Last year I tried to give sure-fire evidence from a single chapter in Exodus supporting the claim that the first 5 books of the Hebrew Bible were arranged from multiple sources. One negative response to my three-part post came from Julie M. Smith over at Times and Seasons.* She said, essentially, that if this was the extent of the evidence for the Documentary Hypothesis, she’d lost her faith in what she’d been taught during her Master’s Degree in New Testament studies. In the discussion of her post that ensued, the disagreement centered on the existence of possible alternative explanations for the material I was discussing. I argued that we have clear evidence that Deuteronomy knew only one version of Moses’ re-ascent of Sinai (Horeb) preserved, in integrated fashion, in Exodus 34. The only logical explanation for this is that the author of Deuteronomy (not Moses) had before him two distinct sources (or one of the two), which he incorporated into his own source. On the other side, people including Julie argued that this is not the only (nor, I assume they meant to say, the best) explanation of the phenomenon. After all, because our literary sensitivities differ from those of the ancient world, we don’t have license to go carving up the text, especially not one that, tradition tells us, was penned by Moses himself. This is a common (if vernacular) type of critique of the DH, the misconceptions of which need to be dispelled.
The Documentary Hypothesis (DH hereafter), it must be remembered, is a theory that arose because people couldn’t make sense of the text as it stood in front of them, and were no longer willing to find in these seams, repetitions, and non sequiturs some cryptic lesson that God wanted the reader to learn, as the Second Temple and Rabbinic interpreters did (see Kugel, The Bible as It Was). People often lose sight of these origins (what is now called “source” criticism was originally called “literary” criticism because of it) and attack the theory on the premises that a) one can’t really know how these things have come to be, b) people have found multiple sources in places like Melville, c) authors can build in literary seams, d) the argumentation is circular, etc. All of these might be true (I’m tempted here to use Mike Myers’ rejoinder from Wayne’s World), but do they make sense for this particular text? And, more important, do these explain the weighty mass of problems in these “books”?
We often hear that the reliance upon doublets (multiple stories narrating the same event, see below) or upon the varying names of God has been shown to be problematic. Partially true. But what hasn’t been acknowledged by many (especially lay) critics is that scholars have long since moved on from relying just upon the names or on the doublets. It’s easy to come up with an explanation as to why humans are created twice in Genesis 1-3, (the Rabbis certainly did), but when that is only one very small part of an ocean of evidence, it’s much harder to discount. And it’s that ocean that most critics of the DH fail to reckon with. In the words of Richard Friedman, “it must come down to evidence.” So while last year’s post was a careful source-critical reading of one (highly relevant) example, today’s is an effort to take a panoramic view. If the critiques are correct, and the DH is so totally wrong, here’s an incomplete list of what you have to explain better than the DH. (Feel free to add more in the comments. I don’t have a complete list of doublets here, because, well, I think we all get the picture that lots of events are repeated.)
- Why humans are created twice (Gen. 1:27, 2:7ff.).
- Why Noah is commanded to take two of every animal (Gen 6:17) and then seven of every clean and two of every unclean (Gen 7:2).
- Why the seams between the two creation accounts and the two flood accounts also divide perfectly the similar language that is used in the creation and flood accounts.
- Why in one text (Num 12:4-15) the tent of meeting is outside the camp, and anyone can go to it, and in an earlier one (Num 2), it is in the center of the camp, and only the priests can enter it. And why these two texts preserve similar language and textual allusions to other similarly divided texts.
- Why in the flood account it is proclaimed that God limits man’s days to 120 years, but in Genesis 47:9 Jacob says he lived 130 years and Abraham and Isaac lived even longer.
- Why Hagar is banished twice (Gen 16 & 21)
- Why Jacob is named twice (Gen 32 & 35)
- Why Beersheva is given two different etymologies (Gen 21 & 26).
- Why Abraham is born in Ur Kasdim (Gen 11:27-32), i.e., in southern Iraq, then moves to Haran, i.e. North Syria, and while in Haran (apparently) God tells him to leave his birthplace to go to Canaan (Gen 12:1ff.), but of course he’d already left his birthplace—in Ur!
- Why Jacob finished a long discourse about why Ephraim and Manasseh will be adopted as Jacob’s sons, and then in the next verse doesn’t know who they are (Gen 48:3-7, 8), and why if you remove that discourse, the narrative makes perfect sense. And why that discourse matches with language of other similar discourses that are divided by similar literary seams.
- Why in one text (Gen 35:23-26, (=P)) Benjamin is born in Paddan Aram, but in another (35:16-19 (=E)) he is born near Bethlehem, in Canaan.
- Why in one text (Gen 37:36) the Midianites sell Joseph to Potiphar, but in another (Gen 39:1) it was the Ishmaelites.
- Why, in Genesis 37:21-22 it’s Reuben who wants to save Joseph, and in 26-27 it’s Judah.
- Why Moses’ father-in-law has two names (Reuel/Jethro), and Sinai has two names (Sinai/Horeb), and why the use of each of these names corresponds to other linguistic, thematic, and narrative distinctions.
- Why Abraham tries to pass his wife off as his sister in two similar instances (albeit with important distinctions that would indicate different sources) and then Isaac does the same thing. (Genesis 12, 20, 26). Incidentally, even our own Bible Dictionary admits (s.v. “Abimelech”) that these were probably originally a single story that got passed down in different streams of tradition (or what would later become “sources”).
- Why the name of God is revealed as if for the first time in Genesis 3, then again in Genesis 6, and why in these distinct texts there are linguistic similarities that track back to other texts distinguished by linguistic and literary divisions, and why in those divisions the one that hasn’t used the tetragrammaton up to this point (P, for those of you who are with me) now begins to use it, but in the others it had long been used.
- Why God’s hardening of Pharaoh’s heart so that he chases the Israelites (Exod 14:8-9) follows Pharaoh’s decision to chase the Israelites (14:5-7).
- An angel moves to the rear of the fleeing Israelite camp (Exod 14:19a) but in the same verse it was not the angel but the cloud pillar (Exod 14:19b).
- Why it seems that God blew the sea back, drying the ground, in one text, and in the next split the waters through which the Israelites walk. (Exod 14:21a, compare 14:21b-22). And if this seems scant, why in Exodus 15, which recounts the mighty acts of the deliverance, there is no splitting of the sea.
- Why linguistic formulae that are never supposed to be interrupted (the command-fulfillment pattern in the Pentateuch) are interrupted around source divisions or insertions of other material.
- Why Exodus 34 can be separated into two narratives, one about God appearing to Moses and making a covenant, and the other about Moses going up the mountain to receive the second set of tablets with the same writing on them. And why these two episodes preserve the language and terminology of J and E, respectively, and why when these two episodes are separated their narratives match up perfectly with where the other separated J and E narratives left off and where they continue. And why Deuteronomy 10 quotes only one of these tightly integrated stories.
- Why God is immanent in some texts and transcendent in others (compare the Tent of meeting, where he is transcendent, to the Tabernacle, where he is immanent).
- Why in some texts the ark is just a plain wooden box with two tablets in it, and in others it’s an ornately-decorated chest containing several relics.
- Why we have the Decalogue and Covenant Code, (=the set of laws beginning in Exodus 20), and the Deuteronomic Code (Deut 5; 12-26, which is supposed to be a repeat of the Covenant Code, but it’s not; it introduces many, many changes. Compare, for example, the Sabbath laws in Exod 20 and Deut 5).
- Why in Exodus 12 we have pesach (passover) celebrated followed immediately by the seven-day matzah festival, but in Deuteronomy 16:1-8 there is only a seven-day festival (pesach and matzah have been conflated), with no specific date assigned in Deuteronomy. Also, why Deuteronomy requires the paschal lamb to be slaughtered at the central shrine but in Exodus it’s done in a family setting.
- Why in some texts Moses’ staff is used to perform miracles, and in others it’s Aaron’s. (And why in these different cases there is other linguistic and narrative evidence pointing toward source divisions.)
- Why the differentiated texts exhibit particularized geographical foci. Why, for example, do the spies in Num 17-20, 22-24 only see sites in the kingdom of Judah? (These happen to correlate with other J texts, whose focus is clearly on Judah). Abraham, in J, lives in Hebron, a capital in Judah. There are only four birth stories in J: Reuben, Simeon, Levi, and Judah, and in J the first three are disinherited, leaving Judah to rule. These trends can be extended to the other sources.
- Why all of these discrepancies are differentiated along linguistic, terminological, and substantial lines. E.g., the doubled stories will show different names for god, treatments of the priesthood, use of individual words (where such words are excluded from the other sources), sacred objects, etc. And, here’s a kicker, why the separation of these differentiated texts and rejoining to texts seemingly hewn from the same quarry results most often in a continuous narrative. Were one to spell out these individual differentiations, they’d number in the thousands. A replacement theory would have to explain this.
I should add, too, that we know that other narratives were composed from sources, such as the Deuteronomistic History (Joshua, Judges, 1-2 Samuel, 1-2 Kings), which was composed of documents likely from the Temple library. As scholars have long noted, it seems that the principal reason for a rejection of this theory isn’t at all due to disagreement over literary approaches, but rather to theological discomfort.
I think one should conclude with Friedman here: “Above all, the strongest evidence establishing the Documentary Hypothesis is that several different lines of evidence converge.”  He cites as evidence for the theory: the convergence of the thirty-one doublets (repeated stories, which, he says contra others, is never attested elsewhere in ancient literature as a rhetorical device), the contradictions in the text, the different styles and terminology, and the fact that when separated, these form distinct, coherent narratives. No one of these constitutes the bedrock of the documentary hypothesis, without which the theory would crumble. It is the fact that they correlate in toto that makes the hypothesis nearly impossible to refute.
1. The Bible with Sources Revealed, 28. Emphasis his.