Grammar as the Method of Philosophy

The grammar of a word is not its definition but its use: the whole range of contexts in which it appears, the inferences it licenses, the substitutions it permits, the moves in the language game that are open to speakers who employ it. Grammar in this sense is more like a collection of permissions and prohibitions — a logical geography — than a set of mechanical rules. To know the grammar of "pain" is not to know how to spell it or parse the sentence but to know what follows from "I am in pain," what one can ask, what one can doubt.

Wittgenstein's diagnosis of philosophical problems is consistently grammatical: we have been misled by the surface form of an expression into assuming a logical structure it does not have. "The mind" looks like "the body," so we suppose there is an inner realm structurally analogous to the outer world. "I know I am in pain" looks like "I know there is a chair in the room," so we suppose inner knowledge works like outer knowledge. Scrutinising the grammar of these expressions dispels the illusion without leaving anything in its place — as dissolving a false problem leaves not a true problem but no problem at all.

Grammar-investigation does not improve or reform our language; it leaves it exactly as it is. Philosophy, Wittgenstein says, simply puts everything before us and neither explains nor deduces anything. What is hidden is not something of depth but something of breadth — spread out before us but unnoticed because we are always looking elsewhere. The grammatical investigation is an act of attention, not excavation: it brings into view what was always in view, and that is enough to dissolve the problem.

The concept of philosophical grammar is developed across the Blue Book, the Brown Book, and §§90–133 of the Investigations. The remark that "grammar tells what kind of object anything is" appears in §373 of the Investigations, and is one of the most compressed statements of Wittgenstein's method.