2012-01-08

Who wrote the Bible? (2) Challenging the Documentary Hypothesis

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

by Neil Godfrey

Русский: Распределение документов Йахвист, Эло...
Image via Wikipedia

This post continues from my post some weeks ago in which I covered primarily Philippe Wajdenbaum’s account of the rise of the Documentary Hypothesis. At that time in one of the comments I explained I had paused to take stock of how best to address the challenge that has arisen against the Documentary Hypothesis. This is a study I undertook some years ago and so thoroughly enjoyed that it is easy for me to cover way too much detail. Maybe I will have to return to address some of the specifics in separate posts later. Once this is out of the way I would like to post another explaining how political anthropology offers a cogent explanation for the character of the biblical books as Hellenistic productions.

First, to recap the Documentary Hypothesis. This is the idea that the Old Testament was essentially a result of four separate sources that were originally written over a span of some centuries:

  • a Jahwist/Yahwist (J) written in the southern kingdom of Judah around the time of Solomon – 10th century bce / later shifted to the Babylonian Exile period:
    • Gerhard von Rad in 1944 “considers the time of Solomonic enlightenment to contain all the prerequisites for literary production, including history writing. It was first of all a time of political stability and economic prosperity. On top of this came the need of a new state to provide a history of its past. Finally the creative impetus following in the wake of the establishment of an Israelite state created this new literature.”
    • Subsequent scholarship revised this, arguing that “External circumstances were thought to provide the most likely background for this kind of literature.” (pp. 158-9 of The Israelites in History and Tradition, Niels Peter Lemche)
  • an Elohist (E) composed in the northern kingdom of Israel – 9th or 8th century bce
  • a Deuteronomist (D) in the southern kingdom of Judah at time of Josiah – late 7th century bce
  • a Priestly source (P) during the Babylonian Exile – 6th century bce

The dating of the sources is central to the hypothesis:

Essential to the history of scholarship expressed in Wellhausen’s synthesis [the DH is the result of W’s synthesis of two generations of OT historical-critical scholarship] was that these four discrete sources of the pentateuch were to be understood as literary documents created at the time of their written composition, and hence as compositions reflecting the understanding and knowledge of their authors and their world. (p. 2 of Early History of the Israelite People from the Written & Archaeological Sources, by Thomas L. Thompson.)

This meant, for example, that the Pentateuch was not a reliable source for the events it narrates, such as the Patriarchal period and Exodus.

But in recent decades biblical scholars are not so united in their acceptance of this explanation for the Bible or “Old Testament” portion of it.

Basically, the old consensus that had developed around the Documentary Hypothesis has gone, though there is nothing to take its place (Rendtorff 1997; Whybray 1987). Some still accept the Documentary Hypothesis in much its original form, but many accept only aspects of it or at least put a question mark by it. There has also been much debate around the J source (Rendtorff 1997: 53-5) and the P source (Grabbe 1997). It seems clear that the Pentateuch was put together in the Persian period (Grabbe 2004:331-43; 2006). (p. 44 of Ancient Israel: What Do We Know and How Do We Know It? by Lester L. Grabbe)

So where have the cracks appeared?

Cracks were being pointed out at least as early as the mid 1970s but there has been a remarkable explosion of new ideas and challenges to the consensus since 1992. Lester Grabbe identifies three major publications in that year:

  1. The Anchor Bible Dictionary with its multiple-author entry on ‘Israel, History of’ (ABD III, 526-76) with Lemche on the Pre-Monarchic Period, Dever on Archaeology and the ‘Conquest’, R. P. Carroll on the Post-Monarchic Period;
  2. Early History of the Israelite People from the Written & Archaeological Sources, by Thomas L. Thompson;
  3. In Search of Ancient Israel, P. R. Davies.

On this third publication Grabbe writes:

but the book which in many ways made the greatest impact in that year came from a surprising quarter: this was P. R. Davies’ In Search of ‘Ancient Israel’. I say ‘surprising’ because Davies made no claim to originality; his aim was to translate and expound some of the recent trends and their basis for students . . . . Yet this book caused a storm: eliciting reviews, such as that by Iain Provan published in the Journal of Biblical Literature (1995), along with responses from Davies and Thompson. (p. 33)

I have picked out some of the key points of this work and set them out at vridar.info.

Other works followed:

  • The Tel-Dan ‘Ben-David’ inscription was published (1993) with a great flurry of claims and counterclaims
  • The History of Ancient Palestine from the Palaeolithic Period to Alexander’s Conquest, 1993, by G. Ahlström.
  • The Invention of Ancient Israel: The Silencing of Palestinian History, by Keith Whitelam (1996)
  • Several works by Lemche and then Silberman and others.

A central theme of the criticisms of traditional views of the Bible from most of these authors has been to point out the circularity of consensus arguments for the datings of the various biblical works. Another has been to reject the assumption that the biblical works should be accepted as reliable historical sources or frameworks for historical narratives without external confirmation. Whereas archaeology was once a tool to flesh out the “truth” of the Bible, it was now being thought more valid to study the archaeology of Palestine in the same way as archaeology is employed to uncover the history of any other civilization: understand the remains in their own right and only secondarily seek to see where the literary work best fits within what we learn from those finds.

As a result some of these scholars have come to view “biblical Israel” as a theological construct with a range of meanings rather than a fixed historical reality. The united kingdom of David and Solomonic empire has always been an assumption through which archaeological remains were interpreted. The more evidence that has surfaced, however, has indicated to many of these scholars that there is no room in the primary evidence for a united kingdom or empire in the tenth century based at Jerusalem or Judea. The evidence that has surfaced has forced upon such authors the conclusion that such a power was an economical, demographic and technological impossibility. The evidence in the ground pointed to the kingdom of Judah not emerging to any significance until after the destruction of the kingdom of Israel (a kingdom that first emerged under Omri’s dynasty), and then the economic power-house city of Lachish, by the Assyrians.

Where do the works of the Bible fit in here?

The dating question is important:

It is surely extremely naive to believe that the meaning of biblical books can be properly exposed without knowledge of their date of composition, about the ideas current in that age or the beliefs common to their audience; and it is of no consequence whether the subject is a narrative as a whole or parts of it or just single concepts and phrases. (p. 295, Niels Peter Lemche, ‘The Old Testament — A Hellenistic Book?” in Did Moses Speak Attic?)

Traditional dating of the works of the Old Testament has been faulted for the circularity of its approach. It has been assumed that dates for various parts of the Bible must be found that belong to the narrative told by the books themselves, that is, somewhere within either the kingdom of Israel discussed or the exile or among those who returned from the exile. The Biblical narrative has been assumed to be indicative of historical reality and that its authors — J, E, P, D — must belong to that very historical period that is the topic of narration. As Wajdenbaum says, such an approach is merely a modern adaption of the myth told by the Bible itself.

The result of this approach has been a quite complex product:

In the end, the higher criticism ended up with an enormously complicated structure, sometimes ridiculed because of its seeming absurdity . . . [T]he narratives of the first four books of Moses have been dissolved into a kind of jigsaw puzzle of isolated elements, passages, verses, and half-verses. In its mechanical form the documentary hypothesis is monstrous . . .  (p. 158, The Israelites in History and Tradition)

And this has meant that scholars have been blocked from reading the books as literary wholes as they focus on narrow segments seeking some evidence of an historical core to each part. The works have ceased to be read as coherent literature.

It is significant that the first biblical definite evidence of the existence of the biblical texts is in the Hellenistic period — after the conquests of Alexander the Great and the spread of Hellenization (Greek culture and language) throughout the Middle East.

Lemche stresses that a justifiable dating of a text must come from the latest material found within it even if it does contain some earlier material. So if the Pentateuch contains passages that can only be explained as of Hellenistic provenance then the Pentateuch as we have it surely must date from the Hellenistic times even if it contains some Mesopotamian material (e.g. the flood story) that is earlier. Lemche points out, for example, that Karl Ilgen first told us 200 years ago that Numbers 24:24 can only possibly be a reference to the Macedonians and that Martin Luther had likewise made this clear in notes to his translation — hence the poem must be dated to the late fourth or third centuries bce. Yet despite this clear pointer to a late date scholars have generally dated this chapter as some of the most archaic of poetic literature, certainly pre-monarchic. The chapter was later moved to the monarchic period when the name Balaam turned up in an eighth or seventh century Aramaic inscription. Meanwhile, the poem itself keeps murmuring, “Hellenistic”. (p. 160, Israelites in History and Tradition).

In a recent post I re-used in another context Lemche’s resurrection of Maurice Vernes’ 1889 advice that the starting point for dating analysis must be the time where there can be no doubt that the biblical literature existed and only from there move back to where we have more questions. Since the earliest period in which we have concrete evidence for the existence of the biblical works is the Hellenistic era, we must start from that point and consider what, if any, evidence we have that they had an earlier origin.

Linguistic arguments have sometimes been hailed as keys to identifying sections of the Bible separated by decades but they have proven to be falsified when they have been shown to be unable to distinguish books that have turned out to be centuries apart.

To sum up, the Documentary Hypothesis has been judged by these scholars to have been little more than a rationalization of the Bible narratives themselves. Past scholars, they say, forgot to read the texts as evidence of themselves and as evidence of the intentions of the authors and not of the tales they told. An Italian scholar, Mario Liverani whom I have quoted several times on this blog, chastises “the indolence” of many historians who all too easily fall into the temptation to merely paraphrase or rationalize (remove the miraculous bits) a narrative that is ready-made and at-hand.

“Biblical Israel”, for example, has been shown to be more a theological construct than a historical reality. The stories do contain some names and events that do belong to real history, but they are not themselves historical narratives. The narratives are said to reflect the theological and political interests of either Persian or Hellenistic governments and peoples in Palestine. Many of the tales are said to be mythical reiterations, “midrashic adaptations”, of one another and not discrete events at all. Example, the dividing of waters at the Creation is reiterated in the narratives of the Flood, the Exodus, the entry into the Promised Land, and later in the accounts of Elijah and Elisha — always with the same theme of new creation or the creation of a new people of God. We see repetition in the Patriarchal narratives and Wajdenbaum sees the character David being an inverse reiteration of Jacob.

Read as a whole the Bible’s books are tied together with a common theological and literary theme. The history of Israel opens and closes with Babel. The theme of God’s ways being alien to the thoughts of mankind is there from Eden to the poetic literature, and the lesson to submit to the will and mind of God overrides each successive “new Israel’s” desires for land and kings.

Significantly, these historians argue strongly that it is only by the time we reach the Persian and/or Hellenistic eras that we eventually find the cultural, economic and infrastructural conditions in Palestine that would be required to produce the biblical literature. Some have suggested the period of King Josiah, but this is judged by others also to be problematic. Certainly this period saw the rise of a literate class required for administrative writing. But there was no historical heritage that could explain the particular sorts, and diverse richness of, the narratives we read in the Bible. Finkelstein and Silberman have not produced any evidence actually tying this literature to this time period in their popular work The Bible Unearthed.

It must also be kept in mind that the Hebrew Bible was not known to have existed until well into the Christian era. Before then we only have evidence of the Greek language Biblical books, and in this Greek language version we encounter several books that are undeniably Hellenistic in origin: Ecclesiastes, Daniel, Song of Songs, Maccabees, Ben Sira.

Many of the concepts and themes are Hellenistic, and some have argued that the final collection of biblical books was based on the works of the Greek historian Herodotus. I have addressed some of the arguments for this in older posts here. One of the arguments related to this urges us to consider that the frequent contradictory narratives set side by side (e.g. Creation, the rise of King David) Lemche sees several strong (including structural) parallels between the Bible and the works of the Roman historian Livy as another testimony of a common Hellenistic background.

In short, the documentary hypothesis has given expression to what scholars have traditionally merely assumed — the historicity of the united kingdom of Israel that began in a pre-monarchic period and that has been identified by its unique religion. Historians have approached this literary construct as if it were a historical precursor of the Church of their faith.

If the biblical works were not composed until the Persian (Davies) or Hellenistic (Thompson, Lemche) era, an increasing number of scholars have found more satisfying solutions to matching the themes of the books with the facts of archaeology. The stories of David and Solomon become in one sense romantic legends like the tales of King Arthur in a Camelot that really was fictional no matter the occasionally genuine geographic locations found in the adventures. But the biblical stories had a strong theological theme (not unlike Herodotus who wrote his Histories to illustrate the overriding will of the Delphic God Apollo against the hubris of mankind — first the Persians and then (imminently) the Greeks themselves.

I realize that this outline has omitted many details — the themes of exodus, migration, landlessness, inheritance, the people of the land and Canaanites, the Temple — but this has to be of necessity a blog outline. Some are addressed in the vridar.info pages on Davies’ book.

My own aesthetic preference is for a hypothesis that reads the biblical literature as a coherent whole if that can be justified at all. And the historical books — and prophetic ones, too (and these, too, often refer directly or indirectly to the Persian era) — do come together to tell a coherent story. The evidence for the books being products of the Hellenistic (or Persian) era is, I believe, strong. I don’t know how this can be well reconciled with a hypothesis that the whole has been stitched together from a J, E, P & D.

The following two tabs change content below.

Neil Godfrey

Neil is the author of this post. To read more about Neil, see our About page.

Latest posts by Neil Godfrey (see all)



If you enjoyed this post, please consider donating to Vridar. Thanks!


9 thoughts on “Who wrote the Bible? (2) Challenging the Documentary Hypothesis”

  1. Judah became a state in the Late Iron IIa (Omride or Aramean period), and did become the most powerful state in Palestine only after the fall of Samaria. If becoming the most powerful state in Palestine counts as having “any significance”, you (Vridar) have some very high standards for that term. Also, the book of Joshua, one of the easiest books to use for comparison with archaeology, seems to derive its information from the late 7th century district lists, not the Perso-Hellenistic period district lists (most notably, Ekron was of no significance in the Perso-Hellenistic period). The idea of revival of a lost, non-existent empire could easily stem from the Babylonian period. I find the idea of a Hellenistic OT (besides Ezra-Nehemiah, Ecclesiastes, and Daniel, which are obviously of Selucid-Hasmonean date) unlikely at best. Genesis, for example, in some details (e.g., the deity/ies smelling the food, the sending out of the birds) is more similar to the Gilgamesh epic than Berossus. Also, there is some debate as to whether the Song of Songs should be dated to the Persian or Hellenistic period.

    1. Certainly (as I think I pointed out somewhere in the post) historical names are recycled in the Bible narratives and earlier stories, such as that of the Flood, are used. But the principle stressed by Lemche and others (Lemche appeals to Vernes) is that a date for a text should be set by its the most recent contents and not its earliest. The Flood story was well known for centuries in the Middle East and Plato himself speaks of Greek mythical equivalents. Scholars have long acknowledged that an older story has been at the very least “redacted” in places by a relatively late “Priestly” editor.

      1. What you said is correct, but, to date a text, one must first focus what is integral to the narrative. For example, the oracle against Kition is obviously a bit of Hasmonean redaction, though the Baalam tradition clearly dates back to the 8th C BC, and the verses just preceding v. 23 are highly unlikely (at best) to date any later than the Assyrian era, probably the late 8th C BC. However, many times it is hard to distinguish between the original text and interpolations.

        1. Instead of explaining the sprinkling of Septuagint anachronisms as later interpolations is it not simpler to suggest that someone in Hasmonean times wrote the whole Balaam prophecy to meet the interests of his own day?

          If we take it as a very early poem preserved and redacted until its inclusion in the Bible then when and where did this preservation take place? There is no evidence of a literary elite in the area of Judah until very late. If in Samaria then what ethnic or religious history could have lent any relevance to the poems?

          By assigning the poems to a very early date and rationalizing the evidence for Hellenism in them (further weeded out by the Christian era Massoretic text) how can we say we are not simply taking the Bible’s myth and merely creating another version of it?

    2. I think there is a textual layer in the Deuteronomistic histories that is possibly pre-Persian Period. However, it has been sufficiently demonstrated, for example, that several passages in Genesis are literarily dependent on stories in Samuel, and that the creation store in Genesis 1 is literarily dependent on Psalm 104. Early works like Hosea have a vague notion of migration from Egypt (perhaps included in the Yahwistic liturgy at Bethel) but no knowledge of captivity, an exodus, or Moses. Ezekiel appears to be abundantly familiar with the Holiness Code portion of Leviticus and no other part of the Pentateuch, and the story of primordial Eden in Genesis probably post-dates Ezekiel’s version of Eden as a royal garden.

      There are many literary clues that the Pentateuch is a late work (albeit compiled from numerous source documents, some old) and is probably late Persian or Hellenistic.

      1. The argument of Lemche and others is that it makes most sense to date the texts by the evidence of what must clearly be their most recent referents. Are these authors working with existing texts and adding bits to them or are they simply drawing on their wider knowledge of earlier literature and composing something afresh?

  2. I’d be curious whether you’ve interacted with the notion that a J/E/P/D scheme might be supportable independent of the date that the texts were written. I’m not really in a position to have a strong opinion of the documentary hypothesis, but I found this quote by Joel Baden interesting (from /The Composition of the Pentateuch/, 2012, p. 31):

    “Indeed, attempts to order the documents chronologically (that is, to date them relatively) and situate them temporally (to date them absolutely) with any specificity are based more on a given scholar’s a priori historical beliefs than on the texts themselves. We must be careful not to confuse the literary question with the historical one. Like thematic and stylistic considerations, the dating of the documents can be accomplished only after the sources have been isolated on other grounds. And at that point, the various datings of the documents have no effect on the literary analysis: if it could be demonstrated somehow that J is from the tenth century BCE and that P is from the third century BCE, while E is from the second millennium BCE and D was written during the Hoover administration, the literary evaluation of the text and the isolation of the sources on the grounds of narrative flow would be precisely the same.”

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Discover more from Vridar

Subscribe now to keep reading and get access to the full archive.

Continue reading