What Wittgenstein Got Right About Language

On private language and its stakes for consciousness studies

There is a passage in the Philosophical Investigations that philosophy of mind has not yet finished digesting. It concerns what Wittgenstein called the private language argument, and while the argument has attracted an enormous secondary literature, most of that literature debates its validity as an argument rather than attending to what it reveals. What it reveals, I want to suggest, is something that changes the terms of consciousness studies entirely, not by offering a solution to the hard problem but by showing why certain formulations of the hard problem rest on a grammatical confusion so deep that solving it would be the wrong kind of achievement.

Wittgenstein is one of those philosophers who gets cited often and read carefully less often than he should be. In the philosophy of mind, he is typically invoked in one of two ways: either as the source of the private language argument, which is then treated as a puzzle about rule-following, or as a precursor to functionalism, which misses almost everything important about his later work. What tends to get overlooked is the use to which the private language argument can be put in thinking about consciousness, and specifically about the relationship between phenomenal experience and the concepts we use to describe it.

Child communicating

The Private Language Argument, Stated Plainly

Wittgenstein invites us to imagine that someone tries to give names to their own inner sensations independently of any public criteria. The person concentrates on a particular sensation, writes the sign S in a diary each time it occurs, and supposes that by doing so they have established a private reference for the term. Wittgenstein's question is whether this constitutes genuine naming, whether the term S means anything in the way that words in a public language mean things.

His answer is that it does not, and the reason is not merely that no one else can check to what S refers. The deeper problem is that the person themselves cannot check it. In a public language, the correctness of a word's application is determined by criteria that can, in principle, be assessed independently of any particular act of applying it. What makes red mean red is not simply that I call this colour red; it is that there are standards of correct application that my usage is answerable to and that exist independently of my current inclination. Those standards are what make it possible for me to misapply a word, and the possibility of misapplication is what makes meaningful application possible at all.

In the supposed private language, there are no such standards. Whatever seems right to the person is right, and that means, Wittgenstein says, only that we cannot talk about right here. The diary entry is not the application of a concept to a sensation. It is a gesture that mimics concept-application without achieving it. The private linguist has produced something that looks like naming from the outside but is not naming from the inside, where inside means: governed by a practice that distinguishes correct from incorrect use.

Whatever seems right is right: that only means that here we cannot talk about right. The point is not sceptical. It is that the grammar of meaning requires a background

of practice that pure interiority cannot supply.

What the Argument Is Not

The private language argument is frequently misread as a behaviourist conclusion in disguise, as if Wittgenstein were saying that there are no inner sensations, or that talk of inner states is just talk about dispositions to behave. This reading is wrong, and Wittgenstein says it is wrong explicitly. He is not denying that people have sensations. He is questioning the assumption that a word gets its meaning by being correlated with an inner object, the way chair might be thought to get its meaning by being correlated with physical chairs.

The target of the argument is not the inner life but a particular picture of how language works, what Wittgenstein calls the Augustinian picture: the idea that words name objects and the meaning of a word is the object it names. If meaning were correlation between sign and named object, then the private linguist would succeed in establishing a meaning just by concentrating on a sensation and writing a sign. What Wittgenstein is trying to show is that the Augustinian picture, whatever limited validity it has for describing physical objects, fails entirely when applied to sensation terms, and that its failure here illuminates something important about how sensation terms actually work.

How they actually work is not by naming inner objects but by functioning within practices of expression, response, and correction. I learn what pain means not by being shown my own pain and given a label for it but by acquiring a form of life in which expressions of pain, responses to pain, reports of pain, and questions about pain are all interwoven with behaviour, circumstance, and the responses of other people. The meaning of pain talk is spread across all of this, not concentrated in a single act of inner ostensive definition.

The Stakes for Consciousness Studies

Here is where the argument becomes important for consciousness studies in a way that has not been sufficiently appreciated. The hard problem of consciousness, as standardly formulated, asks why physical processes give rise to subjective experience. It typically presupposes that we know what we mean by subjective experience, that the target of the explanation is clearly identified even if the explanation itself is lacking. The private language argument puts pressure on precisely this presupposition.

When consciousness researchers speak of qualia, of what it is like, of the phenomenal character of experience, they are using terms that purport to refer to purely private objects of inner acquaintance. The reference is supposed to be secured by introspection alone, independently of any public criteria. Thomas Nagel's bat argument works by stipulating that there is something it is like to be a bat, where the force of the stipulation derives entirely from each reader's ability to identify, by inner acquaintance alone, what something it is like means. Frank Jackson's Mary thought experiment similarly supposes that Mary, upon seeing red for the first time, comes to know something, where the something in question is a purely phenomenal item whose nature is given directly to experience and requires no public criterion for its identification.

Wittgenstein's argument suggests that this is exactly the wrong model for how such terms work. If the meaning of experience and what it is like were determined purely by inner ostension, independent of public criteria, then those terms would be in exactly the position of S in the diary. They would look like referring expressions without actually referring to anything, because nothing would distinguish correctly applying them from merely seeming to correctly apply them. And if that is right, then the hard problem, as usually stated, is not a genuine problem with a missing solution but a question that cannot get started because its key terms have not been given a meaning that would allow it to get started.

The hard problem presupposes that we know what we mean by phenomenal experience. The private language argument asks us to demonstrate that we do. That is a different kind of pressure on consciousness studies than any empirical result could supply.

The Standard Responses and Their Limits

Philosophers of mind have several standard responses to this line of argument, and it is worth taking them seriously before pressing further. The most common response is that the private language argument shows only that a term cannot mean what it means solely in virtue of a private baptism, but that this leaves open the possibility that terms like pain refer to inner states whose nature is given to the subject by acquaintance, even if those terms acquire their meaning through public use. The public use fixes the meaning; the inner acquaintance is then the mode by which the subject knows what the term refers to. On this view, Wittgenstein's argument is compatible with the existence of qualia and with the hard problem as typically formulated.

This response is not unreasonable, but I do not think it survives careful attention to what acquaintance is supposed to achieve. If the role of inner acquaintance is simply to provide an instance of what a publicly-defined term applies to, then there is nothing distinctively problematic about it. But the role that acquaintance is supposed to play in consciousness studies is stronger than this. It is supposed to reveal the nature of phenomenal experience from the inside, to provide access to what experience is in itself that cannot be captured by any third-person description. And this stronger claim is exactly what Wittgenstein's argument targets. If the nature of experience could be fully characterised from the inside by acquaintance alone, and if that characterisation were independent of all public criteria, then the terms used to express it would be in the position of S. The acquaintance would give us the illusion of meaning without the substance of it.

A second response is to accept that phenomenal concepts are public in their meaning but to insist that they nonetheless refer to genuinely private objects. The meaning is public; the referent is inner. This is a philosophically respectable position, but it faces the difficulty that Wittgenstein himself presses: what determines that a term with public meaning, when applied by a subject to their own inner states, has successfully latched on to a private object rather than simply been applied in the usual public way? The inner object seems to be doing no work that the outer circumstances of application were not already doing.

What Wittgenstein Leaves Open

I want to be careful here about what I am and am not claiming, because the argument so far might seem to collapse into a denial of consciousness altogether, which is not my intention and not Wittgenstein's conclusion. The private language argument does not show that there are no inner states. It shows that a certain picture of how inner state terms get their meaning is incoherent. This leaves entirely open the question of what inner states are and how they relate to the physical processes that realise them.

What it does show, and what matters enormously for consciousness studies, is that the concepts we use to describe experience are not transparent to their referents in the way that the hard problem presupposes. The terms qualia, phenomenal character, and what it is like do not come with their reference already fixed by inner acquaintance, independently of the public practices in which they are embedded. This means that when we ask whether Mary learns something new upon seeing red, we cannot simply read off the answer from our introspective sense of what knowing what red looks like means. We have to understand the meaning of the expression within its practice, and that understanding may well reveal that the question is not as clear as it seemed.

This is not scepticism about consciousness. It is a demand for conceptual clarity that consciousness studies has been largely reluctant to meet. The field has proceeded as if the target of inquiry were clearly identified by each researcher's inner acquaintance with their own experience, and as if the philosophical task were simply to explain how that clearly-identified phenomenon could arise from or relate to physical processes. Wittgenstein's argument suggests that the identification is not as clear as it seems, and that some of the difficulty in consciousness studies is difficulty that the field has generated for itself by adopting a picture of meaning that does not work.

The Positive Upshot

It would be a mistake to end here, as if Wittgenstein's contribution to consciousness studies were purely destructive. There is a positive upshot, and it points in a direction that is, I think, more promising than the standard approaches.

If the meaning of our experience talk is embedded in public practices rather than secured by private reference, then the nature of experience cannot be fully separated from the forms of life in which experience is expressed, reported, acknowledged, and responded to. Experience is not a purely inner event whose character is given by inner acquaintance and whose relation to the outer world is a secondary question. It is always already shaped by the practices through which it is articulated, and those practices are essentially social in character.

This does not dissolve the hard problem but it relocates it. The question is no longer how a purely physical process gives rise to a purely private phenomenon. It is how the forms of life in which experience is embedded constitute the experience as the experience it is, and what the relationship is between that constitutive embedding and the neural processes that are clearly implicated in experience. This is a harder and more interesting question than the standard formulation, and it has the advantage of not presupposing a model of meaning that Wittgenstein showed to be untenable.

The specific contribution that the private language argument makes is to establish that consciousness studies needs a philosophy of language that it has not had. It cannot simply take for granted that its key terms mean what they seem to mean from the inside. It needs to ask, for each of its central concepts, what practice it belongs to, what makes its application correct or incorrect, and what view of experience is being presupposed by the way the concept works. That is the philosophical work that Wittgenstein's argument demands, and it is work that remains almost entirely undone.

Language Games and the Study of Mind

Wittgenstein's concept of language games is often treated as a sociological observation about how language varies across contexts, but this domesticates what is genuinely radical about it. A language game is not merely a regional variation in usage. It is a complete form of activity in which language and practice are constitutively intertwined, such that separating the language from the practice would leave neither intact. When Wittgenstein says that to imagine a language is to imagine a form of life, he is making a point about the conditions of sense: what a word means is inseparable from what people do with it, and what they do with it is inseparable from the context of action, response, and purpose in which the word is at home.

For consciousness studies, this matters because it means that experience, sensation, and phenomenal consciousness are not neutral descriptive terms that pick out a phenomenon independently of any theoretical commitments. They are terms embedded in specific language games, carrying specific presuppositions about what it means to have an inner life, what makes inner life reportable, and what relation reports of inner life bear to the inner life they purport to report. Unpacking those presuppositions is not a preliminary to the real work of consciousness studies. It is part of the real work, and until it is done, the field will continue to mistake grammatical puzzles for empirical ones and to look for scientific solutions to problems that are, in Wittgenstein's sense, philosophical through and through.

That is what Wittgenstein got right about language that no one talks about. Not that inner life is an illusion, not that the hard problem dissolves on inspection, but that the concepts through which we approach inner life are themselves in need of investigation, and that the investigation is of a kind that neither empirical science nor standard philosophical analysis has yet been equipped to carry out.

The author is Chair of The Philosophical Society: Oxford and a researcher in philosophy of mind. This essay connects to ongoing research on consciousness as a relational phenomenon.