Biblical scholars build models.
A model is a way of accounting for as much of the available data as possible in as coherent and persuasive manner as possible, producing along the way as little cognitive dissonance as possible.
A model is a hypothesis of what the “big picture” looks like. Models do not focus on biblical issues in isolation, but are after the big picture. All biblical scholars–fundamentalist to liberal and everything in between–have models that form the intellectual parameters within which they handle the particulars of biblical interpretation.
Ideally, biblical scholars understand that the model and the data (the forest and the trees) are in dialogue. They are self-consciously aware of the paradox that models can both guide and distort biblical interpretation. A good biblical scholar will embrace that tension, which means being on the lookout for when the model moves from help to hindrance.
When biblical interpreters need to massage the data in order to maintain the model, or need to resort to specious argumentation, or find too many exceptions, it is an indication that a new model is needed–or at least a serious refinement of the existing model.
If I can switch metaphors, biblical scholarship is like building a picture puzzle. The box says 1000 pieces, but there are only 200 in the box. Biblical scholars, working with this limited data set, take those pieces and try to come up with an overall picture of what the entire puzzle looks like.
They begin by sorting the pieces out by shape and color. Pieces with straight edges form the periphery of the puzzle. After grouping the pieces together, one can see a picture forming: a patch of grass here, a path to what looks like a barn (or is it a farmhouse?) over there, a sky with patches of blue and clouds up to the left, mountains off to the right.
This looks like a compelling overall picture, and puzzle experts generally agree. Some point out, however, that there are some pieces that don’t seem to fit the scene very well. Two non-joining pieces are gray and shiny and look like two sections of a fighter jet. Some puzzle experts write books on what kind of fighter jet it is. Some suggest that, if it is a fighter jet, it is out of place in a farm scene and so abandon that model. Others think it is fully compatible with the farm model, although some adjustments need to be made (e.g., the farm scene is post World War 2, not nineteenth century as was first asserted). Some reject the fighter jet hypothesis entirely because it is so out of place with farm the model, that otherwise seems so certain.
Biblical scholars debate over how best to explain all the pieces.
Occasionally someone finds ten puzzle pieces under the sofa and adds them to the picture. The result will either corroborate the farm hypothesis, disconfirm it, or more often than not, answer some questions but raise others.
I could go on like this, but you get the point. All biblical scholars work with a fairly limited data set and by it try to explain the bigger picture. The models that catch on and stick around are those that do the best job of explaining the data in the minds of people who spend their time and energy working with the data.
The fact that the data set is limited does not mean that any explanation is as valid as any other. For example, an anti-farm fringe group may have a vested interest in interpreting the puzzle pieces as an urban scene. They do so through a clever manipulation of some pieces and discarding others. This model will fail to persuade those outside of this group, and so will likely not catch on in the long term.
Because biblical scholars are, ideally, open to migrating to news models (or at least modifying old ones), they tend not to be persuaded by arguments that rest on the authority of theological tradition. In other words, the mere presence of theological tensions resulting from a model is not automatically an argument against that model’s value. It may mean that the presence of theological tensions is an indication of a theological model that is itself in need of adjustment of some sort. (Theologians have models, too.)
“Moses wrote the substance of the Pentateuch in the 2nd millennium BC” is a model of Pentateuchal authorship. “The Pentateuch is a product of postexilic scribal activity working with older oral and written traditions and adding new material” is another model of Pentateuchal authorship.
The question is which of these models (or other models) does a better job of accounting for the data in as coherent and persuasive manner as possible, producing along the way as little cognitive dissonance as possible.
The fact the mosaic authorship is the traditionally accepted model does not automatically validate the mosaic authorship model or invalidate the postexilic scribal activity model.
What complicates the matter is that one must decide at some point what actually constitutes ”data” and how precisely does the interpretation of those data affect one’s assessment of the overall picture.
What makes it even more complicated is discerning how debates over models of Pentateuchal authorship square up with other issues of biblical scholarship. To return to the puzzle metaphor, Pentateuchal authorship is one group of puzzle pieces in the lower left of the partially constructed puzzle. Does this grouping of pieces “fit” with what we see in the puzzle as a whole or are we forcing pieces together and laying them out in a way that just doesn’t work?
This is why it’s sometimes difficult to answer questions directly like, “Why do you interpret Genesis 36:31 as evidence of postexilic authorship rather than slight editorial updating during the monarchy?” or, “Why do you read Gen 1 and 2 as two separate creation accounts?” Whole models lie in the background of any answer one might give, and models can’t be laid out quickly, and certainly not in a moment of interrogation. That takes time and patience.
So, that’s what biblical scholars do. Aren’t you glad you asked? I hope you sleep better tonight.