Z Wenham, The Priority of P.pdf

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Vetus Testamentum.

http://www.jstor.org

The Priority of P Author(s): Gordon J. Wenham Source: Vetus Testamentum, Vol. 49, Fasc. 2 (Apr., 1999), pp. 240-258Published by: BRILLStable URL: http://www.jstor.org/stable/1585645Accessed: 28-08-2014 11:02 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of contentin a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.For more information about JSTOR, please contact [email protected].

This content downloaded from 192.231.59.35 on Thu, 28 Aug 2014 11:02:57 UTCAll use subject to JSTOR Terms and Conditions

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=bap

http://www.jstor.org/stable/1585645

http://www.jstor.org/page/info/about/policies/terms.jsp


THE PRIORITY OF P

by

GORDON J. WENHAM Cheltenham

For the last hundred years it has been the received wisdom in Old Testament scholarship that P is the latest of the pentateuchal sources or redactional layers. Yet before Wellhausen' established this as the consensus view, it was generally held that P was one of the oldest

sources, the so-called Grundschrift, e.g. de Wette, Hupfeld.2 Dillmann3 continued to maintain the priority of P even after Wellhausen had written. But since Dillmann few scholars4 of stature have maintained the priority of P. Yet in other areas the Wellhausenian consensus has been fiercely attacked. The analysis into sources has been challenged, so that many doubt the existence of an independent E source. The date ofJ has been questioned: as many now argue for a sixth-century J as for a tenth-century J. But the relative order ofJ and P has gone unquestioned: that J was written before P is still generally accepted.5 The modern debate is only about how much earlier J is than P, not whether J is earlier.

J. Wellhausen, Geschichte Israels I (Berlin, 1878). 2 W.M.L. de Wette, Beitrage zur Einleitung in das Alte Testament (Halle, 1806-7); H. Hupfeld,

Die Quellen der Genesis und die Art ihrer Zusammensetzung von neuem untersucht (Berlin, 1853). 3 A. Dillmann, Die Genesis5 (Leipzig, 1886), pp. xv-xix. 4 Two recent exceptions are J.L. Ska, "El relato del diluvio: un relato sacerdotal y

algunos fragmentos redaccionales posteriores", Estudios Biblicos 52 (1994), pp. 37-62, who confines his discussion to the flood story and argues that the J elements are secondary. He does not think this is true elsewhere in Genesis. J. Blenkinsopp, The Pentateuch. An Introduction to the First Five Books of the Bible (London, 1992), pp. 61-97, suggests that at many points in Genesis material traditionally ascribed to P antedates so-called J material (e.g. "the Priestly strand (P) is basic... It has been filled out at several points with narrative expansions-corresponding to the J of the documentary hypothesis" [p. 93]). However Blenkinsopp relies too much on vocabulary to establish relative dates, which is a slippery argument.

5 Representative of this school of thought are: J. Van Seters, Abraham in History and Tradition (New Haven, 1975); E. Blum, Die Komposition der Vdtergeschichte (Neukirchen- Vluyn, 1984); C. Levin, Der Jahwist (Gottingen, 1993).

? Koninklijke Brill NV, Leiden, 1999 Vetus Testamentum XLIX, 2



THE PRIORITY OF P

Here I propose to re-examine this axiom of pentateuchal criticism, at least as it applies to the book of Genesis. The issues raised by the cultic laws in P are of quite a different order and have been discussed at length elsewhere.6 However I do not propose to treat these questions historically, simply from a literary/redactional standpoint. I am not trying to determine the absolute dates of the J and P material, just their relative ages. I shall look at the P passages in Genesis and

attempt to discover how they relate to the nearby J passages. It is obvious that as the texts stand at present P passages comment on and illuminate J passages and vice versa. But can one say which was written first? Are the P passages a commentary on the J material or are the J passages expansions of the P material?

The P material in Genesis covers a variety of genres. There are

genealogies (chs. v, xi), a table of nations (ch. x), straight narrative (ch. xxiii), an extended promise (ch. xvii), a semi-poetic creation account

(i 1-ii 3), sundry chronological notes (e.g. xvi 3, 15-16), and headings to sections (e.g. vi 9).

Usually the P material appears quite distinct within Genesis: genealogies for example interrupt the ongoing story line. But in two or three instances P and J are held to be interwoven, in the flood story (chs. vi-ix), in the table of nations (ch. x) and possibly in ch. xxxiv as well. These supposedly composite passages are the most difficult to evaluate source critically, so I shall begin by examining those passages where the P material is most distinct before looking at the more complex sections.7

Genealogies

Within Genesis the genealogies stand out as distinctive blocks, par- ticularly v 1-32 and xi 10-26. They are characterised by exact repetition of standard formulae. "When A had lived x years, he became the father of B. A lived after the birth of B y years, and had other sons and daughters. Thus all the days of A were x + y years and he died". This recurs six times in this exact form in vv. 6-8, 9-11, 12-14,

6 Among recent discussions I. Knohl, The Sanctuary of Silence: The Priestly Torah and

the Holiness School (Minneapolis, 1995) and J. Joosten, People and Land in the Holiness Code (Leiden, 1996) break new ground in these discussions.

7 Within the space of an article it is not possible to deal with every passage. For further discussion see my commentaries Genesis 1-15 (Waco, 1987) and Genesis 16-50 (Dallas, 1994) usually under the heading "Form, Structure, Setting".

241



GORDON J. WENHAM

15-17, 18-20, 25-27. There is a slight expansion of the first formula in v 3-5, which adds "a son in his own likeness, after his image and named him". There is also expansion in the comments about Enoch in v 22, "walked with God" for "lived", and in v 24 "Enoch walked with God; and he was not for God took him". However it is not clear that these differences are to be ascribed to a different hand. But in v 28-31 the regular formula is expanded by "and he called his name

Noah, saying 'out of the ground which the LORD has cursed this one shall bring us relief from our work and the toil of our hands"'. The final formula in v 32 is slightly altered and cut short. The expected completion of the formula reappears in ix 28-29.

The genealogy in xi 10-26 shows less variation. In translation the standard formulae appear to be the same as in chapter v with the third element, "Thus all the days of A were x + y and he died", omitted. But it should also be noted that the opening part of the formulae differs slightly. In ch. v it begins "A 'nl" (i.e. waw consecutive +

imperfect verb + subject), whereas in xi 10 it begins "When A (= Shem) was x", and in xi 12 "'n Al" (i.e. waw + subject + perfect verb). But from verse 16 it follows the verbal pattern of chapter v. In the last verse of the genealogy only half the formula is present "when A

(= Terah) had lived x years, he became the father of.. ." (xi 26) What could be the close of the formula, albeit in the style of ch. v, is found in xi 32 "The days of Terah were 205 years; and Terah died in Haran".

The similarity in formulation between these two genealogies makes it likely that they both come from a common source. The introduction to ch. v, "This is the book of the generations of Adam", instead of the usual, "These are the generations of Adam/Noah etc.", (cf. vi 9) seems to imply that v 1-32 is based on a written source, and it is

tempting to suppose that the same source also underlies xi 10-26.8

Certainly the regularity in the formulae allows us to identify editorial changes to the source, on the assumption that it was more regular than the present text of Genesis. This assumption may not be warranted, but if the deviations from the regular coincide with the known ten- dencies of the redactor elsewhere, we are likely to be right in ascribing the deviations to the redactor rather than his source.

Two of the deviations in chapter v clearly fulfil these criteria and

may confidently be ascribed to the redactor. The first, v 29 "And he

8 G. von Rad, Genesis2 (London, 1972), p. 70.

242



THE PRIORITY OF P

called his name Noah, saying 'out of the ground which the LORD has cursed this one shall bring us relief from our work and from the toil of our hands"', not only uses the divine name Yahweh and alludes

quite clearly to the curse of the ground in iii 17 pronounced by the

LORD, but it also makes a pun on his name just as vi 6-8 does.9 The

opening clause in v 29, "and he called his name A" is also found in another deviation from the usual formula in v 3, "and he called his name Seth". Identical phraseology occurs in iv 26; xxv 25, 26; xxxv 8, 10; xxxviii 3, 29, 30; and similar terminology occurs in about twenty other passages. This makes it likely that the clauses in v 3 and 29 should be ascribed to the editor. That these deviations occur towards the beginning and the end of the putative P passage also enhances the case for seeing them as redactional.

It is clear that there has been editorial intervention at the beginning of each genealogy and also at the end of them. Not only is there a heading "This is the book of (These are) the generations of. .", but both genealogies seem to have had the last formula dismembered, so that the expected conclusion of the formula occurs much later at ix 28 and xi 32. In the first case the flood story vi 1-ix 27 separates the first and last parts of the formula, and in the second xi 27-31 does the same.

It is universally agreed that v 29 is a J passage. It makes no sense on its own, but it does illuminate the genealogy highlighting Noah who is to be the key figure in the following section and making the first

play on his name, of which there are to be several more in vi 5-8, another J passage.'0 The nearest parallels to v 3, "he called his name Seth", occur in iv 26 and v 29 ("he called his name Enosh/Noah"). Both are ascribed to J, so it seems likely that v 3 is J gloss too.

Thus looked at in isolation, v 3-32 appears to be a genealogy with

J glosses (v 3 and v 29). If this genealogy is ascribed to P, then it follows that P is one of the sources used by J rather than vice-versa. The close of the genealogy with two major elements v 32 and ix 28 sep- arated by the flood story is also compatible with this conclusion, though not absolutely required. It looks more likely that the flood story was inserted into the genealogy than that the genealogy was added later to the flood story. Only the heading in v 1-2 could suggest the opposite conclusion, and we shall return to this later.

9 A. Strus, Nomen-Omen (Rome, 1978), pp. 66, 158-9. 10 Cf. GJ. Wenham, Genesis 1-15, pp. 128, 144-5.

243



GORDON J. WENHAM

The genealogy of Ishmael in xxv 12-18 consists of a list of Ishmael's twelve descendants (vv. 12-16), a note about his death and burial (v. 17) and then a comment on their territory, "They dwelt from Havilah to Shur... he settled over against his people" (v. 18). Most commentators recognise that this is a fulfilment of the promise in xvi 12b, "He shall dwell over against all his kinsmen", which is part of a J narrative. It therefore appears that xxv 12-18 is a genealogy with aJ comment tacked on the end. Because it is a genealogy and prefaced by the toledot formula, xxv 12-17 is traditionally assigned to P. If this is

correct, this suggests that the P material has been supplemented by J,

just as in chapter v. The genealogy of Esau (xxxvi 1-43) poses problems for every source

theory. Its very length is the first: generally genealogies of the non- elect lines (e.g. Cain, Ishmael) are dealt with briefly in Genesis (cf. xxv

12-18), but this runs to 43 verses. xxxvi 1-8 would have been long enough for an Esau genealogy. The second oddity is the double toledot formula in vv. 1 and 9. Elsewhere in Genesis, each new section is introduced by just one toledot formula. Why should Esau's genealogy have two? Furthermore the presence of two toledot formulae here brings the total in the book to eleven, whereas ten would seem a more likely figure. It is therefore agreed that there is a diversity of sources here, and that xxxvi 9-43 is a later supplement to xxxvi 1-8. The amount of detail devoted to Edom is surprising given later animosity between Israel and Edom, so that Westermann's suggestion" that vv. 9-43 were

perhaps part of the Edomite archives brought to Jerusalem when David

conquered Edom is not impossible. At any rate it is likely that vv. 1-8, usually assigned to P, represent an earlier part of the tradition than vv. 9-43.

In xlvi 6-27 we have the genealogy of Jacob's sons which is usu-

ally assigned to P. As elsewhere in Genesis this P material is prefaced by a J introduction, a divine call to Jacob to leave his home and go to a foreign land, a call that in content and phraseology echoes the call of Abraham. Then from v. 28 J resumes: so once again we meet a genealogy topped and tailed byJ material. Furthermore within the

genealogy itself there are signs that an older document has been accom- modated to the J tradition. For example Dinah is mentioned in xlvi 15 (cf. ch. xxxiv) and the death of Er and Onan in xlvi 12 ( cf. xxxviii

1 C. Westermann, Genesis. 2. Teilband: Genesis 12-36 (Neukirchen-Vluyn, 1981), pp. 683-4.

244



THE PRIORITY OF P

2-11). Thus xlvi 6-27 is a good example of an earlier source incor-

porated by J, who has added a preface and conclusion and modified or glossed the content slightly to suit his own scheme.

Chapter x, the table of the nations, is not strictly a genealogy, nor is it usually regarded as a substantially homogeneous text with just a few glosses, but as a complex mixture ofJ and P. This makes it more difficult to determine which is the primary source. Conventionally vv.

la, 2-7, 20, 22-23, 31-32 are assigned to P, the rest vv. lb, 8-19, 21, 24-30 to J. But Savasta'2 has plausibly argued that vv. 4c-5, 20, 31-2 should also be assigned to J. Whether the conventional analysis or Savasta's modification is accepted, it is clear that in P we have a short

compact genealogy giving roughly equal space to the three branches of the Noahic family. But the J material is characterised by long digres- sions about certain of the sons of Ham, especially Babel, Egypt and Canaan, and then a similar expansiveness about the sons of Shem, but nothing at all about Japhet.

In other words the P material could stand alone, whereas the J material looks like long glosses on the P list. This seems to be confirmed when the Tendenz of these glosses is considered. They show an interest in the flood, the dispersal of the nations into different language groups, in the origins of Babel, Egypt, the Canaanites, and the Arabian tribes descended from Shem. More precisely these J additions to the P table show the fulfilment of Noah's words in ix 25-27 (J) cursing Ham's descendants and blessing Japhet and Shem's. In short it is easy to see the point ofJ's additions to a P genealogy, but difficult to see why an editor would have wanted to modify J material by adding P.

Burial Notes

Passages dealing with the burial of the patriarchs and their wives are usually assigned to P (xxiii 1-20; xxv 7-10; xxxv 27-29; xlix 29-

32). Chapter xxiii, the purchase of the cave at Machpelah, is generally assigned to P and consequently all those passages mentioning the interment of the patriarchs there are assigned to P as well. But apart from some rather distinctive vocabulary e.g. "Hittites, prince, holding" which may be prompted by the topic rather than the source, there is little that requires this chapter to be assigned to P. It has been described

12 C. Savasta, "Alcune considerazioni sulla lista dei discendenti dei figli di Noe", Rivista biblica 17 (1969), pp. 89-102, 337-63.

245



GORDON J. WENHAM

as "untypical of p",'3 as pointing "strongly to J",14 or as "an older narrative" taken over by p.'5 Rendtorff'6 declared there were "many reasons against" regarding this chapter as P.

If the assignment of this chapter to P is not proven beyond doubt, do the obvious allusions to it in subsequent chapters clarify the situation? xxv 1-11 gives a summary account of the last days of Abraham

including his death and burial in the cave of Machpelah (vv. 7-10). The latter verses are assigned to P because of their links with ch. xxiii. But the surrounding verses (1-6, 11) have commonly been given to J. However Westermann'7 noted that vv. 1-6 presuppose the death of Abraham (in vv. 7-10) and therefore must come from a post-P writer. This divergence of opinion would be resolved if it were admitted that P precedes J chronologically.

Burials of the patriarchs occur in xxxv 27-29 and 1 12-13, both in contexts where source-critical analysis is very difficult. According to the classical source analysis of S.R. Driver18 1 12-13 is framed by the

surrounding J material (1 1-11, 14). Westermann's idiosyncratic belief in a non-J Joseph story comes to a similar conclusion.'9 According to his view vv. 12-13 are P verses surrounded by the end of the Joseph story.

Most source critics ascribe xxxv 27-29 to P, but Coats20 argues it

belongs to J. This would mean that the list of Jacob's descendants

(xxxv 22b-26), ascribed by all to P, would again be framed byJ material in xxxv 21-22a and 27-29. Even on the usual analysis J material introduces the P section. Thus all the short summaries of patriarchal burials seem to consist of P material framed by J.

However we may go further. The diversity of analysis of these chapters is an index of their lack of coherence as perceived by most readers. The hodge-podge of topics appears strange after the well-ordered

3 S.E. McEvenue, The Narrative Syle of the Priestly Writer (Rome, 1971), p. 22. 4 E.A. Speiser, Genesis (New York, 1964), p. 173.

15 G. von Rad, Genesis2, p. 249. 16 R. Rendtorff, Das iiberlieferungsgeschichtliche Problem des Pentateuch (Berlin, 1976),

p. 130. 17 C. Westermann, Genesis 12-36, pp. 482-6. Cf. J. Skinner, A Critical and Exegetical

Commentary on Genesis2 (Edinburgh, 1930), p. 349. 18 S.R. Driver, The Book of Genesis3 (London, 1904), p. 397. '9 C. Westermann, Genesis. 3. Teilband: Genesis 37-50 (Neukirchen-Vluyn, 1982), pp.

222, 228. 20 G.W. Coats, Genesis with an Introduction to Narrative Literature (Grand Rapids, 1983),

p. 245.

246



THE PRIORITY OF P

accounts of patriarchal lives in Genesis xxv-xxxiv, and xxxvii-xlvii. However the parallels between the close of the Abraham cycle, the

Jacob cycle, and the Joseph story suggests there is a method in the redactor's madness.

Call to Journey xxii 1-2 xxxv 1 xlvi 2-4 Patriarch's Obedience :3-14 :2-8 xlvi 5ff. Promise Reaffirmed :15-18 :9-14 xlviii 4 Journey :19 :16 :5 Birth of Sons :20-24 :17-18 :5-6 Death and Burial of Wife xxiii 1-20 :18-20 :7 Son's Marriage xxiv 1-67 :21-22 :8ff.

xlix 3-4 List of Descendants xxv 1-6 :22-26 xlix 3-28 Death and Burial of Patriarch xxv 7-10 :27-29 xlix 29-114

This repeating pattern at the close of each patriarchal cycle suggests that the editor is trying to draw attention to the similarities in the careers of Abraham, Jacob and Joseph. According to traditional source critics J is the major component and J's narrative frames the whole.

The picture becomes even clearer, if with many recent critics we abandon E as a separate document and regard it as part ofJ. Then

J frames the account of Abraham's purchase of the cave at Machpelah as well as subsequent references to it. J also frames the promise section in xlviii 4 f. and the list of Jacob's descendants in xlvi 6-27. As we have already observed, the latter appears to be an old list modified to reflect narratives assigned to J.

Promises in P

In xlviii 3-4 we have a promise passage which echoes xxxv 9-12 which in turn echoes xxviii 3-4 which in its turn looks back to ch. xvii. Because of the distinctive vocabulary in each passage e.g. El

Shaddai, be fruitful and multiply etc. they are all ascribed to P. It is not my purpose here to contest this attribution, though it may well be asked how far the genre of promise as opposed to a different author, is responsible for the distinctive vocabulary. What should be noted is that in each case the promise passage seems to be embedded in J material,21 and indeed in chapter xvii seems to be cast in patterns

21 Cf. GJ. Wenham, Genesis 16-50, pp. 203-4, 323, 461.

247



GORDON J. WENHAM

attested in neighbouring J sections. Furthermore the content of that

chapter is presupposed in succeeding chapters. Chapter xvii resembles the preceding chapter in the following ways.

First, both close with references to time xvi 16, cf. xvii 24-27. Second, both chapters are built round a long central scene. In chapter xvii this consists of five divine speeches (vv. lb-2, 4-8, 9-14, 15-16, 19-21) inter-

rupted by two questions from Abraham. In chapter xvi the central scene consists of four angelic speeches (vv. 8a, 9, 10, 11-12) with two comments by Hagar (vv. 8b, 13). In both chapters some of the intro-

ductory quotation formulae, "And God/the angel of the LORD said", strike the modern reader as redundant, serving only to break up the

speeches.22 Chapters xviii-xix resemble the preceding chapter in the following

ways. First, xviii 1 "Then the LORD appeared to him" echoes xvii 1 "Then the LORD appeared (to/Abraham and said) to him".

Secondly, both sections exhibit a double structure, a palistrophe super- imposed on parallel panels. McEvenue23 noted this in chapter xvii.

The palistrophe consists of la Abraham 99 A lb-2 1st speech B 4-8 2nd speech C 9-14 3rd speech D 15-16 4th speech C' 19-21 5th speech B' 24-25 Abraham 99 A'

The parallel panels arrange themselves as follows:

1-2 Oath about progeny 16 Sarah and progeny 3a Abraham falls on his face 17-18 Abraham falls on his face 4b-6 Abraham a father 19 Sarah a mother 7 Eternal oath 19b,21a Eternal oath 9-14 Sign of oath 23-27 Sign of oath

I noted a similar arrangement in chs. xviii-xix. xviii 16-xix 28 may be read in eleven scenes arranged palistrophically with the first match-

ing the last, the second the penultimate and so on.24 But it is also cast in two parallel panels with the visit with xviii 1-32 paralleled by xix

22 For an explanation of this feature see EJ. Revell, "The repetitions of introduc- tions to speech as a feature of Biblical Hebrew", VT 47 (1997), pp. 91-110.

23 S.E. McEvenue, The Narrative Style of the Priestly Writer (Rome, 1971), pp. 156-59. 24 GJ. Wenham, Genesis 16-50, pp. 41-44; idem, VT 41 (1991), pp. 103-5.

248



THE PRIORITY OF P

1-22. This combination of two literary patterns in the same section has been noted only in chs. xvii and xviii-xix.

These similarities in literary structure between chapter xvii and the

surrounding chapters suggest that the material is much more closely integrated than is usually recognised. But who is responsible? Has the

priestly writer remodelled chs. xvi, xviii-xix, or has J, the Yahwist, remoulded ch. xvii? Both hypotheses are possible, but the evidence is better explained by the latter. First xvii la "The LORD appeared to him" appears to be J's editorial introduction to the chapter. It is not

merely the mention of Yahweh that marks this out as J, but the verb

"appear" (Niphal of ;7Il) as in xii 7; xviii 1; xxvi 2, 24. Second, that this introductory remark is secondary to the rest of

the chapter is suggested by the use of "El Shaddai" (God Almighty) in the latter part of the verse. It is typical of J's method to use an El name in the direct speech of God and human actors and then to use Yahweh in narrative comments e.g. xvi 11, 13. It fits in with the the-

ory of Exodus iii 13-15, vi 3 that the patriarchs knew God as El

Shaddai, but that the name Yahweh was first revealed to Moses. The

interpretation of these passages has been much discussed, but it at least shows that the redactors of Exodus regarded El Shaddai as the older name and Yahweh as the younger.25 This makes it easy to take xvii la as the later gloss on the earlier tradition enshrined in xvii lb.

Third, the contents and phraseology of chapter xvii are presupposed in subsequent chapters which are usually ascribed to J. Most obviously the names of Abraham and Sarah are used in ch. xviii without com-

ment, whereas in ch. xvi their names were Abram and Sarai: only xvii 5 and 15 explain the change. Further xviii 9-15 appears to be aware of the etymology of Isaac of xvii 17, 19 and quotes the phrase, 'I1r2_ "at the appointed time" in xviii 14 from xvii 21.26

xxi 1-5 recording the birth and circumcision of Isaac looks back to

chapter xvii as does xxiv 36. The source-critical analysis of xxi 1-7 is controversial. But Westermann's reading of the situation is at once the

simplest and most exact: "w. 3-5 is an untouched piece of P which is really the genealogical conclusion of ch. 17, this is set between two

25 R.W.L. Moberly, The Old Testament of the Old Testament (Minneapolis, 1992), pp. 5-78.

26 McEvenue, pp. 153-5 and Westermann, Genesis 12-36, pp. 321-2 note the parallels and suggest xviii 4-15 is the source of xvii 15-22. But it is easier to suppose the borrowing is in the opposite direction.

249



GORDON J. WENHAM

J passages vv. 1-2 and 6-7".27 But whereas he sees this arrangement as the work of a post-P redactor, I think it is more economically explained as the work of J himself, who often tops and tails earlier source material with his own comments.

Finally T.D. Alexander28 has argued that xxii 1-19 presupposes chapter xvii. He holds that ch. xvii is the promise of a future covenant not the ratification of it, which is found in xxii 16-18, the divine oath to Abraham made after the sacrifice of the ram. He observes a similar structure in the flood story, which contains a promise of a covenant

(vi 18), the fulfillment of its obligations (vi 22; vii 5), the offering of a sacrifice (viii 20), and the establishment of a covenant (ix 9-17). The cited parallels between chapters vi-ix and xvii and xxii are again more easily explained if chapter xvii (P) is earlier than chapter xxii

(J or J/E).29

Narratives combining J and P

There are only two lengthy narratives where according to traditional source critics P material is interwoven with another source, namely the flood story in chs. vi-ix and the account of Dinah and Shechem in ch. xxxiv. In the flood story in vi 9-ix 17 fourteen sections of P are

interspersed with thirteen snippets ofJ. On the whole the P passages are

longer than the J passages, which may consist of mere half-verses such as vii 16b, "and the LORD shut him in". If the division of the flood

story into two sources is fundamental,30 the obvious conclusion from this evidence has been drawn by Westermann who states: "At first glance it is clear that... J has been worked into the basic P document".31 While elaborating at length on this observation Westermann fails to note that if J is being worked into P, rather than vice-versa, P should be regarded as the earlier document that is being modified by J.

27 C. Westermann, Genesis 12-36, p. 405. 28 T.D. Alexander, "Genesis 22 and the Covenant of Circumcision", JSOT 25 (1983),

pp. 17-22. 29

According to conventional source criticism chapter xxii is a blend of J and E, but here as elsewhere the modem trend is to see it as all J or even a post-J editor. For a brief synopsis of critical standpoints GJ. Wenham, Genesis 16-50, pp. 101-2.

30 Here I do not propose to restate my doubts about the validity of the source division in Gen. vi-ix or in Gen. xxxiv (see VT 28 (1978), pp. 336-48; 41 (1991), pp. 84- 109). I regard the attempt to distinguish P material in these chapters as much less secure than say in chs. v, xi, or xvii.

31 Genesis 1-11, p. 533 (cf. ET p. 396).

250



THE PRIORITY OF P

This point has been well grasped by J.L. Ska in a recent study, in which he argues the source analysis of Gen. vi-viii should be slightly modified and the priority of the P material acknowledged. Then it is clear that most of the flood story comes from P and that it has been

supplemented at a few points by J.32 There is also then no need with

many commentators to invoke an additional redactor to explain the

appearance of P material in J contexts such as vii 7-9. J never con- tained a full account of the flood: it merely supplemented P at a few

points. On Ska's analysis J contains no mention of the building of the

ark, no entry into the ark nor exit from it, no account of the waters' rise or fall, no description of the earth drying out. "This proves sufficiently that aJ account of the flood parallel to P's does not exist".33 "The J source in Genesis vi-ix is a series of late fragments, and not an independent account, complete and earlier than the P source".34

They are designed to supplement P with details from the Mesopotamian version of the flood story and demonstrate the importance of the cult.

Even if the older division ofJ and P is retained, two further observations support the priority of P. First, the flood story narrowly defined in vi 9-ix 17 is prefaced and rounded off by passages universally ascribed to J, viz. vi 5-8 and ix 20-27. As already noted above, it is characteristic ofJ to top and tail earlier material with his own comments in this way.

Secondly the flood story finds a very close thematic and structural

parallel in Gen. xviii-xix, the destruction of Sodom and Gomorrah. Both narratives tell of a universal destruction in which one family was saved by divine intervention. Both narratives are arranged as extended

palistrophes.35 There are also numerous allusions in Gen. xviii-xix to the flood story. Many key terms in the flood story are echoed in Gen. xviii-xix.36 Yet while Gen. xviii-xix is ascribed almost in its entirety37

32 J.L. Ska, "El Relato del diluvio: un relato sacerdotal y algunos fragmentos redac-

cionales posteriores", Estudios Biblicos 52 (1994), pp. 37-62. Ska assigns vi 9-22; vii 6-9, 11, 13-16a, 17-21, 24; viii 1-2a, 3-5, 13-19; ix 1-3, 7-17 to P, vi 5-8; vii 1-5, 10, 12, 16b, 22-23; viii 2b, 6-12, 20-22 toJ and a few phrases in vi 17; vii 6, 17, 22 to the redactor (p. 51).

33 Ibid., p. 52. 34 Ibid., p. 60. 35 See GJ. Wenham, VT 41 (1991), pp. 103-5; Genesis 16-50, p. 41. 36 E.g. nnr "destroy" In tr "find favour" 'rnn "preserve life" irt "produce descend-

ants" CtDlL "rain" Dt "remember". For fuller discussion see Genesis 16-50, pp. 42-43; VT 41 (1991), pp. 108-9.

37 Only xix 29 is conventionally ascribed to P in Gen. xviii-xix. If this assumption

251



GORDON J. WENHAM

to J, the verbal allusions to the flood story involve both P and J passages. This is difficult to account for if P were composed after J, but is natural if the J source presupposes P.

Gen. xxxiv as analysed by the traditional source critics presents a similar picture to Gen. vi-ix, namely a fairly full P version comprising about 22 verses with a much briefer J version of nearly 10 verses.38 As in the flood story the P material constitutes a complete tale that can be understood on its own, whereas the J material is not intelligi- ble by itself, but looks like a series of glosses on the P story. Thus it would be natural to conclude that P is here the earlier source expanded byJ. However since this runs counter to the normal order of sources, other critics have come up with quite different suggestions. Some39 have denied that the P material comes from P, but suggested that it comes from E instead. This is unlikely in that xxxiv 15, 22 quote xvii 10 and xxxiv 24 echoes xxiii 10, 18 (both chapters usually assigned to P) and other terms (e.g. t'fl "prince", ntR "acquire property") are

typically ascribed to P. Yet others have argued that much more or even all of the narrative

should be assigned to J, while Westermann,40 although he noted the

affinity of the basic narrative with P, denied that either source in Gen. xxxiv should be identified with P or J. As is the case with the flood

story a unitary reading of Genesis xxxiv is quite possible and may eliminate the need to assign the narrative to two sources.41 This is the most economical hypothesis of all here, but next to it for simplicity and clarity is the old source analysis into P and J, where P is seen as the basic narrative supplemented byJ. The other hypotheses are much

is correct, it is another pointer to P's antiquity, for it is quoted by Amos iv 11. That Amos is quoting Genesis is shown by its mention of "God", the only passage in Amos to do so. Westermann's suggestion Genesis 12-36, p. 364, that Amos may be quoting a tradition independent of Genesis is special pleading prompted by his failure to recognise that P is early.

38 According to S.R. Driver, An Introduction to the OT5 (Edinburgh, 1894), pp. 10, 14. P = 1-2a, 4, 6, 8-10, 13-18, 20-25, 27-29, while J = the rest.

39 E.g. J. Wellhausen, Die Composition des Hexateuchs und der historischen Biicher des ATs2 (Berlin, 1889), pp. 49, 318. H. Gunkel, Genesis9 (Gottingen, 1977), p. 374. J. Skinner, Genesis2 (Edinburgh, 1930), p. 418.

40 C. Westermann, Genesis 12-36, pp. 651-4. 41 E.g. M. Sternberg, The Poetics of Biblical Narrative (Bloomington, 1984), pp. 445-75;

D.N. Fewell and D.M. Gunn, "Tipping the Balance: Sternberg's Reader and the Rape of Dinah", JBL 110 (1991), pp. 193-211; P. Noble, "A 'Balanced' Reading of the Rape of Dinah: Some Exegetical and Methodological Observations", Biblical Interpretation 4

(1996), pp. 173-204.

252



THE PRIORITY OF P

more convoluted, and probably would never have been suggested had not the late date of P been regarded as axiomatic.

This completes our discussion of the genealogies, promise sections and narratives in Genesis that traditionally have been assigned to P. In all cases it has been argued the P material looks as though it has been subsequently reworked by a redactor like J or incorporated into J contexts. Thus the genealogy in Genesis v is glossed byJ comments, and the covenant promises in Genesis xvii are prefaced by a J introduction. Where J and P are interwoven, the P material appears to be the more basic and the J material supplementary (e.g. Gen. vi-ix, xxxiv).

But there are two elements of P not discussed so far which constitute the strongest prima facie case for dating P after J: they are the toledot formulae and the first creation account in Genesis i 1-ii 3. These must be considered now.

The Toledot Formulae

The toledot formula nlblnl F,-l "This is the family history of.". occurs eleven times in Genesis (ii 4; v 1; vi 9; x 1; xi 10, 27; xxv 12, 19; xxxvi 1, 9; xxxvii 2). Everywhere in Genesis it serves as a head-

ing to a cycle of narratives (e.g. vi 9) or to a genealogy (e.g. xxv 12). Attempts to argue that in ii 4 nl5ln fSl introduces a colophon to i 1-ii 3 are misguided, as Stordalen42 has shown. Furthermore -nlrn comes from the root 1'", "to bear", hence the traditional translation

"generations". However this can mislead for the toledot of x are not x's ancestors but x's descendants. Hence the toledot of Terah (xi 27) is the family history of Terah and his descendants, chiefly Abraham, and the toledot of Isaac (xxv 19) is the history of Isaac's sons, Jacob and Esau. The same logic applies to ii 4a, where the "toledot of heaven and earth" is the heading of ii 4-iv 26: these toledot tell what emerged from the heaven and the earth. Furthermore ii 4b "in the day that the LORD God made" parallels similar expansions of the toledot formula in v 1 and Num. iii 1, which show the two halves of the verse cohere and should be ascribed to the same redactor. But whereas the first half of the verse looks forward to what is going to happen in the fol-

lowing chapter, the second half of the verse looks back to what has

already been achieved in the creative acts described in chapter 1. In

42 T. Stordalen, "Genesis 2, 4: Restudying a locus classicus", ZAW 104 (1992), pp. 163-77.

253



GORDON J. WENHAM

other words the editor responsible for ii 4 was familiar both with Gen. ii-iii and with Gen. i 1-ii 3. "Gen. ii 4 is a literary unit bridging Genesis i and ii-iii".43

It may therefore be taken as proved that the toledot formulae are editorial. But who is the editor responsible? In Gen. ii 4 it is usually held that P was responsible for the first half of the verse and J for the second. When the close connection between the two parts is recognised, it seems more likely that only one writer was responsible. In themselves the formulae tell us nothing about their originator. It is

merely the assumption that P is the last source or editor that has led them being ascribed to P. There is nothing distinctively P about the word "toledot". The phrase "by their generations" does occur in x 32 and xxv 13. But while xxv 13 is generally accepted as P, x 32 may well be part of J's editorial conclusion to the table of nations.44

The prefacing of genealogies ascribed to P by the title "This is the toledot of" does not prove the title itself originates with P any more than the nearly as common prefacing of narratives ascribed to J by the same title proves it originates withJ. However the observation that elsewhere J tends to top and tail P material with his own glosses or comments e.g. xvii 1 etc. does make it more likely that these head-

ings come from his hand rather than P's. Furthermore the tightly inte-

grated chiastic structure of ii 4, whereby the toledot formula (ii 4a) is tied to the opening of the Garden of Eden story (ii 4b-iii 24), which is always ascribed to J, reinforces the likelihood that J is responsible for the toledot formula here, and by analogy elsewhere in Genesis.45

Genesis i

Genesis i 1-ii 3 is sui generis. Westermann has aptly characterised it as a "festive overture".4 Its Hebrew style is distinctive. Its elevated

prose style and its use of repetition with elegant variation set it apart from what follows and show it is a carefully composed introit to the rest of the book.47 But from ii 4 the main story line begins with ten main sections,48 each introduced by a toledot formula.

43 Ibid., p. 173. 44 Cf. footnote 12. 45 Cf. Stordalen's cryptic comment "a 'redactor' closer to 'J' than to 'P"'. Ibid.,

p. 176. 46 C. Westermann, Genesis 1-11, p. 129. 47 P. Beauchamp, Creation et separation (Paris, 1969). 48 On the double formula in xxxvi 1, 9 see discussion above.

254



THE PRIORITY OF P

The opening chapter of Genesis thus stands right outside the main narrative structure of the book. It therefore should not be unthink-

ingly assigned to one of the sources used in the main body of the book. Its elaborate structure and elegant style suggest rather that it was specially composed as a preface to the book of Genesis if not the whole Pentateuch.

But can the author of Genesis i be identified? He could of course be quite different from any of those responsible for editing the rest of the book. Scholars though have rightly been wary of multiplying enti- ties beyond necessity and have preferred to associate the composition of Genesis i with one of the sources or editorial layers of the rest of the book. Because P is the last of these, he is usually credited with the composition of Gen. i 1-ii 3. Various other subsidiary reasons are

given for attributing Genesis i to P. It speaks of God not Yahweh. It offers a different account of creation from chapter ii. It is interested in the Sabbath, a priestly concern. Its vocabulary resembles that of other P passages.

None of these arguments are conclusive: indeed some point in quite a different direction. The divine name criterion has now been aban- doned by many writers as a means of distinguishing J and E sources in the patriarchal narratives.49 It is admitted that J could well choose to speak of God as Elohim in appropriate contexts, e.g. where for-

eigners speak or are spoken to about the divine.50 Indeed J seems to be aware that Yahweh is the name of God first revealed to Moses, so whenever possible he avoids using it in the direct speech of God or human actors.51 If historical or theological reasons dictate the choice of divine epithets in the patriarchal stories, may not this be the case in Gen. i-xi as well?

In fact some such explanation is required in Gen. ii 4-iii 24, for

though it is universally ascribed to J, it never uses Yahweh by itself but only Yahweh Elohim or Elohim alone. L'Hour52 has offered a

plausible theological explanation of the variation in divine names in Gen. ii-iii, which could well be extended to Genesis i. He argues that

J has used the unusual title Yahweh Elohim to express his conviction

49 E.g. J. Van Seters, Abraham in History and Tradition (New Haven, 1975), p. 156. R.N. Whybray, The Making of the Pentateuch (Sheffield, 1987), pp. 64-72.

50 E.g. chapters xx, xxxix-xli. 51 See R.W.L. Moberly, The Old Testament of the Old Testament, pp. 70-8. 52 J. L'Hour, "Yahweh Elohim", RB 81 (1974), pp. 524-56.

255



GORDON J. WENHAM

that Yahweh is Israel's covenant partner and the God (Elohim) of all creation. At the moment when God's partnership with mankind is

being doubted in iii 1-5, Yahweh is omitted and Elohim alone is used. This suggests J could use Elohim by itself in the context of creation.

The contrasts between the accounts of creation in Genesis i and ii have also been overplayed too. Genesis i gives a broad overall picture of God creating the world in six days and resting on the seventh, so

relatively little attention is given to the creation of man. However that is where the focus is in chapter ii: God's perfect provision of all mankind's needs is the starting point of the story and provides the

backdrop to the centre of the story, the act of disobedience, as a result of which mankind lost many of the privileges granted to him at creation. Similar transitions between kindred but distinct materials are also marked by the use of the toledotformula.53 Clearly the final editor of Genesis whether he be P or J found no irreconcilable clash between Gen. i and Gen. ii, so modern exegetes should not create problems where he saw none. If we postulate different sources behind chapters i and ii, this hardly decides who incorporated them into Genesis.

However it is argued that the emphasis on the Sabbath in Genesis i 1-ii 3 must point to P. It is certainly correct to see the sabbath as central to the intention of this chapter. Here God is portrayed as work-

ing for six days and then resting on the sabbath. Though it is not

spelled out, the implication is clear that man made in God's image should imitate his creator and work for six days and rest on the seventh

(cf. Exod. xx 8-11.). But is the sabbath a peculiarly priestly notion? It is of course mentioned in the decalogue, usually held to be one of the earlier texts, but it does not figure very prominently in the cultic laws and calendars of Leviticus and Numbers.

In Genesis itself observation of the sabbath is only implicitly en-

joined in ii 1-3, and only implicitly referred to in the flood story, where both God and Noah work on a weekly cycle. The LORD an- nounces the flood will begin in seven days (vii 4) and accordingly after seven days it does (vii 10). Similarly at the end of the story, Noah sends out a raven, waits seven days and sends out a dove, then waits another seven days and despatches the dove again (viii 7-12). Now it is not clear on which day of the week God and Noah act, though if

53 See v 1; vi 9; x 1; xi 10 and the discussions of R.S. Hess, "Genesis 1-2 in its Literary Context", TynB 41 (1990), pp. 143-53 and T. Stordalen, ZAW 104 (1992), pp. 173-4.

256



THE PRIORITY OF P

the Jubilees calendar is presupposed,54 these acts take place on a Sunday the first working day after the sabbath. But whether or not this calendar is presupposed, the parallels with Genesis i surely suggest that God is observing a weekly cycle of work and rest and that Noah, the

perfectly righteous man, is doing the same (cf. vi 9; vii 1). However on the usual source analysis of this chapter all these references to the week and its implied rest day occur within the J material. If J is interested in the sabbath in the flood story, why should not the same be true in Gen. i 1-ii 3?

Finally we must consider the question of vocabulary, which is alleged to demonstrate Genesis i comes from P. As is now recognised, the old-

style vocabulary lists of P words, J words, and E words are of little value in determining authorship. The vocabulary of a passage is more a reflection of its content and genre than authorship, so the affinity of Genesis i with some P passages proves little. Nevertheless since some source critics still make crude use of vocabulary counts to determine sources analysis, it warrants further discussion. Driver55 lists the fol-

lowing terms as characteristic of P in Gen. i 1]' "kind", r"it "swarm", "swarming thing", WrZ' "creep, creeping thing", l3'1 'fB "be fruitful and multiply", fil*R "for food". Most of these terms occur in Genesis

only in chapter i and the flood story. This could be expected from the subject matter: the flood is essentially a story of de-creation and re-creation. The land emerges from the flood waters just as it emerged from beneath the primeval ocean in Gen. i 2, 9-13. Noah is a second Adam figure, who like his predecessor is bidden to be "fruitful and

multiply" This command (albeit in the singular) occurs in only one other passage outside Gen. i and viii-ix in xxxv 11: this is hardly enough to demonstrate it is source-specific especially since the roots FTZ' and flT1 are used in other pentateuchal sources too. These vocab-

ulary items thus fail to prove that Gen. i must have been composed by P. They show that chapters i and vi-ix are dealing with similar

topics, not necessarily that they are by the same hand. However if for the sake of simplicity we grant that the common

vocabulary does point to the same authorial hand, it could be either P or J. As discussed above, conventional source analyses of the flood

54 A. Jaubert, La Date de la cene (Paris, 1957), p. 33. GJ. Wenham, VT 28 (1978), pp. 343-5.

55 S.R. Driver, Introduction, pp. 123-4.

257



GORDON J. WENHAM

story suppose that Gen. vi-ix is composed of a P-base with a J preface, J glosses and aJ conclusion. However some of the J glosses within the body of the story have shorter P glosses within them e.g. vii 3a, 8-9, 23b. This is curious, so a redactor is invoked to explain these

glosses, though quite what his purpose was in adding these elements is not clear.

Ska's solution to these glosses is to deny that they are later additions: rather they belong to the original P story of the flood.56 This makes J's contribution to the flood story much less. But it may be asked whether it is right to characterise them as P glosses. This is done

simply because they parallel phraseology in Genesis i: vii 3a parallels i 27, vii 8-9 echoes i 24-28, and vii 23b harks back to i 24-25. How- ever were Genesis i classified as a J composition, this anomaly would

disappear. There would be no grounds to regard vii 3a, 8-9, 23b as P glosses on J material. vii 1-5, 7-10, 23 would be pure J.57

For the purposes of this discussion it is not necessary to decide whether Genesis i originated with P or with J. It is sufficient to note with Stordalen58 that the clearly redactional ii 4 knows both Gen. i and Gen. ii-iii. Whether the editor of Genesis created chapter i himself or drew it from some other source does not affect our main argument, that, wherever P material has been identified, it appears to antedate the J material. This seems clearest in the genealogies in chs. v, xi, xlvii, where we are dealing with originally independent material con-

ventionally attributed to P. The same may be true of chs. xvii, xxiii, xxxiv, vi-ix, though here the evidence is not so clear cut. What I have tried to establish is that the P-material throughout Genesis is not a late insertion into an essentially J-composition, rather that it is one of the sources used by J to form our book of Genesis. It is widely admitted thatJ did not create his material from scratch, but that he drew on earlier oral or written sources including genealogies cf. iv 17-26; xxii 20-24, and narratives e.g. ii-iii; xviii-xix; xxii to produce the J document. It is the contention of this paper that the so-called P material in Genesis was another source which J drew on to create his work.

56 See above footnote 32. 57

Unitary readings of Gen. vi-ix and Gen. xxxiv still seem to me preferable to the source analyses proposed. My argument here is that there is nothing in Gen. i or vi-ix to contradict the obvious priority of the P passages elsewhere in Genesis.

58 See above footnotes 42, 45.

258



Documents

Z Wenham, The Priority of P.pdf