Michel de Montaigne was opposed to rewriting. Thus he found in his Essais, "as far as I am concerned, I fear to lose by the change: my understanding does not always go forward, it goes backward too. I distrust my thoughts hardly any less for being second or third than for being first, or for being present than for being past. We often correct ourselves as stupidly as we correct others." In other words, writing (or rewriting) for him does not amount to sitting in judgment of ourselves, as Ibsen and Sontag thought. 
This makes sense, at least to some extent, because it's not a good idea if the judge and the accused are one and the same person, and it's not even clear what kind of crime is being committed (since writing is still going on).
But that is not Montaigne's point. There are several people involved. He thinks that "myself now and myself a while ago are indeed two, but when better, I simply cannot say. It would be fine to be old if we traveled only toward improvement. It is a drunkard's motion, staggering, dizzy, wobbling, or that of reeds that the wind stirs, haphazardly as it pleases." 
This is much too Heraclitean for me. I'd prefer to think that there is something like a (semi-permanent and roughly identical) self that sits in judgment of what it has wrought. In other words, I disagree with both Ibsen and Montaigne. I also think that Sontag was mistaken in thinking that she was following Ibsen.
1. This does not mean that Montaigne never changed his text, but only that the changes were mainly additions: "I add, but I do not correct" because "my book ... is only an ill-fitted patchwork" that isn't changed by "extra ornaments." Accordingly, there are clearly identifiable strata of text in the Essais.
2. Michel de Montaigne, The Complete Works. Essays, Travel Journals, Letters. Tr. Donald M. Frame. Intro. Stuart Hampshire. New York: Alfred A. Knopf, 2003, pp. 894f.