I don't know on what basis you say that the expected utility loss is "effectively zero". There's a utility gain to the person who takes an illegal copy of the book instead of buying it, because they have more money that way. There's a utility loss (which I'd have thought is obviously approximately equal in general) to the people who'd have profited directly from the sale of the book: author, publisher, distributor. And then there are second-order effects, less localized and therefore harder to see and harder to assess, from (e.g.) the slightly reduced incentives for others to write, publish and sell books, the increased social acceptability of getting books in this way, etc.

It looks to me as if what we have here is: first-order effects that cancel out exactly when expressed in terms of money, and therefore probably cancel out approximately when expressed in terms of utility, and second-order effects that are hard to get a handle on but look clearly negative to me.

Could you justify your position further on this point?

As for the second claim, note that this also needs to be true to make your "doing something wrong" assertion correct -- and ought to be obvious to justify your having made it so baldly. I'm glad you agree that it isn't.

No one was claiming or suggesting that anyone should go straight from "I'd find it interesting to read that" to buying the book, without any consideration or weighing of consequences in between. So if my last sentence is equivalent to your main point, it seems to me that you were attacking a straw man.