Giorgio Buccellati

Digital analysis

Home

The conceptual dimension
The data
Cybernetica Mesopotamica
Programming
Programs
Websites
The masthead's image


     First season in Mozan, November 1984

Introduction: The conceptual dimension

     I was exposed to computers the very first year I arrived to UCLA, in 1966, and what followed was a long itinerary that matched the many rapid changes in hard- and software. The excitement of constant innovation was tempered by the frustration of not being able to fully complete a project before the techniques had become obsolete.
     But, in the process, I developed a strong appreciation for the value of method over and above technique. I came to see more and more how the basic intuitions relating to digital analysis rested on principles that lasted even while the technique was becoming obsolete. This conceptual dimension of my research was one of the great benefits I derived from the long experience. Let me refer to two points.
     The first is that a highly integrated, systemic approach to categorization transcends the level of tagging and coding, and creates a proper grammar. This affords us a powerful control of the data through its otherwise almost inextricable maze of connections. The distributional analysis that is possible as a result gives us a unique handle on the life of a culture to which we have no bridge of direct experience: a broken tradition.
     The second point of interest is the notion of digital thought. Based on a sharp distinction between electronic and digital scholarship, this notion proposes a mode of relating to the art of the argument in ways that are both perfectly in line, and yet at variance, with the received wisdom of the humanities and the social sciences.
     And yet there was, of course, a direct and constant involvement with the realia, and the consequent effort at constructing appropriate organisms for inclusion of the data and mechanisms for their fruition. This I have done in the areas of linguistics and archaeology, within a larger framework that I called from the beginning "Cybernetica Mesopotamica," which has now found its home within a set of interrelated websites.
Back to top

The data

     At the beginning, my work was directed at the textual material, and as far back as 1968 I received one of the first major grants for computer applications in the humanities: the Old Babylonian Linguistic Analysis Project (OBLAP). It was aimed at the graphemic and morphemic study of Old Babylonian letters, and from it several other projects dealing with cuneiform texts originated.
     Following the start of my excavations at Terqa in eastern Syria I began to plan for digital applications in archaeology as well. In 1978 we brought the first computer to be used in a field situation in Syria, and a long confrontation began not just with the practice, but also with the theory of digital analysis. The attendant conceptualization resulted in a determined effort at establishing a grammar of stratigraphy and at developing a new kind of digital publication, that was as much in line with the database approach as with the interpretive narrative.
     Thus, along with the conceptual effort, there developed a vast program aimed at the data, both linguistic and archaeological. The continuous concern for digitizing existing texts (already published) and for documenting the entire flow of new material (from the excavations) resulted in a vast data base, that could not however be effectively used until the World Wide Web came into existence. I deal with the details of this program and its history in the Cybernetica Mesopotamica website.
Back to top

Cybernetica Mesopotamica

     The concept of an overriding project by this name was a central interest of mine from the beginning. It was to be more than a repository, however rich. It was meant to serve the more ambitious intellectual goal of providing a coherent digital framework where data could be given new life, as it were. A grammatical approach was its basic presupposition, and its eventual fruition would be in the ability to develop a dialog with the data similar to what we have with cultures for which living carriers are still extant. The digital dimension would bring us as close as possible to the ideal of having a living informant for a "dead" culture, the ideal of appropriating the original competence.
     Cybernetica Mesopotamica was then as much a theoretical construct as it was a practical data container. But the strong theoretical bent of the project meant that its full realization would exact a longer time span than I would ever have wished. One of the recurrent obstacles was the need of cross adaptation between grammar and data, and the attendant need to explain this to the staff that was working on the various aspects of the project. The positive side of this was that the growing architecture of the system maintained a rigorous inner coherence, in spite of the constant need to add new data not in an ad hoc fashion, but through a faithful respect of the inherent grammar. The downside was the frustration of not being able to enjoy full access to the data.
Back to top

Programming

     In the first phase, when I was working primarily on textual data, I relied exclusively on professional programmers. The one who most closely identified with my goals, and produced all the initial programs, running on a mainframe computer, was John Settles. Through my close collaboration with him, I developed a sense for the nature of programming, and when it came to start work on the archaeological material, I decided that I would get more directly involved. My intention was to write only the basic code, that could then be properly implemented by a professional programmer. But two factors emerged that changed my perspective.
     The first was that the initial reason for undertaking the programmign on my own turned out to be more than just initial. My aim was to set up the basic parameters for the overall project. But I soon realized that the interaction between programming and grammar was so close as to make it impossible to work on one without working on the other. And so I kept on going.
     The second reason was that the programs actually ran quite well. At first it was difficult to let others run the programs on a routine basis: the hardware, too, was rudimentary (as shown in the photo in the masthead of this page), and limited. The only one who for several years was able to assist was my son, Federico. But as the hardware became more affordable, so that every staff member, or at least every unit, could in fact have a computer, everyone took to running the programs, with relatively little difficulty.
Back to top

Programs

     As of 2012, the programs in archaeology that I have written are still the ones we use regularly in processing the data. They were written in various versions of BASIC, which proved to be more than adequate for handling even large quantities of string data. The rudimentary, and yet efficient, dimension of the early stages is documented by the paper publication of one of the early programs, that I wrote together with Olivier Rouault (the plotter to which the article refers is the one shown in the masthead photograph on this page).
     What I consider an important implementation of this program is the publication of the find spot of the tablets of the archive of Puzurum (TPR 10), which may be viewed as a rough preliminary attempt to deal with the problems that are now addressed by GIS systems. Interestingly, I also used the program to produce the statistics that I published in the introduction to the volume on the Middle Assyrian Laws by Saporetti (which were produced, in their printed format, on the same plotter). Understandably, no other publication came of the extensive work I did on programs over the years. I have only included the description of the various programs within the digital book on Grammar found in the Urkesh website.
     Clearly, these programs are now obsolete in terms of the changed environment that supports a DOS system only with great difficulty. Accordingly, I am planning to do now what was intended from the beginning, namely to work with a professional programmer in order to adapt the programs to the new platforms, within the scope of the system that has by now been firmly established on solid grounds.
Back to top

Websites

     The browser approach provides a perfect venue for the presentation of the results from the various projects I have undertaken over the years. Accordingly, I have desigend and produced a number of websites where these projects are now housed.
Back to top

The masthead's image

     In Terqa we had been operating with bulky CP/M computers. In 1983 I purchased our first notebook computer, a Tandy Model 100, which had just become available and which we took to the field for our first season in Mozan. Two units are shown in the masthead photo, along with a line printer (on the right) and a multi-pen plotter (on the left).
     In the photo (taken in the house we were renting in the town of Amuda) I am shown at the keyboard: the candle is symbolic of the fact that electricity was still irregular, and so the battery option for the computer (not for printer or plotter) was indispensable.

Back to top