I agree, but I wouldn't necessarily state that setting up auxiliary hypotheses necessarily makes a research program degenerative, in Lakatos' sense (it can still have a very powerful explanatory heuristic. Einstein's relativity, for example, may still require auxiliary hypotheses to explain anomalous data, but it still has a powerful predictive capacity and remains progressive, in this sense).
Agreed. My point is that any functional scientific programme must contain both progressive and degenerative tendencies simultaneously, but the progressive tendencies must dominate (in some sense), otherwise what you're doing becomes gradually less predictive and hence gradually less useful. In Marxism, the degenerative tendencies are now dominant, and have been for some time.
Then again, Lakatos' conception is extremely problematic only because he thought that it wasn't necessarily intellectually dishonest to take up a degenerative program (he was, after all, a Marxist - although a revisionist in many ways). This is what prompted Feyerabend to think that Lakatos was an epistemological anarchist and was a 'rationalist' only in name - he did not have a solid basis for deciding when to abandon a 'degenerative' program.
I don't see how it is possible to know when a degenerative programme should be abandoned. In practice, degenerative scientific theories just fade away rather than being refuted at some definite moment in history (though that can happen). There are still some people who believe in the phlogiston theory, for example; it's just that nobody pays any attention to them. The point is that there's no telling when a degenerative programme might suddenly revive and become progressive again. Wegener's theory of continental drift might be an example of this.
You are right, however, that science is mostly 'puzzle solving' (in the Kuhnian sense). Lakatos' theory was an attempt to save rational progress in science (since it seemed lost with Kuhn).
It's important not to apply a quasi-Hegelian teleology to the development of science - scientists are not trying to move collectively towards some sort of 'rational progress', but are merely trying to 'save the phenomena', defend their own theories, and establish a career for themselves.
It is clear, however, that Marxism is not a progressive program in any sense, but this does not mean that it may not be made progressive (its auxiliary hypotheses, for one, could be the basis for a powerful heuristic and predictive success. Although, as it stands, I do not think Marxist scholarship is headed in this direction).
I think it should be borne in mind that Marxism was an unfinished project when Marx died. I believe he had planned out a 12-volume magnum opus, but had only published the first volume (
Capital vol. 1) when he died, leaving the next two volumes in an unfinished state and not having even begun the other volumes. Marxists inherited an unfinished palace, and have been busy adding more wings and outbuildings and trying to put a roof on the whole edifice, sometimes not in a very coherent manner. And frankly, some of the builders were cowboys who didn't follow proper building regulations. It's also arguable that the theoretical development of Marxism was derailed by the Russian Revolution and the need of the Soviet state to pragmatically use Marxism as a tool to maintain themselves in power; Marxism thereby became primarily an 'ideology' rather than a science. The collapse of the Soviet Union, while a temporary setback to the Marxist cause, may have actually cleared the ground, so to speak.
"Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong remedies." - Marx (Groucho)