That is exactly what I suggest. The contradictions are something that bother us, but by their very presence in the text, it is quite readily apparent that even the most recent of the editors through which the text has passed did not take issue with them. This indicates a serious difference in our perception of narrative logic and that of the ancient Hebrews. Similarly, the oddities encountered in the numerical/genealogical framework have remained, in their odd discordance, in the Masoretic tradition, while they have certainly been tampered with in the Samaritan and Septuagintal traditions, which both tried to straighten them out a bit, showing a concern for the narrative logic which is still yet closer to our own than the earlier Hebrew writers, but still not quite identical. This difference is crucial.

You missed my distinction of the context of P and D. I didn’t write that they were “meant for different time periods,” but the narrative context of the elements differed greatly, in a way which explains the differences quite sufficiently. There’s quite a difference in those two perspectives. The skill of the author (or editor, if you will) of the materials at hand was such that they are contextually appropriate. Thus the “citified” stuff of D is not presented in Leviticus, which was set in a wilderness setting and as coming directly from God in the Tent of Meeting, but in Deuteronomy, where the additional instructions were narratively arranged as a series of regulations on what to do once having settled in the Land. The specific differences between the two which you mentioned are more easily explained by context than by alternative theologies of suppositional source materials.

You also say this: If he does not think that we can say something about those sources, he will need to explain why different scholars are able to reproduce the same results by looking at the data in the text. Ah, but do they produce precisely “the same results”? You cannot pretend that they do. I therefore can have nothing to explain. Throughout the history of the project, the lines have always moved, and no two scholars (unless, as the sheep you posit, in their copying of one or another) have ever achieved an identity of equivalence in their separation of those sources. If it were such an objective process, there would be no such differences.

And yes, text critical data, as I described them, are still more objective. Variants are verifiable, while your “sources” are not. That is one of the requirements of objectivity. Has anyone found a manuscript of J/E? Of Q? Then you could verify your sources. Until then, it is all supposition.

In the end though, I’m happy to have had this interesting discussion, and to agree to disagree. As I said, I’ll be going more in depth on the matter later in the year, and we can revisit it in more detail then. For now, we seem to be going in circles.