B İ L D İ R İ L E R | Page 126

KONGRE BİLDİRİLERİ general administration in which classification by caste as well as religion became paramount. For the gradual production of this knowledge reports were required from districts and provinces in every department of administration to feed the imperial archive in London, and from the 1830s those reports concerning the social and economic condition of the population, including reports generated by land revenue settlements, were published. In the imperial imagination efficient administration was tied to accurate information. Reports at the general level were the complement of records at the local level, and the two levels can be said to have been mediated by manuals designed by those at the upper level for use at the local level but in most cases actually written by Indians in between. The techniques of survey, both general and cartographic, fed into the new science of statistics, whose first learned societies in the 1820s and 1830s were manned by former imperial administrators in India as much as by those concerned with the industrial urbanization of England. European sociology itself developed from this nexus. But for a comparable construction of knowledge about Arab, Balkan or any other lands under Ottoman rule, where is one to look? In the newspaper Suriye or the Salnames of 1868 to 1900? Modern Turkish sociology appears not to have been generated from within Ottoman administrative knowledge but imported from without. In particular the Durkheimian tradition fitted well with the construction of a Turkish national sociology. As a technology of rule, censuses provide a good comparative illustration of the different imperial regimes of knowledge under Ottoman rule in Bilad-i Sham and British rule in India. Mention of the term Census of India elicits uniform admiration by demographers and sociologists today, for its scale – a synchronic census of all of India dates from 1881 –, for the categories under which the population was enumerated, and for the abstract level for which information was given (I am speaking here of the censuses conducted during British rule). The decennial census reports of each province were an important source of knowledge about religion and caste after the 1880s. But if one looks for local uses of the census a different impression may be gained, the actual census returns having been destroyed soon after enumeration and no record of households or the names of householders having been retained for local administrative use. Before the idea of an all-India census had taken root in the second half of the nineteenth century, however, a household census (khana shumari) had been conducted as part of early land revenue settlements, including the 1853 settlement of Ludhiana district where for the first and only time maps of the residential sites of villages and towns were prepared which remain in use to this day. Household censuses were superseded by the all-India census for which maps of residential sites, if prepared for enumerators, were not preserved. The value of all-India statistics overrode any local value. Yet what prevented the maintenance of a local register of village residents by village headmen for purposes of civil registration, as was done under Ottoman rule at the beginning of the twentieth century with the civil register or nüfus? It seems that information about Indian subjects of British rule at an abstract, gross level was more important than the provision of a means by which the vital statistics of a population could be kept locally up to date. The relation between citizenship and subjecthood varies between empires. As an instrument of rule, the census became divorced from land registration under British rule in India. But towards the end of Ottoman rule any new entry in a tapu register had to have a reference both to the nüfus and to the vergi registers, showing that coherence in local administration was more important than feeding the desire for knowledge about subjects and the new social science of demography. Yet if knowledge of populations of Arab lands under Ottoman rule was not generally disseminated, there was surely great continuity in the knowledge of experts in fiqh through the whole period of Ottoman rule. Regarding agrarian conditions it is possible to trace subtle changes in the understanding of such terms as sahib-ı arz or hakk-ı karar over the centuries (see Governing property, Chapters 2 and 3). Agrarian legislation under the Tanzimat in the nineteenth century built upon this legal tradition. 112 Tapu ve Kadastro Genel Müdürlüğü Law. There was also great continuity in the idea of tapu registration, wh