Remarks on Perception and Other Minds

It is a simple truth about the ways in which we speak about others that we can see or hear or feel what others are thinking or feeling. But it is tempting to think that there is a deeper sense in which we cannot really see or hear or feel these things at all. Rather, what is involved must be a matter of inference or interpretation, for instance. In these remarks, I argue against a variety of ways in which that thought, the thought that we cannot really see or hear or feel what others are thinking or feeling, might be developed. I. Perception and Other Minds It is a simple truth about the ways in which we speak about others (in English and in many other languages) that we can see or hear or even feel what others are thinking or feeling. But it is tempting to think that these ways of speaking must be mistaken in this respect. It is tempting, that is, to think that there is a deeper sense in which we cannot really see or hear or feel these things at all, that what we can observe or perceive in this respect is limited to the movements of others’ bodies and the sounds that they make for instance. What we ordinarily refer to as cases of seeing or hearing or feeling what others are thinking or feeling are really a matter of something else, of inference or interpretation for instance, rather than of seeing or hearing or feeling, strictly speaking.


I. Perception and Other Minds
It is a simple truth about the ways in which we speak about others (in English and in many other languages) that we can see or hear or even feel what others are thinking or feeling.But it is tempting to think that these ways of speaking must be mistaken in this respect.It is tempting, that is, to think that there is a deeper sense in which we cannot really see or hear or feel these things at all, that what we can observe or perceive in this respect is limited to the movements of others' bodies and the sounds that they make for instance.What we ordinarily refer to as cases of seeing or hearing or feeling what others are thinking or feeling are really a matter of something else, of inference or interpretation for instance, rather than of seeing or hearing or feeling, strictly speaking.
The idea that we cannot observe the mental states of others is one important premise in the traditional problem of other minds, and as such it is both widely held and deeply entrenched, so much so in fact that it is often not explicitly argued for at all and sometimes not even explicitly stated in discussions of that problem. 1 Bertrand Russell, for instance, takes it to be so obvious that we cannot observe others' mental states that for him it (literally) goes without saying: instead, he proceeds straight from noting that we observe such occurrences as "remembering, reasoning, feeling pleasure, and feeling pain" in ourselves to wondering what postulate could be involved in ascribing such states to others (Russell 1948: 482-483).By contrast, Alvin Goldman, for instance, is helpfully explicit, asking how we go about forming beliefs about others' mental states, states that he claims "aren't directly observable", but he too does not offer any reasons in support of that claim (Goldman 2012: 402). 2 My aim in the following remarks is to reject the idea that we cannot see or hear or feel what others are thinking or feeling by exposing what I take to be the confusions involved in a variety of different reasons we might have for wanting to say that and, in doing so, to reject also the scepticism about other minds that is premised upon that idea. 3My aim, however, is not straightforwardly to defend instead the thought that we really can see or hear or feel these things.Part of the point of my argument is that the idea of perception invoked in rejecting the idea that we can perceive others' thoughts 1 The traditional problem of other minds, as I refer to it here, is the problem of justifying our belief in the existence of other minds, but the idea that we cannot observe others' mental states is of course also often a premise in the further problem of justifying our belief that others' mental states resemble our own.
2 Further examples of this claim from the recent literature are given in Overgaard (2017:  743-745) and Varga (2017: 787-788).The idea that we can, in one way or another, observe or perceive others' mental states is defended in (for instance) Austin (1946), Cassam (2007), Dretske (1973), McDowell (2001) and Overgaard (2017). 3Rejecting one of the central premises upon which that scepticism is based is one way of rejecting that scepticism, but I nevertheless do not think that knowledge of the existence of other minds is a straightforward consequence of our seeing (e.g.) the anger in another's face, or our seeing that they are angry.(For the contrary claim, see Cassam 2007: 62.)That suggestion, I think, mistakes the nature of our belief in other minds, and in doing so threatens to trivialize it.I argue for an alternative conception of belief in other minds in terms of what Wittgenstein calls "an attitude towards a soul" (Wittgenstein 1958: Part II,  178) in Dain (forthcoming).and feelings is confused: it involves a kind of myth of perception that should itself be rejected.So I do not want to argue that we really can observe others' mental states in the same sense in which it is claimed that we cannot.I do not think that there is a clear sense of what it is to perceive something involved in rejecting the idea that we can perceive others' thoughts and feelings.With that qualification, then, I am going to argue that we can (and do) in many cases perceive others' thoughts and feelings.In doing so, I shall be focusing firstly on the case of seeing as opposed to other sensory modalities, partly for convenience and partly because the case of seeing brings out some of the issues especially clearly, 4 and secondly on the case of seeing (as in 'seeing the anger in so-and-so's face') as opposed to seeing-that (as in 'seeing that so-and-so is angry'), since much of the resistance to the latter can be explained in terms of resistance to the former. 5My aim, however, ultimately, is to defend the possibility of perception of others' mental states generally (and not only in the case of seeing), and both perception of what they are feeling and that they are feeling whatever it is they are feeling (that is, both perceiving and perceiving-that). 4There are substantive differences between the senses in relation to our perception of others' mental states in some respects: for instance, it is, I think, true to say that we can see (in (e.g.) someone's eyes, face, or movements) a greater range of feelings than we can hear in their voice, and that we can also hear a greater range than we can feel.Some of the different things we might mean by perceiving apply more readily in the case of seeing than in relation to other senses in a way that in some respects assists the thought that we cannot see others' thoughts or feelings: for instance, whereas there is a clear sense of seeing in which one sees whatever passes before one's (open) eyes, whether one notices it or not, I think we are more unwilling to admit of such a sense in relation to hearing, and even more so in relation to feeling.Those differences might be thought to assist the argument that we cannot really see someone's thoughts or feelings insofar as that sense of seeing might be thought to be in some sense basic, and so definitive of what seeing really is in some sense, in contrast with those senses of seeing that involve some further interpretation of what is seen in that first sense by the mind.The focus on seeing then might be thought in some respects to be prejudicial in favour of the kind of view that I am here rejecting.In defending the possibility of perceiving others' mental states, and our ordinary forms of expression in this respect, however, I do not want to defend every case in which we speak of perceiving others' mental states: for instance, though people do sometimes talk of "smelling fear" (for instance, in a packed exam hall), I do not find it tempting to take such talk literally. 5See Dretske (1973) for further discussion of this distinction, and of the relation between seeing and seeing-that.Dretske argues that we can see that someone is, e.g., angry, but he nevertheless rejects the idea that we can see their anger.
Since my target is not a single view, but a number of different, often quite simple thoughts that might lead us to think that what is involved in seeing or hearing or feeling what someone else is thinking or feeling must in some respect be fundamentally different from certain paradigm cases of seeing, such as seeing the colour or the shape of an object for instance, I have found it useful to adopt a style at least superficially like that of Wittgenstein's later philosophy, responding to particular trains of thought across a series of short, numbered remarks consisting sometimes of just a sentence or two, or of one or two short paragraphs.Some of my remarks also make use of what I at least take to be grammatical remarks, of reminders of the uses of words, rather than substantive philosophical claims, and I have in various places drawn on points that are familiar from Wittgenstein's work too. 6The result, however, is not a polished piece of philosophical therapy such as one finds in the Philosophical Investigations, and it has more of the character of a dialogue than anything one finds in Wittgenstein.I have used double quotation marks to enclose an idea that I try to respond to within a remark, and I have used single quotation marks as scare-quotes.
I have found this style of philosophical writing to be extremely helpful in addressing the various reasons one might have for wanting to say that we cannot see or hear or feel what others are thinking or feeling.But there are of course also drawbacks to this style of writing, not least in terms of the effort it requires of the reader and the potential both for misunderstanding (on the part of the reader) and obfuscation (on the part of the author).Perhaps those drawbacks are enough to explain why those who share Wittgenstein's diagnosis of the nature of philosophical problems have not taken on more of his philosophical style in dealing with such problems themselves.But given the intimate connection many of those same people see between that diagnosis and the manner of Wittgenstein's writing, the fact that others have not adopted more of his philosophical style might all the same seem surprising.
The remarks are written as one continuous train of thought, beginning with the basic objection that mental states are not the kind of thing that can be seen at all ( § §1-4),7 before turning first to inference ( §5), and then interpretation more generally ( § §6-15), and ending with some more general remarks on the relation between feelings and their expression ( § §16-18) and on the possibility of pretence ( § §19-20).I argue that the idea that mental states are not the kind of thing that can be seen misidentifies the object of perception in this case, that the idea of an internalized inference would not be an inference at all, and that if we want to exclude all cases of interpretation from qualifying as cases of genuine seeing, then we are left with very little that would count as seeing at all by this criterion.In doing so, I undermine several of the main reasons we might have for rejecting the idea that we can perceive other minds in these ways.

II. Remarks on Perception and Other Minds
1.I can see so-and-so, see the colour of their hair, the expression on their face, the clothes they are wearing.Can I also see the joy in their smile, or the happiness in their eyes?-"Youcannot see what someone is thinking or feeling; you can only see the movements of their body, and hear the sounds they make, for instance."-Inwhat sense can another's thoughts and feelings not be observed?I can, for instance: see the delight in someone's smile; see the recognition in their eyes; see the concentration (or the pain, fear, grief, despair, hurt, joy, happiness, care, love, amusement, attentiveness, etc.) in their face; see the purpose or deliberation (or the lack of it) in their actions, or the hesitancy or uncertainty in their movements; hear the happiness in their laughter; hear the joy (or sadness, agony, confusion, love, anger, impatience, contempt, concern, regret, grief, despair, doubt, insincerity, disbelief, spite, treachery, etc.) in their voice; hear the fear or the exhaustion in their breathing; feel the fearfulness in the trembling of their body; feel the concern or the affection in their touch; feel the determination in their grip.
2. "You cannot really see, or hear, or feel these things.Mental states are not the kind of thing that can be observed."-Butif it is the ordinary uses of these expressions that we are talking about, then we can correctly be said to see and hear and feel these things on certain occasions.So in what sense can one not?In what sense are these things not observable?
3. "Only what can be taken in by the eyes is really a case of seeing."-Don'tI see the anger in their face or the joy in their smile with my eyes then (and don't I hear the sadness in their voice with my ears)? 4. "But the anger itself cannot be seen."-Ican see you, and see your anger rising or subsiding, for instance.-"Theanger itself is not what you see.Anger is a feeling, and you cannot see a feeling at all.Anger itself is not something that can be seen."-Thethought here is not that there is some practical obstacle that prevents us from seeing feelings, but that there can be no such thing as 'seeing a feeling' at all.But then it is not clear that that is true: we do, after all, talk of seeing feelings in some contexts.So what we need is rather the thought that there is no such thing as 'seeing a feeling' in a certain sense.But then our problem is that if this is true, then there will be no way of specifying what that certain sense is, and so no way of specifying what it is that we cannot do here.
There is, all the same, I think, both something right and something wrong about this thought.We could put what is right about it in this way: feelings, just as they are, are not even potential objects of sight.(I do not, for instance, see my own anger in this sense.Either I am angry or I am not; there is no room for seeing in this case at all, unless what is meant is that I catch sight of my reflection, or see my actions mirrored in another's, and suddenly see my own behaviour in a new light as if through the eyes of another, for instance.)But that only tells us that the object of sight in this case is different, is not what we might be inclined to think that it is: it is not a bare feeling, but rather a person who feels something.
Insofar as it makes sense to say that one sees a feeling at all, one can see the anger in another's face, or in another's behaviour: this is what it is to 'see a feeling'.(We do not for instance typically infer, from someone's cry of pain, the existence of pain; rather, we hear the pain in their cry.) "But that contrast is not between an indirect inference and a direct perception, but between an inference that made explicitly on the basis of someone's behaviour, and one that is implicit, one that has been internalized, for instance."-Animplicit or internalized inference here is the idea of an inference that has become so natural, so immediate, as to be automatic, unthinking.One simply does it.One simply concludes.As a result, it does not seem to us as though we are making an inference at all.
Is this still an inference?We can compare it with what we might call an internalized calculation.If someone asks me to add five and five, I do not calculate anything: I simply answer without hesitation.The process of calculation has been replaced by the automatic answer.The calculation has not been carried out implicitly somehow, unless that means simply that I no longer calculate at all, but just give the answer.An internalized calculation, in this sense, is no longer a matter of calculation at all.
In such a case, there is no longer any room for calculation, and it is not clear what it would even mean in this kind of case to say that I calculate, knowing the answer as well as I do.I could, for instance, take two groups of five objects and, beginning from five, say the next number in the series of cardinal numbers as I move each object from one group to the other.I may in this way show someone, a child, how to calculate the answer.But if I were to try to calculate the answer for myself in this way the process would be a sham.I already know what the answer is, and there is no question of my discovering that my initial answer was wrong: there is, for instance, no hypothesis for the process to confirm, or doubt for it to resolve.
An internalized inference in this sense would, then, be no inference at all.But no inference, internal or otherwise, is necessary even for an infant to hear the anger in an adult's voice even if they have not yet learned to call it 'anger'.(Just as a dog does not need a course in elementary logic in order to hear the anger in its owner's voice.)An inference only seems to be necessary here because we are suspicious of the idea that we can see or hear or feel these things directly.So the question is: why should one be suspicious of these ways of speaking here?6. "What someone thinks or feels simply is not there to be seen in the way that the features of their face, or the colour of their lips, are there to be seen."-Butagain, if it is the ordinary use of these expressions that we are talking about, then the joy in their smile very often is simply there to be seen, no less than the colour of their lips or the lines around their eyes.What they are thinking or feeling may be 'written all over their face'.So in what sense is it not 'there to be seen'?What are the differences supposed to be here?I interpret what I see when, for instance, I am unsure of its significance.If I am unsure of the significance of your smile, of the reason for it, I may interpret it as joyful, or embarrassed, or whatever.And if I am unsure of the reason for your joy, I may interpret it as a sign of relief, for instance.I may see your smile, and wonder if it is really joyful.And I may see your joyful smile, and wonder what makes you so happy.Interpretation may be involved in either case, but so too it might not be involved in either case as well.
8. "But you must interpret it in some sense, otherwise how could you know that it is joy (or anger or grief…) that you are seeing."-Comparethis: 'but you must interpret it in some sense, otherwise how could you know that it is red that you are seeing?'So do I also not see red?
Look at a ripe red tomato: you no more interpret what you see as being red than you interpret their joyful smile as joyful.But redness is also a concept, and is not simply given to us by perception.9. "You must interpret it or conceptualize it in some way to know that it is red, but what you see-the precise colour or shade-is not interpreted.(You can see the red without knowing that it is red, but you cannot see the grief without knowing that it is grief you are seeing.)The colour itself is just there to be seen."-It is possible to see someone behaving in all kinds of ways without recognizing what their behaviour expresses.(I may see the despair in their face, for instance, without knowing that they are grieving.)The sense in which you "cannot see the grief" without knowing that it is grief that you are seeing is just the sense in which what you see is what is taken in by the eyes alone: it is what is received by the senses, uninterpreted or unmediated by the mind in terms of concepts.10.How much of what we see, or of what we say that we see, is uninterpreted-and therefore seen-in this sense?For instance, look at a tree in the autumn with the light and shadow playing over a thousand leaves of different shades or red, yellow, and green.Do you see the tree?Not in this sense.What you see in this sense is just the ever-changing array of different patches of colours.But in this sense we do not really see anything except the play of light on our eyes.We do not, for instance, see the behaviour of a person in this sense either, for we do not really see a person at all; we simply see the changing patterns of colours passing before our eyes, which we then interpret as the movement of a body.

"
But what we receive in perception, the raw data as it were, is already a matter of colours, shapes, objects, for instance.Our concepts of colour, etc., do not add anything, any new substance, that is not already present there; they merely organize or interpret the raw data that we receive, unmediated, from our environment."-Butif it is true that these things are already present there, isn't it true for smiles of joy and sadness too?After all, a smile of joy or of sadness is there to be seen too, whether one notices it or not.The raw data is all there for us to see in this kind of case too.
"But the raw data of seeing is not a matter of smiles of joy or of sadness, for instance.The raw data is a matter of patches of colour, of light that strikes the retina, and smiles of joy or sadness or whatever cannot be reduced to that, to differences in the patches of light upon the retina."-Inone sense of the word 'seeing', and in one sense of the words 'raw data', this is the raw data of seeing.But this is not the only sense of those words, and it is not necessarily the most important or the most natural sense of them either.We might, for instance, just as well say that the raw data of seeing is a matter of objects, of bodies, of people, plants, animals, etc., since that is, after all, primarily what we see.The experience of seeing patches of colour that stand in need of interpretation in order to be understood, that we do not immediately see for what they are, is not by any means a typical experience.In the typical case, what we see is, as it were, already interpreted: it can be broken down into patches of colours; it is not built up out of them.Smiles (smiles of joy or of embarrassment for instance) are lost from view in breaking down what we see into patches of colour; they are not added in the process of building up.
12. "What you see is the changing arrangement of their features, the shape of their eyes, the curve of their lips, the lines on their forehead, for instance."-Ifyou ask me to describe how someone looks on a specific occasion, this is not typically what I will describe, and if you ask me nevertheless to describe this, or to describe what I see in these terms, I may be able to give only the most rudimentary account of my perceptions.8(Likewise, if you ask me to draw the precise arrangement of the features of their face and the various shades and colours of them, I may well be at a loss to do so.)Emotion, rather than geometry, one could say, forms the raw data of perception in this sense: what we perceive is typically already conceptualized in these terms."But what you see in this case must in some sense be the same before and after."-Nothing in the conscious experience is the same, and so seeing in this sense becomes something prior to our conscious experience.

"
But what you see must be the same in these two cases even if you do not notice the same things in each case.How you perceive it is different, but what you perceive must nevertheless at some level be the same."-Atsome level, perhaps: although in the one case, the eyes passively take in what is in front of them, and in the other, they focus on different aspects of what can be seen, thereby changing the data even at this basic level.But what is seen in this sense is a very limited part of all that we say can be seen.
15. "As well as noticing what we see or recognizing what it is that we can see, we can also simply see things without being aware of it at all, as for instance when we drive a car on 'auto-pilot' without being aware of what we are doing, and what we see in this sense, the data available to such seeing, is all that can really be seen, all that is open to sight alone as it were.Everything else requires the operation of the mind in interpretation of or inference upon what is seen."-Weneed special reasons for saying that someone did see, or must have seen, what they were nevertheless not aware of having seen at all, such as, in the case of driving on auto-pilot, 9 their having kept the car on the road for several miles in spite of their not being aware of having seen anything at the time.But if we only say that they must have seen x because, in spite of being unaware of doing so, they nevertheless responded appropriately in the situation, because (e.g.) they kept the car on the road, then we will have to say that even this kind of seeing must involve interpretation, and so that this form of seeing is also not really basic, or simple, in the way that we wanted, since without interpretation why should we have responded one way to the stimuli rather than any other, or rather than not responding at all?This kind of case will not exclude the kind of cases we might want to exclude in appealing to it, since insofar as those other cases, seeing the anger in a face for instance, must involve interpretation, so must this kind of case too.So even here we do not find the kind of pure perception in comparison with which our seeing or hearing or feeling what another is thinking or feeling could be shown to be not really a form of seeing or hearing or feeling at all.16. "What can be seen is merely the expression of the anger; it is not the anger itself.You must assume, or infer, the presence of the feeling behind the behaviour that you observe."-Butanger is not a single thing, as this implies: it is a complex, including not only various sensations that may be characteristic of feeling angry or of being angry, but also various states or dispositions, as well as actual behaviour too.(In some cases, for instance, how someone behaves may function as a criterion for their being angry: they insist that they are not angry, and it is their behaviour that contradicts them.)To insist that you cannot see the anger, that you must infer the presence of the feeling behind what can be seen as the cause of the visible behaviour is to isolate one aspect, a single, uniform feeling, and to treat that aspect as, always, the essential thing.17.Moreover, the expression of anger may be more or less closely bound up with the feeling of anger in different cases: the expression of anger may be its manifestation, the visible face of the feeling or its audible aspect for instance, and not simply an indicator from which the presence of anger might be inferred. 1018.That you can in some cases hide your feelings or conceal what you are thinking presupposes the possibility of your not hiding them but letting them show, or of your trying to hide them and failing, for instance.Even in ordinary cases, it is not always so easy to conceal one's feelings, and then there are cases where we do not try to conceal them, and cases where there is no such thing as concealing them at all: cases of overwhelming rage or inconsolable grief, for instance.("The anger itself must remain hidden."Just try hiding it, then, on some specific occasion.)19."But you cannot see the pain in someone's face, for instance, unless you can distinguish between their being in pain and their merely pretending to be in pain, say, by means of what you see alone.But those two cases might appear exactly the same as far as what you can see goes.They may, in qualitative terms, be identical."-Thisidea would rule out much of what we ordinarily say that we see along with feelings.But we do in fact often in the course of our normal interactions with other people simply see whether or not someone is merely pretending to be in pain, whether it is real or feigned, and in many cases too the possibility that they might be pretending does not even arise such that it would need to be ruled out, even though we cannot always rule it out, even though we are sometimes uncertain or mistaken about what we see.If there were not forms of behaviour or expressions characteristic of certain feelings, we could not even pretend to have them when we do not.20."But this simply ignores the possibility that we might be mistaken about everything, that all of our perceptions may have a source quite different from that which we assume."-Ifthis is in fact a possibility, then it is not only what someone else is thinking or feeling that we cannot really see.If it is a possibility at all, then it is not simply one more contingent possibility along the lines of the possibility that they may be pretending, for instance.It puts the entire framework into doubt.Within that framework, however, we can very often tell what it is we see. 11 7. "To see the joy in their smile, you must also interpret what you see."-I may interpret what I see as joyful, but I also may see it without interpretation.I may interpret your behaviour as a sign of discomfort, or grief, or alarm, or whatever.But I may also see that you are grieving, for instance, without interpreting anything.
13. "What is seen must be what is common both to the person who interprets what they see and so understands or recognizes it for what it is and to the person who sees without understanding."-Whatdo they in fact have in common?Their perceptions may be totally different.Imagine this: you come round from a period of unconsciousness, blurs of various colours swim indistinctly before your eyes.Suddenly, everything takes shape: you are on the floor looking up, with white coats, and faces of various shades above you.The sounds in your ears transform into intelligible speech: people are talking to you, asking you questions (what your name is, whether you can hear them, for instance).Here, almost nothing is the same from one moment to the next.(Similarly, what I hear when I hear the words of a language I do not understand is not what I hear when I hear the same words later, with understanding, knowing what they mean.My experience is totally different.In one case, for instance, I may be able to parse the sounds into distinct units, or to repeat what I heard, and in the other not at all.) 5. "You cannot see someone's anger, for instance, directly.You must infer the presence of anger from what you can see."-Wedo infer what others are thinking or feeling in many cases.But those cases stand in contrast to those cases where we simply see, or hear, or feel what another person is thinking or feeling without inferring anything.