Databases as a Research Tool

Body: 

 

 

 

For a long time, databases, due to their technical nature and their complexity, were reserved for the hard sciences and excluded from the social sciences due to misinformation or mistrust.

Venturing individually or as group into this labyrinthine domain, full of risks and difficulties, still seems to be a challenge. The objective of the initiation workshop on Arabic manuscripts in 2010 was to dissipate these ideas, which are widespread in the scholarly world. This is why the members of the Arabic section of the IRHT (Institute for the Research and History of Texts), which organizes a yearly one-day workshop on Arabic manuscripts, proposed as a topic for reflection “databases as a research tool”. 

After a general introduction by Anne-Marie Eddé, director of the IRHT, the speakers presented their own experiences.

First, Christian Müller emphasized the diversity of possible database types and their growing usefulness, as long as one keeps in mind three fundamental “rules”, which he summarized with the acronym U.F.C.:

●   consider the usefulness of the database that one wishes to create to be sure that it does not become a “tomb for data”;

●   examine the feasibility of the projected database; 

  ●   finally, evaluate its cost-effectiveness.

Vanessa Van Renterghem, who created a database as part of her doctoral work on the Bagdadi elite under the Seljuks (5th-6th / 11th-12th C), was the first to describe her experience. She adopted a historiographical and methodological approach in order to emphasize the pertinent questions that can come up in a literary field such as her own. She had to sort through the Arabic sources related to literary life, and in particular the approximately 2000 biographical notes found in Ibn al-Ḥaṭīb al-Baġdādī’s Tārīḫ Baġdād. It very quickly became apparent to her that the treatment and analysis of such a mass of data would be impossible without the help of a database. The user-friendly program FileMaker Pro was used for its technical development. She received help in order to adapt the program to her own needs and, despite the defects of her “home-made” database, this tool unarguably made it possible for her to complete her dissertation.

This personal experience, “almost without any computer science training”, sparked the curiosity of a number of young doctoral students who were present at the workshop. They were then able to discover how databases can be created with more sophisticated technical and logistical means.

The afternoon session opened with a presentation of the database project “with four lives”, as its creators described it. Jacqueline Sublet, Muriel Rouabah and Zouhour Chaabane recounted the history of this international undertaking entitled Onomasticon Arabicum, which goes back to the early days of computer technology, in 1966. The idea of creating a database germinated when the analysis of Arabic biographical sources intensified thanks to the commitment of the many scholars who became involved in the project.

Jacqueline Sublet, a specialist of Arabic onomastics and the originator of this project, remarked that the team had inventoried over 28,000 biographical notes on the lives of eminent figures in various fields (scholars, religious and literary men from all walks of life, men of power, etc. …) from the beginnings of Islam to the 16th C, and had included them in a first database. But the lack of means for recruiting computing specialists and doctoral students in the field of Arabic prosopography was an insurmountable obstacle to the fruition of this project, despite the support of the IRHT and the commitment of the researchers. Nonetheless, the rapid evolution of information technology in recent years made it possible to transfer the data into a database that is now entirely in Arabic using Access; part of the data will be made available online in the near future. For the moment is can be consulted in the Arabic section of the IRHT.

Another recent database was conceived of as part of the innovative European ILM  (Islamic Law Materialized) project, which began in April 2009 and will continue until the end of 2013. Unlike the preceding case, the way the project would precede and its subventions, research group and objectives were settled in advance. The CALD (Comparing Arabic Legal Documents) database was thus created in under a year with the objective of studying and analysis the entire body of Arabic legal acts from the Medieval period, with a comparative perspective on the law and legal practice in the Muslim world.

The three speakers (Christian Müller, Lahcen Daaif and Moez Dridi) wished to demonstrate the efficiency of this database for research, both for comparing documents with one another and for analyzing them individually.

The development of this database opened up a vast field of investigation entirely devoted to legal practice. Each text is apprehended as a complete “discourse” that owes its coherence to its legal reason, and on which its constitutive elements depend. It is thus only a prototype of a coherent institutional practice, containing its own internal logic. This is why the documents were cut into numbered sequences called SQN (Sequence Unity Numbers) with the aim of making it easier to find them and to identify their contents.

After a basic technical presentation, Christian Müller gave the audience an overview of the project’s history and of the analytical phase of the sequences. He demonstrated how documents are dissected by means of “analysis by abstraction” so that, once they are translated into numbers, they can be apprehended differently. Once this dismemberment of the document into various sequential units is complete, the comparative reading of legal acts will be within reach for all researchers thanks to this database.

The workshop led to a very animated debate between participants and audience members, who were very attentive to their arguments. Professor Abdallah Cheikh Moussa, who presided at a round-table discussion on the theme “which scholarly project for which technical solution?”—along with other participants—expressed their mistrust of databases, which make fascinating promises but are sometimes disappointing in terms of the results obtained.  

Even if all were in agreement that creating a database is a difficult mission, requiring precise analysis of scholars’ needs as well as a degree of technical competence, databases remain nonetheless necessary and promising tools for certain types of research.

 

                                                                                                                                                                                                                                                                                                                                                                                                                                                                  Moez DRIDI