Because the methodology of historical research gives primary importance to the study of the context in which events occur, it is often seen as antithetical to theory. Theory implies an abstraction from the details of historical reality—it is constructed on the basis of stylized facts—and seems therefore incompatible with the historian's approach. The latter purports to interpret the raw facts in all their richness and complexity. Interpretation is needed not only because all the facts are not known with certainty but also because the manner in which their effects interact is often open to debate. In the eyes of most historians, however, the ambition to interpret the raw facts in all their richness and complexity excludes any recourse to the artifact of a model, on the grounds that it is always a simplified version of reality that adds an unnecessary second-order elaboration of the facts.
I argue that the tension between the need for contextualization and the need for theorization is not as strong as it is usually thought among historians and that historical research would immensely benefit from accepting the guidance of theory. Without using the word, some historians actually propose a theory, or the sketch of a theory, to explain the facts that constitute the object of their inquiry. Yet the general pattern remains one in which grand theoretical designs of the mechanisms of long-term change are laid down by other social scientists and by economists, sociologists, and political scientists in particular.
Two features of a (good) theory deserve to be stressed. First, if a theory simplifies reality (in this sense, it is an abstract construct), then it is not disconnected from the same. Many theoretical endeavors are actually sparked or motivated by a fact, be it an empirical observation on a field of study, a transformation process discovered by reading a history book, or any sort of knowledge acquired inadvertently or purposefully. Once the intuition has been elaborated into a theoretical structure, it can be put to systematic testing by collecting additional evidence with which it can be confronted (evidence regarding other regions, countries, or periods). Second, a theory does not yield a unique prediction but typically highlights the conditions under which a given outcome is predicted to occur. In no way does it presume that reality is uniform. As a matter of fact, the fecundity and relevance of a theory are not demonstrated up until the research community has found evidence that confirms its central prediction when the appropriate conditions obtain and evidence that disproves it when such conditions are absent. This requires materials that lend themselves to comparative analysis across regions or across time. The latter, dynamic analysis is particularly appropriate for the historical discipline. There is thus a need for historians to have available theoretical structures that can transform a succession of events into a meaningful and logically articulated story.
I now illustrate the previous idea by an example involving Islamic study material. A recent theory of legal dualism proposed by Aldashev et al. purports to explain the influence of the statutory law on the custom followed in rural communities.Footnote 1 The theory describes strategic interactions between a customary authority (the informal judge), the modern judge, and the plaintiffs who belong to the disadvantaged groups of the society (the commoners). Elite people do not complain because their interests are protected by the custom. Commoners decide whether to appeal to the formal law, which is more favorable to them, but this involves two costs: the transaction costs of using the modern court system and the opportunity cost of losing the benefits of participating in traditional social networks (questioning the customary authority invites serious retribution at the local level). When settling a case, the informal judge must choose the extent to which he departs from the custom, which is costly to him, in order to keep the commoner–plaintiff within his jurisdictional realm, which yields him a benefit. In such a setting, a “progressive” statutory law may cause the custom to shift in the direction of the law, the so-called “magnet effect.” Such an effect does not necessarily take place, however. Under some conditions, the custom remains constant. At the other extreme, the custom may disappear in the presence of the law because all plaintiffs seek litigation in the modern court system.
An interesting application is provided by the transformation of women's rights of inheritance in Muslim rural areas. In many such areas, women do not traditionally inherit from their parents in spite of the Islamic injunction that women should inherit half of the share of their brothers. Behind the customary rule lies the fear that ancestral lands might fall into strangers’ hands or be excessively split, especially when marriage practices follow the rule of virilocal exogamy.Footnote 2 In this initial situation, no “magnet effect” of the Islamic law is thus at work.
In terms of the theory, the opportunity cost of referring to the formal law (the Qurʾan, in this instance) and appealing to the formal judge (the local marabout) is sufficiently high, and the outside options sufficiently unattractive, to prevent any woman from calling erstwhile norms into question. By antagonizing their male relatives, women would lose important social protections they have traditionally enjoyed. Under the custom, women are insured against various contingencies: in the event of separation/divorce, they have the right to return to their father's land, where they are allowed to work and subsist until they find a new husband.Footnote 3 The psychological cost of taking a land dispute to the (formal) religious authority was also perceived to be large insofar as, in the women's view, open disputes between close kin “are to be avoided at all cost.”Footnote 4
Over the last decades, however, as I could observe in the Senegal River Valley,Footnote 5 the value attached by women to their participation in community life—in particular, the value of the customary system of social protection—has fallen. This is a direct result of an increase in female education and a concomitant expansion of non-agricultural employment opportunities. Under such circumstances, as predicted by the theory, the custom has started to evolve. There is no evidence, though, that the custom has gone so far as adopting the Islamic prescription. What we observe instead is an evolving practice of transfers aimed at compensating women for their de facto exclusion from inheritance of a portion of their father's land. The same evolution has been detected in Niger, where women, in recognition of their ownership rights, may receive part of the crop harvested on the family land by their brothers under an arrangement known as aro.Footnote 6
How can we account for Timur Kuran's contention that, in Ottoman Turkey at least, Islamic law—in particular, its specific injunctions regarding inheritance, commercial partnership, and the waqf—was strictly followed?Footnote 7 The situation corresponds to one of the two limit cases of the theory: the custom has vanished because all plaintiffs belonging to the disadvantaged group are willing to refer to the law in the event of dispute. As a consequence, members of the elite are no longer able to activate the custom. A plausible reason is the egalitarian nature of Islamic law, which creates a large constituency of people ready to demand its application, provided that the costs of doing so are not too large. In the urban context that underlies Kuran's argument, not only are the transaction costs of using the associated court system low (distances are shorter, people are more educated, etc.), but also the opportunity cost of parting with traditional social networks declines with the rapid expansion of new income-earning possibilities. In rural areas, by contrast, these costs tend to be all the larger, because outside economic opportunities are limited and a strong patriarchal authority structure makes any appeal against the custom appear as a gesture of intolerable defiance.
The main merit of a sound theory is to clarify a problem by highlighting the underlying mechanisms and to yield predictable outcomes. Such outcomes are often multiple, and their realization depends on specific configurations of parameters that reflect the peculiarities of situations, as illustrated previously by the difference between urban and rural settings. There is thus an obvious need for collaboration between historians and economists (particularly development economists). Unfortunately, this potentially fecund interdisciplinary cooperation is currently much below expectation.
