What value is there in the counterexample method in philosophy? For sure, a well-constructed counterexample is an effective means of revealing the deficiency in a proposed analysis of a concept. But where do we go from there? If, for example, as is clearly the case in the recent history of analyses of “S knows that p” (hereafter just “knows”), there is no consensus on an adequate definition, does the churning out of ever more clever counterexamples simply devolve into a competition to knock contenders out of the ring? It can certainly seem that way, which leads many philosophers to despair of the very idea of analysis. In my view, however, drawing that inference misses the point and beauty of the proposed-analysis-and-counterexample method. We learn a lot from the failures of analysis, which drives us to original thinking about the concept at issue. Nowhere is this clearer than in the history of epistemology after Gettier’s (1963) refutation of the Justified True Belief analysis (JTB) of “knows.” In this chapter, I aim to show how the ground beneath Gettier’s demolition has proved to be remarkably fertile.