Julius Wellhausen goes to Copenhagen: Towards a more progressive commentary on Exodus

Introduction

A survey of commentaries on Exodus reveals a striking uniformity in outlook. Commentaries on the book, almost without exception, are variations on a theme: that the Pentateuch was composed over a period of centuries, by several editors, drawing on four distinct, independent sources. These commentaries do differ over minor points—exactly when these four sources were written, the extent to which those sources relied on earlier oral traditions, and even whether we should properly speak of a Pentateuch or a Hexateuch—but no alternative theory of pentateuchal composition is seriously entertained.

This is surprising given that, over the last thirty years, other writers in pentateuchal studies have contributed new and compelling theories as to how the Pentateuch assumed its final form. Many of these writers have supported and developed their ideas extensively and asked probing questions of the validity of the documentary hypothesis. As a result, the documentary hypothesis no longer seems as unassailable as it once did and there is a rich seam of alternative thought available for commentators to mine. And yet, no commentary writer has yet broken ranks to do so.

Although I believe the documentary hypothesis has significant weaknesses, my primary aim for this essay is not to assess that debate. Nor will I speculate on why commentators seem so reluctant to abandon the documentary hypothesis. Instead I will focus on what is, to my mind, a more interesting discussion: how the disconnect between commentaries and revisionist scholarship is affecting the field of pentateuchal studies.

After briefly sketching out the most prominent theories of pentateuchal composition, and demonstrating the commentaries’ very limited engagement with these theories, I will highlight three effects arising from this state of affairs: i) both conservative and revisionist positions are becoming ever-more entrenched; ii) commentaries are imposing unnecessary and detrimental limits on themselves and iii) confessing readers of commentaries are being sold short. Finally, I will suggest a new way forward for an Exodus commentary: a commentary which would take seriously alternative theories of how the Pentateuch came to be, embrace a range of opinions, grapple with complex theological questions, and still draw out aspects of the book of Exodus of relevance for modern readers.

Wellhausen’s Documentary Hypothesis: A Brief Summary

The documentary hypothesis for the composition of the Pentateuch has shown impressive endurance. Although no longer commanding the widespread acceptance it enjoyed in the early-to-mid twentieth century, it is still defended vigorously in some quarters. The tenacity of the hypothesis is all the more remarkable, bearing in mind that its adherents cannot reach consensus on a number of issues relating to its precise nature.

The hypothesis has its roots in the eighteenth century; with Jean Astruc’s attempts to prop up the traditional view of Moses as author of the Pentateuch by suggesting that Moses worked from two source documents, thus accounting for the contradictions and repetitions within the finished text. During the nineteenth century, this theory was refined by Kuenen, De Wette and others but achieved widespread recognition in the work of Julius Wellhausen. ¹ It was under Wellhausen that the documentary hypothesis began to assume the form by which it is now recognised: that there were four major sources—denoted J (Jahwist or Yahwist), E (Elohist), D (Deuteronomist) and P (Priestly)—woven together to form the Pentateuch. ² The idea of Mosaic authorship was eventually abandoned almost universally amongst historical-critical scholars.

That said, the early formulations of the documentary hypothesis were inextricably linked to the defence of Mosaic authorship. It is therefore an inherently conservative position; an attempt to provide a credible framework for the historical reliability of the Pentateuch. The hypothesis is further underpinned by an assumption that the Pentateuch’s authors were seeking to create a religiously normative piece of literature; a work which represented Israel’s lived experiences of the deity. ³

Proponents of alternative theories of pentateuchal composition (including Lemche, Thompson, Davies and others) have been grouped into the ‘Copenhagen school’; labelled ‘minimalist’ or, slightly more kindly, ‘revisionist’ by their opponents. Broadly speaking, revisionist scholars tend to start from a point of scepticism and accept the biblical version of events only when it is supported by extrabiblical evidence. In contrast, the documentary hypothesis starts from what might be referred to as a ‘maximalist’ position, assuming the biblical accounts are generally reliable and deviating from that only when it is necessary in order to accommodate extrabiblical evidence which contradicts the biblical account.

Wellhausen was one of the first to present the documentary hypothesis in a simple, coherent and elegant way in his 1878 work, Geschichte Israels I. For this reason, his work deserves special attention as representative of the documentary hypothesis in its early incarnation. Wellhausen relies largely on stylistic and linguistic grounds for ascribing portions of the pentateuchal text to different sources. J, for example, is apparently in step with the language, style and themes of other literature thought to have been composed during a purported united monarchy period, so Wellhausen reasons that J would have been written during the same era and is the earliest of the four pentateuchal sources. P is distinct from J in terms of language and style and also in its thematic focus on the Temple cult and the Mosaic law. ⁴

Wellhausen is quite candid about some key unanswered questions relating to his theory. For example, while positing that P was the last of the four sources to be written, he acknowledges problems in dating it and mentions disputes amongst his colleagues as to which sources built on which. ⁵ However, Wellhausen also articulates important principles behind the documentary hypothesis which are still widely endorsed by its adherents today. Apart from the idea of four major source documents, each written by an individual author, Wellhausen argues that these sources were deliberately compiled to form the Pentateuch ⁶ as we know it: ‘. . .continuous narratives which can and must be understood each by itself are woven together in a double or threefold cord’. ⁷ He therefore implies the existence of a redactor or redactors, making ‘additions and supplements . . . on the most extensive scale’. ⁸ Also key is Wellhausen’s suggestion that the later sources were informed by the earlier ones. Above all, Wellhausen makes it clear that he believes the Pentateuch (or Hexateuch) was not composed by a single author or group of authors, over a short period of time. Rather, it ‘came into existence during a longer period, and passed through various phases’. ⁹

As the twentieth century progressed, Wellhausen’s hypothesis gained more and more support, until its accuracy was largely assumed, and opinion verged on consensus. Since the mid twentieth century, his theory has begun to fall out of favour again, with alternatives gaining traction amongst Bible scholars. However, the documentary hypothesis is still seen as axiomatic by some: not least commentators on the books of the Pentateuch.

Recent Commentaries and their Engagement with Wellhausen’s Theory

Despite the growing body of work supporting alternative theories of pentateuchal composition, the documentary hypothesis still looms large over the Exodus commentaries written within the last twenty years. For example, Scarlata and Smith endorse the documentary hypothesis and offer no comment on alternative theories. ¹⁰ Bruckner, while lamenting that questions of source criticism have detracted from the interpretation of Exodus in its final form, still views the documentary hypothesis as the most likely means of its formation. ¹¹ And, in the commentary accompanying his magisterial translation of the Pentateuch, Alter not only mounts a spirited defence of the documentary hypothesis but even attacks the ‘extremists’, who propose a Persian or Hellenistic date for its composition. ¹²

There are, admittedly, a few recent commentators who demur: the robustly conservative Stuart insists on Mosaic authorship, and Enns sidesteps the question of authorship altogether, expressing dissatisfaction with the documentary hypothesis but claiming that, to the reader of Exodus, the book’s divine inspiration is more important than its human author. ¹³

Carol Meyers is one of the more progressive of Exodus commentators. For example, she argues that Exodus should not be read as a literal historical account, and advances some innovative ideas about the nature of the Pentateuch, including suggesting that Moses is more a figure of Israel’s collective memory than a depiction of a historical individual. ¹⁴ She is also mindful of the important contributions to pentateuchal studies from revisionist scholars, and offers an even-handed and comparatively lengthy (two-and-a-half pages) treatment of their conclusions. ¹⁵ But despite her open-minded attitude on these points, Meyers still seems unable to turn her back on the documentary hypothesis. She envisages numerous authors and redactors, drawing on a range of source material of disparate genres. ¹⁶

Close attention to three particular publications should illustrate the enduring influence of the documentary hypothesis over Exodus commentaries. Below, I summarise three recent commentaries and their interactions with the documentary hypothesis and alternative theories. I focus on these specific commentaries because they were all published more than fifteen years after John van Seters’ In Search of History (1983). All three writers could therefore have drawn on at least one of the most influential book-length works advocating a departure from the documentary hypothesis. These commentaries are also exhaustive in length (Propp’s is shortest at 680 pages but addresses only Exodus 1º18), and they all belong to influential commentary series, so offer a useful litmus test of the extent to which alternatives to the documentary hypothesis have percolated into the mainstream of the commentary genre.

In his Anchor Yale Bible commentary, William Propp gives every impression of being a thoroughgoing Wellhausen disciple. He envisages several writers for the Pentateuch, based on the repetitions and internal contradictions it contains, although he is not precise on when he believes J was written—merely that it was combined with E following its arrival from the Northern Kingdom in the late eighth century. ¹⁷ Further, he envisages multiple ‘editors’ who combined the component sources at various stages, including a ‘Redactor^JE’ who combined J and E, and a ‘final Redactor’, who performed the last round of editing work in shaping the text of the Pentateuch into the form in which we now find it. The final Redactor was, apparently, a scribe but nothing more: certainly not any kind of ‘genius’. ¹⁸

So invested is Propp in the documentary hypothesis that he favours not only source analysis but also redaction analysis; positing what exact contributions Redactor^JE and the final Redactor made in combining the sources and refining the resulting text. ¹⁹

Propp’s engagement with alternative theories of Pentateuchal composition is minimal. Although his commentary is extensive, it has no room for descriptions or critiques of these alternatives. Works by van Seters and Thompson appear in his bibliography, ²⁰ but only those works which support—or appear to support—the documentary hypothesis. Five of van Seters’ publications are mentioned, but not In Search of History. Three of Thompson’s works are included, but Early History of the Israelite People (1992) is overlooked, and Thompson is directly referenced only in relation to the structure of the text of Exodus and the exact translation of particular Hebrew phrases. Indeed, Propp cites Thompson’s 1987 essay in support of the purported existence of two sources for Exodus 5, but does not mention that his 1992 volume disavows the documentary hypothesis. ²¹ The outcome is a commentary that endorses the documentary hypothesis without any apparent caveat and presents no clear alternative to Wellhausen’s theory.

‘I . . . depart from the documentary hypothesis’, insists Thomas Dozeman (in the Eerdmans Critical Commentary). ²² However, despite expressing reservations about some aspects of Wellhausen’s theory, Dozeman remains conservative in his views on pentateuchal composition. Although he references John van Seters to an extent that indicates a strong influence on his work, Dozeman does not engage with van Seters’ theory of a late Yahwist, nor other scholars’ theories of Persian or Hellenistic dating. The Pentateuch is, Dozeman affirms, the product of four written sources (composed between the ninth and fifth centuries bce), redacted at various stages within that period. Echoing Propp, he holds that the redactors themselves did little, if anything, in shaping the content of the Pentateuch: they were ‘editors’ who merely combined the sources into one document. ²³

Dozeman shows refreshing candour in outlining several weaknesses in the documentary hypothesis, specifically: the doubt over whether an E source ever existed, the debate over the dating of J and its relation to D, and questions over whether P included legal material as well as narrative. ²⁴ But he still appears to favour the documentary hypothesis as the most convincing theory of pentateuchal composition, highlighting the repetitions and contradictions within the text of Exodus and arguing that these imply multiple authors. ²⁵

He is keenly aware of the slight differences in understanding of the designations J, E and P among pentateuchal scholars. (For example, van Seters appropriates the ‘Yahwist’ for a theory entirely distinct from the documentary hypothesis.) Dozeman therefore employs the term ‘Non-P’ to describe what is generally known as JE, and ‘P history’ for the narrative elements of P, distinct from legal material which may or may not have once formed part of a P source. ²⁶ His exactness is admirable but unfortunately only serves to further confuse an already complicated subject.

Finally, Dozeman differs from other adherents to the documentary hypothesis in dating the completion of J and E comparatively late—the post-exilic period. ²⁷ (It would have been helpful if Dozeman had been more specific about the time scale he envisages: ‘post-exilic’ could mean the early fifth century, the late fourth, or anything in between.) For all his caution, though, Dozeman still appears to endorse the documentary hypothesis as the most persuasive theory available, and his engagement with alternative theories of pentateuchal composition is very limited. He references van Seters, Lemche and Thompson, especially in the section of his introduction which addresses ‘Ancient History Writing and Exodus’, ²⁸ but he does not offer even the briefest description of their views on pentateuchal composition.

Desmond Alexander’s very recent contribution to the study of Exodus (Apollos Old Testament Commentary, 2017) focuses on the final form of the book, reasoning that it is neither possible nor necessarily helpful to reconstruct the exact provenance of each literary unit. Alexander attributes the definitive form of Exodus to a single redactor-editor, who drew together this disparate pre-existing material and shaped the text into a coherent whole. ²⁹

He is wary of the documentary hypothesis and acknowledges the extent of opposition to it, ³⁰ but is equally wary of revisionist alternatives. He engages briefly with Thompson and with Lemche but summarily dismisses them both. ³¹ Indeed, Alexander concludes that, ‘Almost nothing can be said with any certainty about the authorship and date of composition of Exodus.’ ³² Alexander’s caution impedes him from making any clear suggestion for the time and place when the ‘author’ (as he describes the editor) was operating. Although he is wary of the documentary hypothesis in its traditional form, he still sees the text as composite and he rules out extensive authorial work on the Pentateuch following the exile, ³³ so perhaps we can infer that he tends towards a similar dating schema to the documentary hypothesis, with a single editor-redactor operating during the exile. But none of this is explicit. And although Alexander distances himself from the documentary hypothesis, the idea of the Pentateuch as a composite work, shaped by a redactor, retains important similarities to it.

Alternative Theories

While commentators on Exodus almost uniformly endorse the documentary hypothesis (albeit with some reservations), there is much more diversity in opinion amongst scholars outside of the commentary genre. These alternative theories range from extensive revisions of the documentary hypothesis (what might be described as ‘documentary hypothesis 2.0’) to arguing that the whole Pentateuch was composed as late as the Hellenistic period.

John van Seters, while distancing himself from the documentary hypothesis, still advocates a ‘Yahwist’. For van Seters, the Yahwist is not the author of one of four source documents for the Pentateuch, but a virtuoso editor, gathering myths, legends, genealogies and other source material and redacting it all into the Pentateuch as we know it. ³⁴ The Yahwist is thus, ‘a historian of quite remarkable scope’. ³⁵

However, van Seters himself never uses the word ‘editor’ to describe the Yahwist. Editors are a hallmark of the documentary hypothesis; redactors who compiled existing material, using a copy-and-paste technique and minimal written additions to smooth over breaks in the text. An editor may correct errors from his sources, but will add little, if any, new material of his own. van Seters suggests that the Yahwist is far more than an editor of this kind: there is a creative and ‘compositional’ aspect to his work. ³⁶ For van Seters, the Yahwist is an ‘antiquarian’, in the same mould as Herodotus and Dionysius. ³⁷ The Yahwist, as an antiquarian, has not only combined disparate sources (both from Israelite tradition and the wider ANE context) but changed and developed these sources and invented material of his own to fit the major themes he wanted to emphasise. ³⁸ The Yahwist thus created a history: a story to help his own people understand their origins and identity. ³⁹ This is a significant shift from the documentary hypothesis. Instead of the Pentateuch answering theological questions and having a religiously-normative purpose, van Seters understands it as focusing on questions of Israel’s origin and identity. He sees it as primarily a national history.

As noted above, van Seters departs from the documentary hypothesis by suggesting that elements of it may be completely fabricated. In van Seters’ 1994 volume, concentrating on Exodus specifically, he uses the language of ‘historiography’, describing the book as a collection of traditions, structured so as to form a biography of Moses and story of Israel’s origins. ⁴⁰ The category of historiography in itself implies at least some historical basis but, in van Seters’ understanding, this category does not preclude an element of invention on the Yahwist’s part. For example, Israel’s desert wanderings are apparently ‘fictitious’, and the Reed Sea episode is supposedly based on the very similar description of the crossing of the Jordan, composed earlier as part of the Dtr. ⁴¹

van Seters seems to have been influenced in his more recent writings by Halpern and Thompson, who also propose antiquarian scribes as a model for biblical writers. ⁴² van Seters’ antiquarian Yahwist, preserving and combining his source material, while creating some of his own, echoes Halpern’s idea of the pentateuchal redactor (not the ‘Yahwist’) working to ‘preserve the flow of the received materials [. . . which] did not preclude the introduction of fable’. ⁴³ And van Seters appears indebted to Thompson for his conclusion that the Yahwist assembled traditions in such a way as to speak into the deepest concerns of his own time and context. ⁴⁴

In van Seters’ line of thought, the time scales proposed under the documentary hypothesis have been collapsed, the Elohist has disappeared, and the Deuteronomist is never mentioned in connection with the Pentateuch. Briskly and summarily, the documentary hypothesis (or at least Wellhausen’s conception of it) is terminated.

Also fundamental to van Seters’ thesis is the possibility that both the Yahwist and the Deuteronomistic Historian would have read and consciously imitated early Greek historiography; especially that of Herodotus. ⁴⁵ This prospect has an initial ring of plausibility to it. There are plenty of similarities between Herodotus and the Pentateuch. Both contain genealogies, etiological myths, deluge narratives and the use of parataxis to connect apparently unrelated literary units. ⁴⁶ The parallels between the two works also shed light on one or two narrative puzzles in the Pentateuch. Herodotus’ repetition of key events to emphasise important ideas raises the possibility that the Yahwist may have done likewise, and the ‘repeated motifs’ in the Greek work (such as that of the wise advisor) are paralleled by, for example, the repeated wife-as-sister trope in the Pentateuch. ⁴⁷ Thus, the repetitions and contradictions within pentateuchal narrative can be explained without recourse to the theory of four differing source documents.

There are, of course, weaknesses in van Seters’ argument. As Blenkinsopp observes, maybe the most significant of these weaknesses is its reliance on a Greek influence to the Pentateuch. It is currently impossible to prove conclusively that Greece and Israel might have been linked and shared cultural artefacts through Phoenicia and although a link through Mesopotamia is possible, this is far from certain too. ⁴⁸ Blenkinsopp also counters van Seters’ insistence on parallels between the Pentateuch and Herodotus by pointing to significant differences between the two: for example, Herodotus’ History is directly attributed to a specific author, and is more secular in tone than the Pentateuch. And the Pentateuch is more ambitious in scope than History, referring not just to a national history but to universal history—the creation of the world. ⁴⁹ With the possibility of a Greek influence on the Pentateuch left as a mere possibility and nothing firmer, van Seters’ concept of an antiquarian Yahwist has little to offer beyond the antiquarian proposed as a biblical author by Thompson and Halpern, and the scribes favoured by Davies.

Just as the Yahwist is pivotal to van Seters’ arguments, the role of a scribe is central to Philip Davies’ understanding of how the Pentateuch took shape: he posits a thriving scribal school, established in the Persian period. Davies explains that, in ANE societies, the scribe was employed by the ruling elite as an archivist and composer of legal documents, but was also a diplomat and would have had sufficient freedom and creative ability to write according to his own interests. According to Davies, it was scribes of precisely this sort who were responsible for composing the Pentateuch. ⁵⁰

Like van Seters’ antiquarian Yahwist, Davies’ scribes would have had access to archives of existing sources, and both are credited with adapting these sources to meet their own agendas, and with creating material of their own. But there are significant differences between these two theories. For Davies, a guild of scribes, ⁵¹ rather than an individual antiquarian, was responsible for composing the Pentateuch, and such scribes would have been more susceptible than the antiquarian to pressure from the political establishment to write in step with a certain political agenda.

Extensive scribal activity could only have taken place during the Persian period or even later. ⁵² Davies argues against the idea of a Solomonic ‘golden age’, contending that any kind of large-scale writing project would have required a scribe and would have been extremely unlikely in a context when Jerusalem itself was nothing more than a small town. ⁵³ Since it was only following the return from exile that Jerusalem had the Temple facilities, the established ruling elite and the scribal school to make large-scale works of literature possible, Davies concludes that the Hebrew Bible ‘was at least largely compiled into its present form, and at most almost entirely written . . . during the rule of the Persians and then the Hellenistic monarchies’. ⁵⁴

The content of the biblical narrative supports this theory. Biblical accounts of arrival in the land, establishment of the law, a strong, prosperous, influential Israel and a glorious Temple all reflect the concerns of the immigrant elite, settling in Yehud and seeking to establish themselves. ⁵⁵ However, it must be noted that there is little extrabiblical data available from the Persian period. This is an important issue and Davies appears to gloss over it. He admits the dearth of information from the Persian period but insists that this need not prevent us from dating the production of the Hebrew Bible to this period. After all, the same can be said of the exilic period, and there is no shortage of scholars dating extensive writing activity to then. ⁵⁶ Instead of providing a strong argument for the Persian period as the most likely setting for the composition of the Pentateuch (despite the lack of available evidence), Davies deflects the issue back to his opponents.

A particularly valuable element of Davies’ work is that it brings to the fore the fundamental question of why the Pentateuch was written. Adherents to the documentary hypothesis tend to assume that at least one of the putative four sources (P) was intended to be an authoritative document for cultic worship, just as the Pentateuch is interpreted by confessing readers today. For Davies (similarly to van Seters), the Pentateuch—and the Hebrew Bible as a whole—is primarily a national history; a political document, first and foremost, not a religious one. ⁵⁷

While Davies dismisses the idea of significant literary works being produced in exile, K. L. Noll believes that the exile was precisely what prompted the collection and subsequent compilation of biblical traditions: ‘One can imagine that the crisis of deportation to Babylon, the disruption and loss that it entailed, motivated [the scribes] to raid those archives and gather as much traditional lore as possible, lest it be lost forever.’ ⁵⁸ So the writing and editing work performed on the Hebrew Bible was catalysed by the exile, continuing through the Persian period and into the Hellenistic.

The Hebrew Bible is a Persian-period document then, in that the majority of it was written in the fifth and fourth centuries. However, Noll also believes it was completed in the Hellenistic period and sees a distinctly Hellenistic aspect to its content. In Greco-Roman histories, ‘evidence’ meant whatever seemed reasonable and believable. When the ludicrous and incredible (and divine) was stripped away from myths and folktales, what remained was what was believed to have happened, even when one historian had simply copied another’s manuscript and tweaked key details, as was common. ⁵⁹ The result was a ‘history’ which almost certainly contained elements at odds with actual events and personalities and yet, this was not and should not be understood as deceptive on the part of the writer: ‘the ancient historian was not telling lies, he was presenting the past in a more “truthful” way than how it actually happened’. ⁶⁰ Noll applies the same understanding of history to biblical narratives: ‘Fiction is a common . . . kind of literature in the Bible.. . . It is not illogical to extend these observations to the entire biblical anthology.’ ⁶¹

Thus, biblical history almost certainly includes a good deal of material which deviates from the actual course of events but, as far as the authors were concerned, they were providing a ‘true’ account of events, people and the deity. The Bible’s contradictions, logical impossibilities and multiple versions of the same story with different protagonists are all folkloric elements—evidence that the authors had no intention of recording a purely factual account of the past. ⁶² As for biblical history having divine inspiration or any kind of inherent authority, ‘If the scribes who complied these folktales were alive today, they might be surprised, perhaps bewildered, to find that Genesis has become a sacred document attributed to divine inspiration or revelation.’ ⁶³

To my mind, the composition of the Pentateuch by scribes in Persian-period Yehud is an important possibility to keep in mind, especially as this theory opens up vital questions as to the very nature of the Pentateuch. Should the Pentateuch be understood as a national history, over and above fulfilling any kind of religiously-normative function? If so, what definition of ‘history’ does it represent? And what other ancient cultures might have influenced the Pentateuch’s form, or even its content? To ignore this theory is to gloss over these fundamental questions.

Thomas Thompson is in agreement with Davies on a good deal, relating to the Pentateuch’s composition but parts ways with Davies on questions of the Pentateuch’s dating, adopting an even more radical position by placing the completion of the Pentateuch in the Hellenistic period. According to Thompson, even the earliest biblical texts were not fully edited until the third century bce. His rationale is that the Dead Sea Scrolls represent the earliest biblical sources still available to us, and we therefore cannot assume that anything approximating a complete or nearly complete biblical tradition existed prior to this. ⁶⁴

Thompson’s position on this matter appears to have evolved. In his 1992 work, he posits that the cohorts resettling Yehud after the exile were sent by the Persian authorities, and implies that these same authorities sponsored the production of supposedly historical elements of the Hebrew Bible. ⁶⁵ Thus, the majority of composition work on the Hebrew Bible would have taken place in the Persian period, and texts relating to the fall of Israel and Judah depict the Babylonians and Assyrians as ‘barbaric oppressors of the people’, functioning as propaganda for the Persian regime. ⁶⁶ However, in his 1999 book, Thompson insists that the Hebrew Bible was not completed until the third century bce. ⁶⁷ This then suggests a lengthy period of composition, beginning in the Persian period and extending into the Hellenistic. The result is that Thompson articulates a position not far removed from Noll, but with a little more emphasis on the Hellenistic period as the setting for the shaping of the Hebrew Bible and a later date in mind for its completion.

In line with van Seters, Thompson envisages a single Yahwist, not a school of scribes, as the creator of the Pentateuch. Similarly to van Seters’ antiquarian Yahwist, this composer produced the Pentateuch by compiling pre-existing material: ‘classifying, associating, and arranging a cultural heritage that is both greater than the compiler or any single historiographical explanation’. ⁶⁸ Yes, the Pentateuch—and other biblical narratives—almost certainly absorbed ‘collections of fragmented traditions that survived the past’, ⁶⁹ but attempting to reconstruct those traditions from the completed text is futile. It is impossible to say what these existing traditions were, when they were written and the extent of any alterations to these traditions as they were subsumed into the final text. ⁷⁰ The Hebrew Bible offers us understanding only of the world of its authors, not of any earlier ‘historical’ Israel. ⁷¹

Thompson rejects the view of the Yahwist as a historian, in the strictest sense. That is, he was not interested in literal descriptions of historical events. Instead, his priority was ‘classifying, associating, and arranging a cultural heritage’, preserving stories of Israel’s origins in order to give the nation a common identity. ⁷² Such a late dating for the Pentateuch’s composition makes it difficult to escape the conclusion that its primary function was to do with national history and collective identity—as Thompson and Davies, for example, would have it—rather than religious norms. According to Thompson, while the Pentateuch includes theological interpretations of Israel’s past, and God is presented as a character in the narrative, it is not a religiously-normative document, should not be understood as a literal depiction of historical events and has little, if any, verifiable basis in earlier written or oral traditions.

Clearly, this is a radical departure from the documentary hypothesis. And these radical conclusions derive from a radical starting point: a contestation that the Pentateuch’s historical reliability cannot be assumed and can only be accepted where it is in step with archaeological findings and extra-biblical written sources. The revisionist position is inherently sceptical. If we accept this position, even insofar as dating the Pentateuch to the Persian or Hellenistic period, that is enough in itself to make a literal, uncritical reading of the Pentateuch impossible, and to open up fundamental questions on its very nature. That said, while Thompson argues that it is impossible to know the form and contents of pre-existing traditions, he does not dismiss their existence; he merely points out that we have no access to them as individual sources.

Niels Peter Lemche set out his stall in 1993, with what is now widely considered a seminal essay: ‘The Old Testament: a Hellenistic Book?’. Gone now were Lemche’s previous adherence to the documentary hypothesis and endorsement of a pre-exilic date for the first collections of pentateuchal traditions. ⁷³ Instead of assuming the earliest possible date for the composition of the Pentateuch, then moving the date forwards as literary and archaeological evidence demand, Lemche now begins with the latest possible date—the date at which we know particular biblical texts existed—and then moves the date backwards, according to the evidence available. Thus, as an extreme example, Lemche starts with a date of 350 ce for Samuel. ⁷⁴ In the case of the Pentateuch, his terminus ad quem is the first century bce; the approximate date of pentateuchal fragments of the Dead Sea Scrolls. ⁷⁵

Having established the latest possible date for the completion of the Pentateuch, Lemche offers literary and sociological evidence to suggest its earliest possible dating. Noticing Greek ideas of the creation of the four elements in Genesis 1, he reasons that this text must have been written in the sixth century at the very earliest. ⁷⁶ And since it is doubtful that Greek writers were widely read in the Persian Empire, Genesis 1 and other biblical texts bearing Greek influences must have been composed following the exile. What is more, in the absence of conclusive evidence that, following the return to Yehud, the province was well overseen and administered by Persia and Jerusalem re-established on a large and prosperous scale, we must assume that the prevailing socio-economic crisis immediately preceding the exile would have continued. In this context, any significant amount of writing and editing of biblical texts was unlikely. ⁷⁷ By process of elimination, we are therefore left with a Hellenistic date for the bulk of the composition of the Pentateuch and indeed of the rest of the Hebrew Bible.

Lemche’s move away from the position he articulated in 1985 and 1988 is startling and his new line of argument has prompted no small degree of debate. His opponents accuse Lemche and like-minded scholars of conducting ‘revisionism with a vengeance’, deliberately seeking to undermine Christian beliefs ⁷⁸ and of basing their conclusions on ‘naked assertion, rather than argument’. ⁷⁹ Scholars with more conservative leanings take particular issue with Lemche’s ‘verification’ principle: the insistence on empirical proof for a supposition and the rejection of that supposition in the absence of such proof. Provan attacks this as a ‘maximally suspicious stance’ and a ‘narrow view of reality’. ⁸⁰ Why should we not prefer a ‘falsification’ principle, ‘by which ancient texts are given the benefit of the doubt unless compelling reasons to distrust them are apparent’ ⁸¹ ?

Since conservative and revisionist scholars begin from such different starting points – the conservatives assuming the historical reliability of the biblical text, the revisionists assuming its unreliability—it is no surprise that they reach such different conclusions. And this offers a possible explanation of why the documentary hypothesis continues to dominate commentaries on Exodus: why would a commentator try to include two radically different theories of pentateuchal composition, when s/he is bound to disagree strongly with at least one of them? I also fear that the differences between the conservative and revisionist positions might be so stark as to make co-operation impossible between members of the two camps.

What are the Effects of this Disconnect?

The disagreements between conservative scholars and revisionists—especially those known as the ‘Copenhagen School’—have been well publicised and often vociferous. I can sympathise with those on both sides. Several of the conclusions drawn by revisionist scholars strike at the very heart of the conservative position. For example, suggestions that the Pentateuch bears little or no relation to actual historical events are extremely difficult to accept for most readers who hold that the Bible is in any way inspired by God. On the other hand, I am willing to believe that revisionists consider the evidence on their side to be overwhelming. If so, the conservatives’ refusal to accept this evidence must be highly frustrating for them. The conservatives may seem intractable or even positively stubborn.

When the two sides begin from such markedly different positions and with opposing sets of assumptions, it is perhaps impossible that they should ever reach an agreement. But mutual understanding at least may be a realistic goal. Even this modest goal, however, remains distant as long as commentaries remain tethered to the documentary hypothesis with, at best, a cursory discussion of a possible Persian or Hellenistic dating for the Pentateuch. While revisionist scholars extend and develop their cases for Persian or Hellenistic dating, the conservatives remain rooted in the documentary hypothesis. The battle lines are drawn and the trenches on both sides are dug ever deeper.

If the Pentateuch was indeed composed or compiled in the Persian or Hellenistic periods, then this has exciting implications, in terms of revising our understanding of the nature and character of the Pentateuch. The Pentateuch could represent a national history; an attempt by the Yehud community to articulate its shared identity, following the return from exile. The five books might be overshadowed by collective trauma, perhaps arising from the exile and the fall of Jerusalem preceding it. These and other possibilities are discussed at more length by progressive Pentateuch scholars ⁸² but commentaries remain closed off to these ideas, as a result of their reliance on the documentary hypothesis alone. Commentaries would be richer for removing these self-imposed limits and exploring the possibilities opened up by alternatives to the documentary hypothesis.

Anyone who reads commentaries but no other academic texts relating to the Pentateuch would be forgiven for believing there is no credible alternative to the documentary hypothesis. Given the near-absence of references in commentaries to the Persian or Hellenistic periods as possible contexts for the composition of the Pentateuch, anyone who does not read more widely would have no awareness of these alternative theories.

Scrutiny of the bibliographies and ‘suggested readings’ lists within commentaries makes it seem unlikely that many readers will go on to read revisionist scholars. For example, Bruckner makes no reference to revisionist ideas on the composition of the Pentateuch, and no revisionist scholars appear in his bibliography. Propp’s bibliography runs to nearly 60 pages ⁸³ and still overlooks all prominent revisionist scholars but Thompson. As we have seen, Dozeman references van Seters, Lemche and Thompson, but does not engage with their theories of pentateuchal composition. Alexander nods to Lemche and Thompson but soon brushes them aside. ⁸⁴ Even the more progressive Meyers allows Davies, Thompson and Lemche only one footnote each, and does not mention revisionist scholarship at all in her extensive ‘Suggested Readings on Exodus’ list. ⁸⁵

I am aware of no hard data on the demographics of commentary readers. However, the majority of commentary series have exegesis as their aim and confessing readers as their target audience. For example, Dozeman’s commentary forms part of the Eerdmans Critical Commentary series, which favours confessing authors and emphasises ‘the original meaning of the text’. ⁸⁶ Alexander’s commentary belongs to a series for ‘preachers, teachers and students of the Bible’ ⁸⁷ , with the publisher’s website proclaiming, ‘The Apollos Old Testament Commentary (AOTC) aims to take with equal seriousness the divine and human aspects of Scripture.’ ⁸⁸ The Anchor Yale Bible series (to which Propp’s commentary belongs) is singular in offering ‘excellence in biblical scholarship and a commitment to advancing biblical understanding’, making no mention of any specific doctrinal allegiance. ⁸⁹

Thus, it seems a reasonable assumption that a good number of the readers of commentaries are thinking, confessing readers, including clergy. For such readers, commentaries represent a vital bridge to the academic study of the Pentateuch so, while commentaries continue to ignore revisionist thinking, these confessing readers and their congregations will remain unaware of the important work being done by revisionist scholars, and they would not know where to look to find it. For as long as commentaries do not give a full and rounded picture of the works available, confessing readers will be limited to a one-dimensional understanding of the field of study.

What Could an Alternative Commentary Look Like?

I propose a commentary with a fresh approach, to address these problems. A commentary of this kind would make important recent research available to a wider audience, while remaining accessible and useful to confessing readers of the Pentateuch:

An editorial board including conservative and revisionist scholars would ensure that the documentary hypothesis and alternative theories were presented fairly and accurately, throughout the series. This combination of scholars with different viewpoints would also lend the series vibrancy and ensure it continued to be open to new ideas in biblical scholarship. This vibrancy would also be enhanced if the editorial board was ethnically diverse and demonstrated a balance between male and female scholars.

The publishers should approach scholars with well-established expertise in pentateuchal studies in general, as well as some with strong publishing records on individual books within the Pentateuch (assuming the series will expand beyond a commentary on Exodus). The editorial board would also benefit from scholars with more theoretical backgrounds. For example, I would be especially interested to see how feminist, womanist or queer perspectives might add new dimensions to the series. It would also be vital for the board to consult with practitioners of Christian ministry, to ensure that the commentary is relevant to live pastoral concerns (see point iv, below).

The majority of commentaries on Exodus include a section which addresses the composition of the Pentateuch, but this section is usually limited to one or two pages within the introduction. A more progressive commentary would include an extended ‘composition’ section, outlining alternative theories of pentateuchal composition alongside the documentary hypothesis, even if the series editors still decide on the documentary hypothesis as the most convincing of the available theories. All theories of composition which are mentioned would be reflected in the bibliography.

This ‘composition’ section would address the documentary hypothesis, the late Yahwist theory and Persian and Hellenistic dating in turn. Although the idea of Mosaic authorship has largely fallen out of favour among scholars, I would advocate for including that, too, for the sake of offering a broad sweep of theories of composition. The writer should outline the origins, key proponents, strengths and weaknesses of each theory, as well as its implications for interpreting the text.

It would not be problematic for the commentary writer to favour one theory of composition over the others—indeed it would probably be inevitable—as long as s/he were sympathetic to alternative positions and willing to outline them in an even-handed way.

It may be appropriate, at some points, to delve into source criticism, especially when this approach might offer an answer to a textual difficulty. But the commentary would avoid protracted discussions of source analysis (and even redaction analysis) in favour of a focus on Exodus in its final form.

Commentaries too often gloss over a debate, an ambiguity of translation or a theological difficulty, for the sake of brevity. For example, in Exodus 14:4, YHWH pledges to ‘harden Pharaoh’s heart’ so that he will pursue the Israelites, in order that YHWH might be justified in destroying Pharaoh and thus ‘gain glory for myself’. Indeed, in v. 17, YHWH hardens the hearts of all the Egyptians, not just Pharaoh. This picture of a deity manipulating the hearts of his enemies, so that they will threaten his people and he will therefore be justified in destroying them, is highly unsettling. Yet, very few commentaries acknowledge the theological difficulties raised by this picture. Bruckner and Hyatt suggest that YHWH merely allowed Pharaoh to continue in a course of action on which he had already decided. ⁹⁰ In a similar vein, Propp argues that YHWH ‘strengthens’ Pharaoh’s heart, rather than hardening it, ⁹¹ suggesting that the deity merely increases Pharaoh’s resolve on a course of action to which he was already committed. Other commentators (e.g., Childs, Durham) don’t mention this theological problem at all. ⁹²

Where such an uncertainty arises, it can be explained in an excursus and multiple possible answers identified. Informed by this material, readers can and should be allowed to draw their own conclusions. Meyers’ commentary provides a rare example of this, on the implications of YHWH hardening Pharaoh’s heart, ⁹³ and I applaud her for it. Any new commentary series should feature discussions of this sort.

For example, the plagues narrative—and especially the plague of the death of the firstborn—makes for uncomfortable reading. It raises knotty theological questions around what kind of god would carry out a mass killing of this sort, and what could justify it. An excursus on this point would be very helpful. The commentary should also acknowledge that there is some doubt as to the historicity of Israel’s sojourn in Egypt and of the exodus itself. The writer should outline different perspectives on the question, signpost the reader to other scholars who have addressed it in more depth, and comment on how a confessing reader might reconcile the doctrine of the divine inspiration of scripture with a narrative which might not represent a literal historical account.

Looking back over this article, I am aware that I could perhaps be misunderstood as denigrating the entire commentary genre. This absolutely is not my intention. The existing commentaries are invaluable. I have benefited immeasurably from the insights offered by commentaries on Exodus, in particular, and I have no doubt that numerous other readers have done so, too.

One of the features I appreciate and admire most in the existing commentaries is the writers’ skill in drawing out aspects of the book of Exodus which have especial relevance to the reader’s own situation. For example, a number of commentators, including Fretheim, Bruckner and Sarna, ⁹⁴ observe that the Israelites’ freedom from slavery resonates deeply, even today, with other victims of oppression. Identifying and articulating these parallels are skills to be prized and I would hope to see them enshrined as central characteristics of any new commentary series.

Since there is no shortage of injustice and abuse of power in the world, Exodus’ themes of oppression and liberation will continue to strike a chord with readers for the foreseeable future. I am also convinced there is something to be drawn from Exodus, to do with how a group of people (or an individual) experiences and recovers from a traumatic experience. ⁹⁵ And a reflection on the ecological implications of YHWH’s manipulation of nature (in the plagues and Reed Sea narratives, for example) would seem timely, given the present climate emergency. Close consultation between the editorial board and ministry practitioners would help ensure that important pastoral concerns are addressed in the commentary, whenever the text includes something of relevance.

Conclusion

It may be naïve to call for co-operation between conservative scholars and revisionists. I fear the differences between the two schools may be simply too great for meaningful collaboration to be possible. However, greater mutual understanding might lead, at least, to tolerance of each other’s positions, if not acceptance. A commentary of the kind I have suggested would represent a step towards greater mutual understanding. How long it will take for such a commentary to see the light of day remains to be decided. All I can do is to commend it to the academy and await my colleagues’ response.