Jump to ContentJump to Main Navigation
Genesis as DialogueA Literary, Historical, and Theological Commentary$

Thomas L. Brodie

Print publication date: 2001

Print ISBN-13: 9780195138368

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0195138368.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 26 February 2017

(p.495) Appendix Four Sources: The Theory of Four Hypothetical Documents (J, E, D, and P)

(p.495) Appendix Four Sources: The Theory of Four Hypothetical Documents (J, E, D, and P)

Genesis as Dialogue

Thomas L. Brodie

Oxford University Press

Abstract and Keywords

The documentary theory of four hypothetical sources (JEDP) is immensely attractive. It gives a ready explanation to difficult data, and it is flexible (for instance, J – the Yahwist – can be redated or extended). However, though the four letters correspond to data in the text, the theory as a whole is radically confused and misleading. It is a literary theory that is built not on genuine literary criticism but on some relatively superficial data, and especially on considerations of history and theology.

By the time extant sources of Genesis began to emerge in 1872 (cf. Appendix 3) several hypothetical sources had already occupied the imagination. The emerging extant sources gradually gained a foothold, but only in a secondary position: they were generally seen as sources not of Genesis but of the hypothetical documents. The hypothetical documents retained their primary place.

Yet as already stated (beginning Appendix 1), it is not possible—as a general literary principle—to take a finished text and reconstruct diverse sources that otherwise have never been seen, sources that are hypothetical. Such a conjectural process is a last resort, to be undertaken only if there is no prospect of identifying sources that are known, and if, through unusual circumstances, the hypothetical documents emerge clearly and withstand prolonged testing.

In the case of Pentateuchal studies this conjectural process was virtually the first resort, thus causing centuries‐long confusion. The detailed history of such literary theorizing is rich and complex (for surveys of scholarship see esp. Westermann, 1984, 567–574; 1985, 86; Knight, 1985; Rogerson, 1991; Moberly, 1992; Campbell and O'Brien, 1993, 1–20; Houtman, 1994; Whybray, 1987; 1995, 12–28). What is important, however, in assessing its lasting value is not its rich complexity but its underlying logic, the pivotal dynamics that have enabled it to function. The purpose of this appendix is to summarize leading aspects of that logic.

The Runaway Principle

In 1678, Richard Simon's pioneering study (Histoire critique du Vieux Testament) concluded that Moses was not the only author of the Pentateuch. Thus he launched the quest for the Pentateuch's origin, for its sources.

Seventy‐five years later an answer came from a French medical doctor working at the court of Louis XIV. Jean Astruc, though he held for Mosaic authorship, saw that Genesis uses two divine names (Elohim; and YHWH), and (p.496) so he drew a simple conclusion: Moses' Pentateuch is based on two different sources.

Earlier rabbis, who had also noticed the variation, had regarded the different names as expressing different aspects of God (Sandmel, 1978, 329). Astruc did not refute the rabbinic tradition, but his conjecture—his book was called Conjectures (1753)—had a double advantage: it is easier to imagine two documents than two aspects of God; and, above all, whether Astruc wanted it or not, the idea of two documents filled the developing vacuum in the theory concerning the sources of the Pentateuch. In commercial terms, the market needed sources, and Astruc's documents, however conjectural, however simplistically conceived, seemed to meet that need. The idea of two documents stayed.

Implicitly, a central literary principle was thereby established: variation in the text may be used as an indication of diversity of sources. There was no litterateur at hand to say that the principle was utterly unreliable, that for several reasons authors may change style and point of view.

Over the next two hundred years, application of the variation principle engendered a proliferation of claims to sources and editors. Every variation or complexity could be seen as evidence of diverse sources. The principle of variation went out of control. Applied to the Pentateuch it tended to divide it into fragments.

History: A Decisive Role

There were essentially two possible ways of bringing equilibrium to the runaway literary principle: (1) genuine literary criticism, a literary appreciation that would set the phenomenon of variation in literary context; or (2) historical criticism, a picturing of Israel's history around a few main lines or events, thus attracting the scattered fragments into a few main clusters.

Of these two possibilities, what was needed was the first. The problem, the runaway principle, was literary, and needed a literary solution. But literary help was scant. Most biblical scholars were preoccupied with theology, history, or apologetics, and another two centuries would pass before full‐fledged literary criticism would influence biblical research deeply.

In the interim, the prevention of literary proliferation was carried out by historical criticism. Without really clarifying the literary principles, the historians effectively tied the discussion of sources to a specific historical framework, including the historical hypothesis that Deuteronomy constituted the program for a seventh‐century reform by Josiah (2 Kings 22–23). (This hypothesis was formed without engaging the question of the possible literary unity of Genesis‐Kings or without considering how well the account of finding the lost Law fits into the Primary History's overall plot.) Wellhausen (1883) provided the linchpin. He linked a plausible reconstruction of the history of Israel to a theory of four hypothetical sources (JEDP). The plausibility of the historical reconstruction (p.497) lent plausibility and stability to the idea of the four hypothetical sources. JEDP became standard.

The Attraction

The JEDP theory is extremely attractive. It not only provides three sets of answers—concerning history, sources, and variations—but does so with a minimum of difficulty. The essential dynamics are easy to grasp: diverse historical situations produce diverse documents; these diverse documents are assembled into one text; and, almost inevitably, the one resulting text contains tensions or variations.

Within this attractive theory there is still room for great movement. The history may be adapted (for instance, by changing dates); the sources may be redefined, or reduced (say, to three; or, for Genesis, to two); and some of the variations may be explained in some other way. The dating of the hypothetical J provides a good example of variation: estimates range from Solomonic times (von Rad, 1962, I, 55; Wittenberg, 1988) to the exile or later (van Seters, 1975; Schmid, 1976; 167; Blenkinsopp, 1995, 1). Or one may place both J and E in the eighth century (Seebass, I, 34).

This flexibility—this room for movement—makes the theory all the more attractive. The JEDP hypothesis, therefore, forms a congenial place in which to live and work.


As already suggested, the JEDP theory has two aspects—historical and literary. It implies a reconstruction of Israel's history; and it also claims to account for a literary text, that of the Pentateuch, including Genesis.

At the historical level, the theory is coherent. This does not make it true. The idea, for instance, that Deuteronomy provided a program for a seventh‐century reform has now been countered (McConville and Millar, 1994, 141). Ultimately the historical reliability of the account of finding the Law is akin to the historical reliability of the account of writing it. But at least the theory is not self‐contradictory. True or not, it has an advantage: in discussing distant centuries, it fills a void. This is probably its primary strength, the decisive reason it has endured.

At the literary level the JEDP theory is a mixture of insight and confusion. It is insightful insofar as it distinguishes, at least implicitly, between two of the basic stages in the composition of the Pentateuch—an initial stage of writing an overall account (J or JE), and a final stage of composing law (much of D and P) (see Introduction, Chapter 9).

(p.498) But there is confusion also, particularly because of the principle that variation is a clue to distinction of sources. The problem is twofold.

First, the principle does not work well. The effort to apply it appears, in practice, to result in changes and inconsistencies. As far as Genesis is concerned the theory meant three sources: JEP. Then, under closer scrutiny, some proponents found the distinction between J and E unconvincing (e.g., van Seters, 1992, 4), and so, for these, the theory, as applied to Genesis, now means just two sources: JP. But the criteria for distinguishing J and P show multiple inconsistencies (see esp. Rendtorff, 1990). A so‐called pedantic feature of P occurs in a classic J text (Rendtorff, 1990, 149–150); likewise, a verb usage that is supposed to be a mark of P occurs in J (ibid., 151–152); P is distinguished at one stage by being rambling, at another by being brief (ibid., 153). And so on; many of the supports for the theory turn out under inspection to be made of sand (see also Blum, 1984; Radday and Shore, 1985, 8). The confusion is complete when those who hold for the JP distinction disagree on its basic meaning—on whether P supplements J (van Seters, 1992, 4) or J supplements P (Blenkinsopp, 1995, 1) (see also Vervenne, 2001). At times, the disagreement seems based on views of history rather than on literary criticism. Overall, the contradictions in the explanation are greater than the apparent contradictions in the text.

Second, and more importantly, the principle itself is emerging as wrong. Variation in the text need not indicate variation of source. Inch by inch, as literary understanding develops, the variations are emerging as expressions of deliberate art, literary art which is theology‐oriented.

Thus the conclusion begins to dawn: like a confusing myth, the JEDP theory has created an unreal world.

It is symptomatic of the unreality of the JEDP world that instead of using the term “editor” it prefers to speak of “redactors.” Editors are indeed of various kinds, yet the word “editor” has a reasonably clear meaning, one generally associated with bringing texts into order. The perverse glory of the word “redactor” is that it is essentially meaningless, and so it can be used to fill any gap in any theory. It may be connected with order; or it may just as easily be connected with confusion. It may refer to minimal activity, but then again it may be associated with thorough authorship. Hence Vervenne's question (2001, 65): “What is meant by the terms ‘redaction’/‘redactor’?”

The distinction between J and P is something like the distinction between East and West, the distinction that excluded the influence of the Greeks; it appeals to a superficial differentiation of people(s). As the East‐West distinction appealed to caricatures of Greeks and Asians, as though they had nothing in common, so the JP distinction has appealed implicitly to caricatures of two supposedly distinct authors: one, full of humanity (J), but curiously incapable of handling some basic features of life, including genealogies; the other, a priest (P), competent with genealogies, rituals, and some other features, but lacking humanity.

(p.499) The problem involves a paralysis of imagination.1 For complex reasons, historical and psychological, there is difficulty sometimes in imagining a human priest. In this situation, priest, by definition, means a lack of humanity. Humanity, almost by definition, has little to do with the world of priests. The problem is akin to others, such as the difficulty of grasping the unity of the transcendent and the earthly (Gen. 1 and 2), or the difficulty in times past of imagining a woman author. People are boxed within narrow categories.

The truth is the opposite: people who are fully human break all sorts of categories. A woman writer such as Annie Proulx of Wyoming—author of Close Range and The Shipping News—can enter fully into the reality of masculine life. People who are thoroughly human make the best priests; a true priest is fully human. Priest authors/historians include Berossos of Babylon, Manetho of Egypt, Josephus, numerous chroniclers, and, later, thinkers such as Luther, Copernicus, and Teilhard de Chardin.

This does not necessarily mean that priests were largely responsible for the Pentateuch or Primary History. If the Athenians could commission an outsider, Hellanicus, to write their history, then the Jerusalem authorities could do likewise. It simply means that the division between P and J, particularly as applied to Genesis, has more to do with modern constricted categories than with the rich reality of ancient litterateurs. “P,” as used in the JEDP theory, is a caricature. Ironically, van Seters and Blenkinsopp are both essentially right. The texts—all the Genesis texts—supplement each other. Or, more simply, they form a unity.

As for who wrote these texts, maybe it was a priest. Or maybe the Jewish authorities put their trust in a great prophetic writer, who, while well acquainted with worship, was able to place worship in the context of a larger world. It may seem unlikely that a prophetic writer would have such knowledge of detail, including detail of worship. But again, greatness of spirit does not exclude gathering sources and attending to detail. Annie Proulx may build primarily on insight; but she takes extraordinary pains to get every detail right.

Dividing the Unity

It can be objected of course that one source (J) has none of the characteristics of the other (P). Therefore the two are distinct. They are indeed distinct, but only because modern scholarship has made them so. If, from some complex (p.500) object I remove all those components that have a specific series of traits, I will effectively divide the object in two—one part with those traits, the other without. I can then claim they are distinct; one part has none of the distinguishing features of the other.

With this kind of reasoning one could conclude that a Mercedes is not one object. Rather it is a combination (a mechanical redaction) of a small house (the outer frame of the car, including windows) with an open‐air many‐wheeled traveling machine (the lower part). The proof: the house has no wheels. True enough, the top of a Mercedes does not make a wonderful house, but neither do some of the conjectural sources make wonderful narratives. The J source, of course, makes a fairly good narrative. Indeed; and the bottom half of a Mercedes could make a powerful driving machine.

Most complex objects can be divided in two (or three or four). Some of Picasso's paintings, particularly those with overlapping faces, could be taken as reflecting two artists. The human body can be deboned, thus giving two distinct entities—a complete skeleton, and a body that, apart from being boneless, is near complete. Neither has any of the distinctive traits of the other, but that does not prove that they are of diverse origin. One may argue about which supplements which, but in fact they are a living unity; together they come into being, and they live through each other; there is only one body, a body that is greater than the sum of its parts. To treat the skeletal frame as separate—to comment on P as a distinct source—is to treat what is living as if it were dead. And to treat J without P is to create an animal that is spineless.

Further confusion occurs when the alleged diversity of JEDP is compared with the actual diversity between versions of certain ancient epics, such as the Atrahasis Epic or the Epic of Gilgamesh (see particularly the nuanced presentation of Carr, 1996, esp. 16–40). These epics existed in diverse versions, and therefore, according to this argument, so could the Pentateuch—in J, E, D, and P (Carr concentrates largely on P). The Jesus narratives are another example; there are four versions.

But in the cases of the epics and Jesus the versions are real; the diverse texts or manuscripts exist. For JEDP, however, there is not a shred of evidence, not a fragment of a manuscript of any one of these alleged versions. The comparison with Atrahasis, Gilgamesh, and Jesus serves not to confirm JEDP but to disqualify it even more thoroughly, to underline its unreality.

The issue is not whether something can be divided in two (or three or four) but whether it is more intelligible when taken as a unit. When Genesis is taken as a unit it is indeed perplexing, but ultimately it is supremely intelligible—great literary art, with a magnificent vision of the struggle and richness of life and of a transcendent dimension surpassing human calculation. Its variations emerge increasingly as features of literary artistry, including artistry that is diptych‐shaped. The text is complex, but it is orderly—a delicate tapestry where essentially every piece fits. Some aspects still await explanation, but their proportion is diminishing.

(p.501) And all the while, while literary explanations are gradually blurring the distinctions underlying the hypothetical documents, the presence of extant literature is coming closer and closer to the center of the Primary History, closer especially to the center of Genesis. Extant epics, including the Odyssey, are no longer the sources of hypothetical sources. Rather, they are among the sources of Genesis itself. Such are the prophetic books, especially Ezekiel, Isaiah, and Jeremiah. The old principle applies: entities are not to be multiplied unnecessarily. JEDP are not only increasingly unfounded; they are also increasingly unnecessary. Rendtorff's judgment (1993, 44) is truer than ever: “I believe that the traditional Documentary Hypothesis has come to an end.”

In Kuhn's analysis of scientific revolutions (1970, 144–159, esp. 151), the old paradigm lives on after the basic revolution has happened and even after an alternative theory has achieved essential credibility. It is likely then that some form of the JEDP theory will continue for many years. (p.502)


(1.) For Vervenne (2001, 41–42), the P debate sometimes “underestimates creative forces. . . deal[s] with a narrow concept of ‘redactors’. . . [uses] clichés. . . tend[s] towards some form of ‘ideologization’. . . [and] is fraught with a great deal of emotion.” The paralysis of imagination is particularly powerful when the imagined division between a Yahwist and a priest is identified—consciously or unconsciously—with a real division between modern religious groupings. The division becomes a matter of faith.