Virtually all multibook contracts—regardless of their nature, and even including fiction!—include "completion by another author" clauses. From the publisher's perspective, this is to guard against precisely what happened to Manchester's multivolume biography of Churchill: partial publication of early volumes, followed by noncompletion of later volumes. From the author's perspective, it essentially allows publishers to give their children plastic surgery so extensive that they end up a different species. (No matter how professional the writer, every book is in some sense the writer's offspring.)
As a rule, these clauses are non-negotiable. The best case for the author is restricting noncompletion to literal noncompletion (that is, the manuscript is not turned in within five years of the due date); but only a handful of authors have the kind of clout necessary for such a change. However, what is essential is to make sure that only the publisher's editorial judgment may be used to determine whether the manuscript is "satisfactory"—or you'll end up in court, with no good solution for anyone. Unfortunately, it is becoming more and more common for publishers to struggle to find an editorial justification for rejecting or modifying a manuscript because the publisher discovers that it may have made a "bad bargain." (Yes, I have several matters running behind the scenes right now with exactly this problem.) When the publisher's general discretion is an acceptable reasons, one must turn to Chodos and expensive litigation. It's actually simpler and cheaper for everyone to leave all of the discretion to the editors, and not allow the S&M folks to stick their noses in—especially when, as in Chodos and perhaps most similar disputes, the editor who acquired the book is out of the picture at the publisher.
All of which is a roundabout way of saying "It depends (but probably so)."