How large a proportion of the population was Muslim in different periods of history? And how did conversion to Islam unfold across the various regions of the Islamicate world?
The answer most often given invokes Bulliet’s famous conversion curve, a model that has exercised considerable influence but has also been widely misunderstood. For decades, it has frequently been treated as an estimate of the proportion of Muslims in the population as a whole. That, however, was never what Bulliet intended it to measure.
Even so, when scholars and general readers speak of the gradual conversion of populations to Islam, they often have Bulliet’s chronology, or something much like it, in mind. Yet the evidence for conversion proceeding at anything like this pace is far from convincing. Indeed, there are good reasons to believe that the process was considerably slower than is commonly assumed.
Although this particular discussion concerns Muslim Spain, the criticism may be extended to other regions and historical questions where people believe the answer to be settled on the authority of premodern writers. Such accounts are often treated as reliable records of entire societies, despite the fact that their authors had little reliable knowledge of ordinary life beyond the circles in which they moved. They were not conducting fieldwork, nor were they attempting to establish the composition of a population in the modern sense.
What emerges from this discussion is that some familiar assumptions about these societies cannot be sustained. The notion that entire populations converted to Islam within a few centuries, for example, is far from established. In many regions, the process appears to have taken considerably longer. The same caution should be applied to other attempts to reconstruct the composition of historical populations as a whole.
One might, for instance, apply a similar line of inquiry to the familiar distinction between semi-nomadic and settled Kurdish communities. If the criticism of historical sources is valid in the case of mass conversion—and I believe it is—then it may also be worth reconsidering the assumptions upon which our understanding of Kurdish settlement patterns has been built.
Before turning to these questions in greater detail in future articles, I would like to summarise a few of the paper's principal arguments. They provide a useful starting point for the discussion. I would also encourage anyone interested in the subject to read the paper in full.
1. Muslim authors frequently underrepresented, or simply omitted, non-Muslim communities in their accounts. This makes it difficult to reconstruct the religious composition of earlier societies from the written sources alone. What an author chose not to mention may be just as important as what he recorded.
2. Arabization should not be confused with conversion to Islam. An increased interest in learning Arabic, or the adoption of Arabic names, does not in itself demonstrate that a person had become Muslim. Arabic was the language of administration and elite culture across much of the Islamicate world. Conquered populations could therefore adopt Arabic linguistic and cultural practices without abandoning their previous religious traditions.
3. The surviving written evidence tells us remarkably little about the religious lives of ordinary people. This is particularly true of rural communities, which constituted the greater part of the population. Our sources tend instead to reflect the concerns and experiences of rulers and the literate. The lives of the people who made up the greater part of society are much harder to recover.
No comments:
Post a Comment