Continuing our observations on this volume, we have just a few notes from Chapters 5-9.
There are some valuable (but familiar) observations in Chapter 5 about the production of books. Here we find some of the same points made by Randolph Richards in his book The Secretary in the Letters of Paul, with respect to how much authority a scribe could be given in composing material for the authority for whom he worked. 
We get to a problematic point where van der Toorn attempts to justify what I have called “veg-o-matic production” of texts of the sort required by the JEDP theory (as well as by the Q/Markan priority hypothesis). The blending of texts in that fashion has been argued by myself and others to present unrealistic views of scribes juggling multiple texts at once. Attempting to justify that this did indeed happen as a practice, van der Toorn finds what he thinks is an actual example of such integration of texts, in a Babylonian textbook series titled Sakikku. Of that text, it was written:
The lore from that old time had not been given a new [authorized] text, but that was like twisted threads for which no master copy was available…[the scribe]…using the splendid intelligence which Ea and Asahullu had granted him, deliberated with himself, and produced a new text of Sakikku, from head to foot…
From this van der Toorn somehow gets the idea of a new text being composed from old ones after the manner seen by JEDP theorists. But this is far from sufficient, not the least because there is no specification about to what extent this alleged “weaving” was done – as some JEDP theorists see “weaving” done to the point where alternating verses come from different sources. In contrast, it appears that the Sukikku corpus, from van den Toorn’s description, was a collection of well-defined topics (such as “behavioral omens”) as opposed to (per JEDP theorists) narrative material.
Moreover, taking van der Toorn’s report at face value, it is the older manuscripts that are seen as “twisted threads” – not the scribe’s new version. In the end, this provides little in support of JEDP composition theories, since we are given no idea from van den Toorn to what extent “integration” of the Sukikku texts was done – whether it was an integration of large blocks of well-defined topical material (requiring little to no juggling of texts), or (as in JEDP) scattered bits and pieces from multiple documents (which would have required extensive juggling).
In chapters that follow, there is little new; van den Toorn appeals to some standard arguments for the disunity of the Pentateuch, and simply presumes such disunity (rather than arguing it) for the bulk of what follows. But this does not mean his more neutral observations are not adaptable to our views. For example, we can easily agree to the idea of successive “editions” of Deuteronomy  as the text was updated over time to make it more intelligible to later readers, even if we do not accept the radical extent to which van den Toorn proposes changes.
A minor point on which van den Toorn is lacking is that he thinks such as this is a valid point: “According to the Biblical account, [Baruch] wrote at the dictation of [Jeremiah]. This part of the tale, however, is historically suspect because it is obviously designed to prove that the collection had the authority of the prophet…”  This does not make it “suspect” in the least, for it assumes without basis that the motive was invention; why is it not rather for verification? Another minor point is van den Toorn’s assumption that stories become historically suspect simply because they are reported similar events in similar ways [197; see link below]. (However, his claims of “doublets” can also be questionable; I cannot see how he thinks Jer. 21:1-7 is a “doublet” of Jer. 37:3-10!)
We’ll conclude with any notes on the last two chapters tomorrow.