OK, I missed the coincidence (by which I mean they are identical, not that it's by chance) or the names of the authors. That clears up my confusion on that point.

Regarding the lack of mention of the counter in the section of the description you mention, I thought it was obvious. In the section before he says: we keep the number of marks set. That means, to me, that if the element in array accidentally/incidentally points to an element in marks that accidentally/incidentally points back, it's not counted as pointing back because it's known (by using the counter) not to be valid.

It is subtle, I agree, but for me the explanation is clear enough already. Adding more detail could simply overwhelm the reader with words, without letting the reader see the wood for the trees.

I agree, now that you've expanded it, that your point is valid, but I, personally, prefer the version as it is.

In this I expect we should simply agree to disagree. One person's "necessary detail" is another's "pointless nit-picking."