Codes in the Bible

In 1997 a book soared to the top of the bestseller lists. It became the topic of conversation and debate from the tv talk show circuit to the kitchen table. The book was "The Bible Code" and its controversial thesis was that secret codes hidden in the Old Testament predicted historical events that occured thousands of years after the bible was written.

The book's author, Michael Drosnin, claimed that he was able to foretell the assassination of former Israeli Prime Minister Yitzah Rabin before it happened. He also predicted a cataclysmic earthquake would hit Los Angeles in the year 2010.

So is "The Bible Code" just another crazy idea that sold a lot of books? Or does this volume rest on more solid scientific ground?

The idea of using codes to extract information from the bible isn't a new idea. In the 1940's and 50's, Rabbi H.M.D. Weissmandel attempted to find hidden words in the first five books of the bible (referred to as the Torah) by using "skip" codes. A skip code works by looking at the text simply as a string of letters (usually with the spaces removed) and picking out a series of letters which are separated by a particular number of other letters. Codes which use equaldistant letter sequences (ELSs), that is the same number letters between each letter of the code, are the most common approach.

ELS Codes

ELS messages are codes found in text documents by skipping an equal number of letters in the text between each letter in the code. For example, look at this code discovered by physicist David Thomas in the King James version of Genesis. The passage reads:

And hast not suffered me to kiss my sons and my daughters? Thou hast now done foolishly in so doing.

Now pick out every 4th letter starting with the R in daughters and counting spaces:

daughteRs? ThOu haSt noW donE fooLishLy in so...

and then pick out every 12th letter (not counting spaces) starting with the U in thou:

ThoU hast now done Foolishly in sO doing..

The message ROSWELL UFO can be found. Is this a prediction in the Bible that there would be an alleged UFO crash in Roswell, New Mexico in 1947? More likely it is the result of random chance.

Of course any large text will have words or even phrases that will show up in ELS codes just by accident. Early in the 1990's, though, three scholars, Doraon Witztum, Eliyahu Rips, and Yoav Rosenberg, decided to see if they could detect codes in books of the Hebrew version of the bible using a scientific methodology. The result of their attempt was published in the journal Statistical Science in 1994. Statistical Science is a peer-reviewed scientific magazine of good repute. The study done by Rips (who is a group theorist), Witztum (who is a physicist) and Rosenberg (who is a computer programmer) was a sincere attempt to put the question of bible codes on a scientific footing.

The approach the three chose was to treat ELS letter sequences found in the books as an unknown foreign language. They hoped to prove that a small group of related words (like screw and screwdriver) would show up in close proximity to one another in the code. If they appeared closer than they should by chance, that might show that the sequences in the code had meaning.

For the test they chose a group of 34 historical figures in Judaism. The search was made based on the name, the birth date and the date of death of the figures. The text used was the first book of the bible, Genesis. The results were surprising. The authors of the paper claimed that there was a 2 in 10,000 likelihood that the closeness between names, birth dates and death dates was the result of chance.

Because of the startling nature of the subject, the paper was put through an especially rigorous review before being published in Statistical Science. In the article the authors stated, "We conclude that the proximity of [equidistant letter sequences] with related meanings in the book of Genesis is not due to chance."

Given that the book of Genesis was written well before the historical figures lived, the study seems to suggest that the names had to be encoded by someone with a knowledge of the future and was taken by some as proof of the supernatural origin of the bible.

The paper had many skeptics. One, Harold Gans, a retired cryptologist from the U.S. government's National Security Agency (NAS), decided to see if he could reproduce the article's results. To his own surprise, he was successful. He even tried new sets of names with a similar outcome.

While the three authors were working on their paper, Michael Drosnin, a former reporter with the Washington Post and the Wall Street Journal, became interested in encrypted messages in the bible. He interviewed and spent time with Rips, then started using similar methods to search through the Bible for word sequences that might be used to predict future events. His most famous find was a "prediction" of the assassination of Yitzhak Rabin which he located in 1994. Drosnin warned Israeli officials, but Rabin was assassinated in 1995.

Drosnin took this prediction, plus other coded messages he found, and wrote The Bible Code. In the book Drosnin claims he has found predictions of the assassinations of Robert F. Kennedy, John F. Kennedy, and Anwar el-Sadat. In addition to mayhem, Drosnin found the name Edison encoded near light bulb and electricity, as well as Newton near gravity. Drosnin's most chilling message was a prediction that World War III would start in 1996 with a nuclear attack on the state of Israel. This didn't pan out, but Drosnin later reported he found the work "delayed" encoded near the prediction.

Many experts denounced the book. Some of the criticism came from surprising sources. Eliyahu Rips and Doron Witztum, authors of the original study, disagreed that the codes could be used to predict future events. "It is literally impossible to make future predictions based on codes. Mr. Drosnin's book does have some examples of codes that are statistically significant, and some that aren't, and the problem is that any layman reading that book will have no way of making a distinction."

Rips, Witztum, and also Gans argued that there would be a number of related words appearing close together in the codes just due to random chance. It was impossible to know which were true predictions and which were just accidents. The Bible codes only worked well in hindsight when the searcher knew what words he might be looking for. Witztum pointed out that in addition to assassinations of Kennedy, Sadat and Rabin that could be found in the code, so could the assassination of Winston Churchill, who wasn't killed at all but died of natural causes.

Other critics found flaws not just with Drosnin's book, but with Rips, Witztum and Rosenberg's paper. Some were unable to duplicate the experiment with the same success as Harold Gans. Others pointed out that the bible in its current form is not letter by letter exactly the same as the original texts that have been lost to history. Even small variations would destroy codes, especially if those codes employed long skips between letters ( Drosnin's Rabin prediction employed a skip of 4,771)

Brendan McKay of the Computer Science Department of the Australian National University, in conjunction with Dror Bar-Natan, Maya Bar-Hillel and Gil Kalai of the Jerusalem Hebrew University, wrote an article which appeared in the September 1999 Statistical Science entitled "Solving the Bible Code." The article refuted the original 1995 paper claiming that the method used to establish statistical significance was flawed.

McKay also demonstrated that any large block of text will yield ELS codes with seemingly meaningful bunches of words. In the U.N. Convention on the Law of the Sea signed in 1982, he searched for words after making the text more hebrew-like (removing the vowels). In it he found the code:

Hear all the law of the sea

as well as:

Nato need an agreement on the sea

The probabilities of finding them in the document he estimated as 95 out of a million and 21 out of a million respectively. The article asserted the authors of the original paper made mistakes in the way they designed their experiment by choosing particular forms of words that "tuned" their method to their data, thus invalidating the test.

McKay also used the text of Moby Dick to find "predictions" of the assassinations of famous figures including Trotsky, Ghandi, Robert Kennedy and others. One prediction was for the murder of Drosnin himself.

The problem with the codes is that if someone was looking for a prediction of a particular subject in any large block of text, it would not take too long before he could find encoded words that seemed to be related to it. If the searcher was looking for a prediction about a flying saucer crashing in New Mexico he would find at least some of these suggestive words:

ROSWELL, UFO, FLYING SAUCER, COVERUP, ALIENS, 1947, DISC, CRASH, AUTOPSY

and others. While the probability of finding at least a few of these is pretty high, the probability of finding any particular one may be very low.

The Bible Code controversy is not over, though. Neither Eliyahu Rips nor Michael Drosnin have backed down on their claims. Professor Rips has stated that he believes that the evidence for the codes was "stronger than ever" and Drosnin has said the critics have "told a lie."

Copyright Lee Krystek 1999. All Rights Reserved.