As the previous chapter closed by noting, thought about language in the early twentieth century had split into a linguistics that describes language as a system, and a philosophy asking about meaning itself. This final chapter follows one turn that happened on the philosophical side, and a further fork that opened in the second half of the century.
The whole genealogy, inside one life — Wittgenstein
As the overview noted in passing, one philosopher compressed this entire genealogy into a single lifetime: Ludwig Wittgenstein.
In his early masterwork, the Tractatus Logico-Philosophicus (1921), he writes:
The logical picture of the facts is the thought.
(Japanese and English renderings of quotations throughout are ToL’s own.) A proposition, on this view, “tentatively assembles” a state of affairs by arranging names — the very arrangement of names becomes a picture of how reality is arranged (article 024). This picture stands on the same side as Aristotle, locating truth’s condition in the sentence (Chapter 1), and as Saussure, treating language as a system (Chapter 4) — the idea that language is a device that accurately mirrors the structure of the world, or of thought.
But in his later masterwork, the Philosophical Investigations (1953), Wittgenstein dismantles this picture with his own hands.
The meaning of a word is its use in the language.
The levers on a locomotive’s footplate all look much the same, yet one varies continuously, one is a binary switch, one grows stronger the further you pull it — looks alone tell you nothing about function (article 026). A word’s meaning works the same way. Picturing the object it refers to settles nothing. What settles it is only how the word actually gets used, in what scene of life.
The method of family resemblance — “don’t look for a definition, look at cases” (article 027) — and the argument that meaning lives not in a private mind but in shared practice (article 028) — both flow from this single turn. Structure to activity: the pendulum this whole feature has traced across 2,300 years happens again, inside one philosopher, in roughly thirty years.
The fork — Chomsky and Austin
Once Wittgenstein had dismantled his own picture theory, thought about language moved down both sides of the fork he’d opened.
One path was walked by Noam Chomsky. In Syntactic Structures (1957), Chomsky argues that grammar is a device “autonomous from both meaning and frequency of use” (article 054). The ancestors he names explicitly are Port-Royal and Humboldt — in Cartesian Linguistics (1966) he calls Port-Royal’s “hidden sentence” a forerunner of deep structure, and flies Humboldt’s “infinite use of finite means” as generative grammar’s own banner (Chapter 2, Chapter 3). But what Chomsky inherits from Humboldt is the word “energeia” alone — its content is reversed. For Humboldt, activity meant something a speaker produces anew each time, never twice the same. For Chomsky, the grammatical device that produces it is an innate, fixed structure built into the mind. In a different way from the picture theory Wittgenstein dismantled, Chomsky pulls language back, once more, into a system — onto the structure side.
The other path was walked by J. L. Austin. How to Do Things with Words, published posthumously in 1962, pushes Wittgenstein’s insight that “meaning is use” a step further. “I do.” “I bet.” — some sentences don’t describe anything; uttering them simply is the performance of an act (article 061). Wittgenstein showed that a word’s meaning lies in its use. Austin shows the step beyond that — that the utterance itself is an act. Saying is no longer even a derivative form of “mirroring the world.” Saying, on its own, already is doing.
What’s striking is that these two near-contemporaries (Chomsky born 1928; Austin born 1911, died 1960) faced in opposite directions. Chomsky cuts language loose from both meaning and situation, studying it as a formal device inside the mind. Austin holds that exactly that “meaning” and “situation” — the felicity conditions that decide whether a performative comes off — are language’s very center. From the same late-twentieth-century, same English-speaking philosophical soil, the one axis running through this whole feature splits cleanly in two once again.
At the end of the genealogy
Seven chapters, and here we are. The question Plato opened hardened, in Aristotle’s hands, into an axiom; became, in Port-Royal’s hands, a mirror of the mind; was released, in Humboldt’s hands, into activity; got pulled back, in Saussure’s hands, into a system; was relocated, in Sapir’s hands, into the middle of culture and history; turned once more, inside Wittgenstein’s own lifetime, from structure to activity; and finally split, with Chomsky and Austin, into two paths.
| Chapter | Thinker(s) | In one line | Position |
|---|---|---|---|
| 1 | Plato → Aristotle | Convention becomes an axiom; the unit shifts from name to sentence | Structure is born |
| 2 | Port-Royal | Grammar mirrors the mind’s act of judging | Structure |
| 3 | Humboldt | Language is not a product but ceaseless activity | Structure → activity |
| 4 | Saussure | A system with no substance, made only of difference | Activity → structure |
| 5 | Sapir | Plural worldviews, none superior | Structure → activity |
| 6 | Wittgenstein → Chomsky / Austin | Picture to use, then to the autonomy of syntax and speech act theory | Activity → structure / activity |
In a nutshell — there is still no final answer to “what is language.” But the sheer fact that this pendulum never stopped swinging, across 2,300 years, teaches something on its own. The urge to fix language as a “thing,” and the urge to release it as a “working,” were probably both written into the object itself from the very start. Which way the pendulum swings next, no one yet knows.