The World’s Oldest Torah Scrolls

By Gary A. Rendsburg

 

A recent announcement by the Library of Congress regarding the purchase of a single Torah scroll sheet dating from approximately 1000 C.E. has generated great interest in the topic of old Torah scrolls. Just what are the world’s oldest Torah scrolls and where does the Library of Congress scroll fit in?

 

Torah Scroll Sheet dated ca. 1000 C.E., containing Exodus 10:10-16:15. Courtesy of the Hebraic Section of the African and Middle East Division, Library of Congress.

 

The Library of Congress scroll sheet contains five columns of text, comprising Exodus 10:10-16:15, a portion extending from the Plague of Locusts to the appearance of Manna in the desert. Included within the text is the Song of the Sea (Exodus 15:1‒19).

According to an inscription in both Hebrew and Russian on the back of the scroll, the sheet was presented by Shelomo Beim (1817-1867 C.E.), Karaite hazzan in Chufut-Kale, Crimea, to Grand Duke Constantine, brother of Czar Alexander II, in the year 1863. One may assume that the scroll sheet emanates from the Near East, based on considerations of text, handwriting, section divisions, and layout of the Song of the Sea.

At some point, the scroll sheet was taken to England , where in 2001 it was offered for sale by Christie’s Auction House. Fortunately, before the sale, Jordan Penkower of Bar-Ilan University was able to study the document closely and described it in a very detailed article in the journal Textus.

In 2017, the sheet was again offered for sale, this time by the 2001 buyer, the noted rare book dealer Stephan Loewenthiel. The Library of Congress purchased the sheet, and the Hebraic Section of the African and Middle Eastern Division now serves as the custodian of this exceedingly important document. I had the opportunity to inspect the scroll sheet at the Library of Congress in October 2017, courtesy of Dr. Ann Brener, head of the Hebraic Section, in advance of the Library’s public announcement in January 2018.

But is this document unique? How many truly old Torah scrolls are there? How many survive from approximately 1000 years ago or more? Readers of The Ancient Near East Today are likely aware of the approximately 220 biblical manuscripts from amongst the Dead Sea Scrolls, dating from 3rd century B.C.E. to 1st century C.E., along with the related documents from Masada, Naḥal Ḥever, Wadi Murabba‘at, and other sites, which date from the 1st-2nd centuries C.E. But what about the ensuing centuries, until we reach the date of the Library of Congress portion at approximately 1000 C.E.? What scrolls, or portions of scrolls, do we possess?

1. The oldest document is the Ein Gedi scroll, which was recently digitally ‘unrolled’ through remarkable micro-CT scanning, revealing the text of Leviticus 1-2. Archaeological evidence suggests the date of the Ein Gedi synagogue is approximately 500 C.E., but carbon-14 testing reveals that the scroll itself is much older, dating to ca. 300 C.E.

The Ein Gedi scroll, containing Leviticus 1-2, digitally ‘unrolled’ and revealed through micro-CT-scanning.

 

The scroll was found in the Torah niche of the Ein Gedi synagogue during excavations in 1970, so we may conclude that it was used for the liturgical reading of the Torah. Then, as now, Torah scrolls were sometimes used for centuries.

 

The Ein Gedi Synagogue, with the Torah niche indicated by sign no. 8 near the top of the photo. Photo courtesy of Gary A. Rendsburg.

 

But the Ein Gedi scroll commences with a blank sheet, so we can be certain that this was not a complete Torah scroll, but rather contained one, two, or three books only (that is, Leviticus only, or Leviticus and Numbers, or at most Leviticus-Numbers-Deuteronomy). I mention this because it relates to a parallel question: at what point did Torah scrolls come to contain all five books of the Pentateuch? There is no definitive answer to this question, but the blank sheet offers a clue.

A blank sheet at the beginning of a scroll is known as a ‘protocol’, from medieval Latin protocollum (Greek πρωτόκολλον), literally ‘glued (or attached) before’. Typically, this introductory sheet was not completely blank but contained notes concerning the contents of the document to follow; hence the derived meaning of the English word ‘protocol’. Presumably there was no need for any such writing, and/or there was a halakhic prohibition against it, for a scroll of Torah or a Pentateuchal book. The Ein Gedi scroll shows that by the fourth century C.E. there was not yet a requirement or custom that all five books of the Pentateuch be united into a single scroll.

2. Next in age come the London and Ashkar-Gilson sheets, which derive from the same scroll, dated ca. 700 C.E. Moreover, to our good fortune the surviving sheets contain Exodus 9:18–13:2 and 13:19–16:1, which once again includes the Song of the Sea. But these sheets have a complicated history.

 

The London sheet, now in the private collection of Stephan Loewenthiel, containing Exodus 9:18–13:2.

Ashkar-Gilson no. 2, spanning Exodus 13:19–16:1 including the Song of the Sea.

 

The first is also held in the private collection of Stephan Loewenthiel, though for many years it belonged to the collection of Jews’ College, London, hence, its designation as the London scroll sheet. Its origins and how it came to London are unknown. The second sheet is named for the two individuals who purchased the document, along with others, in Beirut and eventually donated them to the Duke University Library in 1979, where they now reside. This donation, and the size of the resulting charitable tax deductions, however, attracted the attention of the Internal Revenue Service, which litigated the matter for over a decade.

The Ashkar-Gilson sheet, known technically as Ashkar-Gilson 2, can be read only with great difficulty, but with modern photographic technology, namely multi-spectral imaging, the text now has been revealed. Using these new data Edna Engel and Mordechay Mishor concluded that the two Exodus sheets derive from the same scroll.

But the Ashkar-Gilson collection includes other old Torah scroll sheets, including one that contains the Deuteronomy version of the Decalogue in Deuteronomy 5. This document, Ashkar-Gilson 14, was previously totally unreadable, although once again multi-spectral imaging has allowed scholars to read the text.

 

 

Askhar-Gilson no. 14, to the naked eye (left) and as revealed through multi-spectral imaging (right).

 

Does Ashkar-Gilson belong to the same scroll as Ashkar-Gilson 2 and the London sheet? Working only from photos, Paul Sanders of the Protestant Theological Seminary in Amsterdam informs me that Ashkar-Gilson 14 most likely derives from a different manuscript. A task that still lies ahead is a full analysis of all the Ashkar-Gilson documents, to determine their dates, texts, layouts, handwritings, and possible origins.

3.-4.  Amongst the Cairo Geniza documents are portions of two very old Torah scrolls.  We know that these fragments originate from scrolls (and not from codices), because the verso of each document is blank.  All the documents are to be found in the Taylor-Schechter (T-S) Collection of the Cambridge University Library (C.U.L.).

The older scroll is comprised of two sections of the book of Genesis:

C.U.L. T-S NS 3.21 – ca. 800 C.E. – containing portions of Genesis 13-17

C.U.L. T-S NS 4.3 – ca. 800 C.E. – containing portions of Genesis 4-6

Now, whether this was a complete Torah scroll or a scroll of the book of Genesis only cannot be determined. And while I have provided the date of ca. 800 C.E. for these two documents, some handwriting experts would date them centuries earlier, ca. 500 C.E.

 

 

Cambridge University Library, T-S NS 3.21 (ca. 800 C.E.), fragment of a Torah scroll (or at least Genesis scroll) containing portions of Genesis 13-17, found in the Cairo Geniza. Used by kind permission of the Syndics of Cambridge University.

 

Cambridge University Library, T-S NS 4.3 (ca. 800 C.E.), fragment of a Torah scroll (or at least Genesis scroll) containing portions of Genesis 4-6, found in the Cairo Geniza. Used by kind permission of the Syndics of Cambridge University.

 

The second scroll is dated to c. 900 C.E., and may comprise as many as 19 different fragments, stretching from Genesis through Deuteronomy, assuming that all derive from the same original Torah scroll.  Such is the opinion, for example, of Colette Sirat, a renowned expert in the field of Hebrew paleography and codicology. The first extant portion is this one:

C.U.L. T-S NS 4.8 – ca. 900 C.E. – containing portions of Genesis 25-26

 

Cambridge University Library, T-S NS 4.8 (ca. 900 C.E.), fragment of a Torah scroll containing portions of Genesis 4-6, found in the Cairo Geniza. Used by kind permission of the Syndics of Cambridge University.

 

5. Next in age comes the fascinating MS Florence, Biblioteca Medicea Laurenziana pluteo 74,17, a palimpsest. The overtext is a Greek manuscript, dated to the 13th century C.E., but much of the undertext in the second half of the manuscript is comprised of sections of six old Torah scroll sheets, dated to the 10th century C.E., cut up and reused for the production of the overtext.  The undertext, the scribal hands, the direction of writing, and more, have been studied in great detail by Colette Sirat and her collaborators.

[I am grateful to Jordan Penkower (Bar-Ilan University) for calling this unique manuscript to my attention and more generally for his valuable comments to an earlier version of this article.]

 

MS Florence, Biblioteca Medicea Laurenziana pluteo 74,17, a palimpsest, with Greek overtext and Hebrew undertext, in this image, portions of Genesis, chs. 5 and 7.

6. At this point, as we proceed chronologically, the Library of Congress Torah scroll sheet, dated to c. 1000 C.E., with which we began this essay, may be mentioned again.

7.-8.  Finally, I must note two items, our oldest complete Torah scrolls, both from Italy. The oldest complete Torah scroll known is in the Biblioteca universitaria di Bologna, dated via radiocarbon to ca. 1190 C.E.  In a library catalogue dated to 1889, this scroll was mistakenly dated to the 17th century.  Mauro Perani (Bologna) “rediscovered” the scroll in 2011, realized that the document was much older, and arranged for the radiocarbon testing.  The complete scroll reaches 36 meters (120 feet) and is written on sheepskin.

The oldest complete Torah scroll that is still in use is in Biella, Italy, and is dated via radiocarbon to ca. 1250 C.E.  Credit for the recognition of the age and importance of this Torah scroll goes to Rabbi Amedeo Spagnoletto, recently appointed rabbi in Florence.

 

Biblioteca universitaria di Bologna Torah (National Geographic)

 

Biella Torah scroll, with the one column shown here including the Song of the Sea (Exodus 15).

 

Long after the Christians had adopted the codex form for Scripture and for other writings, as early as the 2nd century C.E., Jews continued to write biblical texts on scrolls. Eventually, the codex was adopted by the Jews as well, though not until the 8th or 9th century. The famous Aleppo Codex dates to ca. 920 C.E., while the St. Petersburg (Leningrad) Codex dates to 1009-1010 C.E. But such codices were for study. For liturgical purposes, Jews continued to use the scroll form of the biblical book, a practice that continues until the present day.

Were the Library of Congress’s scroll sheet part of a complete Torah scroll, it would be at home at any synagogue in the world today. In the meantime, though, this precious document will be on display in the Great Hall of the Library of Congress, right next to the Gutenberg Bible from March 27 to April 17.

 

Gary A. Rendsburg serves as Blanche and Irving Laurie Chair in Jewish History at Rutgers University.