Chapter Thirteen

Fiat Lux: The Electronic Word

Just as Johann Gutenberg and other vanguard printers used manuscripts as models for printed books, so those who created early digital forms of Bible texts looked to established traditions, imagining their goal in the form of a book. The Second World War spurred the development of specialized electromechanical and electronic computers. In 1946 the University of Pennsylvania unveiled ENIAC (Electronic Numerical Integrator and Calculator), the first large-scale, electronic, general-purpose computer. Within a decade, two major concordances vital to biblical studies were attempted using electronic computers, one of ancient texts and one of a modern translation of the Bible.

The first biblical concordance created by computer was the Complete Concordance of the Revised Standard Version of the Bible, edited by the Reverend John W. Ellison. When the full Revised Standard Version appeared in 1952, Ellison was already at work doing biblical research using the computer. In 1951, the American Philosophical Society granted him funding to attempt to trace the relationship of Gospel manuscript traditions by collating various versions by computer. Ellison reportedly "deplored the idea that scholars with two or three doctoral degrees apiece should sit around sorting words." (Burton 1981b, 6) He believed that the concordance could be produced by computer, and chose the Remington Rand Company’s Univac, one of the first computers to accept alphabetic input.

By the standards of the time, the task was huge. The RSV contained approximately 800,000 words. This may have been the first major project to use a dual transcription method to ensure accuracy. Two teams transcribed the full text, one to metal tape and one to punch cards. The result was 480 pounds of cards and 80 miles of tape. Both were transferred to magnetic tape and compared by the computer to identify transcription errors. The entire process of transcription and correction took nine months. A full word list was run, and based on that list, Ellison decided to eliminate 131 common words from the final concordance. Univac then processed the final sort, and the results were printed out in about 1,000 hours. Although the computer’s work was completed in 1955, typesetting the results took almost two years. The final concordance did not appear until 1957, accompanied by much ballyhoo in the popular press.

While Ellison was working on the RSV concordance, the patriarch of digital text processing, Father Roberto Busa, s.j., was undertaking a project involving ancient texts. Father Busa was primarily concerned with his monumental work, the Index Thomisticus, a complete concordance of the works of St. Thomas Aquinas. He began the work of analyzing and collating the Thomistic corpus using electromechanical card-sorting machines in 1949. Father Busa pioneered many of the techniques required to encode a complex textual corpus to produce a comprehensive, analytical, contextual concordance. In 1956, he established the Centro per l’Automazione dell’Analisi Letteraria (CAAL) in Gallarte, Italy, to aid scholars in indexing classical and other ancient texts.

CAAL’s first project involving electronic computers was to index the non-biblical Dead Sea Scrolls texts published before 1958. The project required careful design and a great deal of preliminary editorial work. Busa supervised the transcription of some 30,000 Hebrew, Aramaic, and Nabatean words. Besides carefully recording the position of each word, significant characteristics of the text, including lacunae, indecipherable letters, editorial reconstruction, and alternative readings were preserved. Programs and equipment were modified to preserve right-to-left reading and appropriate justification. The cards were prepared in Italy, then transferred to IBM in New York. At IBM, the cards were transferred to magnetic tape and processed using an IBM 705 computer. Paul Tasman of IBM spent three months creating the programs for tracking the frequency, use, and sequence of words in the text, as well as supplying a context. The resulting word lists in alphabetical order, in reverse alphabetical order to collate word endings, in order by word length, as well as analysis of word frequency provided tools for reconstructing words partly or totally obliterated in the scrolls. By Tasman’s own assessment, the computer itself had thrown no new light on the Dead Sea Scrolls, but had "... freed the scholar from the tedious, repetitious clerical work and returned him to his real vocation—scholarly research" (Tasman 1958, 12).

The project was well underway in 1958 and publication of the indexes was planned for 1959 (Busa 1958, 189). Unfortunately, Father Busa had to abandon the project in order to keep the Index Thomisticus on schedule. The full results were never published, and the cards and tapes were later destroyed, according to Busa in a 1997 email message to the author.