This week in Canon: An AJR Forum Eva Mroczek imagines the fluidity of sacred writing.
How did the canon of the Hebrew Bible come to be? Rather than surveying the evidence systematically, I focus on two issues that make this a tricky question to ask. Both have to do with how the cultural iconicity of the Bible deeply structures our thinking and distorts our understanding of ancient sources.
The first issue with canonicity is the problem of teleology, where scholars begin with the Bible as the end result and retrace the historical process in early Judaism that led to it. When we go looking for the Bible, we often find it – even if we are looking at something else. It is tempting—in textual studies as in archaeology—to associate our sources with the Bible and to read them all as if they contributed directly to a history of the Bible’s development and use. But while the text and canon of Scripture are the central concerns of most modern scholars, it is not a given that they were always the central concerns of the ancient writers we study.
One illustrative example from the early days of Qumran scholarship is a manuscript scholars named “Proto-Esther.” The fragment’s title suggests that it was a precursor or earlier version of the biblical Esther, which would be significant—Esther was the only biblical text not found at Qumran, and scholars wondered how to account for its absence. But the fragment’s only real link to Esther is its setting in the Persian court. No names or events from Esther appear, and it is written in Aramaic, not in Hebrew. In this case, the impulse to find biblical origins, and perhaps to find traces of the one book “missing” from Qumran, was so powerful that this text was pressed into service to do both.
While few scholars today connect “Proto-Esther” to Esther directly, this biblical way of organizing our evidence is still central. But we cannot take for granted that Second Temple literature was itself consistently organized around the Bible. Kipp Davis’s recent work on Jeremianic literature at Qumran shows this well. We might assume that non-biblical Jeremiah-traditions are resting on the authority of, and interpreting, the biblical Book of Jeremiah. But Davis demonstrates that the Book of Jeremiah itself is not particularly prominent at Qumran: other Jeremiah texts rest not on the “scriptural authority” of the biblical book but on the “reputational authority” of a well-known character—a character who inhabited multiple textual homes. Texts like the Apocryphon of Jeremiah are not interpreting the Bible or directly drawing on its textual authority; they are new texts about a popular figure.
Even apparently obvious references to the Book of Psalms may point to something other than the biblical book. For example, references to “David” in 2 Maccabees and 4QMMT are typically identified with the Psalms, and used to make claims about the Psalter’s authority. But, as I have written in “The Hegemony of the Biblical in the Study of Second Temple Jewish Literature,” David is associated with a range of traditions unrelated to Psalms as we know them. This includes texts connected with liturgical practice and prayer, such as, in the Book of Chronicles, the Temple blueprint and instructional material for liturgical practice. And when David is connected with Psalms, it is with a broad and uncontained genre of revealed prayer, not a specific scriptural book. It is a double leap to use such references as evidence for a biblical book’s authoritative status.
Many scholars have questioned the uncritical assimilation of early Jewish literary references to the Bible. For example, 2 Maccabees 2:13 (Nehemiah “founded a library and collected the books about the kings and prophets, and of David, and letters of kings about votive offerings”) was long read as a classic proof-text for the canonical process. Most scholars, however, have recognized that the verse is not about the canonization of Scripture, but about collecting writings and archival material of various kinds (John Barton includes a discussion in his pioneering Oracles of God). Others have questioned the presumption that references to “torah” must mean the Pentateuch. References to what is written in the Torah of Moses in Ezra-Nehemiah, for example, do not always correspond with actual pentateuchal texts (the law of the wood offering in Neh 10:34 is a key example). And, in terms of the Torah’s bibliographic shape, the first references to Torah as a Pentateuch—specifically a five-volume corpus—are found no earlier than the first century CE (see Philo, On the Eternity of the World 19). The letter of Aristeas—which relates the legend of how the “books of the law of the Jews” were translated into Greek—describes the beautifully crafted scrolls of the Law presented to the king, but does not describe how many or exactly which ones they were. Using such references to make arguments about the status of the Pentateuch as we know it rest on shaky ground, since “Torah” may refer to a broader idea of authoritative law and teaching, rather than a specific textual unit of the five familiar books.
Throwing off what Robert Kraft has called “the tyranny of canonical assumptions" means being open to discovering different ways that writing was organized and understood in antiquity. While our sources may not always tell us the information we want about the development of the Bible, perhaps we can hear them saying something different—and just as interesting—about their own literary culture.
The second problem is how to imagine a world in which the Bible does not yet exist as a concept. (On this see Bowley and Reeves) Some of the characteristics we associate with a biblical canon—stability, unity and completeness, a sense of boundaries—inform our sense of what sacred texts are supposed to be like. This is reinforced by another anachronism: the idea of the book, which we tend to understand as a sealed container that preserves, once and for all, the essence of its author’s creation. The Bible-as-book, as inflected for scholars shaped by a post-print, post-Reformation culture, is so iconic that it has been difficult to imagine that sacred text could be conceived of in any other way.
But the Dead Sea Scrolls revealed a world where sacred writing did not have a stable essence and was not contained in specific boundaries. It seems dissonant to think of texts as unstable, uncontained, and sacred all at once. But could this hint at a different set of features, assumptions, and values that ancient people associated with sacred writing?
Ancient writers themselves give us some clues of their own scriptural poetics—native theories of what sacred writing looks like. One text in a Qumran Psalms Scroll, for example, says David wrote psalms—3,600—and many more songs for liturgical occasions, bringing the total of his compositions to 4,050. Of course, nobody had ever seen a real scroll of 4,050 compositions: the number does not enumerate specific texts but communicates values about them. It stands for cosmic correspondence (the numbers relate to the solar calendar) and a connection with Israel’s heritage (recall that in the earlier 1 Kings 5, Solomon’s compositions number a close second, 4,005). It also celebrates proliferation and abundance—written revelation so vast, so overwhelming, that only parts of it can possibly be available in any existing scribal collection. Ben Sira imagines the tradition that he transmits scribally through metaphors of shining light, growing, blossoming trees, and waterways of overflowing rivers and streams in constant motion. None of these can be contained or fully grasped at any one moment. Jubilees relates a history of revealed text dating back to Enoch, and makes almost every major figure a recipient of written text that is not exactly identified, or that exists only in legend.
Such native reflections on the shape and transmission of traditions show that it is not enough to say that the canon was not yet fully fixed during the Second Temple period. That suggests that we had something shaped essentially like a Tanakh, but with some blurring around the edges. But before the category of the Bible was the mental container in which texts were kept, sacred writing could take very different shapes: it was prolific, not contained in a single corpus, but found in multiple locations, only some of them readily accessible.
To be sure, something happens in the first century. In Josephus and 4Ezra, we find the first mentions of a particular number of books of scripture: twenty-two and twenty-four. This is evidence of a new concept for imagining sacred writing. But is it what we usually think of as “canon”? Neither source tells us precisely which texts are included; the numbers are clearly symbolic, representing the number of letters of the alphabet; and both authors recognize the existence and authenticity of other revealed writing outside these bounds. Such numbers are typological, representing the values of completion and coherence, and they must precede (rather than reflect and enumerate) a particular list of specific texts. To determine how these references relate to the emergence of a fixed Jewish canon will require more work—both on the meanings that numbers have in religious discourses and on what kind of boundaries canonical lists are really supposed to draw.
Some of these ideas are examined in more depth in my book, The Literary Imagination in Jewish Antiquity (New York: Oxford University Press, 2016), which will appear in April 2016.
Eva Mroczek, UC Davis