> Israel > Jewish Society

Addressing Biblical Criticism: A Critique of the Documentary Hypothesis

April 16, 2020 | by Prof. Joshua Berman

An in-depth examination of the story of the Flood yields numerous faults of the source-critical approach.

The following is adapted from the author’s new book, Ani Maamin: Biblical Criticism, Historical Truth and the Thirteen Principles of Faith (Maggid, 2020).

Perhaps the greatest challenge to traditional belief raised by academic biblical studies has come through what is known source criticism, or as the documentary hypothesis. This states that the Torah constitutes the editing – or redaction – of four separate documents, pioneered in the mid-nineteenth century. Many find source criticism satisfying because it strives to make sense out of passages that are difficult to understand due to what appear to be inconsistencies in the text. But how reliable is this type of inquiry? Increasingly scholars are calling into question whether it is really possible to work back from a received text, such as the Tanakh (the Bible consisting of the Torah, the Prophets and the Writings), and recreate its prior stages of development without the help of any external corroborating evidence.

In the eighteenth and nineteenth centuries, classicists believed that what appeared to them to be inconsistencies in Greek and Roman compositions could be the clues to retracing the original sources that gave rise to those texts. In the twentieth century, however, classicists largely gave up the task as lacking in rigor. The pursuit of the “prehistory” of the text on the basis of the received text before us – the Tanakh – remains widely practiced in biblical studies alone.

The parade example of the so-called achievements of this approach to the biblical text is found in the scholarship to the story of the Flood in Genesis 6–9. And yet it is precisely here that we can see all of the faults of the source-critical approach on display.

A Source-Critical Reading of Noah and Its Difficulties

The attempt to separate the Flood story into two strands stems from seemingly reasonable observations. A linear or synchronic reading of this passage reveals many difficulties, as the Flood account seems riddled with doublets and inconsistencies. To recount the most significant of these, it alternates between two divine names, Hashem and Elokim. Some passages speak of a downpour lasting forty days and forty nights (Gen. 7:4, 12, 17a), while others speak of a cosmic deluge whose waters crest for 150 days (7:11, 24). One passage instructs Noah to gather a pair of every living creature (6:19–20), while another differentiates between clean animals, of which seven pairs are to be taken, and unclean animals, where a single pair of each will suffice (7:2–3).

The source-critical solution to these and other similar inconsistencies and redundancies has been to identify within the account the interweaving of two versions of the story, usually referred to as the Priestly (or, simply, “P”) and non-Priestly (“non-P”) accounts. The strength of the approach is in its cumulative power, because the presence of so many details that coalesce so neatly along these lines suggests that what we see here is more than just a coincidence. (For a color-schemed breakdown of the Genesis flood account along the lines of the proposed P and non-P accounts, see A Historical and Contextual Approach, The devil, however, is in the details; the theory yields to scrutiny on many points of methodology, five of I which I describe below.

1. A Double-Standard for Determining Doublets

The great appeal of the source-critical approach was that it supposedly produced two accounts of the Flood that would read cleanly, without the repetitions and doublets that seem to plague a reading of the full Genesis text. However, neither the proposed P text nor the proposed non-P text achieves this. Consider that within the non-P account we find the following reconstruction of our Genesis text: “(7:12) And the rain was upon the earth forty days and forty nights. (7:16b) And God sealed him therein. (7:17) And the deluge was forty days upon the earth, and the waters increased and lifted the ark so that it rose above the earth.” The note in verse 17a that the deluge was forty days long is glaringly superfluous following the exact same claim two verses earlier in the non-P version.

The fact that source critics are willing to overlook this doublet and others like it calls into question the criterion of doubling that is the basis for the hypothesis of two strands. The criterion does not seem to be applied rigorously and consistently. Rather, it seems that source critics see doublets when these will fit into the procrustean bed of two separate sources, but overlook doublets when they remain within the hypothesized versions.

2. Creating False Doublets

Further, the two-source theory is foisted upon the text; it produces dichotomies and doublets that are of its own creation and not inherent in the text. One such “imaginary” difficulty and contrived doublet concerns the source of the deluge. For source critics, the P version claims that God allowed the waters of the depths and the heavens to flood the earth (Gen. 7:11; 8:2) whereas the non-P source maintains that the deluge was rainfall (7:4, 12; 8:2). The difference and distinction between the two founts of the deluge are presented as if they are mutually exclusive.

Logically, of course, there is no reason why the deluge could not have emanated from both rainclouds and heavenly and earthly wellsprings. There is no contradiction between the two. Moreover, the notion of divine deluge stemming from these two sources is a common trope. In fact, consider the sources of the deluge in the Mesopotamian account of the flood story, which is caused both by rainfall and opened dikes:

I gazed upon the appearance of the storm,
The storm was frightful to behold!...
A black cloud rose up from the horizon,
Inside [the cloud] Adad was thundering…
Erregal tore out the dike posts,
Ninurta came and brought with him the dikes.

Divine deluges that stems from both from cloud rain and from the wellsprings of the earth are a familiar trope in the Tanakh (Ps. 77:17–18; Prov. 3:20). Moreover, the Genesis Flood account mentions these two founts together in two places (Gen. 7:11–12; 8:2). However, were source-critics to adopt a reading whereby the Genesis Flood derived both from cloud rain and from other wellsprings together, it would no longer be possible to bisect the text into two accounts. Source critics must ignore the attested trope in the Mesopotamian version of the flood story and the other biblical sources of divine deluge from rain and from other wellsprings, so that each of the putative versions of the story will have a flood unto itself. When critics separate the founts of the deluge, they do so not because the theory solves a problem in the text; rather a problem in the theory gives rise to an unnecessary and forced distinction in the text.

3. Irrational Non-Sequiturs in the Putative Sources

In addition to creating unnecessary and unwarranted dichotomies, the source-critical reading also produces non-sequiturs in the putative sources that it claims to recover. Consider the Masoretic Text’s version of Genesis 7:15–16: “[The animals] came unto Noah, unto the ark, two by two, from all of the living creatures. They were male and female of all creatures, as Elokim had commanded him. And Hashem closed him in.” The final phrase of verse 16, “And Hashem closed him in,” follows directly from the previous elements in verses 13–16. Noah and his family enter the ark, the animals enter the ark, and, to conclude, Hashem “shuts the hatch” as it were, and closes Noah in. However, in the putative non-P source, the following text is hypothesized: “(7:10) And after seven days, the waters of the deluge were on the earth. (7:12) The rain was on the earth forty days and forty nights (7:16b) and Hashem shut him in.” Source critics splice the text in this fashion because verse 16b refers to God as Hashem, and thus must be assigned to the non-P source, which they reckon refers to God exclusively as Hashem. However, this reading is deficient on two grounds. In the first place, it creates a non-sequitur as it implies that it had been raining already for forty days and forty nights before God enclosed the ark! Secondly, it removes verse 16b, the notice of God shutting in Noah, from the simple context of the verses in which it is organically found in the Genesis text, following the embarking of Noah, his family and the animals.

4. Hiding Problems with the Theory in the Work of the “Redactor”

Other seemingly needless repetitions abound in the Flood narrative, such as the extended repeated report of Noah’s entry into the ark in the hypothesized P version (Gen. 7:8–9, 11, 13–16). Difficulties such as the unnecessary and juxtaposed repetition of the duration of the rain in non-P, and the wholesale repetition of the boarding of the ark in hypothesized P, might have been the types of literary phenomena that could have called into question the very suggestion that we have here two conflated sources. Splitting the text clearly does not provide us with two accounts, each free of repetition and free of incongruities. And yet rather than walking back from the hypothesis, source critics have sought to buttress their hypothesis by resorting to a series of editors, or what scholars call “redactors,” who are the agents responsible for the disruptive passages.

The recourse to redactors and the claims that various words and verses are later additions are made solely so that scholars will be able to preserve the integrity of the two sources, purportedly identified in the remaining verses of the narrative. The strategy is reductive in that it ensures that the underlying premise of two sources will always be preserved. For the source critic, data that complicates the split into two sources is not allowed to undermine the theory. Instead, “bad” data – data that are incongruous with the two-source theory – are isolated from the “good” data, and are assigned to redactors, often with no explanation as to why a redactor would add such sloppiness to the text. The theory of two sources is thus always sustained.

5. The Mistaken Presumption of Preserved Sources

The source-critical approach rests on the foundational assumption that the biblical redactors faithfully preserved their sources and that these sources, therefore, can be recreated by properly analyzing the received, redacted version we have today. However, this assumption is challenged both by contradictions within the source-critical approach itself and by the evidence we now have of editorial practices of scribes in ancient Israel and in the ancient Near East.

The source-critical approach rests on an internal contradiction in its claims. Source criticism does not produce two complete standalone accounts of the Flood when the fourteen snippets of hypothesized P and the thirteen snippets of non-P are separated and reconstructed. The P account may be considered a full account, but not so the non-P account, where two omissions are notable: First, it lacks a command to build an ark. Moreover, the non-P account does not report the exit from the ark by Noah and the animals. Source critics are forced to concede that the final redactor does not retain full fidelity to the putative original version of this account but has borrowed from it selectively. Source critics claim that material from an original source may be missing, but what is preserved is derived word for word from the original source and can be recovered. Yet, if, as we have seen, the putative redactor could violate the integral nature of the original version by omitting sections of it, by what right may we assume that he has not supplemented and otherwise altered other phrases in his creation of the final version of the text before us?

Source critics retort that the redactors sought to conflate the original sources as fully as possible so as to create a relatively seamless whole and tried to tamper with the original text as little as possible. But the very repetitions and inconsistencies noted by source critics in the biblical text before us undermine that claim. For their theory to account for the unevenness of the Flood narrative as found in Genesis 6–9, source critics must uphold three claims: (1) that the redactor worked tirelessly to disassemble the original sources and then conflate them, combining a total of twenty-seven snippets, some no longer than a phrase; (2) that the redactor freely omitted material from the non-P source, and yet with no clear explanation of why he would or could do so; (3) that in the end the redactor(s) had free reign to tamper with the text, and yet performed his (their) task in sloppy fashion, or that later accretions are responsible for the unevenness seen in passages such as the accounts of Noah’s embarking the ark, discussed earlier.

The stakes here for the field of critical biblical studies are enormous. The very enterprise of tracing the history of composition of the biblical texts rests on the assumption that the earlier sources are recoverable solely on the basis of the so-called clues and evidence within the received text, and without supporting textual witnesses or epigraphic evidence. But those putative sources are available only if we assume that redactors and editors never altered or augmented their sources. Were source critics to concede the possibility that earlier sources had undergone alteration or augmentation, their concession would effectively shut down the quest for the compositional history of the text; it would no longer be possible to work backward from the received text and to isolate the earlier source texts. Scholars committed to tracing the history of the text, therefore, have a vested interest in upholding the axiom that original sources were neither altered nor augmented during redaction.

Conclusion: Unity in the Flood Story

One of the cardinal sins of source criticism has long been a scholarly conceit: if something in the text looks inconsistent to our modern eyes, it must have always seemed inconsistent to readers of any age. But the fact of the matter is that the notion that a text must read in linear fashion with no repetitions is an idea that has reached the West from the writings of Aristotle. In fact, it is demonstrable that literary aesthetics – like aesthetics in any field, such as music, art, etc. are culturally dependent. The nineteenth century scholars like Julius Wellhausen who gave us the documentary hypothesis, were unaware of ancient Near Eastern writings, which were only discovered and deciphered late in that century. Over the years these findings have upended many of the underpinnings of this theory. For instance, it was initially thought that the Tetragramatton (often spelled in English YHWH) and the divine name Elokim were two mutually exclusive names for God, and were proof positive of the claim that the Torah’s narratives were interwoven sources. But scholars now know that many ancient writings refer to one and the same god by different names; these merely reflect different aspects of one and the same deity – indeed, just as our midrashic sources always maintained about these two divine names.

Or, for another example, these scholars pointed to what is called the change in address in some biblical laws. The law will commence by addressing “you” in the singular, but midway through the law switch to addressing “you” in the plural. Source critics saw this as evidence that two traditions of the law had been interwoven; after all, we wouldn’t speak or write that way. But that an eigth century BCE letter of a king to his officers was discovered, in which the king vacillates between addressing the officers each in the singular and collectively in the plural.

One of the things we have learned about writing during this time is its penchant for chiastic structure. This is a literary technique whereby two ideas, A and B, together with variants A' and B', are arranged in the text as A,B,B',A'. Chiastic structures that involve more components are sometimes called "ring structures." The Flood account of Genesis 6–9, for all of its repetitions, forms an elaborate chiastic structure:

A. Elokim pledges to Noah to destroy all flesh (6:13)

B. Flood to destroy all flesh (6:17)

C. Covenant to sustain Noah and his animals (6:18–20)

D. Command to gather food while world is destroyed (6:21)

E. Command to enter the ark + fulfillment (7:1–5)

F. Year 600 – beginning of the flood (7:6)

G. Birds enter the ark (7:8)

H. Seven days waiting for Flood (7:10)

I. Rain on the earth (7:12)

J. Birds enter the ark (7:14)

K. Hashem shuts Noah in (7:16)

L. Forty days of Flood (7:17a)

M. Waters increase (7:17b–18)

N. Mountains covered (7:19–20)

O. 150 days when waters prevail (7:24)

God remembers Noah (8:1)

O’. 150 days when waters abate (8:3)

N’. Mountaintops visible (8:4–5)

M’. Waters abate (8:5)

L’. Forty days of receding waters (8:6a)

K’. Noah opens window of ark (8:6b)

J’. Raven and dove leave ark (8:7–8)

I’. Water on the earth (8:9)

H’. Seven days waiting for water to subside (8:10)

G’. Dove leaves the ark (8:10b–12)

F’. Year 601 – the earth dries (8:13)

E’. Command to leave the ark + fulfillment (8:15–19)

D’. Commands regarding food in the new order (9:1–5)

C’. Covenant to sustain all flesh (9:8–10)

B’. No flood will destroy flesh (9:15)

A’. Elokim pledges to Noah to preserve all flesh (9:17)

There are several aspects of this structure that are worthy of note. Note that it centers around Genesis 8:1: “And God remembered Noah and all of the animals that were with him in the ark.” This is the turning point of the story. All that precedes this verse is death and destruction. All that follows is rebirth and recreation – it is exactly the verse we would expect to find right in the middle of the structure. We ought to note as well that this is an extremely tight and elaborate structure. By “tight” I mean that it has no major gaps anywhere in the three chapters it covers. By “elaborate” I mean that it covers matching pairs of seventeen motifs and elements. But most significant is this: Every matching pair is much more than a deliberate repetition. Rather each pair is a report of an event of destruction and its matching event of recreation. Thus element A is a divine pledge to destroy all flesh, while A’ is a divine pledge to preserve all flesh; element B tells that the flood will destroy all flesh, while B’ tells that never again will a flood destroy all flesh, etc. This structure is found only when the entire Flood account of Genesis 6–9 is read in order.

Click here to order your copy of Ani Maamin: Biblical Criticism, Historical Truth and the Thirteen Principles of Faith (Maggid, 2020)

🤯 ⇐ That's you after reading our weekly email.

Our weekly email is chock full of interesting and relevant insights into Jewish history, food, philosophy, current events, holidays and more.
Sign up now. Impress your friends with how much you know.
We will never share your email address and you can unsubscribe in a single click.
linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram