There Were Three Torah Texts—Until We Found the Dead Sea Scrolls

It is hard to overstate the impact that the discovery of the biblical scrolls from the Judaean Desert, especially those found in the eleven caves at Qumran, has had on the entire field of Hebrew Bible textual criticism in the last two generations. Forty years ago, I began working on the Deuteronomy manuscripts from cave 4Q for my doctoral dissertation under the supervision of Frank Moore Cross (the leading North American Hebrew Bible scholar of his day). At the time, I made two assumptions:

Complete Deuteronomy Scrolls

If a manuscript, however fragmentary, contained only text from Deuteronomy, I assumed it must have begun as a complete Deuteronomy scroll. This assumption, shared by all early Dead Sea Scroll editors, was based on the model of the Great Isaiah Scroll (1QIsa^a), one of the first Dead Sea Scrolls to be discovered. It contained the complete book of Isaiah, chapters 1–66, in one large scroll.

I soon realized that that assumption was too simplistic. True, some Dead Sea Scroll manuscripts did likely contain the complete text of Deuteronomy: For example, 4QDeut^c, which contains fragments with text from chapter 3 through chapter 32, must have been a complete Deuteronomy scroll. However, as I studied other manuscript fragments, in particular 4QDeutⁿ, which contains the Decalogue,^[1] I began to suspect that some of these manuscripts were not complete scrolls of Deuteronomy, but most likely contained only excerpts.

4QDeutⁿ is a small manuscript; its first column contains Deuteronomy 8:1–5, while the second through fifth columns contains Deuteronomy 5:1–6:1 (where it breaks off). These three texts are associated with ancient Jewish prayer: the grace after meals, the Decalogue, and the Shema. Why were these three texts gathered together, in that particular order, in a "pocket-sized" manuscript? Was it a personal copy, created for personal study or liturgical use?

This kind of usage made sense to me; after all, we find these same kind of excerpts in Siddurim or in Christian prayer books. Why not a very early example of the same? This type of excerpted text testifies to the significance of scripture in the rich intellectual and worship life of Qumran’s milieu.

Three Core Recensions

The second, more complex and difficult to break down assumption was that the Pentateuch reflected three main text types:

• Babylonian recension, reflected in the Masoretic Text [MT];
• Egyptian recension, reflected in the Septuagint [LXX];
• Palestinian recension, reflected in the Samaritan Pentateuch [SP].^[2]

Working through the 32(!) Deuteronomy manuscripts, however, as well as the so-called Reworked Pentateuch manuscripts (4Q364–367),^[3] I began to realize that Cross’ model did not fit all the evidence.

Some manuscripts did, indeed, reflect a consonantal text close to MT (e.g., 4QDeut^g), or the Vorlage of the LXX (e.g., 4QDeut^q), or the text of the SP (e.g., 4Q364), but the majority could not be neatly aligned with any of the three complete exemplars. At the time, we generally referred to such manuscripts as “mixed” or “independent,” but such nomenclature is circular, starting with an assumption about “the” three main text types and working backwards to describe the (many!) examples that didn’t fit.

What I came to see was that we were working with these four categories—MT, LXX, SP, mixed—not because they were the best exemplars of the text, but because they were complete versions which had been hallowed by communities of faith for millennia and studied by textual scholars for centuries. Nevertheless, this model is built on the accidents of historical preservation, and the adoption by given religious groups of this or that text as their canonical Torah.

Moreover, the model inverts history. The complete exemplars of MT and SP, in the form that we possess them, are medieval, while the great LXX codices come from the fourth or fifth centuries C.E. In contrast, the Qumran biblical scrolls are closer, by at least two or three centuries, to the beginning of the process of the transmission of the biblical books, long before they reached their codified forms.

In short, Qumran manuscripts needed to be understood as the product of their own time and place, the third century B.C.E. through the first century C.E., in Greco-Roman Judea. From there, we can look forward at the eventual emergence of the three complete exemplars.

A Complex Web of Recensions

What I found when I looked at the manuscripts from this new perspective was a complex web of relationships, with many streams of textual tradition. Jewish scribes in the Greco-Roman period were not mere copyists, but textual scholars, with the responsibility of passing down the cultural and religious inheritance of ancient Israel as a living tradition, relevant to their contemporary audience.

Thus, I had to begin any manuscript study by examining each manuscript first as its own unique entity. What was its purpose? Why did the scribe do what he did? How did this manuscript relate to other Qumran manuscripts with the same text? (These questions, of course, tie in with my realization described above that not all "biblical" manuscripts were complete copies of biblical books.) Only after I had answered those questions could I ask what the relationship of the manuscript was to the later complete exemplars, where it was located in the web of textual relationships or in the ongoing streams of tradition.

This approach really changed the discipline of textual criticism for me, from a dry, rather pedantic evaluation of variant readings into an encounter with a living tradition that continues to the present day. I feel privileged that my career-long journey with the Qumran biblical manuscripts brought me to this rich new understanding of the history and transmission of the biblical text and has given me a glimpse into the mind and techniques of Second Temple period Jewish scribes—fellow scholars from the distant past.