抱歉，没有时间翻译了。只好用英文。欢迎指正。

Overall, Popper’s theory of knowledge (as articulated in LSD, C&R, OK) is of limited (if not very limited) relevance for philosophy of (natural or social) sciences today, after the on-slaughter of Kuhn, Lakatos, and many others.

What he had left behind has been two key ideas:

1) A scientific theory must have a deductive component (and only deductive hypotheses are interesting things to be tested and thus falsifiable). [But see below.]

2) Science (or the enterprise of human knowledge) is an evolutionary enterprise (to be discussed next time).

His main theses, however, suffered from fundamental flaws. Briefly:

1)      There is no truth (or a theory of truth) in his original theory (although he tried to integrate what Tarski had to say about truth later, he did not do it well). Without a theory of truth (as the aim of science), Popper’s theory of knowledge, when pushed to its logical conclusion, will hinge on epistemological nihilism (or absolute skepticism). Put it bluntly, if every scientist follows Popper’s advice (i.e., formulating conjectures and then trying to falsify them, like finding enemies to kill), we would have no foundation for believing in anything (see Lakatos 1970[1978]).

2)      Immediately from the first point, if we do not have a foundation for believing in something, there is no foundation for formulating any conjecture (here, Kuhn and Lakatos came in). See the account of the Popper-Kuhn debate by Worrall (2004).

3)      Popper seemed to believe that all sciences are alike, in the sense whatever subjects a science is studying, that science obeys the same kind of epistemological rules. This has been rejected. Mathematics involves epistemological rules that are different from physics and chemistry, and biology differs from physics and chemistry. And naturally, social sciences differ from natural sciences in fundamental ways. These sciences differ because they study different things. [I do not deny that all sciences share some fundamental core principles and properties.] [This point is my own understanding. Once again, I am saying that ontology underpins epistemology.]

4)      Popper’s account of science has been too ahistorical: the real processes of scientific discovery figured little in his discussion. In contrast, Kuhn, following the lead of Ernest Cassiers, Emile Meyerson, and Alexandre Koyre, made the historical studies of science a key part of any philosophy of sciences. And this is to the benefit of philosophy of sciences. (Ironically, Popper actually himself said that philosophers should study sciences more). [This is why I believe it would be helpful if scientists themselves tell the real story of their adventures (not painting them as heroes), and if possible, do philosophy of sciences.][And this will have major implications,as we get to Kuhn.]

5)      Although deductive hypotheses should be a part of any scientific theory, falsification (or testing) is a far more complex enterprise than Popper had conjectured (cf. Kuhn, Lakatos and their interpreters). Popper’s stand on falsification was quite naïve, as noted by Kuhn and more forcefully by Lakatos. [Well, scientists are not angels.]

Lastly, it is important to note that Popper’s defense of “the Open Society” and his critique of “historicism” can stand on their own: that is, although his critique of dialectics was part of his overall theory of knowledge (see his paper, “What is Dialectic?”). His social critique does not have to depend on his overall theory of knowledge. [We will get to this part later.]