Saying what one thinks

01/02/2025

Sometimes one only comes to find out what one thinks by putting the thought into words. In her forthcoming book with OUP, Saying What One Thinks, Dr Léa Salje calls this ‘the phenomenon of articulatory self-knowledge’. In this post she sets out the phenomenon and says something about how we should understand it.

Picture this. You’re walking down the produce aisle in your local supermarket. Much of your attention is taken up with the task of choosing fruit while you lazily listen to the tunes piping through your headphones. Still further back in the shadowlands of your headspace you’re idly shuffling through all sorts of ideas and impressions without any special reason or goal. A snapshot of a hallway chat with a friend this morning, a half-formed thought about the changing season, a faint pang of something that is not-quite-sensation and not-quite-emotion, an earworm of an advertising slogan, and on it goes. None of this is very well formed, or – one gets the sense – very important; it’s just the ordinary background chattering of your mind as you go about your day.

Then, suddenly, you are struck with a thought. It’s a thought that makes you feel immediately sad, but for the life of you, you can’t say what it is. It gnaws at you, tipping your mood into an unfocussed melancholy as you make your way around the supermarket. The music in your ears is irritating now as you strain to bring the thought into focus. What was it? It’s something about your friend from this morning and her work, but why would that make you feel sad? She had been joking about it, about the strains of the job, but that’s what you always do when you get together. This triggers a recollection of the many times you’ve had this conversation – a sort of collection of mental filecards of all the conversations the two of you have had around work. Is it that you are getting bored of having this conversation with her? No, that’s not it – but it’s close.  

You turn into the cereal aisle and wilfully force your attention outwards to the display of boxes. Just as you’re considering the options an uninvited image floats into view – it wasn’t what your friend said, it was a look she gave you as you left her, an unguarded look that lingered several moments too long. Now the thought snaps into view: She is actually deeply unhappy in her job. But she won’t do anything about it, because she can’t see that what she’s describing is much worse than a jovial gripe. That’s it! As this newly articulated thought runs through your mind, you feel terribly sad for your friend, but also a sense of relief at having managed to verbalise the thought that had been dogging you around the aisles. 

It would, of course, be surprising if this particular episode had ever really happened to you! But for many of us, episodes of this kind are an ordinary part of our mental lives. They characteristically begin with a flash of mental content that feels ‘just outside one’s reach’ – one knows it is there, but can’t say what it is. This is followed by a period of mental effort directed towards the thought’s recovery that takes the form of a kind of verbal trial-and-error. Such efforts are not always rewarded with success. When they are, the episode culminates in an ‘aha!’ moment as one is finally able to put the thought into words. If this has ever happened to you then you’re in good company – among others, Einstein and Poincaré both reported going through thinking processes of just this kind in their intellectual work. 

These episodes might be familiar, but they are also baffling. Once the thought has popped into our minds, what could prevent us from saying what it is? Why the need for such a circuitous trial-and-error approach – like a game of self-directed battleships in which the thought flames into conscious view only if one manages to guess at the right sentence?

A couple of kneejerk responses fail to satisfy. First: it might be explained as nothing more than a neural blip, a problem of wetware. This reaction fails to do justice to the rich and extended processual phenomenology of episodes of this kind, or to the fact that the causal route to the thought’s recovery seems to be driven by personal-level mental action. Second: it might be an issue of missing vocabulary – I need the right words if I am properly to express my thought. But this suggestion too falls flat. In the above example, none of the vocabulary ultimately used in expressing my thought was especially exotic or hard to retrieve.

My own proposal is that in cases of this kind the thought that pops into one’s head is not yet apt for expression in a natural language sentence. It is a mere jumble of associated contents. The reason I can’t yet say what the thought is, is not a problem of vocabulary or misfiring synapses – it’s a problem of thought-format. It is not yet a thought of the right form for verbal expression. I must work at the thought, and work on it linguistically, in order to shape it into the right sort of content to put into words. That’s why (or so I argue in my book) we sometimes have to put our thoughts into words before we can know what they are.

Dr Léa Salje is an Associate Professor at the University of Leeds’ School of Philosophy, Religion and History of Science, and Director for the Centre for Theoretical Philosophy. She has been at Leeds since 2016, after completing a PhD at UCL. She works in philosophy of mind, especially in issues around self-knowledge and first person thought.