Missing Link in Hebrew Bible Formation

046

New analysis of a previously known scrap of a Biblical text provides fascinating insight into the formation of the Hebrew Bible. Known as the Ashkar-Gilson Hebrew Manuscript #2, the text is a remnant of a Torah scroll from the seventh or eighth century C.E. and contains a crucial section of the Book of Exodus.

Although the fragment came to light more than three decades ago, it was disregarded by scholars and remained unpublished. But a recent analysis shows that this damaged sheet fills a gap in our knowledge regarding the transmission of the Biblical text.

The earliest texts of the Hebrew Bible—or the Old Testament, as Christians call it—are found among the Dead Sea Scrolls, which include more than 200 Biblical texts ranging from a few words to almost complete books, such as a nearly undamaged copy of the Book of Isaiah (1QIsa^a).^a The dates of these ancient Bible manuscripts range from c. 250 B.C.E. to 115 C.E., so they are much older than the Ashkar-Gilson Manuscript. In this early period, the texts were not yet completely fixed; their transmission was still fluid. Copyists made mistakes, wanted to improve or expand a text, or adapted the spelling of certain words.^b Sometimes the copies could be quite different versions of the same text—for example, the Book of Jeremiah.

047 048

Thus, the Biblical text in the Dead Sea Scrolls is not quite the same as the version that later became official in Judaism and Christianity. The text that is authoritative in present-day Judaism is first found in codices, bound books produced many centuries after the Dead Sea Scrolls. Virtually all modern translations of the Hebrew Bible are based on these relatively late codices.

The earliest Hebrew Bible codices date from only the ninth or tenth century C.E. The authoritative Aleppo Codex (c. 930 C.E., with extensive parts missing since 1947)^c and the well-preserved Leningrad Codex (c. 1008 C.E.)^d comprise all the books of the Hebrew Bible. Other codices comprise only a part of the Hebrew Bible, for instance only the Torah (Genesis–Deuteronomy) or only the Prophetic Books (Joshua–Malachi).

These codices were composed by a group of specialists known as Masoretes, who worked in Tiberias, on the western shore of the Sea of Galilee, but also in some Mesopotamian cities with large Jewish populations. Most of the surviving codices, however, were produced by the Masoretes of Tiberias. They succeeded in their effort to completely stabilize the Biblical text, and their work has remained 049 authoritative to this day.

But the Masoretes even went a step further. To the existing text, which comprised only consonants, they added vowel and accent signs indicating how exactly the text was to be read. They also wrote short (and sometimes longer) notes in the margins to elucidate textual problems and to prevent even the slightest changes.

This version of the Hebrew Bible, known as the Masoretic text, became normative in Judaism—and it still is. The scrolls that since then have been produced for use in synagogues contain exactly the same consonantal text, although the vowel and accent signs are left out because they are not original. Not a single consonant, however, may differ from the normative, Masoretic version found in the earliest codices.

Thus we have a standard text from about the tenth century C.E. and a variety of differing texts (some close to the Masoretic text) from about a millennium earlier (i.e., the Dead Sea Scrolls). But we have almost nothing—just a few scraps—from the centuries in between.¹ The history of the text of the Hebrew Bible is almost blank in this intermediate period. Scholars sometimes refer to it as the “silent era.” The puzzling question is how the text developed during these enigmatic centuries.

From the “silent era,” there are many manuscripts of Greek, Latin, Syriac and Coptic translations of the Hebrew Bible—but hardly any Biblical texts in the original Hebrew. The prevalent conjecture is that the continuing persecution of the Jews, first by Christians and later by Muslims, led to the destruction of their Hebrew Bible manuscripts. Apparently the Christian persecutors did not realize that the Jewish Bible manuscripts they destroyed might be more accurate than the translations that the Christians used, such as the Greek Septuagint.

Knowledge of developments in the “silent era” has increased significantly, however, with the rediscovery of the Ashkar-Gilson Manuscript—which would certainly seem to be an unlikely source. It is a severely damaged and blackened manuscript containing excerpts only from Exodus 13:19–16:1, and it contains only the consonantal text.

In 1972, two American doctors, Fuad Ashkar and Albert Gilson (hence its name), bought the sheet from an antiquities dealer in Beirut. Several years later, Ashkar and Gilson donated it to Duke University. The renowned New Testament scholar James Charlesworth, who was then teaching at Duke University, dated it to between the sixth and eighth centuries C.E. on the basis of a paleographical analysis. His dating was soon narrowed by scientific carbon-14 analyses, showing that the parchment is authentic and dates from the seventh or eighth century C.E.

Since 2007, the Ashkar-Gilson Manuscript has 050 been on extended loan in the Shrine of the Book of the Israel Museum in Jerusalem. However, it will soon return to Duke University, where it will be housed in the David M. Rubenstein Rare Book & Manuscript Library. While the sheet was on display in the Shrine of the Book, two Israeli experts, Mordechai Mishor and Edna Engel, noticed that the handwriting and layout reminded them of a better-preserved sheet of a Torah scroll, known as the London Manuscript, which contains the text of an earlier passage: Exodus 9:18–13:2. The two scholars had seen a photo of the London Manuscript in the Encyclopaedia Biblica (1968). Strangely enough, the London Manuscript too had been disregarded in scholarly research. It had been kept in Jews’ College in London—hence its name—but had been sold to the New York collector Stephan Loewentheil.

Following some additional research, it was established that the London Manuscript was a remnant of the same Torah scroll as the Ashkar-Gilson Manuscript. The two sheets were divided into columns of 42 lines and were written by the same hand. One complete column appears to have been lost between the two sheets. From the size of the surviving sheets it can be inferred that the Torah scroll was originally around 20 inches high.

Although the authenticity of the two fragments is beyond doubt, it is unclear where they came from and how they became separated from each other. One guess is that they came from the famous Cairo Genizah. Although most of the fragments from this hoard—more than 200,000 pieces—ended up in university libraries, some were obtained by antiquities dealers.

In May 2014, I obtained an infrared photo of the Ashkar-Gilson Manuscript from the Israel Museum and soon thereafter published the results of my study.² Both the Ashkar-Gilson Manuscript and the London Manuscript contain the consonantal text of the authoritative Masoretic codices of several centuries later. (Since 1947 the text of Exodus has been missing from the Aleppo Codex, the most accurate of the early codices, but its text can be reconstructed 051 052 with certainty.) The text of the codices agrees completely with the text of the Ashkar-Gilson and London Manuscripts. (In some codices there are a few insignificant orthographic deviations concerning some slightly different spellings.)

This shows for the first time that the Masoretic copyists reproduced an older consonantal text as faithfully as possible and did not allow their copies to deviate from the original. Only meaningless variants were sometimes still accepted. Of course, the Ashkar-Gilson and London Manuscripts display only part of the Hebrew Bible—some excerpts from Exodus—and it cannot be proven that the rest of the Hebrew Bible was also copied so carefully. But it is hardly conceivable that the care of the copyists was limited to these sections of Exodus. This proves that the text of the Hebrew Bible was stabilized earlier than when the Masoretes created their first codices.

But there is more. By some good fortune, the Ashkar-Gilson Manuscript contains one of the most beautiful examples of Biblical poetry, the Song of the Sea (Exodus 15). Moses and the Israelites sang this hymn after God had parted the sea for them and then drowned Pharaoh’s pursuing army. This hymn, it is generally agreed, is one of the oldest passages of the Hebrew Bible.

The Masoretic copyists transmitted this poem with utmost care. Apparently they were aware of its unique quality. They copied it in a special symmetric layout that resembles brickwork, with two blank spaces in the even lines and one blank space in the odd lines. This arrangement was chosen not only for its beauty but also for its meaning, with each of the spaces marking the end of a colon (a small poetic unit that must be sung in one breath). The importance of this brickwork layout is reflected in the fact that it is reproduced in every Torah scroll used in synagogues today. (A similar layout is required only for the Song of Deborah in Judges 5, another old and exceptionally beautiful poem.)

In the Dead Sea Scrolls, the Song of the Sea does not yet have this special layout. But it does appear in the Ashkar-Gilson Manuscript! This marks the first time the brickwork pattern is found, without any deviation from the later arrangement.

In the Ashkar-Gilson Manuscript, some minor details in the column with the song are accidental. These details concern the layout, as well as the column’s coincidental start with an ordinary Hebrew word (הבאים) meaning “that went in” (namely, into the sea; Exodus 14:28).

Remarkably enough, even these insignificant details are also found in the oldest Masoretic codices. But there they are not accidental. The copyists had to make a special effort to reproduce them as faithfully as possible. For instance, they compressed or spaced out the text in the preceding columns to let the column with the Song of the Sea start with 074 exactly the same word as the column in the Ashkar-Gilson Manuscript. Even the copyists of the more recent Torah scrolls did their best to reproduce these seemingly insignificant details.

How can this be explained? I think there is only one convincing solution: The very Torah scroll of which the Ashkar-Gilson Manuscript and London Manuscript are remnants was consulted by the Masoretes of Tiberias. In other words, this seventh- or eighth-century Torah scroll must have figured prominently when they produced the now-standard text of the Hebrew Bible.

The Masoretes were apparently impelled to maintain the “brickwork” layout of the original because of its exceptional beauty. This is hardly surprising. Other scribal features of the ancient Torah scroll are also impressive and conform to the highest standards of the early Middle Ages. The sheets were dry-ruled before being inscribed—both vertically, to demarcate the margins of the columns, and horizontally for the individual lines. Also, the height of the columns conforms to the early medieval rule that a column of a Torah scroll must be 42 lines high. The text was written with a firm hand, and the copyist observed the ruled margins, trying to avoid protrusions beyond the margin line. Such features indicate that the Torah scroll was a first-class manuscript that deserved to be copied. So it is quite understandable that the Masoretic copyists selected this scroll.

It happens only rarely that a direct antecedent of ancient Biblical manuscripts can be traced. That this specific Torah scroll was used by the Masoretes could not have been shown on the basis of the London Manuscript alone. By a fortunate coincidence, the Ashkar-Gilson Manuscript also displays the text of the Song of the Sea and its context, thereby providing crucial evidence for the use of the scroll in the creation of the Masoretic text still used in synagogues today.

Of course, this does not mean that the Masoretic text is the original text of the Hebrew Bible. The Biblical texts preserved among the Dead Sea Scrolls differ from the Masoretic text that ultimately became authoritative. These Biblical manuscripts document the fact that at that time there was still a lot of flexibility in the transmission of the text.

So the Masoretic text is the fruit of a long process of both adaptation and faithful transmission.^e The period between the second and sixth centuries C.E. must have been one of gradual stabilization of the Biblical text. All of the surviving Dead Sea Scrolls that were written after the first Jewish revolt (post-70 C.E.) show a Biblical text that is already relatively close to the Masoretic 075 076 text. For example, the Twelve Prophets’ Scroll from Wadi Murabba‘at, about 10 miles south of Qumran, was transcribed around 115 C.E. and contains excerpt passages from Joel to Zechariah. The text is remarkably close to the much later Masoretic text.

The following centuries must have seen a further stabilization of the normative Jewish Bible text. The Ashkar-Gilson and London Manuscripts prove that this process of stabilization had already come to an end some centuries before the Masoretes started to produce the earliest Bible codices. The Masoretes reproduced a text that had already been stabilized and no longer allowed any deviations. It was not their goal to innovate—but rather to preserve the finest textual traditions that existed at the time.

New analysis of a previously known scrap of a Biblical text provides fascinating insight into the formation of the Hebrew Bible. Known as the Ashkar-Gilson Hebrew Manuscript #2, the text is a remnant of a Torah scroll from the seventh or eighth century C.E. and contains a crucial section of the Book of Exodus. Although the fragment came to light more than three decades ago, it was disregarded by scholars and remained unpublished. But a recent analysis shows that this damaged sheet fills a gap in our knowledge regarding the transmission of the Biblical text. The earliest texts of […]