Theo Todman's Web Page - Notes Pages
Sensations as Perceptions
(Work In Progress: output at 17/08/2017 11:09:35)
(For earlier versions of this Note, see the table at the end)
Are bodily sensations perceptions of one’s body?
- In order to address this question we need to consider two things. Firstly what bodily sensations are and secondly what perceptions are. Then, we can consider the relation between the two. In considering this question I will mainly follow Armstrong’s positive line (see the reference below), given that he supports the view that bodily sensations are perceptions of one’s body, supplying criticisms (including responses to his responses to criticisms) along the way.
- My overall view is that bodily sensations are perceptions of one’s body. They appear subjectively to be this, and I find Armstrong’s analysis to be persuasive. Additionally, this view explains how bodily sensations have come about in an evolutionary sense.
- The essay is particularly indebted to "Armstrong (David) - Bodily Sensations", "Armstrong (David) - A Materialist Theory of the Mind", Chapter 14, as well as to lecture and backup class material3.
- It is also indebted to
→ "Anscombe (G.E.M.) - On Sensations of Position", in narrowing the scope of enquiry, and
→ “The Intentionality of Bodily Sensation4”, "Crane (Tim) - Elements of Mind - An Introduction to the Philosophy of Mind": §24
- So, what are bodily sensations? They include such things as tickles, pains etc, but also sensations of position, movement and balance (or, more grandly, proprioception, kinesthesia and the vestibular sense).
- Armstrong divides what might be considered to be bodily sensations into various categories and sub-categories. He starts off by considering what we mean by the English word “feel”, and this splits roughly into two meanings. The first usage is for touch, which provides information about our body and material environment. The second usage is for what he somewhat prematurely calls bodily perception, where the focus is on one particular material object, one’s own body. The two usages of ‘feel’ don’t just differ in the objects towards which the (alleged) perceptions are directed. The real distinguishing feature is that, in the case of touch, it makes sense to ask what we felt the object with (for instance, the finger, the tongue or various parts of the skin). However, this is not the case for bodily “perception” (eg. feeling the cheeks to be hot or the limbs to be moving).
- Having effectively disposed of the sense of touch, Armstrong distinguishes bodily sensations from bodily feelings, though there are close resemblances between the two categories and intermediate cases. His examples of the bodily sensations-proper are pains, itches and the like, together with sensations of pressure, warmth or movement; while those of the bodily feelings are feeling tired, hungry, fresh, etc. According to Armstrong, bodily feelings involve sensations, but aren’t identical with them. Additionally, bodily sensations are distinguished from bodily feelings by being located in particular parts of the body. Mostly this is clear, but there are fine distinctions and counter-examples; hunger pangs can be located but not feelings of hunger. We do talk of feeling tired in parts of the body, though not, he says, in the precise way that locates sensations. He also says that we talk of some sensations (say of giddiness) without giving them a location, though he seems ambivalent whether to classify giddiness as a feeling or a sensation, and puts all this untidiness down to borderline cases. Location helps individuation – for example of two similar pains.
- Armstrong additionally divides what he calls bodily sensations into the transitive and the intransitive. His reason for separating them is that transitive sensations are easier to deal with and serve as a model for the treatment of intransitive sensations. His distinction is that transitive sensations are such that there is fairly obviously something the sensation is about other than the sensation itself. I can feel that my hand is hot, and the objective hand may or may not be objectively hot – this is a transitive sensation. But a tingle just is a sensation of tingling, and the same goes for pains, itches and tickles, at least before further analysis.
- There can be illusory bodily sensations, in that what seems to be a sensation in a particular part of the body may not be so. This situation can arise either in the case of referred pain, or because that part of the body does not exist - as with phantom limbs, or the extreme case of a brain in a vat. However, the sensation is real; it is just the imputation of its cause that is illusory.
- Is bodily awareness the same as bodily sensation? No. Bodily awareness tends always to involve judgements, whereas bodily sensations do not necessarily. What, though, is included in bodily awareness? It seems to include judgements about the size, shape, extent and speed of our bodies as we navigate through crowds, perform other manoeuvres, or maybe just as we sit. This, I think, is where Anscombe’s paper comes to the rescue. If I understand her correctly, she claims that we have bodily awareness without bodily sensation, or at least denies that our awareness of bodily position and such-like is based on sensory evidence or other cues. So, bodily awareness may involve bodily perception, but does not involve bodily sensation, or at least there may be no necessary connection between the two. We can consequently exclude bodily awareness from further consideration, since our question only asks about bodily sensations, though I admit that space has precluded any discussion of Anscombe’s arguments.
- According to Armstrong, perception involves acquiring states giving the capacity for certain sorts of discriminative behaviour, in this case behaviour towards one’s own body.
- We can distinguish between two sorts of perception, namely between perceptions of objects and perceptions of matters of fact (“that p”). We can also distinguish between qualitative (“what it’s like”) perceptions and quantitative perceptions such as spatial awareness. These distinctions will have relevance when we come to consider the relation between bodily sensations and perceptions of one’s body.
- What is the causal structure of perception? While we can’t give a full answer to the question here, we can take it as read that sense-perceptions depend on the excitation of the organs of sense and the subsequent transmission of signals to the brain, where they are interpreted by the relevant neural structures. The sort of perception involved in bodily sensation is not straightforward sense-perception, but arises in a similar manner – though a more complexly orchestrated one in the case of some bodily feelings.
Bodily Sensations as Bodily Perceptions
- We’re asked in the question to focus on whether bodily sensations are perceptions of the body. However, it is worth reviewing, as a way in to this question, whether our ordinary perceptions are themselves sensations. This seems to be the case for the senses of touch, taste, smell and maybe also hearing, which we can consider as sensations, but we don’t normally speak of vision as a sensation. However, as is illustrated by Armstrong’s rejection of the sense of touch from the class of bodily sensations under consideration, these are not bodily sensations as normally understood, except in pathological cases where noise hurts our ears, light blinds our eyes, someone treads on our fingers, or we encounter acrid smells or vile tastes. Is touch a special case? If I touch my leg, is this a bodily sensation twice (or maybe three times) over? Does the same go for smelling my hand, listening to my stomach gurgle, tasting my own sweat or looking at my own reflection? These would mostly seem to be sensations, and to be perceptions of my body, though not those intended by the questioner.
- Armstrong considers first what is, for him, the simplest case – that of transitive bodily sensations. Taking his favourite example of the hot hand, he claims this to be a straightforward perception of the temperature of the hand. We can analyse transitive sensations as we do other perceptions, as if they are nothing but bodily and tactual perceptions, and as mental events resembling belief acquisition. We become, as a consequence, able to behave in a discriminating way towards our own body and things in contact with it – such as when we determine that parts of our body are at an unusually high temperature and require removal from a source of heat.
- This bears interesting comparison with Locke’s (and Berkeley’s) considerations of primary and secondary qualities. Consider the experiment of dunking the right hand in hot water and the left in cold water, and then, after equilibrium has been reached, dunking both in tepid water. The right hand then feels cold and the left feels hot, though, presumably, the right is still objectively hotter than the left. For the present purposes, we don’t much care whether the secondary quality of being hot is really in the water (or, better, the hand) or in the mind, but the question is, how can bodily sensation, if it is perception of the body, give contradictory perceptions?
- Armstrong’s answer is that all perceptions can be illusory, and that such cases as the one described are no worse than those encountered by perception generally. Indeed he takes this as a virtue of his explanation. According to Armstrong, the locations of bodily sensations are intentional locations, in that the sensations seem to have a particular location. This is, he says, similar to visual impressions where the image looks to be behind the mirror, though we know it isn’t. Treating transitive sensations as intentional offers a simple explanation of transitive sensations in phantom limbs. We can analyse the case of an amputated arm feeling hot as: it feels to me that I have an arm and, on occasion, it feels to me as though the ‘hand’ on this ‘arm’ is unusually hot. These are real bodily perceptions that correspond to nothing in the physical world and of whose hallucinatory nature I’m aware. A phantom limb is a bodily phantom much as an image in a mirror is a visual phantom.
- Personally, despite the merits of the above explanation, I don’t think there’s much of a problem here. What the heat receptors are “for” is to determine temperature gradients, rather than temperatures per se – it’s transfer of heat into and out of the body that’s the problem that requires regulation. Hence, the “hot water” case can be taken as veridical – heat is rapidly transferring out of the hot hand, so it ought to feel cold, and vice versa. Obvious malfunctions of the sensory apparatus can be explained as Armstrong does, or more technically at the physical level in terms of inappropriate nerve firings. Whether we want to count an illusory perception of the body as a perception of the body, or deny this, is a matter of grammatical regulation of no great philosophical interest.
- We saw earlier that Armstrong’s division of bodily sensations into the transitive and intransitive was so that he could pick off the easy transitive cases first. In the case of intransitive sensations, Armstrong notes two difficulties. Firstly, that intransitive sensations generally involve reactions (we either do or don’t like them) whereas perceptions are passive, involving the acquisition of capacities for, but not impulses towards, discriminative action. Secondly, all perceptions involve a distinction between appearance and physical reality. We saw this to be the case with hot hands, and know it to be the case also with the occasionally muddled kinesthetic and vestibular senses. However, doesn’t the intransitivity of sensations of pain defeat our attempts to treat them as perceptions since we cannot distinguish between felt and actual pains? It would seem that a felt pain just is an actual pain, while an un-felt pain is nothing.
- Armstrong’s answer to the first (“reactions”) question is that intransitive sensations are portmanteau-concepts that require analysis, involving both a proper sensational component – the bodily perception – and a reaction of the mind to this sensation. The linkage is causal, with the bodily perception evoking the reaction. One mental event brings about another – with a perception bringing about an affection – and this almost invariable connection accounts for why we normally consider the whole causal sequence as one concept. This bundling of the two concepts, Armstrong claims, fits well with clinical cases whereby chronic pain can sometimes be relieved by severing the connections between the prefrontal lobes and the rest of the brain. Patients allegedly report the curious result that, while the pain is still there, it doesn’t worry them any more – as though they had pain which gave them no pain. The explanation is simply that the perception is there but the reaction has been abolished by the operation. The patient still perceives the same thing to be going on in his body, but this now causes no concern.
- Armstrong’s answer to the second question is that, while there is indeed no distinction between “felt” pain and “real” pain, we can translate without loss of meaning a report of pain into a perceptual statement. For example, a statement such as “I have a pain in my hand” translates as “I have a perception of a disturbance in my hand that evokes in me the peremptory desire that the perception should cease”. This response involves deciding, should bodily sensations be bodily perceptions, of what they are perceptions. In the case of pain, the proposed candidate is the perception of bodily damage, or, as Armstrong puts it, bodily disturbance. Armstrong is correct to point out that it is no objection that the experience itself tells us nothing about the nature of the disturbance, any more than colour-perception tells us what it is about red things that makes them appear red. This analysis also allows Armstrong to account for phantom-limb pain in the same manner as the transitive case of the phantom hot hand.
- Armstrong concludes his chapter on bodily sensations by discussing bodily feelings – feeling tired, fresh, faint, sick, hungry and so on, in which the intentional object is usually the whole body. There are too many cases to discuss here, but broadly these seem to fit into the pattern established for bodily sensations. Some get removed from the realms of sensations or feelings, being more closely allied to propositional attitudes. He thinks that feelings of hunger and thirst are essentially desires to eat and drink, with perceptions of bodily state being relatively incidental, and that they are more like feeling angry than feeling faint. Feeling tired is an intermediate case. Feeling tired in the legs when running seems to involve a, maybe not dominant, desire to stop but it may also indicate a perception of their current state. Feeling fresh, faint or sick do not seem to be desires for anything at all, even though they may evoke such desires. Hence, it is plausible to treat them as bodily perceptions. If so, then being tired (and so on), as distinct from feeling tired, must be an objective physical state of the body, as can be seen where the body or a portion of it comes to be in a state where exertion becomes increasingly difficult to sustain. This is being tired, and similar states constitute being faint or being unwell. Given that, on this analysis, bodily feelings are bodily perceptions, it must be possible to feel tired, and so on, without actually being tired, and vice versa. This is true, as we all know.
- We have no need to be diverted by Armstrong’s discussion on whether the bodily perceptions involved in bodily feelings involve inference or not. His example is whether they are they like hearing a coach or hearing a sound. Either way, they are still perceptions so further discussion is irrelevant to the question at hand.
- If we are concerned that Armstrong’s physicalist account of bodily sensations leaves out much of what it is like to feel them, it is of no relevance in accounting for bodily sensations as perceptions, for the “what it is like” of sense perceptions is also left out of account.
- A problem for considering bodily sensations as perceptions is the privacy of our bodily sensations. It is true that our sense-perceptions are also private, in that only we can see with our eyes and know what our qualitative experience is like, and so on, but the objects of our perceptions tend to be public. However, sensations of one’s own body are often doubly private, in that not only can no-one else feel my tickle as I do, but no-one else can feel the tickle at all. This situation is complicated by the possibility of feeling sensations in another’s body, but that is the topic of another essay.
- Can we account for this privacy by distinguishing internal from external bodily perceptions, with bodily sensations being the internal ones? There are two meanings of “internal”. Firstly, “internal” can mean internal to the mind as distinct from related to a mind-independent world – we can relate to our bodies this way, as being external to our minds and part of the mind-independent world. Secondly, “internal” can mean “within the body”, so that bodily sensations are limited by the boundaries of one’s skin and sense perception relates one to the world outside one’s skin. Usual sensory perception, even if it involves sensation, may not be correctly classified as a bodily sensation.
- What has the view that bodily sensations are perceptions of one’s body got to recommend it, other than the advantages already noted by Armstrong? One such is our evolutionary history and the presumed reason for our sensations. Why did bodily sensations provide an evolutionary advantage to human beings and other animals? The natural response is that sensations are indeed perceptions of our body, and that these perceptions provide useful information to correct or guide action. For instance, sensations of hunger are perceptions that there’s a build-up of digestive juices in the stomach for want of food and that we ought to eat. Similarly, pain as an indicator of bodily damage, whether actual or potential, itches as indicators of irritants on the skin, and so forth.
- Another argument in favour of bodily sensations being bodily perceptions is that motor-control depends on the (maybe unconscious) feedback from sensations. This, however, raises one final problem. Given that some so-called sensations are unconscious, we might ask whether unconscious “sensations” are really sensations at all or just perceptions, much as was the case with bodily awareness? Bodily perceptions tell us about status of the body – its state of health and the various motor-control functions such as, “is it falling?”. However, it is possible that there be pain even if none is felt. We can be awoken by pain – a good case of bodily perception – so maybe there is such a thing as subconscious pain; similarly we can be preoccupied by other matters, and consequently ignore pain, but still be exhausted by it.
Objections to Armstrong
- I must respond to some criticisms of Armstrong given in lecture material. According to Mike Martin, Armstrong thinks that bodily sensations present mind-independent aspects of the body to the perceiver. This, he says, leads to four problems:
- Pain but no damage. We care about how the body part is, not just about the feeling of pain.
- Referred pain – damage is in one place, but felt in another. Martin thinks we don’t treat the hurt of referred pain as if its location is misperceived.
- Pains described as dull, or stabbing, but these are subjective. No obvious objective correlate – a stabbing pain isn’t the pain felt on being stabbed.
- Itches aren’t knowable other than how one feels.
- In response to these points:
- I don’t think Armstrong does say that bodily sensations are mind independent in every sense. He admits that they can be illusory and subject to misinterpretation. They sometimes are mind-independent (real disruption), sometimes not. This is similar to other intentional states; for instance, the person who believes in Father Christmas (or, say, God) and who believes the object of his belief is objective and mind independent; in one case, at least, he’s wrong.
- I’d thought that Armstrong’s treatment of illusions covered (1). Mike Martin’s argument is that, on being told that there’s nothing wrong with the painful part, on Armstrong’s account, while we still might feel pain, we’d lose our concern for the trouble with that part of our body – but (he says) we don’t. I think this is probably empirically false. If told by a suitably garbed and serious medic, I think we do. I’ve just had the standard “pseudo-fishbone down the throat” problem (unflatteringly called globus hystericus). I got referred to an ENT specialist how performed the “camera up the nose and down the throat trick” by way of reassurance. Apparently, the symptoms usually go away on being shown there’s “nothing wrong”, but even if they don’t, one is still convinced that the pain is a misperception. Can’t we defend Armstrong along the lines of “false phobias”? Mike Martin’s argument is presumably that the “misperception” argument breaks down because, on being told I’ve misperceived something, I would then no longer worry about it – but we still do worry about pains (they still hurt, for a start). Well, I’m not sure this is true, but even if it were true, can’t we say the situation is like the ghost on the spiral staircase or the monster under the bed. Even when told there’s nothing there, we’re still scared rigid.
- Armstrong treats (2) referred pain very much along the “pain misperception” lines. A real pain is felt, corresponding to real damage (maybe) but the location of the damage is misperceived. Or is he saying something very different – that the location of the pain is misperceived. The two would be the same if pain is bodily damage, but not necessarily so if it’s a perception of bodily damage or disturbance. Mike Martin’s point is that someone doesn’t think, on being told that the pain he feels in his arm is referred from damage somewhere else, that he’s misperceiving the pain in his arm and that he ought to be feeling a pain somewhere else. Personally, I think he does come round to the misperception idea, but he’s still sure he’s got a pain in his arm, and can’t be argued out of it. To check I’ve got things right, consider the Miller-Light (or whatever it’s called) illusion of the two arrowed lines of the same length. I can be persuaded to believe that they are of the same length, but I cannot be persuaded to perceive them as being of the same length. So, what’s the problem? I can be persuaded that the location of the bodily damage isn’t where it seems to be, but I can’t be persuaded that it feels to be coming from somewhere other than where it feels to be coming from. If pain is a perception of bodily damage at a particular felt location, I can be persuaded that the case of referred pain is a misperception, but I can’t be persuaded to change that perception. Just as I can be persuaded that the oasis isn’t really there just a short crawl away, but can’t be persuaded that it doesn’t look as though it is. I understand, of course, that this illusion is often adduced to show that pains are not beliefs.
- I’m not sure how (3) pain descriptions are supposed to be a problem. Aren’t we just as ignorant about what it is in visual perceptions that gives rise to the subjective experience of yellowness as we are of the particular disruptions, or illusory disruptions, that give rise to the various qualitative experiences of pain? Armstrong uses this argument in the paper.
- I don’t know what Mike Martin means about (4) itches. Does he mean that itches are intrinsically private (there’s no bodily disturbance that anyone else can see) and therefore not the mind-independent entities that Armstrong is alleged to believe in? Can’t a physicalist argue that there must be some disturbance going on at the place we want to scratch, and that this is a veridical case of bodily perception?
- My view is that bodily sensations are perceptions of one’s body, or at least this is the usual situation. They appear subjectively to be this, and I find Armstrong’s analysis to be persuasive. Situations where the sensations are illusory are no worse than the case for non-bodily perceptions, and can be explained as malfunctions in the causal chain linking from the source (maybe would-be) of the perception and the interpretative centres in the brain. Additionally, this view explains how bodily sensations have come about in an evolutionary sense.
In-Page Footnotes:Footnote 3:
Previous Version of this Note:
|Note last updated
||Reading List for this Topic
Summary of Note Links from this Page
To access information, click on one of the links in the table above.
Summary of Note Links to this Page
To access information, click on one of the links in the table above.
Text Colour Conventions
- Black: Printable Text by me; © Theo Todman, 2017
- Blue: Text by me; © Theo Todman, 2017