I keep an index of every book I have read twice, with a note on what was different the second time. It is sixty-one entries long and it has turned into the most interesting document I own, for reasons I did not anticipate when I started it.
The naive theory
I began it assuming the second reading would be about catching what I missed — details, foreshadowing, a structure only visible once you know the ending. This does happen, and it is the least interesting thing in the file.
What is actually in the entries
Overwhelmingly, changes of sympathy. The same characters, the same sentences, and a different sense of who is right. A novel I read at twenty-five as being about a man trapped by his obligations read, at thirty-six, as being about the people he was obliged to. Nothing in the text moved.
Second most common: books that got shorter. Not in pages — in weight. Things that felt profound at twenty-two frequently turn out to have been a single good idea stated with great confidence, and the confidence was what I was responding to.
Third, and rarest, the reversal: a book I thought was minor turning out to have been carrying something I could not receive at the time. Six entries out of sixty-one. Those six are why I keep doing this.
The awkward conclusion
The book is not what changes. Obviously. But the practical consequence is more uncomfortable than it sounds: it means my confident judgement of any book I have read once is really a report on my own condition at the time of reading, presented as a verdict on the text.
I have written a lot of those verdicts. I stand behind fewer of them than I would like.