The book is rather loosely structured. It is arranged thematically, but there is not a lot of logic to the arrangement of the themes. For instance, there is no obvious reason for why the chapter on "Continuing Disputes" is placed where it is, or placed in the volume at all. It is one of the more interesting chapters in the book, but it is more a chapter in which a lot of rabbits are chased than any real issues introduced. In the end, the book is a somewhat rambling affair. The upshot in the end, however, is that Shattuck discusses virtually every theme in Proust.
There are two aspects of Shattuck's approach to Proust that I thoroughly applaud and that anyone coming to Proust for the first time should heartily embrace. First, he adamantly refuses to take the approach developed by Proust's great English-language biographer George Painter, and imitated by a host of his weaker readers, and treating IN SEARCH OF LOST TIME as material for finding all the real life models for Proust's characters. As one of his French readers has pointed out, one can read much of Proust's life into the SEARCH, but one cannot read much of life out of it. In the end, focusing on what character inspired who tends to take one away from the novel and back to a gossipy preoccupation with Proust's life. Shattuck simply refuses to do this. Secondly, Shattuck, although himself an academic, refuses to acknowledge that Proust is primarily the property of the academics. He eschews editions of Proust's text weighted down by largely unusable critical apparatuses. Likewise, while he writes in a highly literate fashion, he refuses to get bogged down in any of the more arcane literary debates concerning Proust.
Speaking of not focusing on Proust's biography in reading Proust, I would like to take issue with the reviewer who found fault with Shattuck's tacit acceptance of the Narrator's affair with Albertine being a heterosexual one, and for three reasons. First, the reviewer assumes in making that statement that a reading that takes as primary the real-life sources of Proust's characters. The problem with this approach is that it overlooks the fact that it is a work of literature, and Proust did not leave his real life sources alone, but remolded them into fiction. Proust composed Albertine as a female, and a quite convincing one at that. Unless one happens to know that Albertine is modeled on several homosexual relationships, one isn't likely to think of the Albertine affair as a homosexual one. Second, "Albertine" is not based on any single, or on just a couple, of primary characters. For instance, Proust wrote many of the Albertine sections before meeting Alfred Agostinelli (commonly regarded as the most significant model for "Albertine"), who caused him to expand the character and the plot. Third, many present day readers don't seem to realize that Proust completely feminized homosexual relationships. At no place does he conceive homosexual affairs in current terms, as two males engaged in an affectionate relationship. In all Proust's affairs, one of the two involved took the "female" role. Likewise, throughout his work, he views male-female love as basic, with one or other of the male participants in an "inverted" relationship (his term) substituting for a female.