Brink Messick’s long-awaited Sharīʿa Scripts is finally out at Columbia University Press:
Multi-century approaches of the sharīʿa have regrettably transformed law into a banal history of ideas without much connection to practice. Messick’s Sharīʿa Scripts takes the sharīʿa right from the economy of the local, that of central Yemen, and places research at a micro level. Historical anthropology makes possible the tracing of genealogical lines of power relations, and the depiction of narratives and discourses in relation to local practices. This book, which takes the logic of texts and their practices to new heights, stands out as a masterful contribution to sharīʿa studies worldwide.
Brink Messick’s book-manuscript on sharia discourses in modern Yemen comes as a welcome change to a literature predominated by studies of long durations of “Islamic law” and the sharia. Messick’s main contention that the sharia is only meaningful as a “local enterprise” (212) in the space and time of a particular territory and political economy is a major endeavor to understand the sharia in all its concreteness as a discursive reality within a régime of truth. Thanks to Columbia University Press for publishing the manuscript in its integrity, without major cuts or changes. This is a much needed book that will open sharia studies to new horizons, and would serve as reference for scholars and students. Even though Messick’s preoccupation centers on “legal” practices and discourses, the book could nevertheless serve as a template for an understanding of Islamicate societies in terms of micro-discursive genealogies of power and régimes of truth.
For years I have been using Messick’s Calligraphic State in my seminars as a must-read reference to understand how sharia texts in all their diversity are constructed in a space and time of competing discourses—Bakhtin’s chronotope model of a space-time configuration of competing pretenders. Messick’s present book provides an even more refined optic for reading texts as x-rays of the power-relations in the archival layers of historical formations to which they belong. This new optic of reading texts in their space and time dimensions demands refined micro-hermeneutical techniques for the power relations to come fully to light, as neither space nor time as categories of knowledge are privileged, since they are utterly interdependent.
Most studies of Islamic law take as their point of departure the postulate that the sharia binds together fifteen centuries of rules and regulations for the societies that are governed under such precepts. In this multi-century approach what is favored, besides the overall coherence of the enterprise of the fiqh (the interpretations of the sharia, which are distributed into a multitude of madhāhib, or law schools), is discourse over practice, or theory over the contingencies of the terrains that are governed by sharia law. Such approaches, which predominate the research in Arab and Islamicate societies and in the West as well, leave us with all kinds of problems and unsolved issues. First of all, they assume that our understanding of sharia law is limited to (or predominated by) the doctrinal level, that is, the discourses that are generated in the libraries of the fiqh manuals across centuries, which in their totality have a level of coherence that cannot be achieved in practice. Second of all, the archival material, whenever available (in particular for the Ottoman centuries), is supposed to be an application of the grand theory as generated by the various madhāhib. There is thus undeniably a “precedence” of the doctrinal over the archival, in the same way that there is a “precedence” of the written text over the oral, or what is said over the seen.
Brink Messick’s book on Yemen in its pre-Republican period comes as a much needed enterprise that challenges the precedence of the doctrinal over the archival, or the library of the fiqh over the archive, or the universal concept over the regional. This is a rare and sophisticated endeavor which points at how much work and patience are needed once we move from the macro to the micro-historical in all its textual complexities (322). Messick has already outlined his method in his Calligraphic State, published in 1993, and if the present manuscript on sharia scripts in Yemen took so much time to materialize it is because micro analysis is more demanding in its execution than anything that the multi-century approach would dare to accomplish.
To begin with, Messick operates within a broad division that places the fiqh manuals in a “library” framework, which stands on its own in the face of the various “archival” materials. The latter are comprised for the most part of the texts produced by the sharia courts and what the parties in conflict or in a notarial act keep in private in the sanctity of their own homes as evidence that a transaction has been accomplished. Such documents therefore “register” the contractual act that the parties must attend to.
However, Messick is not satisfied with such broad division of “library” and “archive” on its own. Following Mikhail Bakhtin he conceptualizes three levels of textual practices (or discursive practices). There is one that is “primary” and which consists of all those document-texts that are kept in the “privacy” of their holders, followed by the “secondary” material of the sharia courts and other “public” instances, and, finally, the “tertiary” level (sphere) of the fiqh manuals in their various genres (tafsīr, shurūṭ, and fatāwa). Those various levels are obviously not separate, they are mutually interdependent, and rely on each other’s existence for their overall organization. Thus, the “library” is the “tertiary” sphere following Bakhtin’s classification, while the “archive” is formed by the “primary” and “secondary” texts, or the “private” and “public,” following a modern civil-law classification. There is no “primacy,” however, of one level over the other. Thus, if the fiqh happens to be the “tertiary” textual discursive level (312), the implication here is that it does not necessarily feeds itself on the practices of the courts, nor are the courts obligated to use the “upper” doctrinal corpus as their framework of reference; what they in fact do in most instances is allude to the doctrinal works rather than cite them directly, even though this possibility is, of course, not to be excluded. For example, when it comes to the practice of writing fatwas, which is commonly assumed to be the most “practical” aspect of sharia law, “there is no such direct connection between local fatwā-giving and court processes” (159). Yet, the fatwas are somehow “needed,” a need that remains elusive at best, particularly in pre-Republican imamic Yemen where the presence of an interpreting imam roots the fatwa in a régime of truth—and provides it with a much needed legitimacy—though of a much higher symbolic authority than that of the mufti in Sunni Islam.
Because in the interplay between writing and the oral and aural, there is no primacy or formal hierarchy (hence an absence of logocentrism which would have pleased Jacques Derrida), “documentary evidence does not stand by itself” (134). Presenting a document as evidence is a complex program of “inscribed writing” (144), which goes through the various stages of dictation (imlāʾ), witnessing, oral reading, prior to creating an authoritative form of oral and aural transmission (130). For instance, dictation involves on its own, first, a retention of the dictated text in memory, and, second, the inscription of a transformed version in writing, both of which constitute a form of “knowledge.” What is therefore at stake in oral reading and dictation is memory, the material trace, and expression (133). I look at the latter as various topoi of practice, which bypass the rigidity of formal hierarchies, and which make the analysis of discourse possible. The traditional macro studies of the fiqh corpuses, which tend to bracket practice (which by definition operates at a micro level in relation to a territory and political economy), are unable to analyze the fiqh texts as discourses—the latter would eo ipso assume their operation only in relation to certain practices.
We thus have a multitude of juxtaposed texts and archival formats operating at different discursive levels with no primacy as such, which makes Bakhtin’s classification of primary, secondary, and tertiary, which in this instance translates as notarial documents, courts processes, and fiqh doctrinal corpus, a bit problematic, even though it serves at delineating discursive series that could be viewed as autonomous in their own right. What in effect Messick is attempting is a direct immersion in practice, without the need to make false delineations like theory versus practice, or the doctrinal versus the court practices. For example, when in Chapter 10 contracts are examined, the question of contact-law versus contract is not portrayed as operating within a formal hierarchy, say, that a contract drafted in court must obey clauses of contract-law. The reason for this absence of a formal hierarchy is that once we establish practice—hence discursive practices—as our main entry into the system, the traditional hierarchies, which generations of scholars have been operating with, receive a new meaning, if at all. We may also not need them at all. For example, when it comes to contracts and the laws derived from the fiqh, Messick avoids the rigid dualism and operates instead in terms of composition, modeling, and models (340). Practice here means “understanding the textual properties” (299) of a text which could be a document emanating from a sharia court, a Zaydī fiqh manual, or an official bureaucratic document. Moreover, this approach has to account for evidential texts that could be written or spoken. What is crucial here, when it comes to the written and spoken, is, again, there is no primacy of one over the other. The same could be said about the spoken and the seen: the say is not to see, parler n’est pas voir, as Foucault would say, following Maurice Blanchot.
By taking practices and their discourses as his entry point, Messick is able to extract from the various discursive layers under analysis themes (or topoi) that may not have been apparent at face value. Of particular importance in this regard is the theme of custom (understood as ʿāda or ʿurf), which traditionally receives the treatment of an “outside” to written law, that is to say, how much of the latter has been “affected” by, or acknowledged as, custom or customary law? What kind of margin does written law tolerate within its corpus as far as custom is concerned? Messick’s approach in contrast looks at the place of custom “not outside, but inside such textual formation” (240), which implies an attentiveness to the “internal duration to the act” of drafting documents. To elaborate, in the grand division between the abstract non-historical “model” texts which serve as ready-made templates, on the one hand, and the historical documents which have been drafted by notaries, judges and their scribes on the other, it is indeed custom that plays on that internal duration to the act of writing (which in turn is an outcome of the oral and aural); hence it is custom that configures the historicity of a document (361).
The immersion into practice is demanding, first of all because it operates primarily and makes sense at a micro-level, and second, because a genealogy of texts is necessary. In other words, unlike the multi-century-macro approach which is historicist, in the sense that it is a general history of ideas that sees each text the product of its own period (even though the roots to this period in relation to practice remains by and large unexplored), genealogy in contrast goes further than that, as it looks at texts as operating in clusters. In the case of the Zaydī fiqh, for example, the late fourteenth-century Book of Flowers was still operative in the pre-Republican period, thanks in part to the interpretations and commentaries in the 1930s and 1940s, of what became The Gilded Crown, with the nineteenth-century interlude of the predominantly critical work of the “Sunni” Shawkānī in between the original matn text (which represented the condensation of the views of the five early imams) and the subsequent tafsīr and sharḥ.
What is crucial here is the awareness that a twentieth-century work like The Gilded Crown is itself a work of previous centuries—a product of archeological layers under different historical formations. The other side of genealogy is an attention to the author–writer paradigm, which Messick borrows from Foucault.
14–15: Where library discourse embodied a set of culturally and historically specific “author-functions,” archival discourse, in contrast, comprised distinct writer-functions. I adopt the plural both for Foucault’s well-known term and for my proposed archival counterpart to account for the distinct genres that existed within both the library and the archive. Following Foucault, both conceptions should be understood as defining functions that provide a basis for what he termed “a typology of discourse.” In this discursive sense, but in different genres, it was possible for the same individual to be both an author and a writer.
I want to question the author-function which is attributed to the library discourse (the fiqh doctrinal manuals). The assumption here is that both library and archive are practices and discourses (or discursive practices): should we then limit the archive to a writer-function, while the library is endowed with the “privilege” of the author?
There are all kinds of writing roles in those texts, the library and archival material: that of author, writer, editor, narrator, signatory, but they are not all present in one text. For example, a contract has a guarantor, but not an author; a letter has a signatory, but not necessarily an author. But a letter drafted by a Zaydī imam has an author, because it could be inscribed within a broader corpus, hence could be disseminated and quoted on the basis of its authorship, due to the status that this particular imam enjoys within the community of scholars. The idea here is that for a text to have an author it must be inscribed within a broader corpus, say, the œuvre of a particular author as the totality of his works. A court verdict signed and sealed by a judge has a writer but does not have an author on the basis that the judge cannot attach his text to a broader “work” of his own making. In Islamic courts verdicts do not serve as precedent as they do in common law, hence a decision by a judge remains localized without authorship. Doctrinal texts in contrast have that authorial quality, because each work is attributed to an author who is more than the writer of the text. Moreover, besides being attached to an author, say, a Zaydī imam, a fiqh manual is part of a larger ensemble of texts belonging to tradition. Messick is right when he traces the genealogy of the “texts that matter” in pre-Republican Yemen to their fourteenth- and nineteenth-century roots: it is such genealogy that makes both discourse and authorship possible. What is probably unique about this kind of authorship where the “school” (madhhab) is predominant is the system of cross-referencing among the “authorities” of the madhhab which spans across many centuries. In the scientific and literary European genealogies analyzed by Foucault, the discovery of common discursive layers is much harder, due primarily to an absence of direct cross-referencing.
I find a limit to this argument, in particular when it comes to the “library” of doctrinal texts. Not that there is anything wrong with attaching authors to such texts, but I question whether the fiqh manuals are intended to be referred to specific authors in the first place. Let us recall the tripartite division which is borrowed from Bakhtin, where “primary” and “secondary” coincide with the archival materials used by individuals and courts alike. Besides that such “archival” material is timely, it operates only by attaching names, signatures, and seals to a document. Moreover, the concerned parties and their witnesses are the names that matter, not the writer of the document. The library texts—the tertiary level—share an a-temporality which dissociates the text from its nominal writer-author.
15: Such usage separates a named author or writer from a discursive function, historical agency from textual form. For library texts, this is to distinguish attributions to (and also in-text citations of) specific authors from the patterned avoidance of the proper name. By the same token, the possibility of dating a book or an opinion is distinct from the a-temporal nature of its textual discourse. Among archival texts, in contrast, we enter a realm defined by identified handwriting and the signed name of the court or notarial writer and, in certain periods, a personal seal. Yet the discourse of the proper name that was characteristic of an archival text pertained not to such third-party secretaries or notaries, who in fact wrote and signed, but rather to the principal parties, the individuals who entered litigation or a contractual undertaking but who (usually) did not sign the resulting documentation.
What I find of value in this claim is the “separation” between author and discourse for the library doctrinal texts, but we need to find out why this is the case. The other matter of contention is whether between library and archive there is another “separation” regarding the proper name attached to the text.
Messick follows a close reading of Foucault on the notion of the author, but he misses a distinction that could be useful between statement (énoncé) and sentence (phrase). A text is obviously composed of sentences which refer explicitly or implicitly to a grammatical “I,” even if the text is drafted in a third-person mode. This “I” which stands on behalf of the writer, and could as well refer to an author, is what brings the text together. When a series of texts are juxtaposed together as belonging to a school of thought or way of thinking it is by virtue of their homogeneous nature. Because anyone can attach the grammatical “I” to a sentence, the claimed authorship is an external variable.
Foucault’s main contention is that statements are different in this regard. First of all, statements are rare because they straddle between several heterogeneous discursive layers. Statements make discourse possible precisely because they are based on intrinsic variables. If the phrase derives from a subject of enunciation (sujet d’énonciation), the subject that pronounces it, the statement by contrast does not derive from its subject: it is indeed the place of the subject that derives from the statement (le sujet de l’énoncé). The place of the subject is in turn an anonymous “we” (le “on” anonyme). In sum, we need to distinguish between a subject of enunciation for sentences and phrases, and a subject of the statement, which tends to be anonymous.
Which brings us back to the doctrinal fiqh manuals. There is a common perception that the fiqh manuals are only nominally authored, since they represent “compilations” (tajmīʿ) of well-known opinions from previous generations of scholars. Each generation compiles and re-arranges according to the relevant criteria of the period: one has to be faithful to the tradition of the madhhab, but at the same time one has to adapt to new situations, as new problems arise which would trigger the hermeneutical circle. The author-faqīh becomes therefore a “compiler” who re-interprets and re-organizes the old texts, which become “his” own, as well as belonging to the madhhab, based on the criteria needed for his period. The notion of “compilation” is at times looked upon condescendingly, as if it is uncreative, lacks authorship, leading to an infernal repetition of the same, hence to a disconnect with the social and economic reality of the time. This is particularly true, we are told, of the Ottoman period, as researchers typically shun the doctrinal manuals in favor of the more “real” sharia courts and other bureaucratic documents.
Messick’s book is a direct response to such condescending views: all discursive levels have realities of their own, and those realities overlap and share languages and grammars, albeit they play different roles. However, in light of the above proposals, we can reframe our take on the various discursive levels a bit differently. Compilations do not in their essence refer to an author or a subject of enunciation (sujet d’énonciation), but to a position of the subject (sujet de l’énoncé). They are therefore, together with the primary and secondary levels, embedded into an archive within a particular historical formation. The latter is inscribed within a diagram of power-relations which produces the institutional apparatuses of knowledge. Strictly speaking, therefore, I would argue that the primary, secondary, and tertiary discursive levels, in their institutionalized organization between library and archive, are all “archival,” whereby the “archive” is inscribed with stratified historical formations. They are all discursive practices with nominal authors and a collective “we” as reference.
Which brings me to my last point, regarding the presence (or lack) of a third-party adjudication in the Zaydī system and in the fiqh at large. By this I mean the presence of a universal process that would accommodate any person as member of the community (or “nation”) irrespective of religion, rank, and status. When the delivery of truth by any member of the community is accepted for what it is, seeing becomes of primordial importance in the process of verification and validation of witnessing. That’s the kind of transition that Yemen went through—a process that is by far from over—when the Zaydī imamate, as a polity, was over in 1962 and the country became a civil-law republic. In a civil-law political economy, where the reference to imamic traditions is not anymore the norm, labor becomes finite, as its reference is a competitive market economy where the traditional notions of equity and fairness are not normative, at least not predominantly so. The modes of truth that produce knowledge are based on a process of verification where the market economy serves a reference.
Even though Messick’s book is mostly devoted to pre-Republican Yemen, it nonetheless prepares us well for the transition that Yemen has been going through since the Zaydī imamate was lost. To wit, a hidden key notion in Messick’s analysis is the concept of governmentality, or gouvernementalité, understood as the “mindset” (mentalité) of “government”—not government in the strict Anglo–American sense of the term, but as the management of relations of truth and verification, and the power relations that they generate in the space of a political economy. Under the Zaydī imamate, and in contrast to the land tenure systems of the Ottomans, land was mostly private (milk), and the document of sale-purchase, known as baṣīra, was predominant (345). The wager here is to follow up such contracts under civil law when the régimes of truth and governmentality are altered. Maybe this could be the topic for another book which is already implied in this one.
Professor of history and social sciences
Loyola University, Chicago