Now that I'm officially "Emeritus," although not fully "retired," I've been asked to write an autobiographical retrospective on my involvement with computers and textual studies, which began in earnest around 1970. I was 36 years old at the time and an associate professor with tenure at the University of Pennsylvania. Before that I knew, somewhat vaguely, about the use of computers in Father Busa's Aquinas project (begun in 1949!)[1] because my office mate, a Patristics scholar named Robert Evans, was taking a computing course and occasionally mentioned such things as he shuffled his stacks of IBM punched cards. This would have been in the late 1960s (he died unexpectedly in 1974). Even before that, as an assistant lecturer at the University of Manchester (1961-63), I had also been intrigued by reports of the use of computers in analyzing the problems of Pauline authorship (A. Q. Morton, 1961-64).[2] I was a good candidate for such temptations, having done quite well in math and science in high school, even winning the Renselaer Polytechnic Institute Achievement Award my senior year. But it took much more to stir my latent curiosity to action. (Did you know that the University of Manchester and the University of Pennsylvania were rivals for the distinction of having developed the first operational computer?[3] Neither did I. Retrospection can sometimes be cool.)

When, at a meeting of the steering committee of the newly formed International Organization for Septuagint and Cognate Studies (IOSCS) in 1970 or soon thereafter, the question of creating a new lexicon of "Septuagint" Greek was discussed, I offered the opinion that this would be a good project for the use of computers. Risky move. The presiding officer, John Wevers, immediately commissioned me to look into the possibilities, which I gradually was able to do with the help of then graduate students Jack (John) Abercrombie (now deceased) and Bill Adler (now at North Carolina State University). Emanuel Tov (Hebrew University) was appointed editor of the desired lexicon, and I began to work with him on creating concordance type tools to facilitate the lexical work. He wasn't much "into" computers at the time and perhaps thought it odd that I showed little interest in producing hard-copy tools in the project, since I felt that electronic access would be far superior to traditional static representations. Somewhere along the line he also came to agree with that vision—or got too busy with the other tools—and abandoned the idea of editing a conventional lexicon. Fortunately Johan Lust and his Belgian team picked up that ball.[4]

By 1978, a small feasibility grant from the National Endowment for the Humanities (NEH) was obtained to explore the possibility of getting the Greek Jewish scriptural materials (LXX/OG) into electronic form, with an eye to creating "Computer Assisted Tools for Septuagint Studies" (CATSS). Jack Abercrombie was sent on a trip to the sites in the western world where we knew or suspected that relevant work was underway, most notably Tübingen (Wilhelm Ott) and Oxford (Susan Hockey), with other stops along the way. Meanwhile, David Packard, using Hewlett-Packard equipment, developed the IBYCUS mini-computer system and we were fortunate enough to be able to obtain the necessary hardware (used) and to have several terminals installed in the Religious Studies offices in Duhring Wing at the University of Pennsylvania. By 1980, we were up and running through a dedicated telephone line connection to the Princeton Institute for Advanced Studies, where an IBYCUS central processor was available for access—its primary function at Princeton was publication of an academic journal, so its potential computer power was seldom taxed. As our work became more computer intensive, however, our demands on the Princeton processor became excessive (they would phone us to investigate why processing had slowed to a crawl, and we would back off a bit), and we finally were able to obtain our own mini-computer and add a few more terminals and a larger hard drive. By the late 1980s, thanks to continued NEH funding as well as support from the Packard Foundation and other smaller grants—not to mention the generous contributions in time and technical expertise from David Packard and his associates—the CATSS offices included a Kurzweil scanner, 9 track tape drive, 400 megabyte hard disk drive, and a number of IBYCUS terminals attached to the H-P mini-processor. We had become a real center!

The CATSS project proposed three goals: (1) morphological analysis of the Greek Jewish scripture texts, starting with the Rahlfs text that had already been encoded for computer use by the Thesaurus Linguae Grecae (TLG) project at the University of California at Irvine, but with a view to updating that text in accord with the ongoing work of the Göttingen Septuagint Project in Germany; (2) parallel alignment of the Greek with the available Hebrew BHS text, encoded under the direction of Van Dyke Parunak at the University of Michigan with help from Richard Whitaker and support from the Packard Foundation; and (3) encoding all published variants to the Greek texts. Over the years, a half a million dollars or more from federal and private sources went into supporting the projects. Part of my task, as principal investigator at Penn and co-director with Emanuel Tov, was to manage the purse strings and the grant proposals/reports—something that always seemed to me somewhat counterproductive, since it kept me from working more in the trenches, for which I was better prepared. Indeed, I thought the textual variant module could probably be completed in a year or two of concentrated labor; that was 25 years ago, and I'm hoping to get back to it in current retirement mode!

We concentrated on the Morph module at the University of Pennsylvania, with the help of the automatic analysis program created by David Packard to run on the big IBM mainframe at UCLA and installed by him personally on the similar machine here at Penn. It might still be running at the University of Manchester, where it was also installed some years later. Many graduate students learned the intricacies of Greek grammar (and computer editing) while correcting and/or verifying the computer output. Bill Adler was an early associate in directing these activities, along with Benjamin Wright (now at Lehigh University) and Ted Bergren (now at University of Richmond). The initial public release of the resulting "Morph" files (version 1.0) came late in 1987, with the appearance of the "Demonstration" CD-ROM #1 from the Packard Humanities Institute (PHI) in Los Altos, California. A corrected version of the Morph materials appeared around 1994 (version 2.0) under Ted Bergren's direction, and further corrections and modifications have been made subsequently under the direction of Bernard Taylor (now at Loma Linda University), a relative latecomer to the project staff. With such competent staff members in place, I never got personally involved in the Morph nitty-gritty, beyond making a policy decision here or there where the route wasn't obvious. A "Morph Bible" was put together, offering guidance to those who labored. Not everyone was equally talented at such detailed work, verifying and/or correcting the automatic output, so some things slipped through the cracks. But when David Packard asked for CATSS materials to include on the projected 1987 PHI CD-ROM, Morph seemed ready for a test drive.

The parallel Hebrew // Greek module was developed by Emanuel Tov and his staff at the Hebrew University in Jerusalem, using a suite of programs written by Jack Abercrombie. One of the coded fields in the parallel text files ("column B") contained the conjectured retroversion into Hebrew of the Greek translation where that seemed to provide a probable explanation for the divergence. In one of our requests for renewal (or continuance) of funding with the NEH, this feature met with sufficient objection on the part of at least one reviewer that we were instructed to drop it from the published data. We did, of course, although somehow over the years it seems to have reappeared in some versions. We considered it to be a strange (unreasonable might be a truer reflection) request, given the context of our project. Again, the first public release of this material came in association with the 1987 PHI/CCAT CD-ROM, although only part of the parallel text made it onto that disk (version 1.0). Currently, an updated and corrected version (3.0?) is being prepared by Frank Polak (now at Tel Aviv University), a member of Tov's original team.

The third goal, encoding textual variants, proved to be the most difficult (or perhaps, the least prioritized) and is still underway, under my direction at Penn. We started with relatively easy pilot projects—the little book of Ruth, encoded by direct keyboarding from the Cambridge larger Septuagint by Benjamin Wright and myself back around 1985,[5] and the Greek Minor Prophets, with the Göttingen apparatus scanned in at the Oxford Centre (before we had our own scanner). We developed a rather cumbersome format, one Greek word per line (computer record), at a time when technology did not easily enable viewing and editing of blocks of material. The format remains valuable for its flexibility—it can easily be transformed into a "data base" type file—but the learning curve and the drop-out rate for student workers is high. Not everyone gets excited about this sort of detailed work. It's difficult for me to understand why! Significant funding has probably been wasted, and much time consumed, on inefficient production with inadequate results. But we press ahead. Scanning technology has never really caught up with this type of project, so easy solutions are still evasive.

Various peripheral developments accompanied the progress, and funding success, of the CATSS project at the University of Pennsylvania. Around 1984, the "Facility for Computer Analysis of Texts" (FCAT, later reorganized as a Center, thus CCAT) was founded at the University of Pennsylvania, with Jack Abercrombie as technical director and the associated CATSS project as one of the main items under this organizational umbrella. Because the various relationships into which we had been drawn through the CATSS project made it possible for us not only to develop the CATSS modules, but to collect and distribute related materials (e.g., English Bible Versions, Latin Vulgate, Aramaic Targums, Syriac samples, New Testament text and tools), we became a repository for such materials and soon were deeply involved in marketing things not only to interested individuals, but also to secondary distributors, some of whom were in the "for profit" side of things. With the acquisition of our own scanner, we also accepted jobs that would produce some additional income to fund the operation. Some of this became troublesome, and at one point the University director of computing threatened to close down CCAT because of complaints about unfilled requests, etc. The arrival of the Internet came none too soon for that aspect of CCAT/CATSS activities, although of course it brought with it other problems!

The main watershed in the history of CCAT/CATSS came with the request from David Packard for materials from us to include on the "PHI Demonstration CD ROM #1," which was mastered late in 1987. We stood near the beginning of CD-ROM development, which promised to revolutionize data storage and transmission. The label on the disk identifies two groups of material: "(1) PHI Latin Texts (partially corrected)," and "(2) CCAT Biblical Materials (University of Pennsylvania)." At the bottom of the printed disk label is the warning, "many of the texts stored on this CD ROM are in a partially corrected state, which will be improved in future releases." True and false. The same range of CCAT/CATSS material never saw a future release from PHI, although selected items were included with updated Latin Texts in 1991 on PHI CD ROM #5 (Bible Versions, Coptic Texts) and on #5.3 (Bible Versions). My own role in these developments included a couple of weeks in July 1987 at the PHI, when I drove an older Volvo borrowed from David Packard (my skill with mechanical things came in handy) and spent long hours at PHI getting things ready for the CD-ROM. It was a potpourri from the CCAT side—various biblical versions or samples of such that we had collected (English, Greek, Latin, Hebrew, Aramaic, Syriac, Coptic), different ways of setting up parallel texts and textual variants, and the like. Much of this material had been in circulation from CCAT on diskettes—at first, true 5.25 inch "floppies" and, as technology progressed, the new 3.5 inch firmer diskettes. Several generations of graduate students had helped support themselves by filling orders for such texts, which was both a blessing (student survival!) and a curse (bookkeeping, queries and complaints, etc.). The CD-ROM technology promised to make all this easier, especially since licensing and distribution was being handled by PHI, not by us. But we still get occasional requests.

Another interesting development involved the presence of CCAT/CATSS staff and equipment at the annual SBL meetings, in conjunction with the Computer Assisted Research Group (CARG). For several years, Jackie Pastis (who entered our PhD program in 1984; currently at La Salle University) coordinated the exhibits at the SBL CARG room, and we sometimes transported CCAT hardware to such events, as well as obtaining cooperation from other scholars and vendors. At one annual conference, we even posted our own security guards, to protect the equipment. I think some of us slept in the CARG room at night. By then, the micro-computer was on the rise, and David Packard's development of the IBYCUS Scholarly Computer, with built-in Greek, Hebrew, and Coptic fonts, made it easier to show what we, and others, were doing with the technology. Unfortunately, the micro-IBYCUS was not set up to communicate on the rapidly developing networking web, so the need for computing diversity was important. Back at the CATSS offices, we were at one point running the original IBYCUS mini-computer and terminals, the newer IBYCUS micro-computer (we have the third that was made), a networked Mac (or two), and a networked IBM PC (or two). We did BITNET communications with the IBM line editor and wondered when it might be possible to transmit and receive non-English fonts, and perhaps even images. On the font front, Unicode was the promised messiah; it still is. Some systems can read the TLG Greek Unicode on the Internet, diacritics and all. I regularly use two computers, the newer one in my office, and a slightly older portable in the classroom. In the office, the Greek is beautiful; in the classroom, we still wait for the promise to be fulfilled. Both can receive beautiful color images, I'm happy to say.

For ten years, from 1984 to 1994, I published quarterly "Offline" columns in SBL and CSSR newsletters as a service to the scholarly community. That tells me something about technological progress, since the initial issue (April 1984, "In Quest of Computer Literacy") was indeed written in a situation in which it would have been futile, even if possible, for me to publish the column "online." But within a year, I was introducing a BITNET bulletin board set up by Jack Abercrombie for scholarly communication (Offline #4). Soon thereafter, "Offline" became a paradox by becoming an "online" publication as well as hard copy. By the time my swan-song column appeared (Offline 44, February 1994), the Internet was flourishing and the era of color imaging on the Web was rapidly gearing up with all its added excitement and complexities. I remember thinking that it was a good time to step back and let fresher minds, and pens, take over. I had nursed the column along from offline into a flourishing, if still mainly two dimensional black and white (and amber and green), web world; someone else could deal with the psychedelic 3-D developments! And I became famous in that period for computing expertise, at least among those who knew little about the new technology. At a conference at Princeton one summer, I recall strolling across campus to the dining hall with a young graduate student I'd just met, on whom it was gradually dawning that she had read some things of mine on early Christianity as well as the then current "Offline" materials. This apparent paradox led her to blurt, in spontaneous innocence, "Are you ever going to become a scholar again?" or the like. I probably laughed—hopefully, not unkindly. In any event, I think there has been an affirmative result, combining both worlds in the process.[6]

Of course, my computer world didn't end in 1994, even if I consciously withdrew from the "Offline" overexposure. Somewhere in that period, I decided that electronic publication was here to stay and that, as a fully tenured publishing faculty member, my imagined audience would be best served by my henceforth issuing my work in electronic form as the primary medium. If someone wanted to put my contributions into hard copy, I had no objection, as long as it was understood that the main publication was the electronic one, over which I retained control. I have been happy with this decision, although the increasing demand of contributing to Festschriften for aging colleagues has produced an unanticipated wrinkle. They're still producing conventional books. Still, as I nudge along into more leisurely "retirement," I hope to have time and energy to complete the task of electronifying all my past, present, and future contributions (an electronic auto-Festschrift of sorts to myself) and to participate in the new waves of productivity and imagination that this technology permits—and hopefully encourages. Let the one who has ears to hear, and eyes to see, take note and join in the parade![7]

Robert A. Kraft is Professor of Religious Studies at the University of Pennsylvania.

[1] Thomas Aquinatis Opera omnia [computer file] : cum hypertextibus in CD-ROM / auctore Roberto Busa.—Milano: Editoria Elettronica Editel, 1992.—1 computer laser optical disk; 4 3/4 in. + 1 user's guide (64 p.; 18 cm.) The Introduction states: "This CD-ROM contains the complete works of Thomas Aquinas ... all in Latin, and divided into 118 units or writings ... They are followed by another 61 medieval Latin writings, connected in some way with, and congruous to, the former. All of the texts were put in this CD-ROM to enable its consultation by means of hypertext techniques made available by computer. The author is Roberto Busa SJ ... To each word of the texts, information has been added in order to obtain an 'inner hypertext.' ... These two levels of hypertext navigation were planned from the beginning of the project: 1) the inner hypertext, aimed at text understanding, lemmatises all of the words, signals various textual typologies, as well as morphological, onomastic and homographical codes, plus other codes for referencing and sequencing; 2) the outer hypertext is aimed at evaluating [Thomas'] writings by comparison with other authors from the same cultural milieu and on similar subjects. The corpus contains 10,631,973 individual words ... coming to a total of 1,616,060,048 bytes altogether ..."

[2] "A. Q. Morton is a Fellow of the Royal Society of Edinburgh. He is best known, perhaps, for his pioneering work in bringing the academy face-to-face with the computer revolution (Christianity in the Computer Age, 1964). He has also written major works on the Gospels, Acts, and Paul (The Structure of the Fourth Gospel, 1961; The Genesis of John, 1980; The Structure of Luke-Acts, 1964; Paul, The Man and the Myth, 1966); and The Making of Mark (Mellen, 1996)" (Mellen book blurb on the Web).

[3] To decide anything these days (2004), I go first to google.com or a comparable web search engine, and see what sort of information is "out there" and from which I can cut and paste. There are various charts and treatments of the history of computing, with various claims. For example: http://www.cyberstreet.com/hcs/museum/chron.htm— "1941: Colossus computer is designed by Alan M. Turing and built by M.H.A. Neuman at the University of Manchester, England." At UPenn, in "1946: ENIAC (Electronic Numerical Integrator and Computer), with 18,000 vacuum tubes, is dedicated at the University of Pennsylvania. It was 8 by 100 feet and weighed 80 tons. It could do 5,000 additions and 360 multiplications per second." For different Manchester claims, see http://www.computer50.org/—"The Small-Scale Experimental Machine, known as SSEM, or the 'Baby,' was designed and built at the University of Manchester, and made its first successful run of a program on June 21, 1948. It was the first machine that had all the components now classically regarded as characteristic of the basic computer. Most importantly it was the first computer that could store not only data but any (short!) user program in electronic memory and process it at electronic speed." Go figure. Definitions are crucial.

[4] J. Lust, E. Eynikel, and K. Hauspie, with the collaboration of G. Chamberlain. A Greek-English Lexicon of the Septuagint (Stuttgart: Deutsche Bibelgesellschaft; Part 1, 1992, Part 2, 1996).

[5] See Kraft and Tov, eds., Computer Assisted Tools for Septuagint Studies I: Ruth (Atlanta: Scholars Press, 1986).

[6] For more organized thoughts on the development of computing in biblical studies, see not only the aforementioned "Offline" columns (available on the Web), but also the expanded electronic version of my article on "Computers and Textual Criticism of the New Testament [and of Jewish Greek Scriptures]" at http://ccat.sas.upenn.edu/gopher/other/courses/rels/735/textcriticism/nttcart.new (the shorter version appeared as "The Use of Computers in New Testament Textual Criticism," pp. 268-282 in The Text of the NT in Contemporary Research, ed. Bart Ehrman and Michael Holmes (Metzger Festschrift). Studies and Documents 46. Grand Rapids: Eerdmans, 1995).

[7] Samples of my dalliance with the technology are readily available by surfing around on my web page, and in the streets and alleyways that lie behind the facade (i.e., without the final address element). Someday, it will all be prettied up in html format—or whatever replaces html as the computing world turns: http://ccat.sas.upenn.edu/rs/rak/kraft.html

Citation: Robert A. Kraft, " How I Met the Computer, and How it Changed my Life," SBL Forum , n.p. [cited April 2004]. Online:http://sbl-site.org/Article.aspx?ArticleID=246