This article investigates the role played by the analysis of nonliteral language in Wittgenstein’s later philosophical method. In the first section, I briefly present his motivations for abandoning the Tractarian method and developing a new one, which is centered on the ideas of family resemblances and the overview or synoptic presentation of grammar. In the second section, I offer an account that attempts to unify the apparent variety of what Glock called “roots of … philosophical confusion” by treating one of the items in his list, “analogies in the surface grammar of logically distinct expressions”, as the most central target of methodological synopsis. I conclude that the figurative use of ordinary-language terms in philosophical discourse generates the majority of our philosophical problems and that its investigation should therefore be seen as one of the defining features of Wittgenstein’s later philosophical method.
Henrique Mendes (Fri,) studied this question.