The Codes and Their Critics
With computers, the search changed character. A computer can check every starting point and every interval in a text of over three hundred thousand letters in moments. In 1994 three Israeli researchers, Doron Witztum, Eliyahu Rips and Yoav Rosenberg, published a paper in the journal Statistical Science titled “Equidistant Letter Sequences in the Book of Genesis.” They took a list of well-known rabbis from later Jewish history and searched Genesis for their names and for their dates of birth or death. They then measured how close the names fell to the matching dates within the text, and compared the result with the closeness obtained when names and dates were deliberately mismatched. Their claim was that the correct pairs fell together more often than chance would explain. The journal’s editor printed the paper with a note explaining that the referees could not find a flaw in it but did not find the conclusion believable. The paper was presented to readers as a puzzle to be solved.
The subject reached a wide public in 1997 with Michael Drosnin’s book The Bible Code, a bestseller that used the method to find what Drosnin described as predictions of modern events, including the assassination of Israeli Prime Minister Yitzhak Rabin. Drosnin’s approach differed sharply from that of the 1994 paper. It did not test a list fixed in advance; it searched freely for striking combinations. Many people who took the original research seriously were unhappy with the book, and Rips himself publicly distanced himself from it.
Critics answered on two levels. Against books like Drosnin’s, they pointed out that in any long text, searched freely, one can find almost anything. When Drosnin challenged his critics to find a prediction of an assassination in Moby Dick, the Australian mathematician Brendan McKay found a number of them. Against the 1994 paper, the reply was more technical. In 1999 Statistical Science published “Solving the Bible Code Puzzle” by McKay, Dror Bar-Natan, Maya Bar-Hillel and Gil Kalai. They argued that the researchers had made many small choices, such as which names to use for each rabbi, how to spell them, and which forms of the dates to include. Each choice could be defended, but together they gave enough room to produce a strong result without anyone intending to. To show this, the authors made their own choices for the same kind of experiment in a Hebrew translation of War and Peace and reported a comparable result there. The original researchers and their supporters disputed this account and published replies. Other experiments followed on both sides, and the debate has continued in specialist circles without a resolution that both sides accept.
Another objection concerns the text itself. A count of letters at fixed intervals depends on every letter being in its place, since one added or missing letter shifts everything after it. Torah scrolls in use today agree remarkably closely. Even so, the traditions differ in a small number of letters (the Yemenite scrolls, for example, differ from others in a handful of spellings), and ancient manuscripts show more variation still. Defenders reply that the effect would be weakened by any change, so a result that survives in the text as we have it is all the more notable. Critics reply that the text the codes require may not be the text that was originally written.
Within the religious community, too, opinion has been divided. Some educators used the codes for a time as a way to draw people toward the Torah. Others, including respected rabbis, were uneasy about resting faith on statistics that could later be challenged, or about turning the Torah into a puzzle for predicting the future, something traditional teaching has generally discouraged.
That is the state of the matter. I have not taken a side in it, and I will not do so here.
How I Came to the Codes
I first heard of the codes while I was working on the relationship between the Book of Judges and the Torah, the work described in the chapters on Sisera, Gideon and the Book of Judges. The word usage was showing me that the two are interrelated. The author of Judges, and the Talmudic sages after him, saw the defeat of Sisera, along with other events, as connected to an event in the Torah. At about the same time, people working on the codes were reporting connections of their own. Some of the codes they found referred to events in later books of the Bible, and they turned up in Torah passages that seemed related to those events in other ways as well.
This gave me a way of thinking about the codes that did not depend on the statistical debate. If such codes were real, they would have to agree with what the plain text shows. A code pointing to an event in the Book of Judges ought to appear in a passage that the words themselves already connect with that event. If the codes did not agree with the plain text in a case like this, there would be little reason to believe them when they claimed to reach events thousands of years further into the future.
I have not spent any significant amount of time on this kind of search. The work in this book seems to me more down to earth, and more valuable, than finding a code that is said to point to the fall of the Twin Towers. I am also aware of how easily the whole enterprise can mislead. As noted above, one can find almost any word if one searches long enough. I have seen codes promoted that prove nothing at all. One example is a three-letter word made of letters that are common in the Torah. Such a word can be found many times on any page and can be attached to any passage one likes. Even if some codes are real, they can become a dangerous tool in the hands of someone who wants to misuse them.
Nevertheless, after only a very short time of looking, I found one code that unsettled me. It is that code, and that code alone, that I want to present here.