Oxford University Press's
Academic Insights for the Thinking World

Thinking about the mind: an anti-linguistic turn

By Bence Nanay

Contemporary philosophy of mind is an offshoot of philosophy of language. Most formative figures of modern philosophy of mind started out as philosophers of language. This is hardly surprising – almost everyone in that generation started out as a philosopher of language. But this focus on language left its mark on the way we now think about the mind – and this is not necessarily a good thing.

Sentences represent the world. As do some of our mental states: thoughts, beliefs, desires. And we understand pretty well how sentences represent the world. So a tempting way of thinking about the mind is that its building blocks are very much like the building blocks of language: that mental states represent the world the way sentences do.

Sentences express propositions. So, again, it is tempting to think of mental states as representing propositions: as propositional attitudes. And, unsurprisingly, this is the standard way of thinking about the mind: that its basic building blocks are propositional attitudes: beliefs, desires, thoughts. The belief that Paris is the capital of France is an attitude to the proposition that Paris is the capital of France. The general suggestion then is that we can capture the functioning of the mind by appealing to this economy of propositions.

It is easy to see that this way of thinking about the mind is based on mirroring language. But are we justified to do so? Language is an important feature of some select subset of minds (adult human minds), but there are non-linguistic minds: animal minds and infant minds. And they can do amazing things. While no-one denies that they lack language, they are still described by (most) philosophers (and even cognitive scientists) as having beliefs and desires – propositional attitudes.


It would be extremely surprising if the way the mind is shaped had anything to do with language as language is such a late addition to our mental life. A much more natural suggestion is that it has a lot to do with the actions the organism performs. We are evolved creatures and what matters in evolution is really whether one performs actions successfully (and not what one thinks). The mind is shaped in a way that would help us to perform actions. What we should expect then is that the structure of the mind is geared towards facilitating actions and not towards representing propositions. Of course, some select minds can also do that – and, may even use propositional thoughts to perfect one’s performance of actions. But it would be a methodological mistake to start with propositions. We should start with actions.

What would then be those representations that have direct impact on the success of our actions? Representations that attribute properties that are directly relevant for the performance of an action. I call these properties action-properties and the representations that attribute them pragmatic representations. Without pragmatic representations, we would be pretty bad at performing actions. And those of our ancestors whose pragmatic representations failed to get things right had very little chance to survive. Correct pragmatic representations have huge evolutionary benefits and incorrect ones make quick extinction very likely.

But what are these pragmatic representations and why do we not have a widespread label for them, like the ones we have for beliefs and thoughts? Suppose that you are trying to drink a sip of water from the cup in front of you. In order to do so, you need to represent the spatial location of the cup. If you didn’t, you would have no idea which direction to reach out towards. You also need to represent the size of the cup, otherwise you would have no information about what grip size you need to approach it with. And you need to represent its weight, otherwise you would have no idea how much force to exert when lifting it. These are some of the action-properties that your pragmatic representation attributes to the cup. But it happens only very rarely that your pragmatic representation attributes action-properties consciously. Most of our pragmatic representations are unconscious mental states, which of course doesn’t make them any less real. But it would explain why they are missing from the conceptual arsenal we use for describing our minds.

How should we resist the mirroring of language when talking about the mind? We should try to identify mental representations that we have independent, language-free reasons to attribute to agents. If we want to avoid falling back on talking about beliefs, desires and other propositional attitudes, we need some genuine alternatives. Pragmatic representations are mental states of this kind and it would be a good idea to take them seriously not only when talking about animal minds and the minds of small children, but also when talking about the linguistically competent adult human mind.

We, adult humans clearly have propositional attitudes — some of us make a living out of them. Nonetheless, our pragmatic representations still play a much more important role in our mental life: they guide and monitor all our actions (including the ones that have to do with propositional attitudes), they determine the way we see the world and shape the way we interact with others and they may even account for our engagement with fictional narratives. Taking them more seriously would amount to an anti-linguistic turn in philosophy of mind.

This doesn’t mean that we should no longer talk about beliefs and thoughts — these are clearly important constituents of the human mind. So the anti-linguistic turn I am proposing is more like an anti-linguistic half-turn. But linguistically structured representations are late, last minute additions to our mental life — in the same way as humans are last minute additions to our planet. And while humans radically transformed the way the Earth looks, it would be a mistake to try to understand the planet merely focusing on human-made features. Similarly, while language and propositional attitudes radically transformed the way our mind works, even for appreciating just how radical this transformation was, we need to be able to understand the pre-linguistic mind.

Bence Nanay is Professor of Philosophy and BOF Research Professor at the University of Antwerp and Senior Research Associate at Peterhouse, Cambridge University. He received his PhD at University of California, Berkeley in 2006. He is the Editor of Perceiving the World, and author of Between Perception and Action. He published more than seventy articles on philosophy of mind, philosophy of biology, and aesthetics.

Subscribe to the OUPblog via email or RSS.
Subscribe to only philosophy articles on the OUPblog via email or RSS.

Image via iStockphoto.

Recent Comments

  1. David Bzdak

    Great piece. I wonder, though, about carrying over “representation” talk from thought to action. It’s not that we should deny that (often subconscious) representations are used in the performance of actions; rather, we should notice that the basic goal/function of language is representational, while the goal of action is not (usually). Put another way, representations are a means to an end, when it comes to most actions, whereas they’re the end itself when it comes to language. This asymmetry makes me wonder about the symmetrical picture you’re suggesting.

  2. Niall

    What is a representation anyway? And as for mental states, are these identical with brain activity?

  3. Bence Nanay

    Thanks a lot David. You’re right about this asymmetry: representation in language can be an end in itself (I do think that it can also be a means to an end), whereas representation in (for) action is almost by definition a means to the end of performing the action successfully. Do you think this goes against my view? I’m trying to stress another (but very much related) asymmetry, namely, that representation in (action is (evolutionarily, developmentally and explanatorily) more basic than representation in language. So I didn’t mean to suggest a symmetry here.

  4. Ricardo Arribas

    Very interesting proposition, Dr. Nanay. I would nevertheless like to know a bit more about how the concept of the ‘unconscious’ (I mean here the Freudian one, the only one that I’m aware of), would play out in your argument for pragmatic, non-propositional representations, since you yourself seem to implicitly acknowledge its existence in the present article. I would particularly like to know a bit more about how Jacques Lacan’s idea of the “unconscious structured like a language” would fit into your anti-linguistic half-turn. Stated otherwise is there a conception of the ‘unconscious’ that runs counter to, or at any rate distances itself from, or at the very least fundamentally modifies Freud’s (and afterward Lacan’s) take/articulation of it? More simply, are the latest developments in neuroscience and related disciplines fundamentally at odds with the unconscious (structured like a language)? Any light on this will be very much appreciated! Suggested further reading to this effect would be even a bigger help! Thank you!

  5. […] There is a further question lurking in the background here concerning non-linguistic concepts and non-discursive thoughts.  What I’ve sketched here is a theory of language, and it is an open question what kinds of concepts are found in non-language-using animals, and also what role non-linguistic concepts play in our own rational cognition.  It is almost certainly the case that focusing solely on language will not give us a firm grip on what it is for something to count as a mind, and that we should build up to an account of a rational or sapient mind from an account of non-sapient, merely sentient minds.   (For more on this, see “Thinking about the mind: an anti-linguistic turn“.) […]

  6. Bence Nanay

    Hi Ricardo,

    Thanks a lot. The questions you raise are extremely complex. The way I (and those empirical scientists who are working on topics like unconscious perception, unconscious attention, unconscious learning, unconscious social cognition, unconscious influences on decision making and unconscious action) think about the unconscious is very different from the way Freud and Lacan used to think about it. And I see no empirically sound reason why the unconscious would be structured in the same way as language. The general framework in these research paradigms is not to figure out what ‘the unconscious’ is, but rather to show that various mental processes (perception, attention, social cognition, etc) can be conscious and unconscious. Behavioral and Brain Sciences just published a nice article by Newell and Shanks on this, so that may be a good place to start looking for references.

  7. Adam Black

    “What would then be those representations that have direct impact on the success of our actions? Representations that attribute properties that are directly relevant for the performance of an action. I call these properties action-properties and the representations that attribute them pragmatic representations…. But what are these pragmatic representations and why do we not have a widespread label for them, like the ones we have for beliefs and thoughts? ”

    We DO. They are called “Intentions”

  8. Bence Nanay

    Adam, what most people mean by ‘intention’ is a kind of propositional attitude and this is exactly what I’m trying to move away from. Some people make a distinction between two different kinds of intentions (distal vs. proximal; prior intention vs. intention in action; etc) – but even the more proximal/immediate/executive intention is different from what I mean here by pragmatic representations. Pragmatic representations represent the world in a certain way. Intentions prescribe actions. So different ‘direction of fit’.

  9. Italo Giardina

    The image, depicted in this article is credited as an istock photo that ostensibly links to a data base of images. This author/s might well be an ‘negative existential’ (philosophy of language) given the viewer believes the author as part of an ‘imagined community’ of artists whose outputs link as it were to a collective source, but also to an imagined ascriptive creative source. Though the ‘death of the author’ entails the death of beliefs and thoughts given the artwork could be generated by randomly generated code, as a primordial substratum to mind, but not consciousness as such.

Comments are closed.