Atonement. I forgot to list this one because it was a very fast reread for school. When I read it last year, I had this to say:

I did not hate it. I enjoyed some parts of it a lot, and some parts of it just went on and on forever. I disagree with Ian [scroll down to his Atonement review] about the metafictional elements; I think they were the only reason for this book to exist. (And I am very glad that people aren’t still just rewriting the works of Jane Austen.) That said, I don’t think the metafictional stuff here is particularly interesting. The long section at the end was too much.

Do not read this next section unless you have already read both Atonement and Life of Pi; if you have read both, drag your mouse over the blank space to see my comments:

These Booker nominees really need a new gimmick. Life of Pi definitely pulled it off in a more interesting and enlightening fashion, but this trick is getting a little old. I might have been more impressed by Atonement had I not seen this coming a mile away. I feel like I’ve read the same book eight times in the last few years. In Life of Pi I felt like he was doing something different with the gimmick, but I don’t think McEwan is doing anything new here and his story is not otherwise very interesting.

Another spoilery bit that you should skip unless you’ve read Atonement:

Over the last week or so I was worried that McEwan was going to give me a cheater ending like he did with Amsterdam, but the fact that Briony was revising history wasn’t what I was worried about. (I figured that out as soon as we got to Robbie’s story. I’m not sure what clued me in, though; it just felt obvious.) I was just afraid that he was going to make Robbie the villain after all, which would have been cheating. I am still not sure if Marshall as the villain was supposed to be a surprise; I don’t think it was because McEwan practically told us that he was the rapist even before the rape happened.

This is probably my last McEwan novel for a while.

After rereading the novel, I tend to give it more credit and I will even go so far as to say that I liked it and think it is probably McEwan’s best, at least of the books I’ve read. The novel is far more interesting on a second reading, knowing what you know at the end: that information gives far more texture to the character development (although you could argue that there is really only one character in the story) and makes the novel more intellectually satisfying. I still disagree with Ian — without the metafictional frame, this book has no reason to exist, but within that framing McEwan is doing something interesting here.