sense construction, as understood by the Greimasian model, are the same for both law and religion. At the same time, the model allows for the identification of differences. But it also problematizes the value of the concepts themselves: who, we may ask, needs to talk about either “law” or “religion” as such? – a very different question from that of the characterization of particular acts or norms as “legal” or “religious.” “Law As Religion: Religion As Law” is thus a secondary (or meta-) question addressing the relationship between institutional concepts rather than human behaviour in either its factual or normative dimensions. It may, however, figure large in the rhetoric of religious politics, whose full understanding requires us to narrativize the pragmatics (speech behaviour) of its participants.