Much early alchemy seems to have been adventure. You heated and mixed and burnt and pounded and to see what would happen. An adventure might suggest … - Ian Hacking

A single observation that is inconsistent with some generalization points to the falsehood of the generalization, and thereby 'points to itself'.

Well, he wasn't a relativist. There's a long and complicated story of the rise of a desire for scientific relativism. Part of it may well be simply sort of rage against reason, the fear of the sciences and a kind of total dislike of the arrogance of a great many scientists who say we're finding out the truth about everything—and here [with Kuhn] there was a way to undermine that arrogance.

Kuhn cannot take seriously that “there is some one full, objective, true account of nature.” Does this mean that he does not take truth seriously? Not at all. [...]
Kuhn did reject a simple “correspondence theory” which says true statements correspond to facts about the world.[...]
In the wave of skepticism that swept American scholarship at the end of the twentieth century, many influential intellectuals took Kuhn as an ally in their denials of truth as a virtue. I mean the thinkers of the sort that cannot write down or utter the word true except by literally or figuratively putting quotation marks around it—to indicate how they shudder at the very thought of so harmful a notion. Many reflective scientists, who admire much of what Kuhn says about the sciences, believe he encouraged deniers.
It is true that Structure gave enormous impetus to sociological studies of science. Some of that work, with its emphasis on the idea that facts are “socially constructed” and apparent participation in the denial of “truth,” is exactly what conservative scientists protest against. Kuhn made plain that he himself detested that development of his work...