I agree. Among other things, I'm pointing out the opposite risk: a beginner may come up with a correct solution to the exercise, check from the book's provided solutions that her solution is correct, and move on to the next section, fooling herself into thinking that she has sufficient understanding (hey, she solved the exercise correctly!), not realizing that “not needing to look at the solutions” and “coming up with (or finding from other sources) lots more exercises” is part of the task of understanding.
BTW, I didn't mention this earlier, but for verifying solutions to exercises which ask to prove something, I've never found it useful to see a solution that is simply a proof. The only way to get more confidence that you've proved something correctly is to find someone adversarial (e.g. a friend with similar or greater mathematical maturity) and try to prove it to them while they keep say “why” or “I'm not convinced”. (A proof is a social process; you cannot get feedback on your proof from a static, non-interactive source like a book.) One's proof may be superficially similar to what's in the book and yet not be a proof, while it can be very different and still be fine.
What provided proofs can help with is getting you unstuck when you've been unsuccessful at proving something. So they help with frustration and motivation, as I mentioned earlier (feedback not so much).
BTW, I didn't mention this earlier, but for verifying solutions to exercises which ask to prove something, I've never found it useful to see a solution that is simply a proof. The only way to get more confidence that you've proved something correctly is to find someone adversarial (e.g. a friend with similar or greater mathematical maturity) and try to prove it to them while they keep say “why” or “I'm not convinced”. (A proof is a social process; you cannot get feedback on your proof from a static, non-interactive source like a book.) One's proof may be superficially similar to what's in the book and yet not be a proof, while it can be very different and still be fine.
What provided proofs can help with is getting you unstuck when you've been unsuccessful at proving something. So they help with frustration and motivation, as I mentioned earlier (feedback not so much).