Do computers confirm - or deny - the Torah’s divinity?

Biblical criticism has been around for centuries, and those who choose to believe have ignored it until now.

By DAVID SHAMAH
Navy Torah 311 (photo credit: US Navy/Wikipedia Commons)
Navy Torah 311
(photo credit: US Navy/Wikipedia Commons)
Between pogroms, disputations, Inquisitions – and now, computers – it’s hard to be a Jew, or a believing one, at least. The latest challenge actually comes from right here in Israel – in the form of a computer program designed by a team at Bar-Ilan University led by Prof. Moshe Koppel, an expert in the field of Authorship Attribution, the analysis of texts to determine information on who the actual writer was.
Together with several other professors, Koppel’s system analyzes word use in a document to determine information about an “unknown author,” figuring out his/her gender, demographics, personality, cultural background, etc. The texts are analyzed by a computer using specially designed algorithms, which the team says has a very high level of accuracy.
In a landmark paper on the subject, Koppel, along with fellow researchers Shlomo Argamon, Jonathan Schler and James W. Pennebaker, wrote that “authorship profiling can help police identify characteristics of the perpetrator of a crime when there are too few (or too many) specific suspects to consider,” or help corporate officials analyze blogs and postings on websites to sharpen their message.
It sounds like a sort of modern version of “biblical criticism,” which has its roots in the 17th and 18th centuries and is accepted by modern scholars as authoritative. So it would make sense that Koppel would apply the Authorship Attribution algorithms to the Tanach, especially the Torah. The result was what any scientist would expect: The software “confirmed” 90 percent of the accepted findings of biblical critics – the three authors of the first chapters of Genesis, the “priestly” and “non-priestly” authors of the middle books, etc.
“We have thus been able to largely recapitulate several centuries of painstaking manual labor with our automated method,” Koppel and fellow scientists Navot Akiva, Nachum Dershowitz and Idan Dershowitz said in a statement after presenting the results at a conference last week.
All that is well and good if you’re a scientist who accepts the idea of differentiated authorship of the Tanach. But if you accept the traditional Jewish belief that the Torah was given by G-d, you have a bit of a dilemma.
Of course, arguably nothing has changed; biblical criticism has been around for centuries, and those who choose to believe have ignored it until now.
On the other hand, it was easier to dismiss biblical criticism until last week; you could attribute the findings of Wellhausen, for example, to his nasty attitude to the Jews, which undoubtedly colored his research. But you can’t really attribute bias to a computer program – especially not one developed by a team from Bar-Ilan University, several of whom are apparently observant themselves! Belief is, of course, a personal thing, and Jewish tradition itself is aware of outward contradictions of text in the Bible. Large swaths of the Talmud are dedicated to resolving textual problems and contradictions.
And for those who accept the historical evidence of Jewish belief – the witnessing of the giving of the Torah in front of millions of people, the unlikelihood that difficult commandments like keeping Shabbat or shmitta (the sabbatical year) could be imposed without a singular historical experience (such as a divine revelation) – a computer’s confirmation, or lack of it, isn’t going to make a difference.
But not everyone’s faith is that rock solid, and if we take the results of advanced computer algorithms seriously in so many other areas, it’s hard to dismiss their findings in this specific area.
But how soon we forget! This is not the first time computers have tangled with the Bible, and the last time they did, the Bible won hands down. I’m talking, of course, about the original studies of the “Torah Codes,” as originally developed by Doron Witztum, Eliyahu Rips and Yoav Rosenberg, and later enhanced by Robert Haralick. The original Codes idea was based on taking passages from the Torah, applying the Equidistant Letter Sequence (ELS) method to find a hidden “code,” containing additional, pertinent information on the subject, as well as hints of historical events and personalities.
According to the original theory, it only works in the Hebrew version of the Torah.
Technically, you could do the same thing by just picking a letter and counting in a random number or sequence backward or forward, but the whole project was made much easier by computers. The most authoritative website on the subject today is Torahcode.net, run by Prof. Haralick and Rabbi Matityahu Glazerson, along with Witztum, Rips, Gans and other scientists involved in the field.
While the authors of the original theory only presented the theory and its application (the site has many examples), many believing Jews seized on the Codes as “proof” of the divinity of the Torah, with computers affirming that the Bible was no ordinary book. The idea was so successful, in fact, that it invited all sorts of other applications that the original team did not see as helpful – and in fact damaged the popular and scientific reputation of the Codes.
First there was the book The Bible Code by Michael Drosnin, which purported to use the Codes to predict the future (Drosnin sent a letter to then-prime minister Shimon Peres warning that he had “foreseen” in the codes a nuclear attack striking Israel in 1996). Then, various Christian groups jumped on the bandwagon, claiming to used the ELS method to find “hints” of the appearance of Jesus in the Torah. Finally, scientists, sick of what had turned into a farce, ran experiments on Shakespeare’s works and Moby Dick, finding “codes” in those as well.
Haralick et al still stand by their findings, and the Torahcode.net website thoroughly discusses each of these phenomena, responding to each criticism and proving why they are scientifically invalid, having nothing to do with proper Codes methodology.
Going through the Torahcode site, believers will come away with the feeling that, indeed, computers “proved” the divinity of the Torah – a perfect antidote to the Authorship Attribution theory of multiple non-divine authors.
So which is it: Do computers prove, or disprove, the Torah’s divinity? That, too, is a matter of faith, and it depends on which computer you believe in!