I'm firmly in the "reform copyright" camp. That is, I think copyright is a useful economic tool for promoting creation and distribution of new work, but the current implementation of copyright law is deeply flawed and no longer fit for purpose in most of the western world.
That said, I want to challenge this statement you made, because I think it's too strong:
However, no matter how optimistic you are, what becomes clear is that if copyright dies in a practical sense, you cannot make a living as an artist.
I don't think this is a black/white question, but rather a matter of probability, scale, and variety of options. Many people do make a living in creative industries without really relying on copyright all that much.
For example, most of the work I do is subject to copyright protection, and in some of my roles I would normally transfer the copyright to clients/customers at the end of a job. However, often neither I nor my customers much care about that, because if we're talking about software that is running on their web server or embedded in their device, it has much more practical protection against someone ripping it than copyright affords, and in any case the software would have limited value in isolation so there's not much incentive for others to copy it.
Not everyone in software works on projects where that would be the case, so for others copyright offers a better incentive. But in those cases, other models might also work. I have some hope for the crowd-sourcing idea, as the likes of Kickstarter have already shown that even quite substantial projects staffed by solid industry veterans can pull in a decent amount of funding to match. Potentially there's a lot of middleman removal as a pleasant side effect, all the while still allowing the overall cost of developing a moderately large project to be amortised over many customers (and unlike typical copyright-and-sale business models, potentially allowing different customers to contribute more or less according to their means, so perhaps better satisfying your "democratic model" criteria). I think we need a few more of the bigger projects to actually deliver before drawing too many conclusions here, and of course even the biggest are still orders of magnitude smaller than what copyright-backed industry has achieved, but the early signs look positive from here.
So while I'd agree that the scales proven so far and the odds of success are not as good without copyright as with it, at least for those kinds of creative work where copyright is fundamental to the existing business model anyway, I think it's too strong to say that you can't make a living as an artist without it. What we should be concentrating on is whether more people wind up making more and better work that is ultimately enjoyed by more people with different variations of copyright or other IP frameworks. The idea is to maximise creativity and productivity for the benefit of society as a whole, IMHO.