According to the critical consensus, 2 Samuel 8:18, 20:26 and 1 Kings 4:5 above show no acquaintance with ?P?s?Enotion that priesthood was restricted to members of the tribe of Levi; from this evidence, inter alia, the conclusion is drawn that P must not then have been in existence, for if it were, the authors of Samuel and Kings would certainly have condemned the presence of nonLevitical priests. In my interpretation of these texts in The Priesthood of the Plebs , I have assumed, on the contrary, that the restriction of priestly ministry to the tribe of Levi predated the early monarchy, a stance that, though raising problems about the texts in question, is necessary to open up a more accurate interpretation of these texts, and a fuller understanding of Israelite priesthood generally. Clearly, I cannot here examine the classic JEDP thesis fully here, but only point to several of the reasons I find it utterly unpersuasive. The notion that a distinct priestly source can be isolated at all is, I think, highly dubious (cf. Cassuto 1941), but for the sake of the following exercise I will assume that a distinct ?priestly document?Eor ?layer?Ecan be abstracted from the Pentateuch, so that what is at issue in the following argument is the date of P, the priestly literature that distinguishes most sharply between layman and Levite. A number of rather technical recent studies have contested a postexilic dating for P (summarized in Zevit 1982: 481-511), but I have no competence for evaluating questions such as the development Hebrew style and grammar or archeological evidence; my arguments do not rely on specialized knowledge. Besides, a late pre-exilic dating for P, such as Zevit and others argue for, does not in the least help me here. In order to present my reasons for rejecting the critical post-exilic date for P, I will follow the arguments of S. R. Driver?s Introduction to the Literature of the Old Testament , whose balanced assessment of the evidence is less strained by prejudice, and therefore more worthy of detailed attention, than Wellhausen?s often tendentious, albeit eloquent, tract. I will, however, address Wellhausen specifically at several points, and wish throughout to demonstrate Driver?s essential agreement with him.
According to Joseph Blenkinsopp, who has recently defended the postexilic dating of P (Blenkinsopp 1996), ?One of the main arguments for Wellhausen?s hypothesis, repeated many times in the Prolegomena, is that no pre-exilic writing betrays the slightest acquaintance with P?E(Blenkinsopp 1977: 58). Thus, by beginning with incontrovertibly early literature, and showing its lack of acquaintance with P, and by gradually moving through later literature, again showing that this literature shows no knowledge of P, P can be removed to a later and later period. If a document can be assigned an absolute date, and if it can be shown that this document does not assume the structures and institutions of P, then it must be that P is later than this document.
Driver endorses something like this procedure, beginning his discussion of the date of P by pointing out that in Judges and Samuel, sacrifice is offered by anyone, anywhere, ?without any hint of disapproval on the part of the narrator, and without any apparent sense, even on the part of men like Samuel and David, that an irregularity was being committed?E(Driver 1913a: 136). Deuteronomy likewise does not presuppose P, but is liturgically much closer to the early Pentateuchal documents J and E (Driver 1913a: 137), and Deuteronomy can be dated with some certainty to the Josianic reforms and centralization of the sanctuary in the seventh century (Driver 1913a; Wellhausen 1885). Hence, P must be later than the seventh century. Finally, Ezekiel 44:6-16, a passage that played a key role in Wellhausen?s reconstruction as well, institutes, as for the first time, a distinction between (Zadokite) priests and Levites. This distinction, however, is pervasively assumed by P; if Ezekiel instituted it for the first time, it must be that Ezekiel predates P, and since Ezekiel was living in the early exilic period, P must be either exilic or later (Driver 1913a: 139).
Insofar as this is an argument from the silence of the historical books, it is not only flimsy but also open to a variety of alternative explanations. Among Wellhausen?s arguments for the late date of P?s regulations concerning the priesthood, for example, was the sharp contrast between the elaborate priesthood described in Leviticus and Numbers and the apparent freedom in worship and sacrifice found in narratives describing Israel?s early life in the land; the suddenness of the institution of the cult, he said with characteristic wit, is matched by the suddenness of the cult?s disappearance after the conquest, so that ?the Book of Judges forthwith enters upon a secular history completely devoid of all churchly character?E(Wellhausen 1885: 127). The absence of the priesthood and cult from most of Judges is quite striking, since the Hexateuch leaves the impression that the high priest was the central figure in Israel at the time. It is, of course, exceedingly tenuous to draw large historical conclusions on the basis of a single book. More importantly, Wellhausen?s analysis ignores the fact that Judges is not written as mere narrative history but as a theological account of Israel?s failures; it is the work of a ?former prophet,?Enot a ?scientific nineteenth-century historian.?E Once this is recognized, the question arises of whether the absence of reference to priests and priestly institutions plays some theological role in the narrative. The plausibility of this line of thinking is increased by the fact that Levites do, in fact, appear in both of the long closing narratives of the book, and in neither case are they presented in a good light. By placing these narratives out of chronological order at the climactic point of the narrative, the author is highlighting the fact that the political failures of the early theocracy are in large measure the product of the failure of the Levites to guard Israel from idolatry (Jordan 1989). The narrative absence of the priests can be constructed as a thematic device to underscore their practical absence from the life of Israel. The book raises precisely Wellhausen?s question, ?Where are all the priests??E?Ewithout at all endorsing Wellhausen?s answer.
Driver?s argument is not, however, simply an argument from silence. He recognizes that the failure of many documents to allude to the institutions of P is evidence of ?less importance: the writers of these books may have found no occasion to mention them?E(Driver 1913a: 137). The argument is stronger when the texts record, without criticism, activities that are condemned by P, such as the permission seemingly granted to all lay Israelites to offer sacrifice. Here also, alternative explanations of the data are possible. A specific example will help. Judges 17-18 tells the story of Micah, who establishes a sanctuary for his ephod, and a wandering Levite named Jonathan, who is hired to manage the sanctuary. Most scholars recognize that this provides evidence, from sources deemed early, of a preference for Levitical priests. But more has been made of the story. From this incident, Kraus, among many others, concludes that during this period ?Anyone who managed to erect a sanctuary for the family or clan appointed a qualified man for the cult?E(Kraus 1966: 93), but to construct theories about what ?anyone?Emight do from a single narrative is building on sand. To be sure, the writer takes Micah and Jonathan as illustrative of a general tendency in Israel during the period of the judges, a tendency for every man to do what was right in his own eyes. Clearly, however, the author is critical of both Micah and Jonathan the Levite, so that it is simply not true that Judges passes over cultic abuses without comment.
Detailed exegetical debates aside, Driver?s thesis betrays serious internal inconsistencies. For critical scholars, the dating of P, it must be emphasized, depends entirely on comparison among the various texts of the Old Testament. Once the claim of the Pentateuch to Mosaic authorship is dismissed as unhistorical, one is left with no internal evidence for dating, and few fixed dates in relation to which other texts may be dated. Merely noting differences between P and Deuteronomy does not provide any information about relative dates. If P is to be assigned a later date than D, it must be shown that P presupposes what D requires, and that the opposite is not the case. If D presupposes P, then the dating would have to be reversed.It is a remarkable testimony both to Driver?s care as a scholar and to his blind commitment to a postexilic date for P, that he provides abundant evidence for Deuteronomy?s knowledge of P?s institutions and regulations. Thus, Deuteronomy speaks of clean and unclean animals, requires sacrifices to be without blemish, uses the language of ?fire sacrifice,?Ealludes to burnt offerings and peace offerings, tithes, heave offerings, vows, and freewill offerings, the sanctity of firstlings and firstfruits, prohibits eating blood and the flesh of animals who died of themselves (Driver 1913: 144), all of which are prescribed (not presupposed) in P. Driver concludes that
It is thus apparent that at least one collection of priestly T?rth, which now forms part of P, was in existence when Dt. was written; and a presumption at once arises that other parts were in existence also. Now, the tenor of Dt. as a whole conflicts with the supposition that all the institutions of the Priests?ECode were in force when D wrote; but the list of passages just quoted shows that some were, and that the terminology used in connexion with them was known to D (Driver 1913a: 145).
Driver draws the same conclusion from the references to P in the historical books (Driver 1913a: 142-144). The institutions described in P, and even some written regulations that were later incorporated into P, predated Deuteronomy, and traces are found in the historical books. P is thus not a pure invention of exilic priests but was ?based upon pre-existing Temple usage?E(Driver 1913a: 143). His claim about the post-exilic dating of P, then, is a claim about the full document, as we presently possess it, not about any of its specific contents.
But the claim, so restricted, simply collapses, dying not the death of a thousand qualifications but the more rapid death of basic fallacy. For starters, Driver?s theory could accommodate evidence that proved that all the institutions of P existed during the first temple period, without giving up the claim that P as a document is postexilic. But then the argument about P is merely an argument about when the regulations were written down, and is comparatively trivial; one could not reconstruct the history of Israel on the basis of this kind of claim, which is, however, what Driver, following Wellhausen, dearly wishes to do. Further, once it is admitted that the institutions and rites described in P are based on those of the first temple period, something that even Wellhausen was willing to concede, then the supposed absence of allusion to these institutions and rites in acknowledged first temple texts, far from being proof for the lateness of P, is rendered problematic. If Deuteronomy is of Josianic provenance, and if at least some of the institutions of P predate Josiah, there is no historical reason why Deuteronomy would not allude to those institutions of P but, on the contrary, every reason to expect Deuteronomy to contain a great deal more priestly material. Given the assumption that P did not exist as a written document, the reform of Josiah would have been a natural moment for the Jerusalem priests to put their practices into some kind of systematic written form.
Driver discovers, however, that references to priestly institutions are not absent from Deuteronomy after all, but explains these references as evidence of a knowledge of institutions described in P but not evidence that P existed as a document when Deuteronomy was written. If Driver can acknowledge that Deuteronomy contains numerous allusions to priestly institutions, and yet maintain a late date for P, the question of whether Deuteronomy displays a knowledge of priestly material is irrelevant to the dating of P (unless ?allusions to P?Emeans specific reference to a priestly document, but that, to reiterate, would be trivial, not to mention quite impossible to prove). But the relationship of P to Deuteronomy is one of the key arguments for the late date of P, so, having conceded that the evidence is not clear-cut, on what basis does Driver cling to this conclusion?
The ruse that priestly material in Deuteronomy and the historical literature is a ?later interpolation?Eis even more apparently an argument of despair, one that Driver largely avoids, though not entirely (cf. Driver 1913a: 143, on Judges 20-21). Again, the absence of allusions to priestly material forms one of the main arguments for the conclusion that P did not exist, and that the rites, officers, and institutions of P are late. To suggest that, when allusions do appear in earlier literature, one can still assign them to a late period, makes the whole enterprise suspect, and renders the theory of a late dating of P practically unfalsifiable. Without endorsing a Popperian falsification criterion, one must question whether a theory so completely resistant to counter-evidence explains anything at all.
In the end, Driver is left with little more than appeals to differences in ?tenor?Eand ?impression.?E It is thus not only the detailed presentation of worship in Judges and Samuel that sets these texts off from P, but rather ?the different tone of feeling, and the different spirit which animates the narratives of the historical books,?Eso that the actors of these books ?move in an atmosphere into which the spirit of P has not penetrated?E(Driver 1913a: 137; cf. 138 [on D]; 139 [on Ezekiel]). As with Wellhausen, the overall impression created by the different documents is finally the decisive consideration. ?Overall impression?Edepends, of course, on who is being impressed. Though Driver is less explicit about it than Wellhausen, it is clear that for both the impression created by P is an entirely negative one. P?s anthropomorphisms describe a ?transcendent?EGod (apparently, a bad thing), and his historical narratives show no ?color?Ebut are rather theoretical and abstract (Driver 1913a: 140-141). Again, as with Wellhausen, this evaluation of priestly religion is allied to an implicit assumption about the development of religious institutions. Describing the divergences between Ezekiel and P, Driver suggests that Ezekiel?s liturgical prescriptions are ?simpler?Eand therefore the idea that P develops from Ezekiel, as Ezekiel developed from D, ?naturally suggests itself?E(Driver 1913a: 140). Even granting the facts are as Driver indicates (which, in the light of Ezekiel?s elaborate description of the temple, is dubious), there is nothing more ?natural?Eabout a development from simplicity to complexity than about an opposite development; they are simply different kinds of developments. One might as well argue that a medieval high mass is a ?natural?Edevelopment from, and therefore later than, the worship of the churches of Zwinglian Zurich. Related assumptions underlie Driver?s claim that ?Ezk. plainly attached a value to ceremonial observances, and is thus the less likely to have introduced a simplification of established ritual?E(Driver 1913a: 140), as if the links of ?complexity=high value=late date?Eformed an unbreakable chain.
Thus, the JEDP theory as argued by Driver must be rejected as illogical and internally contradictory.