Chapter 1
The Stabilization of Texts in Authoritative Collections
It is widely recognized that the discovery of the Dead Sea Scrolls has changed our understanding of the development of Jewish sacred literature. In particular, the vigorous study of the biblical scrolls from Qumran has brought to light a stage in the formation of the Hebrew Scriptures that was unimagined by scholars before the discovery and subsequent analysis of these texts in the second half of the twentieth century. The wealth of manuscript evidence from the caves near Qumran has revealed that our concept of a âHebrew Bible,â with its closed list of books and fixed textual form, was entirely absent in the late Second Temple period. Rather, this period represents a stage in which the biblical texts were still in the final stages of production, and were decidedly fluid both in terms of their position in the recognized collections of authoritative scrolls, and in terms of their literary editions and precise textual forms.
Indeed, scholars of canonical criticism no longer talk of a definitive event, once thought to have occurred in the first century ce, in which the canonization of the Hebrew Bible was finally decided.1 Instead, they describe a centuries long âcanonical processâ which centred around âa collection of authoritative scriptures.â2 Similarly, the terminology that surrounds the textual form of individual biblical texts has undergone a transformation. Text critics, by and large, have become less concerned with the reconstruction of âoriginalâ biblical books, whether an Urtext (Lagarde) or a pristine edition (Kahle), and tend now towards seeing the mass of textual variants attested in the Scrolls as aspects of the biblical text in transmission.3 This represents a dramatic shift away from a position that understands textual pluriformity as the result of textual corruption, towards one that embraces it as an intrinsic part of textual production.
One of the key findings to emerge over the last several decades of Scrolls research is that there was no standard text of the Hebrew Scriptures at the end of the Second Temple period. To all appearances, there was at this time no singularly transmitted edition of any scriptural text, and no ultimately closed list of texts that constituted âHebrew Scriptureâ as a definitive, delimited object.4 Delineations cited in the ancient literature, such as the âwords of the Prophetsâ and âwritings of David,â do not with any certainty refer to the particular textual forms that modern readers would associate with such titles. And, while the collection referred to from the middle of the Second Temple period as the âlaw of Mosesâ5 in all probability equates, more or less, with the collection of five texts we recognize today as the Torah, it is clear that, with respect to the individual books that make up this collection, it was âthe scroll, not its particular wording or literary editionâ which was authoritative in antiquity.6 Multiple editions of scriptural texts, such as Exodus in the proto-Masoretic and proto-Samaritan traditions, or Jeremiah in the proto-Masoretic and proto-Septuagint traditions, were in concurrent circulation, and were simultaneously authoritative, precisely because the notion of textual authority related to the text as a complete entity, and not to its specific form and content.
The claim that the Torah, or the âlaw of Moses,â was an established and authoritative collection in the literary milieu of the mid-Second Temple period is not without its caveats. Long before the discovery of the Dead Sea Scrolls, scholars recognized that none of the traditionally understood authoritative collections of Torah, Prophets, and Writings were consistently demarcated by ancient authors who referred or alluded to the biblical literature.7 The discussion has since been greatly advanced by intense analysis of the wealth of primary data that the Scrolls represent, and, although the problem has not reached resolution, its expression has become significantly more informed and nuanced in recent scholarship.
Armin Lange has shown that the texts of the Torah are by far the most cited and alluded to of any biblical text in the ancient sources in the second half of the Second Temple period.8 Langeâs findings support those of other scholars, and increasingly the earliest authoritative collections are being viewed as the Torah, Isaiah plus Minor Prophets, and some early editions of the Psalms.9 Of these three collections, the Torah stands out as that typically mentioned first in the lists of authoritative collections.10 Importantly, even though the collection of the Torah is referred to by similar titles throughout the literature examined by Lange, its precise boundaries and textual form remain largely undefined. Towards the end of the Second Temple period, allusions to authoritative literature reflect a growing consciousness of the discrete nature of the authoritative collections, but at the same time multiple forms of the individual texts remained in simultaneous circulation.11
In the minds of most scholars, this remained the case until after the calamitous events of the First and Second Jewish Revolts, when powerful social and political forces brought about the crystallization of the texts according to a single Pharisaic prototype.12 This prototype is generally equated with the consonantal text underlying the tradition of transmission perpetuated by the Masoretic scribes of the medieval period.13 This text is properly called the proto-MT.14 As Moshe Goshen-Gottstein demonstrated in his important study of the evidence for the transmission of the biblical text, the proto-MT is the common text-type underlying all of the manuscript copies from the second century ce onward.15 It is this text-type that has been transmitted with minimal corruption for practically the entire Common Era to date. For this reason, it has been said that the textual history of the Hebrew Bible represents a unique development in the ancient Near Eastern textual corpus.16 No other ancient Near Eastern text appears to have undergone quite the same recensional activity.17
While these assertions may be true, they should not be taken at face value. For a start, certain elements of the stabilization process are detectable in the Dead Sea biblical scrolls in the centuries before the ultimate stabilization of the Hebrew Scriptures. In addition, these assertions raise questions about the method by which scholars can meaningfully say that textual stabilization does or does not occur in other ancient Near Eastern textual traditions. Can the process be detected anywhere else, even if an ultimately fixed text did not eventuate, or has not been preserved?
The above statement that elements of the stabilization process are detectable in the Dead Sea biblical scrolls is in need of qualification. Scholars have suggested that evidence in the Dead Sea biblical scrolls points to the existence of the proto-MT as a defined text-type in manuscripts of a particular palaeographical distinction and quality of production. For example, Lange has identified the proto-MT text-type in manuscripts whose palaeographic dates cluster around two periods: the second century BCE and the first century CE. To my mind, Lange interprets the palaeographical evidence in line with Crossâs established dating sequence too uncritically, but his study remains valuable in its identification of proto-MT clusters in terms of the palaeography of the manuscripts.
Previous to Langeâs study, Emanuel Tov identified clusters of manuscripts reflecting the proto-MT text-type in documents of what he called âde luxeâ manufacture. I will return to Tovâs theory of âde luxeâ manuscripts in the Conclusion, but presently it suffices to note that Lange and Tov have independently shown that manuscripts of the proto-MT text-type shared particular qualities in the late Second Temple period. This argues for the identification of the proto-MT text-type as a discrete text-type in that period. The evidence may be interpreted chronologically, as Lange does, or sociologically, following Tov. What is important is that the stabilization of a single text-type, namely the proto-MT, did not occur ex nihilo in the second century CE, but has its roots in processes that were underway centuries earlier. And, while this earlier stage did not see the production of any exclusive textual prototypes, scholars can identify manuscripts with enough characteristics in common to warrant their association with a single textual tradition.
Framed in this way, is it still correct to assert that the textual stabilization of the Hebrew scriptural texts was a singular phenomenon in the ancient Near East? I suggest that the answer to this question is to be sought in a comparative analysis. My approach to the problem will be to compare the textual transmission of the most established collection of texts, namely the five scrolls of the Hebrew Torah in the late Second Temple period, with texts from the rich corpus of the cuneiform literary tradition that existed in Mesopotamia in the first millennium BCE. The position I take here is that this corpus preserves such an extensive and varied collection of literature that, as a source of comparison against which to stand the data for the stabilization of the Hebrew scriptural texts, the cuneiform sources are unsurpassed by any other literary tradition we know from antiquity in terms of their cultural and temporal proximity to the textual milieu in which the Hebrew scriptural texts were produced.
Large collections of manuscripts, preserving multiple copies of established literary texts, such as the âroyal libraryâ at Nineveh, or the various official and private collections from Ashur, provide a parallel to the large collection of manuscript copies of literary texts from Qumran. Such collections allow a rare opportunity to directly contrast the level of exactitude with which manuscript copies of established literary texts were produced in the cuneiform tradition with those produced in the Hebrew tradition. In a few cases, later copies of the cuneiform texts inform our knowledge of their transmission down into the Hellenistic period. In this way the cuneiform sources from the first millennium BCE constitute an excellent source of data for studying the processes of textual stabilization outside of the Hebrew scriptural tradition.
In his contribution to the 29th supplement of Vetus Testamentum, Shemaryahu Talmon commented on the embarras de richesses of variant readings attested in the most ancient manuscripts of the Hebrew Scriptures. The same is the case when scholars turn to the cuneiform literature, only the problem is multiplied manyfold on account of the vast scope of the material available for analysis. Given the sheer volume of cuneiform manuscripts so far uncovered â which numbers in the tens, if not hundreds, of thousands â it is very difficult to treat any aspect of the corpus exhaustively. Rather, every approach must exercise at least some level of selectivity with regard to the evidence.
For this reason, I have taken a small sample of well-attested texts that can be analysed very closely, rather than attempt an overarching summary of the evidence. This book is therefore a series of case studies, and so can only point to areas deserving further investigation. I have selected five established cuneiform texts that formed part of the âstream of traditionâ in Mesopotamia in the first millennium BCE. These texts have been selected from a range of genres, with the idea that examining a diverse set of literary genres will increase the probability of identifying texts which achieved a significant degree of stability in transmission, should they exist.
The Format of This Analysis
I will begin by forming a methodology for approaching the data based on the tools used and developed by scholars working in biblical text-criticism and related fields. The final methodology decided upon in this study is a synthesis of several models, and is presented in detail in chapter 2. An introduction to the cuneiform evidence follows in chapter 3, along with a discussion of the rationale for the selection of the texts.
Chapters 4 to 9 contain the five cuneiform case studies plus the study of the Dead Sea biblical scrolls that contain any of the five texts of the Torah. Each chapter begins with a brief introduction to the text under scrutiny, followed by as full a description of the individual manuscripts as the related scholarship allows.18 Lists of the manuscripts, with the amount of text preserved in each, are provided to give some impression of how much text is being analysed in each case. There are also tables showing the total number of variants for each of the categories established in chapter 2.
All of the variants attested between any two overlapping manuscript copies are presented in the electronic resources that are available online at www.aneapps.com. This website includes six pdf documents, one for each text examined here, jointly listing all of the variants in all of the texts.19 Every variant is marked by a unique identifier, and throughout this book references will be made to variants by this identifier. The reader can then look up variants of interest in the relevant electronic document. The website also contains a searchable database of the non-orthographic variants in Qumran Pentateuchal manuscripts. This presents the same information as is given in the electronic documents, with the ability to construct advanced searches and graph search results.
Within the electronic documents, variants are organized primarily according to the major categories formulated in chapter 2, but some sub-categorizations, particular to each case study, ...