Cybernetica Mesopotamica


Giorgio Buccellati




Grammar Urkesh

OB letters
MA laws

Ebla pers. names
OB royal letters
The backdrop of history
The IIMAS projects
An integrated system's approach
A grammar of space
A universe of forms
The sound of silence
The higher nodes

The backdrop of history

     Five thousand years ago, writing was introduced in Mesopotamia.
     Just over forty years ago, functional computers were introduced in America.
     The stark juxtaposition of these two facts, gives a new dimension to the ageworn dictum that the “Land of the two rivers” is the cradle of our civilization.
     The first scribes provided a whole new crystallization of mental processes, the crystallization which gave to internal, logical brain functions an outer existence, disembodied from the brain. Memory acquired an extrasomatic existence of its own, the mental world of man was accessible as an object apart from the mind. The introduction of writing was in fact so momentous that we mark by it the very beginning of history.
     The introduction of the computer is just as momentous. In comparison to writing, electronic data processing may be described as the extra-somatic extension of active logical brain functions. Data are now not only stored, they are manipulated according to rules or programs. We have projected to the outside one more of our inner functions, and we look at it crystallized in its new, awesome form.
     Think of another related technology, printing. It, too, caused a profound realignment of established conceptual frameworks. A significant aspect of the printing press revolution, not generally appreciated, was the almost simultaneous introduction of the Encyclopaedia and of the scholarly journal. They served two polar counterparts of the same set of needs. On the one hand, the need for a broad generalized control of all the knowable: hence the Encyclopaedia. On the other hand, the need for an unlimited capillary spreading of specialized information: hence the scholarly journal.
      Like the Mesopotamians with writing, or the scientists of the Enlightment with printing, we, too, must live up to the challenge of the new technology. As we bend it to serve a new conceptual world, we may help make the transition possible from history to – post-history!
Back to top

The IIMAS projects

     Since its inception, IIMAS has been intensely involved in the application of electronic-based analysis to the data from the ancient Near East. Most of the analysis served as the needs of other projects, particularly with regard to philology and linguistics, art and archaeology, religion and history. This research has been carried out within the framework of a series called Cybernetica Mesopotamica.
     In addition, we pioneered in the field of mirror digital versions of standard publications.
     The paragraphs that follow present a brief description of the conceptual range of the projects. Some of these concepts are further elaborated elsewhere, especially in the current site on Urkesh.
Back to top

An integrated system's approach

     The Mesopotamians' is a broken tradition. There are no living carriers of their values. In this sense, theirs is a “dead” civilization. But only in this sense. We see it come to life each time we understand a new fragment of its broad range of achievements.

     Computerization of the data can much more easily bring these fragments together than any conventional documentary search. In a way, the computer comes close to serving the same purpose that a living informant does for a contemporary culture. Electronic access to the data combines into one the polarity of the Encyclopaedia and the scholarly journal: we can combine extreme generalization and extreme specialization in one and the same tool. Our data banks can encompass the totality of the evidence in such a way as to allow its being constantly reassembled in response to the researcher's questions.
     Our work on computer aided analysis of Mesopotamian materials goes back to 1967 – an early date for any computer project in the Humanities. In the course of time, we have developed a highly integrated systems' approach to the data. We have developed a comprehensive categorization system which brings to light the inner filaments of the civilization, the web of relationships which held it together then and by which we know it now.
     These new “grammars” are applicable to the data in such a way that massive data storage does not become something like the inarticulate archaeological mounds we excavated in the first place. For it is ironically quite possible to re-bury the data inside the computer if the retrieval channels are superficial and few. Instead, our effort has been to give a highly structured configuration to the input, in such a way that all the various components may hold together more effectively and the resulting yield may be all the more powerful.
Back to top

A grammar of space: the computer in field archaeology

     The fullest utilization of the computer in field archaeology entails much more than just a typological inventory of artifacts. We must understand the dynamic processes of stratigraphy and deposition: how do things relate in the ground? how can we understand space relationships in terms of time sequences? We set out to build a categorization system which would allow us to do just that: the challenge was to construct a grammar of space so that space might be read as time. What we find in the ground are juxtapositions, and the first task is to arrive at a precise morphology of contact, intersection, etc. The sequence implied by the stratified context is the syntax, the process which we assume to be at the origin of spatial relationships.
     The model which we have developed is in itself a powerful explanatory tool, a new conceptual handle on the complex reality of stratification. On the practical level, the application of this model to a computer in the field means a dramatic alteration of decision making patterns. We feel that our model is successful because it is (1) sufficiently differentiated to provide our analysis with real depth; (2) sufficiently agile to allow efficient and relevant retrieval; and (3) sufficiently simple (in terms of both hard- and software) to make it truly operational in even a remote field situation (such as the Syrian desert).
     Our “grammar of space” has ta number of major components. The digital/graphic representation of the data retains the full precision of the measurements while rendering them analogically. Particularly important is the conceptual categorization, with which various programs operate to produce a variety outputs, presented in browser format. Our “grammar” of space generates actual “sentences” which describe all possible relationships of elements in the ground: “the pit cuts the wall,” “this floor builds up against that wall,” “the wall, in turn, cuts into an earlier fill,” and so on. The computer reads stratigraphy in normal human prose: the fine organism of culture, deposited as it was in the ground, is now translated into the logic of a linguistic system.
Back to top

A universe of forms

     The “grammar of space” allows us to define things as we find them in the ground. Once freed from the grip of the earth, they become part of a rich “dictionary of forms.” Whether architectural buildings or simple pots, works of art or plain tools, we have in the physical remains the fossil trace of a once living culture.
     Just as there is a meaning to words, so there is a meaning to artifacts – or a “semiotics,” as we may say more precisely today. But how can we “read” artifacts? The criteria for the decypherment of iconographic and formal rules of artifactual evidence may be viewed as similar to those of the linguistic evidence. The technical term is “distributional analysis.” It means that we observe how elements are distributed or arranged, and from the patterning that we perceive in such distribution or arrangement we deduce as to a value or meaning which might escape us if we see the elements in isolation or – “un-distributed."
     We can begin to derive some firm degree of assurance from a search through our computerized data bases. Such assurance derives from the vast sample we have at our disposal, coupled with the fine degree of differentiation with which we have ordered it. We can begin to say that something does not co-occur with something else, with the assurance (almost) of an ancient user. Such negative statements are virtually impossible with any non-digital type of retrieval system, except for the most obvious (hence, least meaningful) types of associations.
     For instance: on a given cylinder seal, someone is holding a cup and his arm is outstrecthed. How else are cups used? Are there any examples where someone who holds a cup is standing? Is anyone offering a cup to someone else? In other words: what is the “distribution” of a cup in these representations? As it happens, only seated persons hold a cup, and they only hold it for themselves. “Only,” “only": this is the key word. Our computer scans all the available evidence and makes the association of the element “cup” with the elements “standing,” “seated,” etc. and declares what the possible co-occurrences are. From these co-occurrences, we can make well-founded deductions about the distributional range, and therefore the meaning, of each element in question.
     Such a typological use of the computer is by now traditional in archaeology. The value of our work in this area is the range of artifactual classes involved (cylinder seals, figurines, pottery vessels) and the fine degree of typological classification which is applied to them. The result is that we truly lend a new life to the artifactual world: we can “read” the artifacts, we can “talk” to them because we have lent them a voice with which they may respond to us.
Back to top

The sound of silence

     The written texts are the most explicit signature which the ancient Mesopotamians have placed on their history. Through them, we learn their names, the names they gave to the world around them, and even their inner thoughts and feelings. For over a century anda half now, both the script and the languages of Mesopotamia have been “decyphered,” but there are still many questions open. For one thing, there are matters of nuance and style; for another, there are new bodies of evidence and to some extent new languages which are being discovered, such as the texts of Ebla in the last two decades.
     We have by now millions of sign occurrences which are stored in the data banks of our projects, including texts from Babylon and Mari, Ebla and Ugarit, Assur, Nuzi and several other sites. The same pattern of distributional analysis which obtains for the artifactual evidence, lies at the basis of our understanding of the texts. The very concept of grammar is originally a linguistic concept, and we have refined it using the insight of structural and generative linguistic theory as applicable especially to Akkadian. Hence, our encoding manual turned out to be a new venture in grammatical studies. On the one hand, the systemic structure perceived in the language was controlled and verified by the inherent logic of data processing. On the other hand, the substantive arguments could be proven or disproven on the strength of the data base which was being experimentally codified as the system itself grew.
     Short of speech simulation, we are teaching the computer to speak Akkadian, the main Semitic language of Mesopotamia, in two ways. One is interactive: the computer generates the morphological analysis of forms provided as input. The main function of this program is paedagogical: building on its own “understanding” of grammatical forms, the computer quizzes the user at a variety of different levels. Since the forms are generated entirely from rules given as program statements (and not from tables), a study of the programs is in itself a very compact study of the linguistic code of Akkadian. We view this as an important development within the context of the growing trend toward genuine computer literacy.
     The other linguistic use of the computer is to provide a fully articulated documentary basis for specific corpora of Mesopotamian texts. The most intriguing group of texts are those of Ebla, the major royal archive of a third millennium city in Syria. Our project serves as the official channel for the computerization of these texts in collaboration with the University of Rome and within the framework of an international Committee which is responsible for the research and publication of the documents from this archive.
Back to top

The higher nodes

     The processing of the data as carried out so far for publication has yielded extensive sorts based on set clusters of variables. They are basically concordances and indices, built on a totally new categorization system, and with a number of frequency computations on which some preliminary statistical analysis can be performed. All of these outputs are viewed as interconnected components of a larger whole, and their variety adds up to an overall picture that is indeed larger than the sum of its parts.
     Projected for the immediate future is a more explicit analytical effort, where the interconnections are made explicit, and their significance is properly assessed. The distinction made in our publication prospectus between categorization and analysis reflects this basic understanding of our research goals. Categorization deals with the lower nodes of the binary tree which may be used to represent the Mesopotamian cultural system. Analysis deals with the higher nodes in the sense that it establishes multiple connections among all possible branches in the system. Having properly categorized the data in the first place, and having a vast data base accessible, suitable programming will allow us to see distributional patterns of a type which could not even be imagined heretofore.

     As both a new categorization system and a new scholarly attitude toward the documentary effort come to prevail, the conceptual impact of the new technology will indeed go beyond the level of simply performing traditional operations with greater ease: it will affect the very scaffolding and the structure of research. NISABA was the Mesopotamian goddess of writing under whose tutelage the Mesopotamian scribes wrought the intellectual revolution which projected our inner brain functions onto the extrasomatic medium of clay. Some 5000 years later, time has come to go from clay to chips and boards. Nisaba will not mind: we are recovering the past evidence of her art better than her own scribes ever could!