David M. Carr on Reinhard Müller, Juha Pakkala, and Bas ter Haar Romeny’s Evidence of Editing and Juha Pakkala’s God’s Word Omitted

Back in the 1700s, when scholars formed their first theories about the formation of the Hebrew Bible, they had few comparison points. The vast bulk of Egyptian, Mesopotamian, and Ugaritic source documents were still undiscovered, let alone undeciphered. The revolutionary discovery of a mass of early Jewish biblical and non-biblical manuscripts near the Dead Sea in Israel still was far in the future.

Within this environment, the main data that informed theories of the Bible’s formation were found in the Bible itself. The biblical text repeatedly describes events that would have occurred just once, such as the initial creation of humans (Gen 1:26-27; 2:9-22) or God’s announcement to Noah of the flood and instructions about bringing animals into the ark (Gen 6:13, 17-20//7:1-4). In some cases, the doubled narratives contradict each other in crucial details, such as God telling Noah to bring two of each kind of animal into the ark in Gen 6:19 but seven pairs of clean animals and birds in Gen 7:2-3. These are just a few classic examples of doublets, contradictions, and other incongruities that have led scholars, for centuries, to theorize that the Bible was created over time, through the combination of sources and/or layers of scribal insertions and expansionary layers.

What was once thought to be the case, based on internal clues, we now know to be the case, because we have actual copies of multiple editions of the books that are now in the Hebrew Bible. Starting in the 1800s scholars started to gain access to a broad variety of other ancient texts written before or around the time of the Bible. Massive libraries of texts written on clay tablets, mainly in ancient Mesopotamia, have been discovered and gradually published over the last century and a half. Similarly large corpora of texts from other cultural domains were discovered and deciphered as well, from the vast Egyptian cultural heritage to more modest-sized textual collections in Syria, Israel, and neighboring lands. Also important was the collection of ancient Jewish texts found near the Dead Sea in the mid-twentieth century and gradually published in the half-century following. This collection contained not only biblical manuscripts a thousand years older than the oldest manuscript discovered to date, but also multiple editions of other Jewish texts never before known to the scholarly world.

Early on, these ancient texts revolutionized our understanding of ancient Near Eastern religions and the early history of Judaism and Christianity. Only in the last decades, however, have these textual discoveries started to transform our understanding of the Hebrew Bible’s growth. This line of inquiry was anticipated by a classic Journal of Biblical Literature article on Tatian’s Diatessaron by George Foote Moore in 1890, but most would mark a new beginning with Jeffrey Tigay’s 1985 The Evolution of the Gilgamesh Epic, followed soon after by an excellent collection of essays edited by him on Empirical Models for Biblical Criticism. What made these works different was their study of places where we actually seemed to have copies of successive stages in the growth of ancient Near Eastern texts. The Gilgamesh epic studied by Tigay (and others) was one of the best documented examples, but many more soon emerged, such as the other Mesopotamian myths and royal inscriptions studied in Hans Jürgen Tertel’s 1994 volume, Text and Transmission. Gradually scholars used these documented examples of textual growth to explore ancient scribal practices of textual conservation, copying, revision, and combination.

The Dead Sea scrolls were particularly significant because they were Jewish texts. We knew how the Samaritan Pentateuch deviated from the traditional Jewish Masoretic text but attributed these differences largely to the peculiarities of the Samaritan sect that preserved it. We also knew of major differences between the Greek Septuagint and the Masoretic text, but many maintained that most large differences could be explained as abbreviations and other shifts introduced by the Greek translators who created the Septuagint. The Dead Sea manuscripts provided a new perspective. We now realize that many of the large deviations found in the Samaritan Pentateuch were not specific to the Samaritan sect but were part of a rich and varied tradition of expansionist conservation of the Pentateuch in certain streams of early Judaism. We also know that some of the most important differences between the Greek Septuagint and the Hebrew Masoretic tradition do not originate from the Greek translators but go back to differences between early Hebrew witnesses, so that the Greek Septuagint often is a witness to an ancient Hebrew textual tradition quite different from the Masoretic text. As a result of these discoveries, scholars now can draw on a wide range of early Jewish witnesses, such as the Dead Sea scrolls and Septuagint, to explore the growth of the Hebrew Bible. These ancient Jewish manuscript witnesses show that the Hebrew Bible, even in the centuries just before the Common Era, was still being changed substantially by some scribes. Moreover, they show how such scribes typically worked.

This evidence has been important in my own work, starting in a preliminary way with the introduction to my 1996 book Reading the Fractures of Genesis, but particularly in my books Writing on the Tablet of the Heart: Origins of Scripture and Literature and The Formation of the Hebrew Bible: A New Reconstruction, along with a series of articles on method in the study of intertextuality and use of documented cases for study of the formation of the Bible. This line of inquiry has blossomed over the last five years to a subfield in biblical studies with significant works by (among others) Molly Zahn, Russell Hobson, Sara Milstein and Daniel Fleming, and Paul Delnero.

※

Juha Pakkala and Reinhard Müller represent the leading European scholars pursuing this fruitful research trajectory, and their respective volumes, God’s Word Omitted and Evidence of Editing, bring together initial findings. God’s Word Omitted is authored by Juha Pakkala alone, albeit with collaboration from Reinhard Müller and others, and published in 2013. Evidence of Editing is a joint work by Juha Pakkala with Reinhard Müller and Bas ter Haar Romeny, published a year later. Together, these two books provide a very useful synthesis of years of work by these scholars on potential documented examples of growth of the Hebrew Bible as well as a vivid demonstration of the effect, within scholarship, of using a similar body of material (in this case documented cases of editing) to address two different scholarly publics, one in North America (Evidence of Editing) and one in Europe (God’s Word Omitted).

In order to understand the context of these two works, one must know about recent developments in European scholarship, especially German scholarship, out of which they emerge. In the last several decades, numerous branches of Hebrew Bible scholarship in Europe, especially Germany (and related programs), have concluded that larger and larger blocks of the Bible are the creations of scribes working in the post-exilic period. Pakkala, for example, argued (in an earlier work) that the entire account of Josiah’s centralization of worship in Jerusalem in 2 Kings 23 is a post-monarchic creation with no historical value whatsoever. Though Pakkala, Müller, and Romeny still believe that remnants of early Israelite concepts can be located in parts of the Hebrew Bible, such as the Psalms, they believe that such concepts can only be reliably investigated through careful literary-critical analysis of the multiple editorial or redactional layers of such texts.

Meanwhile, North American biblical scholarship has tended toward more confidence in the antiquity and historical usability of the biblical text, combined with skepticism about the feasibility and worth of complicated literary reconstructions of multiple editorial layers in the Bible. Though most biblical scholars outside Europe agree that the Bible was formed over time, there is disagreement about how much we can know about it. Ehud Ben Zvi, an Israeli scholar teaching in Alberta, Canada, is so skeptical about reconstructing editorial layers that he opts to treat the Hebrew Bible as a whole as a product of the Persian period, the period just before the appearance of the earliest biblical manuscripts. As a scholar in the United States, I am more confident about reconstruction of earlier layers of the Bible, though I admit (in a 2011 book quoted in Evidence of Editing) that it is impossible “to reconstruct each and every stage of that [textual] growth” and that therefore “more complicated reconstructions of textual prehistory have not stood and will not stand the test of time.” Finally, some other scholars, such as Tamara Cohn-Eskenazi (also educated and based in the United States), simply choose an explicitly literary approach and focus exclusively on the shape and characteristics of the present form of biblical books.

Pakkala, Müller, and Romeny’s confidence in detailed, literary-critical reconstruction of editorial layers puts them at odds with these streams of North American biblical scholarship. Early in Evidence of Editing they criticize Tamara Cohn-Eskenazi for ignoring “questions about the prehistory of the final texts and the editorial processes that the empirical evidence attests to.” Similarly, they respond to Ehud Ben Zvi’s skepticism about literary reconstruction by insisting that “the attempt to detect the literary history of these books should nevertheless be made, and there are many texts where editorial processes left clearly discernible traces.” Finally, they respond to my arguments against “more complicated reconstructions of textual prehistory” that “the underlying skepticism about the general possibilities of literary-critical reconstructions, evident in Carr’s approach, should be rejected.”

The rest of Evidence of Editing buttresses these arguments against these and other North American scholars with documented examples of editing. The first sentence of the book states its main purpose: “This book seeks to demonstrate that substantial editing took place in the history of the Hebrew Bible.” The subsequent paragraph goes on to elaborate particularly European ideas: “it can reasonably be assumed that editorial reworking of the Hebrew Bible continued unabated for centuries before the texts gradually became unchangeable” and “editorial modification was the rule rather than the exception.”

They make their case by gathering fifteen sets of examples of editing in the Hebrew Bible, most of which have already been studied by other scholars in other contexts. Many of the examples show how the scribes who produced the Bible sometimes added material to it, including some minor additions that would have been difficult to recognize without documentation of the earlier version. These include additional specifications of sacrifices in Lev 17:4 and offerings in the Num 28:16-25 Passover law (cf. Lev 23:5-8), multiple adaptations of the spy story in Numbers 13-14 to better agree with the review of that story in Deuteronomy 1:19-45, an additional few lines in a Qumran manuscript (4QSam^a) to smooth the transition from 1 Samuel 10 to 11, and some minor additions to be seen in both Ezra-Nehemiah and its parallel in 1 Esdras. Sometimes, the authors trace a succession of small additions to an earlier text, such as in narrations of the killing of Gedaliah, the Babylon-appointed governor of Judah after 586 BCE (2 Kings 25:25 and two versions of Jer 41:1-3) or the Chronicler’s treatment of Joash’s reign (2 Kgs 12:1-22; 2 Chr 24:1-27. Finally, the Chronicler’s treatment of Joash’s reign in 2 Chr 24:1-27 does not just feature minor additions, but substantially rewrites and even subtracts parts of its source in 2 Kgs 12:1-22.

A few cases where scribes may have subtracted material from their sources add to their evidence, but the examples they choose are questionable. Pakkala, Müller, and Romeny argue (unconvincingly in my view) that the long version of the law regarding refugee cities found in the Masoretic text of Joshua 20 was abbreviated in the Hebrew precursor to the Greek translation of the chapter in order to better align it with the law about refugee cities found in Num 35:25-28. Following the majority of scholars on this point (cited by the authors themselves), it is more likely that a scribe added elements from the nearby law about refugee cities (Deut 19:4-6) into Josh 20:4-5 so that it coordinated with both Num 35:25-28 and Deut 19:4-6. Joshua 20 is thus best listed as an example of editing through addition rather than subtraction.

In some cases, the authors reconstruct more complex and often unreconstructable chains of editorial changes. They note, for example, that the latter narrations of the burning of Jerusalem in Jer 39:8-9 and 2 Chr 36:18-20 were so thoroughgoing in their revisions of their sources in 2 Kings 25:8-12 and Jer 52:12-16 that it would have been impossible to reconstruct those sources if we did not have them separately. Similarly, they argue that it would be extremely difficult for a scholar to independently reconstruct material reused from the Isaiah 15-16 Moab oracle in Jer 48:29-38a or the exact contours of material adapted from two psalms in Psalm 108 (evident from the parallel of Ps 57:8-12 in 108:2-6 and the parallel of Ps 60:7-14 in 108:7-14) without copies of the reused sources.

In several cases, however, Pakkala, Müller, and Romeny maintain that additions documented by textual traditions could have been recognized, or even have been previously recognized, by scholars without the aid of separate textual evidence. The Greek versions of the book of Esther feature a number of major expansions, some probably composed in Hebrew and others in Greek, that are recognizable as secondary by the way their religious-pious focus contrasts with the secular cast of the rest of the book. They also argue that a Qumran manuscript, 4Q49 (=4QJudg^a) seems to preserve a version of Judges that does not include an account of a prophetic speech (Judg 6:7-11), in this case confirming a supposition by earlier biblical scholars that Judg 6:7-11 was an addition to its context. Finally, they suggest that the oracle to Solomon found in the Masoretic version of 1 Kgs 6:11-14, but absent in several Greek manuscripts, is also recognizable as secondary by the resumptive repetition in 6:14 of 6:9 and its mixture of non-Priestly, semi-Deutoronomistic language (1 Kgs 6:12b, with parallel in Gen 26:3) and Priestly (1 Kgs 6:13a, with parallel in Exod 29:45a) language.

Aside from these three examples of potentially reconstructable editorial additions, Evidence of Editing generally fails in its goal to show the feasibility of accurate and detailed reconstruction of editorial modifications. Instead, the authors competently document that editors and later authors (e.g. the Chronicler) often modified their sources through strategic, often indetectable minor additions (from a word to a few verses), in addition to rewriting their sources so thoroughly that the precursor texts would not be reconstructable. They assert that the “overall methodological skepticism toward the possibilities of reconstructing the literary history of the Hebrew Bible, advocated by some scholars, cannot be justified.” Yet many, if not most, of the cases surveyed by the authors would not, by their own admission, have been reconstructable without documented evidence of the earlier literary versions.

One primary target of Evidence of Editing is use of the Masoretic form of the Hebrew text for historical reconstruction. The main example they use to illustrate the problems with doing this is the case of the (secondary) labeling of Ezra as a “high priest” in 1 Esdras, rather than as “scribe” and “priest” in parallels in Nehemiah. This particular case is problematic because it is not certain that this upgrade of Ezra’s status in 1 Esdras occurred on the level of the Hebrew original of 1 Esdras or its Greek translation. Certainly one must reckon with the possibility, even probability, of editorial shifts in biblical texts used for historical reconstruction (including the precursor to 1 Esdras). Evidence of Editing however, does not succeed in using documented examples of editing to show that such shifts are so widespread and substantial as to undermine historical reconstructions that disagree with the authors’ historical models, and that such editorial shifts can be reliably corrected for through being identified through accurate literary-critical reconstruction.

It is doubtful that Evidence of Editing succeeds in documenting that “editorial reworking of the Hebrew Bible continued unabated for centuries” and that “editorial modification was the rule rather than the exception.” After all, Pakkala, Müller, and Romeny discuss only fifteen or so cases in a vast corpus of Hebrew texts spanning twenty-plus books. To be sure, they know that their discussion could have been expanded with other examples of the same editorial phenomena, yet even a more comprehensive survey of evidence would still only encompass a minority of the chapters and verses of the Bible.

In the end, Evidence of Editing includes no such broader survey of editing, and it fails to establish that the editing found in early biblical manuscripts (all from the Hellenistic and Roman periods) were also characteristic of earlier periods. Instead, we are left with the authors’ assertion that “we can assume that these documented cases attest to merely a fraction of the actual changes that have taken place in the transmission of the Hebrew Bible” and that it can “reasonably be assumed that editorial reworking of the Hebrew Bible continued unabated for centuries.” Where once we had “evidence,” now we have multiple “assumptions,” and we are right back to the particular contemporary scholarly cultures of biblical scholarship (in North America, Israel and Europe), with their varied assumptions about literary growth in the Bible and methodology for reconstructing it.

※

The effect of scholarly culture and addressed audience becomes evident when examining Pakkala’s God’s Word Omitted. While Evidence of Editing, published by the U.S.-based Society of Biblical Literature Press, appears to be aimed at work by North American colleagues who disagree with European redactional approaches and historical models, God’s Word Omitted, published by the European press Vandenhoeck & Ruprecht, seems more directed at critiquing Pakkala’s European colleagues, many of whom confidently reconstruct multiple redactional layers in biblical texts on the assumption that prior stages of the text were perfectly — or virtually perfectly — preserved. To be sure, many of these European scholars allow for more radical changes at an earlier, “compositional” stage of adaptation of source texts. Nevertheless, most hold to the general assumption of careful scribal preservation of precursor texts, particularly for later redactional (editorial) stages in the development of an increasingly authoritative text. This assumption of the lack of scribal omissions in the transmission of the Hebrew Bible (at least in its last stages) has become an important methodological axiom on which whole schools of redactional analysis, particularly in Europe, are based. Only if scribes carefully preserved earlier stages of text growth could such scholars be able to reconstruct layer upon layer of redactional additions.

God’s Word Omitted uses ample documented examples of textual growth to undermine this methodological axiom. Some of the examples of what Pakkala calls “radical editing” are widely agreed upon, such as the radical revision seen in the laws of Deuteronomy when compared with their likely precursors in Exodus 20-23, the further revision of Deuteronomic law seen in the later Holiness Code materials of Leviticus 17-26, minor ideologically-driven omissions and more radical rewritings of Samuel-Kings in 1-2 Chronicles, and radical reuse and frequent omission of biblical source texts in early Jewish texts like Jubilees and the Temple Scroll. He includes a number of excellent examples where text witnesses appear to document the elimination by scribes of aspects of biblical texts that were ideologically or theologically offensive to them. These cases are more persuasive, in my view, than Pakkala’s attempt to argue that the Greek tradition for Esther eliminated a number of textual elements because of concerns about repetition and contradiction or that the author of the precursor to 1 Esdras eliminated Nehemiah in order to focus more attention on a Davidide, Zerubbabel. Nevertheless, whether or not one agrees with all of Pakkala’s cases, he does provide substantial documentation for the idea that ancient scribes sometimes modified their sources so much, including through omission, that reconstruction of those sources would be impossible absent actually having them in hand. This is quite significant for those European approaches that assert that biblical scholars can and must reconstruct the prehistory of the biblical texts they study.

God’s Word Omitted concludes with some fairly substantial proposals regarding methodology in literary criticism. Pakkala argues persuasively for rejecting the typical view of a linear movement from radical revision of sources early in a text’s history (a compositional stage) followed by pure preservation of the text and editorial expansion of it as it gains more authority (a redactional stage). Instead, he borrows the model of punctuated equilibrium from evolutionary biology to suggest that ancient text growth may have alternated back and forth between periods of relative preservation/expansion of texts and times of crisis that precipitated a paradigm shift requiring more radical forms of editing.

With this, God’s Word Omitted, addressed in particular to European colleagues overconfident about literary-historical reconstruction, ends in a similar place to my own methodological position — a position criticized in Evidence of Editing — that “more complicated reconstructions of textual prehistory have not stood and will not stand the test of time.” As Pakkala puts it, “When countless redactional layers are reconstructed, the reconstruction should, at most, be seen as an abstraction of the text’s development.” In Evidence of Editing, Pakkala noted critically that “Carr does not discuss the fact that many, if not most literary-critical scholars distinguish between the composition and redaction phases.” Nevertheless, by the end of God’s Word Omitted, Pakkala himself has noted that literary growth cannot be separated into distinct sequences of “composition” and “redaction” phases. Whatever the differences between Pakkala and myself on the extent of editorial revision in the Bible, we agree more than it might seem on the character and (un)reconstructability of editorial changes, and our (relative) agreement has been informed by our study of documented cases of ancient editorial revision.

Study of actual evidence of editing, then, may contribute to convergence between the different cultures of biblical scholarship in Europe and North America. Though Pakkala and (in Evidence of Editing) Müller and Romeny inveigh against North American colleagues for their skepticism about exact reconstruction of editorial layers, Pakkala himself uses documented examples of editing to argue for more such skepticism in God’s Word Omitted. Pakkala and colleagues remain more confident about literary reconstruction than most North American colleagues, including myself. Moreover, their assumptions about Israelite history remain quite different from scholars like myself, whose conclusions (my own characterized in some American discussions as “European”) Pakkala deems “rather conservative.” Nevertheless, a review of these two works shows both the importance of scholarly audience in framing an academic message, and the possibility that a body of data — in this case documented cases of literary growth — might undermine divides between scholarly cultures on what is methodologically possible and necessary.