Dr. George Hawrysch's speech on concordance book launch (08/04/02)

Dr. George Hawrysch's speech on concordance book launch

Following is the text of the talk delivered by Dr. George Hawrysch at the book launch for the "Concordance to the Poetic Works of Taras Shevchenko."

Concordancing can be a dangerous thing. From the time of their discovery in the 1940s until some 10 years ago, the contents of the Dead Sea Scrolls were largely kept concealed, hidden from researchers and general readers alike by the small international group of academics into whose care these documents were initially entrusted. Over a span of some 40 years, only about a third of the Scrolls were made public. Pressure was building to release the remainder of the corpus, but response to it remained minimal. A number of scholarly articles did appear, however, and among them was a concordance of the Scrolls, prepared (but not published) in the late 1950s by John Strugnell.

Now, fast forward to 1991. Working from some 50,000 index cards, Martin Abegg, a student at the Hebrew Union College in Cincinnati, manually entered Strugnell's concordance into a personal computer - and promptly reconstructed the underlying text from which the concordance had been generated. Abegg now held nothing less than a transcription of the Dead Sea Scrolls, with which he went directly to press. A furor ensued; but the really significant outcome was the immediate release of the microfilms of the Scrolls, followed closely by a publication of a facsimile edition of the originals.

Apart from issues of copyright and intellectual property law, this event confronts us with a remarkable reality: one graduate student, working alone, with no funding and using only consumer- grade electronics is capable of doing something that an entire consortium of the finest minds in the field had not even been able to anticipate.

In dismantling a famous and long-standing academic monopoly, Abegg's achievement was less technical than it was cognitive. The pivotal moment in his approach was being able to locate (and ultimately to extract), an additional, hidden text inside the concordance.

This is exactly the case with the Shevchenko concordance as well. The four volumes are now readily available; and of course we are intimately familiar with the corpus of poetry on which they are based. But there exists another, hidden, text here - a text which none of you will ever see: namely, the files from which the work before you today was actually derived.

I do not mean the machine-readable form of the "Kobzar"; its existence is a given. Rather, I am referring to the fully mapped, indexed and tagged rendering of each of the 220-some poems that comprise Shevchenko's principal legacy. Each of these unique, highly structured data objects was placed inside a binary wrapper - that is, compiled into a stand-alone executable program - after which all such programs were let loose to interact with one another. After several hours, the concordance as you see it today stood compiled, automatically cross-checked against the original printed version, and supplemented with a large superset of the statistical material you can find at the end of Volume 4.

This intermediary, hidden text has a further property that goes beyond its lexicographical capabilities. In a sense, it is formally aware of itself, meaning that it has the capacity to respond to changes made to its configuration, as well as to changes in its linguistic environment - but each poem-module also contains information about its relationship to all the other poems, and also about its function within the total "Kobzar."

Thus if Shevchenko writes "na Vkraini" in one place and "v Ukraini" in another (as he in fact does), then the text itself, as I am describing it, can flag this as an instance of an equivalence class, an editorial variant, an error, or some other polymorphism. The reader is then presented with an informationally enhanced literary object, not merely a static text with annotations or embedded pointers, but an actual real-time representation of Shevchenko's creative process, linear or asynchronous.

Why is this important? Well, consider this: the concordance we are launching today is not exactly a concordance of Shevchenko's work; rather it is a concordance of one particular edition of Shevchenko's work. This distinction is significant, because we do not in fact possess Shevchenko's text.

We have the various texts of his editors and publishers, but not the poet's personal version. The holographs are often a succession of unfinalized variants, or are written under erasure; the printed editions Shevchenko himself collaborated on underwent censorship and other pre-production modifications. We have something like a probability space, more than a bounded set of determinate text objects. However, when taken in its entirety, the Shevchenko corpus does contain the poet's private, closed edition of the "Kobzar" - just as Mr. Strugnell's concordance held the text of the Dead Sea Scrolls.

Having looked deeply at the stylometry, at the mathematical structure of Shevchenko's works both individually and across editions, I can tell you that the exclusive, hitherto unrevealed original text really is latent in what we possess, and that it can be extracted by first constructing the kind of intermediate hybrid object - part verbal, part algorithmic - that I have described.

The concordance we are launching here today is already a part of one such hybrid object. I look now to the scholarly community, to proceed as it sees fit, to collapse that probability space and to present us one day with the long-withheld, original, Shevchenko Scrolls.

| Home Page |