The “Linguistic Turn” – When Philosophers Started Focusing on Language
We all nod along when someone mentions the “linguistic turn,” as if it’s a well-defined philosophical pivot: before, we puzzled over reality; after, we puzzled over language. But honestly, the more I dig into it, the more I think that’s way too clean a story.
This wasn’t one neat turn—it was a series of overlapping moves, sometimes in tension with each other.
Frege, Wittgenstein, Carnap, and Quine weren’t all playing the same game, even if they were all moving their attention toward language.
What really changed wasn’t just “focus on language”—it was what philosophers thought language could do. Was it a mirror of thought? A formal system?
A cultural practice?
A philosophical solvent?
The answers varied wildly.
So in this post, I want to tease apart those differences—starting with Frege and Carnap—and show why this so-called “turn” might have been more like a spiral staircase.
Frege to Carnap – More Than Just a Smooth Continuation
Let’s be honest—when we talk about the linguistic turn in textbooks or classrooms, there’s this almost lazy narrative arc: Frege invents modern logic, Russell gets on board, then Carnap takes the next step by shifting everything into formal languages. End of story.
But if you look closely, there’s a conceptual rift between Frege and Carnap that’s often glossed over, and it completely changes how we understand what “turning to language” actually meant in early analytic philosophy.
Frege’s Logic Isn’t About Language in the Modern Sense
Frege wasn’t “doing semantics” the way Carnap was. When Frege introduced his Begriffsschrift, he didn’t think he was analyzing natural language. He was building a formal language that bypassed ordinary language’s messiness to express pure thought.
In fact, Frege’s central distinction between Sinn (sense) and Bedeutung (reference) wasn’t originally meant to support linguistic analysis. It was aimed at clarifying how the same object could be thought about under different modes of presentation. Think of “the morning star” vs. “the evening star.” Same referent (Venus), different senses. But this was a theory of thought first, not a linguistic theory per se.
Yes, he used language to make that distinction, but only because language is how we access thought in practice. He wasn’t treating language as the fundamental object of philosophical analysis—he was using it as a window into logic and cognition.
Carnap’s Syntax Project: Language as the Very Ground of Knowledge
Now enter Carnap. This is where things really shift. For Carnap, especially in his Logical Syntax of Language (1934), the idea was not just to express logical relations but to reconstruct all of scientific knowledge within formal languages.
This is a big move. Instead of seeing formal language as a tool for logic, Carnap turned it into a philosophical foundation. He proposed that philosophy itself becomes a kind of metalinguistic activity: clarifying the rules and structure of scientific languages.
He even goes so far as to say that metaphysical debates (like “Are numbers real?” or “Is time tensed or tenseless?”) aren’t wrong, they’re meaningless—unless they can be translated into a well-formed language structure. If they can’t be syntactically analyzed, toss them out.
Formal Languages as Epistemological Filters
What’s fascinating here is how Carnap uses language not just descriptively, but normatively. Language becomes a filter for what counts as legitimate knowledge.
Take his Principle of Tolerance. It says, basically, “You can choose any language you like, as long as it’s precise.” Want to talk in intensional logic? Go for it. Prefer Tarski-style semantics?
That’s fine too.
Philosophy’s job is to make those choices explicit and check their consistency—not to tell you which one is ‘right’.
This pluralism sounds freeing, but it smuggles in a key assumption: that philosophy is metalinguistic. That’s a sharp break from Frege, who thought he was uncovering something objectively true about logic and thought. For Carnap, the truth is conditional on the linguistic framework you choose.
Why This Break Matters
So why does this shift matter?
Because it shows that the “linguistic turn” isn’t just “philosophers analyzing language instead of the world.” That’s too vague. Carnap’s project is radically different from Frege’s—and arguably more radical than we usually acknowledge.
While Frege gave us the tools, Carnap weaponized them. He believed that with the right syntax, we could eliminate bad philosophy altogether. But in doing so, he also transformed language from a mirror of thought into a kind of epistemic gatekeeper.
And this move—treating language as the framework that determines what can even count as a philosophical question—sets the stage for all kinds of later tensions: between logic and ordinary language, between formalism and naturalism, and between meaning and use.
Next up: Wittgenstein. And trust me, that rabbit hole goes even deeper.
Wittgenstein’s Two Revolutions
If there’s one figure who gets name-dropped every time someone mentions the linguistic turn, it’s Wittgenstein. And for good reason: he’s basically the Janus of 20th-century philosophy, staring both forward and backward, speaking two entirely different philosophical languages depending on which book you open.
But here’s the thing—we often flatten the contrast between the Tractatus and the Philosophical Investigations into a kind of before-and-after caricature: early Wittgenstein was into logic and structure; later Wittgenstein was into use and ordinary language. But that gloss misses just how deep the split runs—and how each phase disrupts our assumptions about what analyzing language can do.
The Tractatus: Language as Mirror, But a Fractured One
Let’s start with the Tractatus Logico-Philosophicus (1921). The basic idea seems deceptively simple: language represents the world through logical structure. A sentence is a “picture” of a possible state of affairs. So when language is working properly, it maps reality’s logical form.
This is where Wittgenstein still sounds like a cousin of Frege and Russell. But there’s a twist: the Tractatus collapses under its own logic. Wittgenstein tells us that most of what the Tractatus says can’t actually be said—only shown. That includes the book’s own propositions about logic, structure, and representation.
That’s wild. The work is essentially a philosophical suicide note. The ladder metaphor at the end (“You must throw away the ladder after you’ve climbed it”) isn’t a flourish—it’s the central point. The proper analysis of language, according to early Wittgenstein, ends with recognizing that philosophy can’t provide foundational truths about language at all.
So we already see this strange self-negating move: he builds a hyper-rational structure only to show that the most important parts of meaning lie beyond what structured language can capture.
Philosophical Investigations: Language as Social Practice
Fast forward a few decades, and Wittgenstein’s later work blows that early project apart. In the Investigations (1953), he gives up on the idea of a single, deep structure behind language. Instead, he emphasizes how meaning arises from use—not abstract logical form, but everyday human practices.
Now we get language-games, forms of life, rule-following, family resemblances. Philosophical problems, in this later view, arise from trying to force all of language into a single model—usually logic—and ignoring the diversity of how words actually function.
Think about the word “game.” Is there a precise definition that captures all its uses? No. Yet we use it just fine. This leads to one of Wittgenstein’s key insights: language doesn’t need universal definitions—it needs shared norms of use.
That’s a huge shift. He’s not just saying language is messy. He’s saying that philosophical clarity comes not from cleaning up language, but from recognizing how it’s used in real-life contexts.
Is This Still “Linguistic Analysis”?
Here’s the big question: if early Wittgenstein was a language analyst in the strict logical sense, is late Wittgenstein even doing “linguistic philosophy” anymore?
In some ways, yes—he’s still focused entirely on language. But his methods are different. He’s descriptive, not prescriptive. He doesn’t build systems, he shows confusions. He doesn’t seek essences, he explores use-cases.
And most importantly, he doesn’t treat language as a vehicle for representation. He treats it as an activity—embedded in life, shaped by custom, learned by doing.
That’s radical. He’s not just offering a new view of meaning—he’s offering a new idea of what philosophy is. One that dismantles theory-building itself.
Kripke’s Puzzle and the Limits of Use
One more thing: later Wittgenstein is often credited with giving us the use theory of meaning, but is it even a theory? Or is it an anti-theory?
Saul Kripke famously picked up on this ambiguity in Wittgenstein on Rules and Private Language (1982). His “skeptical paradox” says: how do we know we’re following a rule rather than just acting in a way that seems consistent with the rule so far?
If rule-following can’t be grounded in anything beyond communal agreement, then normativity itself—what it means to be right—is up for grabs. That’s both brilliant and terrifying.
So here’s where we land: Wittgenstein didn’t just shift the linguistic turn—he exploded it from the inside. He turned the focus from structure to use, from logic to life, from certainty to practice. And in doing so, he made language philosophy a kind of anthropology of mind.
From Quine to Sellars – Breaking the Spell of Language
By the time we get to Quine and Sellars, the linguistic turn is starting to look… shaky. Or maybe just more complicated than anyone expected.
What’s clear is this: both thinkers pushed back hard on the idea that language, in itself, could serve as the foundation of philosophy. They didn’t reject language analysis altogether—but they reabsorbed it into a bigger, more naturalistic picture. And in doing so, they kind of closed the book on the original ambitions of the turn.
Quine’s Big Bomb: “Two Dogmas” and the Death of the Analytic
Let’s start with the most famous mic-drop in analytic philosophy: Quine’s Two Dogmas of Empiricism (1951). His central target is the analytic–synthetic distinction—the idea that some truths are true purely by meaning (analytic), while others depend on how the world is (synthetic).
Carnap and others took this distinction as the linchpin of linguistic philosophy. If you can clarify the analytic truths via logical syntax, you’ve done your philosophical job.
Quine wasn’t having it.
He argued that there’s no principled way to draw a line between analytic and synthetic. All knowledge, he claimed, is part of a “web of belief” that faces the tribunal of experience as a whole. You can revise any part of the system, including logic itself, if the empirical pressures are strong enough.
That’s not just a critique of Carnap. That’s a tectonic shift. It turns philosophy back toward science, toward a holistic, empirical model of inquiry where language is just one piece of a much larger puzzle.
Language Loses Its Special Status
Quine’s naturalism means language is no longer the privileged access point to meaning or knowledge. It’s just one tool among others for building theories. So while he still uses logical analysis, he doesn’t treat it as metaphysically deeper than biology, psychology, or physics.
This is a big deal: Quine dethrones language from its pedestal. In his hands, the linguistic turn dissolves into something more like a cognitive turn—or maybe just a naturalist flattening of the whole philosophical hierarchy.
Sellars: No Given, No Simple Language-World Link
Enter Wilfrid Sellars, whose Empiricism and the Philosophy of Mind (1956) takes a sledgehammer to the other foundation stone of traditional empiricism: the “Myth of the Given.”
Sellars argues that no sense data, no immediate experience, can serve as a foundation for knowledge unless it’s already conceptualized—already woven into a web of linguistic and inferential norms. There’s no pure perception. All seeing is seeing-as.
But here’s where he adds something Quine doesn’t: for Sellars, linguistic norms are social, inherited, and developed. Language isn’t a transparent medium—it’s a historically layered, norm-governed system. And that makes it both powerful and fallible.
From Linguistic Turn to Normative Turn?
If Quine naturalized language, Sellars normativized it. He kept the focus on language, but not as a formal system or a representation engine. Instead, he saw it as a medium of socially shared commitments. Meaning becomes inseparable from what it’s appropriate to infer or assert.
That’s a very different kind of turn—not toward language as structure, but toward language as a rule-bound, evolving practice.
In a way, it circles back to late Wittgenstein, but with more theoretical bite.
And this move—toward pragmatism, normativity, and inferential roles—sets up the next wave: Brandom, McDowell, even some readings of Habermas and AI alignment theory today.
Final Thoughts
So was there ever really a “linguistic turn”?
Kind of. But it wasn’t one thing, and it sure didn’t stay put.
It started as a shift in method, became a shift in ontology, and ended up—thanks to Quine and Sellars—as a shift in what we think philosophy even is.
What’s left today is a legacy that’s messy, generative, and still evolving. Language is no longer the special key to philosophy—but it’s still one of its sharpest tools.
And maybe that’s the best lesson: language never gives final answers. But it keeps giving us better questions.