DA: "But that would mean for example that Aristotelian thinking - which stood for hundreds of years - would qualify as conclusive. And if we shorten what we mean by 'decent'  Descartes didn't do too badly there for a while, and even Hegel."

True, I retract that criterion then. To express more properly what I had in mind, I would need to tack quite a bit on to that requirement; but I think doing so isn't needed right now, so I surrender that point to you.


Regarding options (2) and (3): thank you for clarifying your misgivings. Now I'm led to wonder, however, about your position on our epistemic relation to truth in general. It seems to me that every--or nearly every--item that we could call "knowledge" is susceptible to the uncertainty that results from seeking a justification which itself needs no further justification (the "unconditioned", Kant called it). This, I take it, is part of the gist behind your criticisms of (2) and (3); I will address (3)'s other point (i.e. showing the mistakes of all previous philosophers) later. Anyway, without one of those mythical self-sufficient justifications, we're left forever asking, "But how do you know that? ...And how do you know that?", etc.
Right, this is old hat: the skeptic's infinite regress. We can also say, hey, anything which we think true now might be judged false by future generations; because we have been wrong in the past, we infer that we might again be wrong in the future. Again, typical skepticism, but now with a stronger bearing on your criticisms of (2) and (3). What certainty do we ever have that we've settled anything conclusively?

In the strictest sense, I don't think we can have that sort of anti-skeptical certainty. However, from a practical point of view we do need to accept certain propositions as true, or at least act as though they're true. That kind of practical necessity leads me to believe that at some point we can or perhaps even must decide that we reach definitive conclusions, even if we simultaneously grant the possibility that those conclusions may be falsified later on. So, in day-to-day life and in non-philosophical disciplines, we accept conclusions even though these domains are not free from the problems in (2) and (3) either. Similarly, to establish anything in philosophy would require the practical acceptance of conclusions that we recognize may one day be disproved.

Regarding the other part of (3) which I previously neglected but promised I'd return to: I deny that it's necessary to show that each and every past philosopher employed wrong methods. If I know as a given that my conclusion is right, and other conclusions differ from mine, then every past conclusion which is not equivalent to my conclusion (assuming the problem has but one solution) is wrong; in which case it follows naturally that all previous efforts employed either bad reasoning or bad premises.