KONGRE BİLDİRİLERİ
the new cadastral surveys of the British Mandate in the 1930s. Although there may be a historical
interest for present villagers of northern Jordan in discovering who held what shares in land during
the last half-century of Ottoman rule, a continuing legal interest would be unlikely at the present day
since the cadastral surveys of the 1930s are the base-line for all subsequent transactions in land. There
are therefore good grounds for preserving the older records of tapu registration in the archives of the
central Department of Lands and Surveys in `Amman (da’irat al-aradi wa-‘l-masaha), to where they
were moved from the regional offices in the early 1990s. Access to these archives is open to bona fides
research scholars. In parentheses it needs to be said that the scale of archiving here is slight, the entire
collection of tapu registers for Jordan being contained in a single medium-sized room. The collection
of surviving registers is, moreover, far from complete, a reason that the originals in Ankara remain of
inestimable value.
In the case of land records that are still live, even if seldom consulted, the question of whether to
archive older records may be more difficult, if only because of the volume of records involved. For
instance, under the system of land registration in the Punjab during British rule of undivided India
before 1947, settlements of rights to land and of liability to land revenue were supposed to be revised
every thirty years. For Ludhiana district in whose records room I did research from 1973 to 1977,
the founding settlement was completed in 1853, a revised settlement was done in 1882, and a further
revision was done in 1909-10 (no further revision being done in the 1940s on account of the second
world war). Records are preserved not only of these Settlements but also of the four-year updates from
1882. The entire set of bundles of records of a single village may occupy many shelves. Although
public inspection of the records of a village usually goes no further back than the last settlement, the
older records are sometimes consulted, for instance to trace where an ancestor lived in the case of
maps of the residential sites of villages or of Ludhiana town which were prepared in 1853 and never
subsequently revised. Moreover some records of the 1882 settlement are definitive: for instance the
genealogy of proprietors (shajra nasab-i malikan) had not been prepared in 1853 and at the 1909-10
settlement the revised genealogies only took the form of updating the genealogies of 1882. Should
the records of at least the first settlement of 1853 be transferred to a central archive although they are
still live? I would argue not, firstly because of the volume of records involved: a district in the Punjab
contains a thousand or so villages, meaning that a central archive would have to cater for the records of
many thousands of villages from all districts; secondly the revenue records room of a district already
has viable procedures both for inspecting entries in a record by members of the public and for regular
checking of records by the administration for damage; and thirdly the revenue records room of a
district is in a sense already an archive for early settlements, since public inspection of the most recent
papers can be done on the second copy preserved in the revenue records room of a sub-district (tahsil)
not on the first copy preserved in the district revenue records room. For research scholars access to a
revenue records room of a district in India is governed by the normal rules of research access, which
in the case of non-Indians has to be authorized by a central ministry. Thus I was given permission to
consult district records on condition that I confined my attention to the period before 1947.
There may be an argument for digitizing the records, including early records, partly for reasons
of space and partly to insure the preservation of their content. But that is a different consideration.
In the case of the first copy of tapu registers of Arab lands done during the Tanzimat period at
the end of the nineteenth century, which are preserved in Ankara, I would plead for some access to
research scholars, firstly because these records are no longer legally live, tapu registration having
been superseded in at least Jordan, Lebanon and Syria by surveys of the British and French mandated
governments, yet second copies of tapu registers preserved in those countries are far from complete.
Secondly, the benefits of extending scholarly access are more general than what concern the tracing
of particular individual cases, as I indicated earlier. I would now like to say something of these
more general benefits by way of the image of records as a window that can be looked through both
110
Tapu ve Kadastro Genel Müdürlüğü
outwards onto socio-economic facts on the ground, and inwards onto the form of rule, or the system
of administration, classification, knowledge and law of the regime concerned. I shall concentrate on
the view inwards by way of comparing briefly land registration of the late Ottoman empire in Bilad-i
Sham and of the British empire in India.
The view inwards.
Administration. A superficial comparison of the records prepared at the initial registration of
land in the two empires might see only an insufficiency of records in the one against an over-abundance
in the other. With the image in mind of annual papers in the Punjab as a photograph, the colonial
regime of the British in India sought to lay bare and record for possible use in courts of law every fact
of relevance to land administration. The basic record of rights of a village included a register of both
proprietors and tenants, a field map, a genealogy of proprietors purportedly from the foundation of the
village perhaps many centuries previously, and a constitutional contract between proprietors and the
government concerning the way relations within a village between proprietors and all kinds of nonproprietors, from tenants and labourers to carpenters and weavers, were to be managed. This was for
every one of tens of thousands of villages in the province of Punjab. It involved an army of surveyors
intruding into the life of a village to collect information, and fitting that information into a grid of
categories that might be used in a court of law. The purpose was to preserve custom on top of the more
obvious imperial aims of extracting revenue and creating a marketable property in land. But setting
the venue for the preservation of custom in courts of law had the unintended effect of freezing custom,
indeed of changing the nature of custom and of its relation to law. And to maintain the records as a
continuously updated “photograph of the actual state of the community” an elaborate administrative
structure had to be created within a district, with a designated official (patwari) responsible for keeping
the registers of three or four neighbouring villages up to date. In such an over-determined system the
records became clogged over time with detail, making their reading difficult except for those who
maintained them on the ground. Their statistical abstraction too became an exercise in maintaining
form at the cost of utility. A Foucauldian reading of the effects of such a system would see power
placed differently from what had been intended.
The form of the first yoklama tapu register of a village in Bilad-i Sham was by contrast minimal.
Of the first four entries in the record of the village of Hawwara in what is now north Jordan in 1876,
two had the father’s name but no family name (one being simply Mahmud al-Ahmad, the other more
readily identifiable as `Abdullah son of Ahmad al-Musa), one had only the family name without the
father’s name (`Abd al-Rahim Barakat), and only one had both father’s name and family name (`Abd
al-Jalil son of As`ad Shuha). For anyone other than someone of the village to identify those named
would have been difficult. Such under-determination left the custom of organizing the distribution of
land for cultivation within the village much as it was, with all power left to the mukhtar and the village
council (al-hai’at al-ikhtiyariyya). Yet reven ue was extracted and a market in land was created.
Classification. The grid of categories into which information collected at land revenue
settlements in British India was squeezed included not only different kinds of proprietors and tenants,
whose relations were governed by laws enacted for each province, but also of religion and caste which
became the governing identities for all colonial understanding of India, the glue that held the whole
system of rule together. It is true that nomads were sometimes differentiated in tapu registers under
Ottoman rule in Bilad-i Sham but no other general category distinguished the residents of villages. The
understanding of Jordanian society as in general tribal is rather an invention of the British Mandate.
Knowledge. It is sometimes claimed that Ottoman rule was both imperial and colonial in relation
to Arab lands. But in comparing regimes of knowledge, what was colonial about Ottoman rule?
British rule in India was premised from the beginning of its administration of territory in Bengal
in the 1760s on the construction of knowledge about India, both for purposes of law, for which the
categories of Muslim and Hindu overrode all other distinctions in cases of personal law, and for
Arşiv Dairesi Başkanlığı
111