Is it because it isn't taking something without permission?
Is it because intellectual property isn't property in some sense?
Is it because the original isn't really "owned"?
Is it be because the owner still has the original?
Given similar reasoning, "identify theft" isn't really theft. After all, if someone "stole" my identity I'd still have the original.
Identify theft is a kind of fraud or perhaps libel, but it's certainly not conventional theft.
Nevertheless, the term has taken on a life of its own, and that's just the way language works. The problem arises when people say inane things like "you wouldn't steal a car, would you?". That analogy is sort of like saying you shouldn't go on strike, after all, you wouldn't want somebody to strike you, right?
So, is it theft? Well, from a linguistic perspective: it's called whatever it's called. But it's simply deceptive to pretend that merely because the same word is used that the meaning must be equivalent too. It's particularly unfortunately that the manipulation of language is probably intentionally deceptive; the notion of intellectual property didn't arise amongst the average English speaker, but amongst a special interest group that stands to gain by causing this particular confusion.
there is some argument that it is conventional theft. It does, essentially, render the original unfit for use until considerable time is spent repairing it.
It really isn't the same thing. It's vaguely related, and that's fine, but pretending the two meanings are identical in an analogy (you wouldn't steal a car?) is still disingenuous.
If someone "steals" your identity, it isn't really theft sure. Really, you both have a pointer to that identity. The problem, which makes it a good idea for it to be illegal, is that you both have a pointer to the EXACT SAME identity. Thus, any destructive(or benevolent) changes they make affect your identity as well.
Contrast this with most forms of "copyright theft", where the original owner still has the original AND any changes you make to your copy have no effect on the original.
Suppose you write a book report for school and another student copies yours and passes their copy in first. You still have your original but it is significantly devalued. Passing it in may actually have negative value as you maybe accused of cheating.
It may not be theft but I think it hard to argue that it is not taking something of value without permission.
That's a preposterously loose requirement. Secondly, even in your example, the damage isn't primarily caused by the copying, it's primarily caused by the misrepresentation. If the copier had been truthful about the source, it's doubtful the original would have been much devalued.
In any case, clearly copyright violation can devalue the copyright. It's certainly not always the case (it may even increase the value of the copyright), but it's a serious risk. However, that still doesn't make it equivalent to theft, any more than vandalism, fraud, competition, criticism, libel, etc. are theft. Some of those are even considered virtuous - as a society we supposedly encourage criticism and competition (even if many sectors manage to pay only lip service to that notion).