Traditionally, the authorship of the Pentateuch has been ascribed to Moses. This is based on self-attribution, intertextual attribution (e.g., Joshua, the Historical Books, the Prophets), and NT attribution.
Genesis is anonymous. However, Genesis is the lead-in to the Torah, and thematic parallels between Genesis and Exodus, to take but one example, evidence their common authorship.
But on the basis of source criticism, liberals deny the Mosaic authorship of the Pentateuch. They offer the following criteria and criticisms:
1. Nominal variants. They assign a given pericope (literary unit) to a given source based on which divine name is used (Yahweh, Elohim).
2. Narrative variants. They treat similar stories ("doublets") as different versions of the same story, based on different traditions.
3. Stylistic variants. They assign a given pericope to a given source based on stylistic differences.
4. Subject-matter. They assign a given pericope to a given source based on different subject-matter (Priestly source, Deuteronomist).
5. Internal contradictions. They draw attention to alleged discrepancies (e.g., two creation accounts, number of animals in the ark) as evidence of divergent traditions.
6. External parallels. They draw attention to apparent or actual external parallels (e.g., the Enuma Elish, Epic of Gilgamesh, Code of Hammurabi), and then attribute the canonical material to the extracanonical material.
7. Num 12:13. They invoke the immodest claim in Num 12:3 to deny Mosaic authorship of the Torah.
8. Exod 6:3. The fact that the Pentateuch is written in the third-person is sometimes cited as evidence of its amosaicity.
More importantly, Exod 6:3 is a prooftext for the Documentary Hypothesis. Since "Yahweh" is on frequent display in the patriarchal narratives, Exod 6:3 has been taken as exposing the fact that P was ignorant of the J tradition.
9. Deut 34. Since the obituary in Deut 34, as well as other alleged anachronisms, is obviously an a-Mosaic postscript, this has been invoked to justify the wholesale redaction and composite authorship of the Torah.
10. Textual criticism. Recent work on the text of the OT, Samaritan Pentateuch, and Epic of Gilgamesh has been taken to justify the application of redaction criticism to the OT, inclusive of the Pentateuch.
By way of reply:
As a general observation, the reason that Documentarians don't believe the Bible is what it claims to be is because they don't believe it could be what it claims to be; and the reason they don't believe this to be the case is because they subscribe to a secular outlook which, if true, would render many events of Scripture utterly impossible. And if the Bible cannot be a divinely inspired record of divine deeds in history, then its authors must have borrowed their ideas from the prevalent mythology of their pagan neighbors. If brief, if the Bible cannot have a supernatural source of inspiration, then, by process of elimination, it must have a naturalistic source of inspiration.
Rarely do they deign to offer up anything resembling an argument for their tremendous assumption. At most we have the barefaced assertion that secularism is unquestionably true. Hence, the Bible is unquestionable false. Specific objections to Mosaic authorship are secondary to this a priori postulate.
So the job of a Bible critic is to play the role of scholarly sleuth who "uncovers" or "discovers" the "true" sources of Scripture. And unlike your average private eye or police detective, who must make a case on the basis of empirical evidence, the Bible critic operates like a medium, graced with the psychic ability to conjure up spectral redactors that are wholly invisible to those of us sadly deficient in their clairvoyant powers of recollection. It's the best show in town since Madame Blavatsky bit the dust.
Moving onto the particulars:
1. Nominal variants:
i) Hebrew is not Esperanto. Moses did not invent an artificial language when he wrote the Pentateuch. He made use of a preexisting language, the language of his people. Hebrew is a Canaanite dialect.
Just as Greek (e.g., NT writers), Latin, Arabic, and English-speaking Christians coop pre-Christian designations to denominate the Christian deity, Moses took over pagan names to designate the true God.
ii) In the OT, God goes by a number of basal names and titles--oftentimes to lay emphasis on a particular role in relation to the world and its inhabitants. "Elohim" is his generic name, whereas "Yahweh" is his covenantal name. These, in turn, each admit a variety of compound names: El Shaddai, El Elyon, El Olam, El Gibor, Yahweh Sabaoth, Yahweh Jireh, Yahweh Shalom, Yahweh Tsidkenu, Yahweh Nissi, &c.
So different forms are used to tag different relations between God and his subjects.
2. Narrative variants.
i) We need to distinguish between similar events and similar accounts of the same event. Two events may be similar, but different. Unless you deny that history repeats itself, there is no reason to deny the historicity of the "doublets."
People, both as individuals and as social entities, are creatures of habit. They tend to act and react in the same way under the same general circumstances.
In order to explain the doublets, one need only keep two things in mind. On the one hand, the exercise of historical writing is necessarily selective; on the other hand, similar things happen--happen fairly often. Life is repetitious. Behavior is stereotypical.
Now, there are differences as well as similarities between like events. If the doublets present an artificially similar aspect, that is only because the author has chosen to select for the similarities, while having less to say about the points of dissimilarity. And that's because he thought that what they have in common is more important than the ways in which they are unalike.
ii) We need to remember that the Bible was written for the ear, not the eye. What is written for the ear contains a certain amount of built-in redundancy (e.g., Gen 24:12-27,34-38) because the eye can go back over the same page whereas the ear cannot go back over the same speech.
Form criticism and tradition criticism are concerned with the orality of Scripture—the oral traditions and process of transmission that supposedly supplied the raw materials for Scripture, while source criticism and literary criticism are concerned with the textuality of Scripture—deconstructing the editorial process by which independent sources were supposedly spliced together or the way a preexisting text underwent creative adaptation and expansion, by uncovering the internal sutures stitching up distinct sources and strata and underlying the surface structure of the document. What both of these lines of criticism fail to take into account is the aural character of Scripture. Scripture was composed for the ear, not the eye. It either consists of speech-acts that were later written down or print media that were composed to be read aloud.
The problem this poses for form criticism and literary analysis is that the modern reader is a silent reader. And this raises a number of programmatic questions I have never seen systematically investigated, must less answered, in the critical literature: is there a difference in the way we process speech and print? Is there a difference in the way a writer or speaker patterns his material for an auditor rather than a reader? Is short-term memory as retentive for speech as it is for print? Has form/tradition criticism tried to isolate the evidence for orality as distinguished from aurality? Does the way we mentally clump material differ depending on its auditory or else visual mode of input? Does the way we register imagery differ if we hear it rather than see it described on the page? What contemporary genres are most analogous to the conditioning of an ancient audience? Homiletic literature? Drama? How does the absence of their original musical setting affect our sensitivity to the internal patterning of the Psalms? What feel does a modern scholar have for the phrasing of a text whose period pronunciation and cadence are irretrievable? Isn’t the analysis of literary divisions dependent on the identification of rhythmic units? What experiments can be conducted to test possible answers?
These are just a handful of questions that come to mind. Any form of literary or source criticism that claims to arise organically out of the life-situation of the author and audience should begin by offering us a detailed psycholinguistic model of aurality and its impact on the compositional techniques of the sacred authors. Absent that, there is no reason why we should give these critics the time of day since they’ve failed to lay the preliminary groundwork that would provide even the most general preconditions the must be met before we ever get to their more specific claims.
iii) Scripture is fond of recapitulation. Bible writers like to draw our attention to recurrent patterns of divine and human conduct--especially the theme of (human) apostasy and (divine) restoration.
iv) "Doublets" actually imply common authorship rather than composite authorship, for it takes a common mind to select for these inner symmetries out of the larger amount of raw material at the disposal of the historian.
3. Stylistic variants. Style is responsive to subject-matter. The same writer varies his style according to the demands of the respective genre (see below).
4. Subject-matter. This criterion is a disguised description masquerading as an explanation. To assign priestly material to a priestly source or legal material to a Deuteronomic source is viciously circular. For there is no prior reason to assume that different subject-matter demands a special explanation or separate source. If Moses wrote the Pentateuch, then he would have ample occasion to write about the cultus and the criminal code, the history of Abraham and the Exodus, &c.
5. Internal contradictions. This is an exegetical issue rather than a source-critical issue. What is the relation of Gen 2 to Gen 1? What is the relation of clean to unclean animals? Regarding the former question, the viewpoint of Gen 1 is cosmological whereas the viewpoint of Gen 2 is anthropological. But these are complementary perspectives, and necessarily so.
Regarding the latter question, the number of unclear animals was limited to a bare reproductive unit, but because the clean animals were both sacrificial animals and foodstuff, one pair apiece would not suffice. Far from indicting the account with error, the practical logic serves to confirm its faultless veracity.
6. External parallels.
i) We need to distinguish between specious and genuine parallels. The fact that two documents may share some generic or incidental points of coincidence does not prove literary dependence. For example, poets in every time and place have recourse to the same stock of natural metaphors.
Again, law codes the world over deal with the same basic issues because human need and nature are the same the world over, viz., life, livelihood, sex, family, property, perjury, personal injury, contract law.
In brief, two documents may share something in common precisely because their commonalties are so very commonplace. It's as simple as that.
ii) We need to distinguish between inspiration and revelation. Whatever is revealed is inspired, but whatever is inspired is not necessarily revealed--in the sense of direct revelation. Inspiration controls the process, and thereby ensures the veracity of the final product, but inspiration does not confine itself to a particular mode of noetic acquisition. There are some things which can only be known by revelation, which is a subdivision of inspiration, and inspiration is, among other things, a quality-control mechanism to verify natural observation or tradition.
The traditional attribution of Mosaic authorship does not rule out possibility or probability that Moses may have had access to various sources of information for some of what he committed to writing (e.g., the genealogies? The Table of Nations?).
iii) Yet another lapse is when the source critics fail to apply their own criteria to the external sources. The late Cyrus Gordon supplies a couple of key examples: "One of the presumed characteristics of P (the Priestly Code, supposedly from the time of the Second Temple, ca. 5C BC), was a preoccupation with details such as the measurements of Noah's ark. However, while at Dropsie I reread the description of Utnapishtim's ark in the Gilgamesh Epic and observed a similar concern with detailed specifications. If this feature obliged us to attribute the Genesis account to P of the 5C, it must, I reasoned, do the same for the Babylonian account, which is absurd. I also found other absurdities in the so-called higher criticism of the Establishment. If Yahweh-Elohim owed its origin to the combination of God's name in J…with his name in E, then every Egyptian inscription mentioning the god Amon-Re must have derived the name from an A-document combined with an R-document. One might also argue the same for Ugaritic documents, which abound with divine names composed of two elements," A Scholar's Odyssey (SBL 2000), 80.
iv) Since Scripture situates the descendents of Noah in Mesopotamia, it is hardly inconsistent with the historicity of the Biblical account that Mesopotamian civilization would preserve an independent account of the flood. Indeed, it would be rather surprising if Abraham were unacquainted with some version thereof.
v) We also need to distinguish between tradition and mythology. There is nothing inherently unacceptable in the idea that Moses may have made use of Jewish tradition or family lore that was handed down from generation to generation--going back to the patriarchs or even the prediluvians. Such preexisting sources could well be historical in substance, while inspiration would correct for whatever errors crept in by process of transmission.
That is quite different from the idea that Moses made use of expurgated pagan mythology. Even if such materials could be redacted in a more orthodox direction, they would be devoid of historical value.
In addition, this is not consistent with ANE practice. As John Currid remarks, "We must question, however, whether the position that the Bible demythologizes Mesopotamian legends takes into account all the critical data bearing on the issue. First of all, the common assumption that the Hebrew stories are simplified and purified accounts of Mesopotamian legends is fallacious, for in ancient Near Eastern literature simple accounts give rise to elaborate accounts, and not vice versa. One can view this evolution from simple to complex in the many recensions of the Sumerian Babylonian flood legends. Ironically, many scholars who accept the model of the complex to the simple would then argue that the Pentateuch is the product of an evolutionary development from the simple to the complex. One cannot have it both ways.
Second, there are no examples from the ancient Near East in which myth later develops into history. Epic simply never transfigures into historical narrative. And, clearly, the creation and flood accounts in Genesis are presented as direct history with no evidence of myth," Ancient Egypt and the Old Testament (Baker 2001), 29.
vi) Besides the Epic of Gilgamesh, the most common parallel alleged to exist between Genesis and pagan mythology is the Enuma Elish. The standard treatment is by Alexander Heidel, The Babylonian Genesis (Chicago 1963). However, the comparison begins to melt away like hot wax under a spotlight:
a) The derivation of tehom (Gen 1:2) from Tiamat has been challenged by linguistically qualified scholars (e.g. Currid, Kitchen, Wiseman). Even Heidel denies the etymology.
b) And even if the etymology were sound, this wouldn’t prove a thing. Our days of the week are named after Roman and Nordic gods, but that doesn’t prove that Ash Wednesday, Maundy-Thursday, Good Friday, and Holy Saturday are pagan holidays--not for us. As I've said before, God didn’t invent a brand-new language for Moses to write in. Hebrew has its share of idioms and dead metaphors that reflect the cultural milieu and prehistory of the language.
c) As to the supposed parallels in narrative order, Heidel has laid these out side by side in a nifty little chart, ibid. 129. This abstract is quite impressive. Where disillusionment sets in is when the reader tries to retrace the process by which Heidel lifted these points of comparison. When you actually go back through the version of the Enuma Elish supplied by Heidel and read these references in context, the comparison falls apart. To begin with, there’s a difference between a ramshackle train of events (Enuma Elish) and a carefully orchestrated order of events (Gen 1). Narrative sequence is only significant if the narrator was mindful of the sequence and arranged the action in order to bring certain events to the fore.
d) In the Enuma Elish, the items highlighted by Heidel are just background details. They don’t frame the action. What Heidel has done is to take the framework of Gen 1, and then superimpose that back onto the Enuma Elish, seizing on some stray detail here or there that is vaguely analogous to the Biblical narrative. There is nothing in the Enuma Elish that brings these elements into high relief. From what I can tell, the Enuma Elish has no thematic plan.
It ought to be unnecessary to point out that there are stock storylines in life and literature. This follows from the fact that life is highly cyclical in character (e.g. day/night, eating/sleeping, the seasons of nature; the seasons of life, &c.). The mere fact that one plot may bear a rough resemblance to another in no way implies literary dependence, a common source, or a collective unconsciousness.
e) To take the most striking of the alleged parallels, Heidel draws attention to the existence of light before luminaries, ibid., 82,101,135. But when we look up the references, what we find is a passing reference to day and night (1:36), an ambiguous allusion to Apsu’s halo (1:68), and a depiction of Marduk as a solar deity (1:102). These scattered asides play no structural role. They are not presented as creative fiats. They don’t follow a logical sequence. They could be omitted without being missed. Heidel has simply contrived a specious parallel by means of free association.
f) Again, Heidel claims that Gen 1 and the Enuma Elish both refer to a watery chaos, ibid. 82,97,115. But Gen 1 doesn’t characterize the primeval deep as a chaotic state. Heidel is importing a connotation into the Biblical text that is absent in the original.
g) To take a final example, Heidel claims that both accounts terminate on the note of divine rest—with Gen 1 capped by a benediction, while the Enuma Elish capped by a celebration, ibid. 127-128,129. But Tablet 5 does not say that Marduk "rested" from his labors. The activity simply peters out. After that, the gods throw a big party. Of course, anything resembling a creation story will have some sort of beginning and ending to the creative activity. So that’s at most an incidental and inevitable parallel. But a bare cessation of creative activity isn’t the same things as "rest." Moses chooses that word because it anticipates the institution of the Sabbath. God is the lord of the Sabbath both by precept and practice. The real parallel is not between Genesis and the Enuma Elish, but between Gen 2:2 and Exod 20:11.
What Heidel has done is to map the thematic design of Gen 1 back onto the Enuma Elish. That design is not internal to the Babylonian document. He factors both accounts down to a lowest common denominator. If you operate at a high enough level of abstraction you can always draw a parallel between A and B. But that fails to establish any kind of genealogical relationship. Frankly, his alleged parallels are so far-fetched that they lack even a coincidental appeal.
Like many moderates, Heidel tries to salvage the dignity of Gen 1 by drawing invidious comparisons between its sublime conception and the decadent outlook of its heathen exemplar. These differences "make all similarities shrink into utter insignificance, Ibid. 140. To be sure, the comparisons are odious enough. But it’s hardly insignificant to claim that the Scriptural story of origins is a retrofitted myth borrowed from pagan tradition. It’s one thing to admit that the Psalmists and prophets many occasionally satirize the pagans by making ironic use of their own mythopoetic imagery. It’s quite another thing to claim that the foundational chapter of Bible history is not historical, but rather a bowdlerized version of a pagan creation myth. The God who called Abraham out of Babylonian idolatry would not go nosing through the same dunghill for a few narrative mushrooms.
h) This sort of source criticism doesn’t really explain the origin of the document. Unless we have independent grounds for believing that B is literarily dependent on A, appealing to A to explain the features of B is only a dodge, for we must then account for the features of A. If A could have arrived at those features independently, then why couldn’t the same hold true for B?—in which case, postulating a literary dependence on A only introduces a gratuitous complication into your explanation.
i) It defies psychological realism to suppose that exiled Jews would have canonized a Babylonian creation myth. When a nation is conquered, its citizens respond in two different ways: on one side are the collaborators who assimilate; on the other side are the patriots who resist. The idea that exiled Jews plagiarized Babylonian mythology must posit a silent revolution whereby the collaborators were able to force this heathen tale on the exiled community at large without leaving any trace of opposition in the historical record. But such an action would have provoked an intense and enduring controversy, represented by opposing schools of thought. We know that Jews split over much less than this!
In the words of O. P Robertson,
"It is somewhat remarkable to note the confident assertions made by the majority of scholars today regarding the literary productivity of this [Exilic] period in light of the meager knowledge that is actually available...This hypothetical picture is quite amazing in light of the condition of the Israelite people during its relatively brief exile. A defeated people living as a small minority in a foreign land, scattered over a vast empire throughout cities, towns, villages, and even among ruined settlements, supposedly raises up a group of anonymous authors and editors who produced the most glorious body of religious literature found in human history. According to the generally accepted critical reconstruction, these exiles produced and/or edited the Deuteronomistic history, involving the shaping of the final forms of the book of Deuteronomy and of the historical books Joshua--2 Kings; essentially reworked the whole prophetic corpus; composed the glorious prophecies of Deutero-Isaiah, along with a thorough editing of the first portion of Isaiah so that it would conform with the message of Deutero-Isaiah; developed the cultic lawcodes of the priestly school, involving the collection and codification of the worship practices intended for the then-nonexistent Jerusalem temple and its priesthood; and integrated the priestly version of the narrative of Israel's most ancient history into the revered documents of the Pentateuch. All this work is affirmed to have been accomplished during Israel's exile, despite the absence of any objective evidence that might support the theory," The Christ of the Prophets_ (P&R 2004), 285, n.7.
Other internal data include the following:
As regards the book of Genesis, it is noteworthy that various social customs and religious practices stand at odds with what developed in the time of Moses and afterward. The freedom with which the patriarchs built altars at different locations and offered sacrifices (Gen 12:7-8; 13:4,18; 22:9; 26:25; 33:20; 35:1,3,7) stands in marked contrast to the religious practices associated with Mosaic Yahwism, with is emphasis upon the role of priests and the importance of a central sanctuary.
According to Gen 20:12 Abraham married his half-sister Sarah, yet this practice is forbidden in Lev 18:9,11; 20:17; and Deut 27:22. Similarly, whereas Lev 18:18 prohibits a man from marrying two sisters, Jacob married Leah and her sister Rachel (Gen 29:15-30).
Esau’s firstborn status in Genesis is unlikely to have been invented by a Jewish writer of the exilic/postexilic period. On the contrary, this would have been a major embarrassment to Jews who viewed the Edomite as archenemies (e.g. Jer 49; Lam 4:22; Ezk 25:12-13; 35:15; Obadiah). The same argument could also be applied to the prominence given to Joseph n Genesis, over against the less important role played by his older brother Judah. If this latter tradition was created by a Judean writer, it is hard to imagine that he would have given pride of place to Joseph, from whom the Ephraimites, associated with the northern kingdom of Israel, claimed a royal lineage.
Whereas the preceding comments have focused on Genesis, A. P. Ross makes a similar point regarding the traditions concerning the tabernacle. If these traditions were created in the exilic or postexilic period, it “yields the improbable scenario in which the nation in exile longs to return to their land but instead receives instructions to build a portable shrine for the desert.”
Dictionary of the Old Testament: Pentateuch, T. Alexander & D. Baker, eds. (IVP 2003), 66.
7. Num 12:3. One danger for readers conditioned by the psychological novel is to treat a remark like Num 12:3 as if it were a clinical profile of his psychological make-up. But we need to guard against reading Scripture through a Freudian or Jamesian lens. The Biblical narrator does not mirror our modern obsession with the subliminal motives and interior moods of the character.
The disclaimer of v3 serves a dramatic function. It is on account of his self-effacing disposition that Moses is vulnerable to challenge by more ambitious figures (vv1-2) and must rely on God’s intervention to vindicate his leadership (vv4ff.). Num 12:3 picks up on Exod 3-4, where Moses resisted the prophetic call. Without this set up in v3, the whole episode falls apart.
The presence of anachronistic place-names in Genesis has also been raised in objection to Mosaic authorship. Actually, this would be more evidence for Mosaic authorship. Moses is writing hundreds of years after the events recorded in Genesis. He is writing about the past from his own historical vantage-point. So you would expect him to use current designations. If I were living in New York in the 21C, and you were asking me for directions, would I mail you a 1950 edition of a New York city street map?
I’d add that I see nothing wrong with the idea that someone like, say, Ezra, might have updated some archaic place-names, just as the scribes may have modernized certain obsolete grammatical forms from time to time.