Except that the whole point of copyright is to enrich the public domain, not keep works out of it. It grants creators a limited monopoly on the work, after which it becomes part of our shared culture. Perpetual copyright and placing the burden of researching and protecting it on those who want to use a piece of culture has a chilling effect on future modification and creation.

Copyright was never intended to be perpetual or to enable large companies to lock down culture so that only they are allowed to create and distribute it.

Copyright isn't perpetual. I'm stating the above argument is weak and hand waving because it is assuming these works are not available strictly because of copyright. It isn't that black and white. How many of these works wouldn't have ever been created if the author had no way to recoup his investment due to copyright?

I haven't seen much serious talk here of abolishing copyright. I think the more relevant question is "How many of these works wouldn't have been created if copyright were limited to a shorter period as it was originally conceived and not extended to be perpetual in practical terms?"

The answer is "basically none", as referenced elsewhere in the discussion: http://rufuspollock.org/papers/optimal_copyright_term.pdf

The answer is probably all of them, and more due to derivative works being easier to create. Copyright has little to do with preserving or incentivizing profits at all, and it certainly doesn't have anything to do with how much someone has "invested" in something.

Copyright is relatively new. Music, writing, art and all the rest flourished before it existed. What copyright added was the ability for non-creators (publishers, producers, companies) to make money off of the creators.

There isn't anything wrong with that. Authors still enter into these deals willingly because it is beneficial for them to do so. I highly doubt JK Rowling would be one of the richest people in the UK without help from her publisher.

What do you mean? The cost of legal uncertainty is one of the reasons software patents are a worse kind of monopoly than patents on chemical compounds. Even if it's a terrible chemical patent, you can tell whether it affects you or not, not so much with software patents.

So I can easily see how similar uncertainty and the associated costs could cause certain public domain works to remain unexploited.