Reconstructing the Bible

Project to correct thousands of biblical mistakes discovered over the years nears completion.

Prof. Menachem Cohen of the Bible department at Bar-Ilan University has no doubt that all Hebrew Bibles sitting on bookshelves in Jewish homes around the world contain errors. Sometimes scores of errors can be found in a Bible, sometimes they number in the hundreds. For the most part these are not dramatic mistakes: perhaps the absence of the letter yod, an incorrect diacritical mark, a mistaken cantillation note, and other such small inaccuracies.

But Cohen says he and his colleagues have produced the most accurate version of the Old Testament published in at least the past 1,000 years, or maybe, ever. Cohen has been working on this Bible for the past 40 years, as the director of the Mikraot Gedolot-HaKeter project. He hopes it will become the agreed-upon standard version of the Scripture that will usher the ancient text from the "age of books" into the digital age.

The last man to take upon himself a similar task was a Spanish Jew named Jacob ben Hayyim who lived in Venice circa the early 16th century, less than 100 years after the Gutenberg Bible was first printed. Like Cohen, ben Hayyim lived in an era of a technological revolution with regard to texts. Handwritten scrolls were being replaced by modern printed and bound volumes. Thus, ben Hayyim thought it necessary to preserve the ancient text.

"I saw that many of the masses, as well as many of the wise scholars among us, in our generation, do not esteem neither tradition nor the mesora [a kind of internal code consisting of symbols and letters written above or alongside the text]. They ask what use the text will continue to provide them; it has nearly been forgotten and lost," wrote ben Hayyim.

He devoted himself to the tremendous project of writing out the most precise Scriptures of his day, which would become the first progenitor of bound Bibles as we know them today. He also added alongside the text a number of commentators on the literal meaning, most notably Rabbi Shlomo Yitzhaki, known by the acronym Rashi.

This was the birth of the Mikraot Gedolot, or the Great Scriptures, which can be found in so many Jewish homes. This edition of the Torah usually includes the traditional Biblical text, notes on the mesora, Aramaic explanations of the text and rabbinical commentaries. It is the pedagogical version of the Bible that has served generations of school children and adults for study and for everyday use. Many editions of this text have been published until now, "all of them offsprings of ben Hayyim," said Cohen.

Ben Hayyim, however, was only partially successful in his ambitious mission. According to Cohen, despite his good intentions, hundreds of errors crept into his text.

The mistakes, which Cohen prefers to call "nonconformities," are measured by comparing the text to the mesora, which offers statistical and linguistic tools to linguists and enables the transcriber to stick as closely as possible to the most original text.

The mesora, for example, can indicate how many times a certain word will appear in a given configuration in the entire Scriptural text. The mesora was formulated more than 1,000 years ago in the land of Israel and with a decades-long intellectual effort.

The best product of this effort was the Keter Aram Tsova, also known as the Aleppo Codex, the most accurate Bible ever written. This version was transcribed by Aharon Ben Asher in the 10th century in Tiberias, and set the standard to which all transcribers of the Scripture after him aspire.

But unfortunately, this legendary version was not in ben Hayyim's possession when, 600 years after the codex was written, he decided to edit and print his version. By then, the Keter Aram Tsova was in Syria, where it was closely guarded by the zealous Jewish community of Aleppo; human eyes were hardly even allowed to see its pages. Therefore ben Hayyim was forced to work with other, error-laden versions of the Bible.

Thus, the mistakes persisted and in the relay race of Bible transcribers, they found their way into subsequent editions. Some meticulous editors managed to correct some of these errors, but in most cases, the redactors then added new mistakes of their own.

"The fact that we skipped over the Keter Aram Tsova at the beginning of print was a huge loss," said Cohen. "This was a blunder, and time after time a situation was perpetuated in which there are about 1,500 differences between the different Mikraot Gedolot that are found in different Jewish homes. In all of them."

Cohen is seeking to redress this great loss at a critical historical moment, the cusp of the transition into the digital age. Cohen's advantage over ben Hayyim is that he is able to use the Aleppo Codex to achieve his goal.

A long journey home

The legendary manuscript was purchased by the Karaite Jewish community in medieval Jerusalem about a century after it was written in Tiberias. It then fell into the hands of the Crusaders, but was redeemed by the Jewish community in Egypt for a high ransom. That's how it reached the desk of Maimonides, who trusted the book and used it as his reference for fine points in the Scripture.

Eventually the document made its way to the Aleppo community, which hid it in a cave at the synagogue for hundreds of years. Some 500 years later, in 1948, the codex was torn during pogroms against the Syrian Jewish community. Eventually, much of it was smuggled to Israel in roundabout ways with the help of the second president, Yitzhak Ben-Zvi.

However, it was an incomplete codex that arrived in Israel; the first 200 pages, which contain nearly all the five Books of Moses, had disappeared. The mysterious story of the codex and the theories concerning the whereabouts of the missing pages have recently been recounted in journalist Matti Friedman's book, "The Aleppo Codex: A True Story of Obsession, Faith, and the Pursuit of an Ancient Bible" (Algonquin Books, 2012 ).

"For Ben Asher, author of the Keter, the conformity was absolute. Well, not absolute. Had it been absolute he would have been an angel," says Cohen. However, according to him it would be difficult to find more than 10 impressions in the Aleppo Codex, as compared to the thousands of corrections in every manuscript of the Scriptures and, as noted, many hundreds in the printed books.

This left Cohen with the task of grappling with the reconstruction of the precise text and the mesora in the vanished parts of the codex. To that end, he and his staff referred to other parts of the codex in order to draw conclusions about words in the list portion. They also used ancient manuscripts that are considered to have a high quality mesora. In addition, Cohen employed a computer program that helped reconstruct Ben Asher's perceptions of the mesoretic notes.

E-Bible to come

The first volume of the Bar-Ilan "Mikraot Gedolot-HaKeter" came out 20 years ago. To date, 16 volumes have been published in all; only four volumes remain to be completed.

But the Bar-Ilan "HaKeter" has a competitor. "Keter Yerushalayim" was published 11 years ago by the late Rabbi Mordecai Breuer of the Hebrew University in Jerusalem. This version is also based on the Aleppo Codex but a philological argument erupted between the two experts concerning the proper way to place a certain kind of accentual notation above the letters.

The second part of the Bar-Ilan project is the preparation of a digital edition of the codex and the building of a program that will be able to analyze the text in the smallest possible units. This program, called the Keter Application, is every philologist's dream. It enables a multi-dimensional search of the text, not only by letters or words but also by diacritical marks and dots or cantillation notes.

The program makes it possible to carry out various segmentations, and for the first time, it enables scientific examination of the ancient mesora. Cohen also believes the program will be a teaching tool that will change young people's perception of the Biblical text; therefore he is providing it to schools at no cost.

Cohen and his team have also included in their work a number of rabbinical commentators and a translation into Aramaic based on ancient manuscripts from Yemen. In this way, they have created the most accurate version of the Mikraot Gedolot ever and, in effect, the most accurate version of the Scriptures.

To the delight of students, the Rashi commentary is printed in regular Hebrew letters. What is known as Rashi writing, which generations of students have struggled to read, is simply a font intended to differentiate the commentary from the Scriptural text. The man who gave his name to the letters had nothing to do with them. "Rashi was not able to read Rashi writing," says Cohen definitively.

Bar-Ilan University recently received a donation of $1.5 million, which will make it possible to issue the remaining four volumes of Cohen's work. Cohen hopes that by the end of next year, the digital and printed project will be completed. "This will be the real Keter Aram Tsova," he pledges.

Alon Ron