So he's accepting the logical possibility that any "falsification" is mistaken, just as the realist might accept the logical possibility of solipsism or skepticism, but it doesn't then follow that he rejects falsification, just as it doesn't then follow that the realist rejects the veracity of our everyday perceptions and beliefs. — Michael

Not sure what the scare-quotes are for, but according to Popper, all falsifications are tentative. And of course he does not reject falsification, but rather offered a methodological approach towards making such a decision. As I said, the Quine-Duhem thesis is true.

Again, provide the full context: — Michael

Why don't you then?

I admit, a conventionalist might say [your emphasis], that the theoretical systems of the natural sciences are not verifiable, but I assert [my emphasis] that they are not falsifiable either. — Popper

And the solution is in the next section (20).

The "I" in that sentence also refers to the conventionalist:

A conventionalist might say "I admit that the theoretical systems of the natural sciences are not verifiable, but I assert that they are not falsifiable either."

Why don't you then? — tom

I did

"These objections of an imaginary conventionalist seem to me incontestable"

Yes, and the realist might say the same about the objections of an imaginary idealist or solipsist or skeptic regarding the logical possibility of Descartes' demon or philosophical zombies or no external world, but to present such a criticism as being the direct words of the realist would be to misrepresent their position. The quotes you provided were taken out of context and so don't give proper justice to what he was actually saying.

So you think the Duhem-Quine thesis is false for Solipsists?

I think that you misrepresented Popper by taking selected quotes out of context.

Well, you're going to hate this:

"If you insist on strict proof (or strict disproof) in the empirical sciences, you will never benefit from experience, and never learn how wrong you are."

I have no idea what Solipsists might make of that though.

Well, you wrote this:

But if your expertise in Popper matched your expertise in physics, then you would know that, according to Popper, it is logically impossible to falsify any theory. — tom

and you could be taken to be claiming that Popper thinks it is not logically possible to falsify a theory full stop, or that it is not possible to falsify a theory logically (by means of logic).

Now, at first glance I think the latter is so obvious that it should be discounted that he would have meant that. On the other hand, it is just the belief that a theory could be falsified by logic that he seems to be targeting in this:

"If you insist on strict proof (or strict disproof) in the empirical sciences, you will never benefit from experience, and never learn how wrong you are." — tom

Only in deductive logic (if anywhere) is "strict proof (or disproof) possible; certainly it is not possible in the empirical sciences.

It is not the 'local' (here, now) claims of hypotheses and/or theories which are not verifiable in principle; it is global or universal (everywhere, always) claims that are not verifiable. The 'local' claims may be either verified or falsified, but the global claims may only be falsified. It is an integral claim in both Relativity theory and Quantum theory that they are universal. This claim would be falsified in either case if a region were discovered where the theory in question did not hold. But then this 'falsification' could only ever be as good as our observations; which could be mistaken, no matter how unlikely that may seem. So strictly speaking. all and any cases of verification or falsification are only as infallible and exhaustive as our observations are.

It is not the 'local' (here, now) claims of hypotheses and/or theories which are not verifiable in principle; it is global or universal (everywhere, always) claims that are not verifiable. The 'local' claims may be either verified or falsified, but the global claims may only be falsified. — John

Nothing can be "verified" including 'local' statements, because all observation is theory-laden and therefore fallible. Also, because of the Duhem-Quine Thesis, (which I have mentioned more than once) there is no such thing as an experimental result logically contradicting a theory.

Perhaps it is best left to Popper to explain to us what his theory actually is:

As I tried to make clear in 1934 (L.Sc.D., p34 ; and sections 10 and 11), I do not regard methodology as an empirical discipline, to be tested, perhaps, by the facts of the history of science. It is, rather, a philosophical - a metaphysical - discipline, perhaps partly even a normative proposal. It is largely based on metaphysical realism and the logic of the situation: the situation of a scientist probing into the unknown reality behind the appearances, and anxious to learn from mistakes

That's from the preface to Popper's "Realism and the aim of Science". And I bolded "normative proposal".

Yes, all observation is conceptually mediated, but I don't think it is appropriate to refer to it as being "theory laden" through and through.

So, within the conceptual framework which is the ground of all experience and knowledge, intersubjectively verifiable observations of the behavior of things, and intersubjectively coherent formulations of those observations, may be made. "Water boils at 100 degrees C at the 'normal' atmospheric pressure and conditions" (to be found at sea level on Earth) is one such example among countless others.

The "conceptual framework" I referred to is not, at base, a matter of theory at all, but is itself the very condition upon which all intelligibility depends. This is not to say that theories cannot come to form parts of that conceptual framework, but they are not essential parts. If water suddenly began to boil at different temperatures then the formulation quoted above would cease to hold. In that case one could rightly say that it had been falsified, as would any theory that had claimed that the formulation was universally invariant.

So I think, contrary to what Popper seems to be asserting in the passage you quoted, that methodology, whether scientific or pre-scientific ('scientific' here being taken in the narrow 'modern' sense) is a natural outgrowth of the "logic of the situation", and not a matter of self-consciously philosophical or metaphysical discipline at all. Adoption, and assertion of the superiority, of particular methodologies of course is a philosophical matter.