Theo Todman's Web Page - Notes Pages
Anscombe - On Sensations of Position
(Text as at 31/08/2017 19:35:02)
(For earlier versions of this Note, see the table at the end)
This note provides my detailed review of "Anscombe (G.E.M.) - On Sensations of Position".
Summary & Aims
- The thrust of the paper is to defend Anscombe’s view that there are various bodily positions, such as sitting cross-legged, that we “just know” about and don’t deduce from sensations or feelings any more than we might from visual clues. We use the term “sensation” in such cases as both an external description of what is the case, and as an internal description of what it feels like. The sensation is not broken down into other more primitive data, which we may not even be aware of, though if we were to attend to we might come to know.
- “Mr. Braybrooke”, while verbally accepting her rejection of something1 with reference to knowledge without observation, seems to have maintained what she rejected by referring to knowledge without clues. Anscombe admits to denying what Braybrooke considers a plain fact of common experience; that we judge that our legs are crossed from clues provided by resistance, weight and pressure. She asks what sensations would be the clues if our legs were crossed differently – not in contact, but with just the calves crossed. She thinks most people wouldn’t know what sensations to expect, but could perform an experiment to find out by putting their legs in that position. She thinks this demonstrates that such sensations are not clues whereby one determines one’s position.
- If we did use the clues Braybrooke suggests in the normal leg-crossing case, we’d have to check that the sensations of pressure, weight and resistance were produced by one leg on the other and not by quite different bodies. We could not attain this assurance from knowledge of the position of our legs because, ex hypothesi, we’ve supposed we don’t know where they are other than from Braybrooke’s “clues”.
- Anscombe had said that it may be that we know our leg is bent because we have sensations. Braybrooke had asked in what sense of because. Anscombe clarifies the situation – she is referring to the case of an anaesthetised leg where it could always be the case that we couldn’t tell its position without looking or feeling with the hand (and, consequently, that the reason we have to adopt this unusual procedure is because the usual sourse of our knowledge – sensation – is absent). She asks us to compare the example of the anaesthetic boy cited by William James2.
- Braybrooke had allegedly been misled by Anscombe’s use of the term sensation one has with respect to the reflex action on having the knee tapped. She thinks the use of sensation in that case is valid, but doesn’t consider it a counter-example to her claim that we don’t know bodily position by observation; the reason being that the sensation is “not separable”. Elsewhere she had stated that a sensation needed to be separately describable if one observed a fact by means of sensation. Braybrooke had assumed she had introduced a class of clues consisting in non-separable sensations by means of which one tells certain things about oneself. Such knowledge, though obtained by clues, would be without observation because it is logically necessary that a clue by which a fact is observed be sharable. She denies holding the view, whatever its merits.
- What she means by saying that the sensation of a reflex kick is not separable is that the description of the sensation-content, the internal description3, is the same as description of the fact that is known. In such a situation, she denies that we can observe that fact by means of the alleged sensation.
- When saying “sensation of X” we need to ask whether “of X” is a description of the sensation-content or whether X is what produces the sensation but not the sensation itself. For instance, if “of X” is “of going down in a lift”, then this is not an internal description of the sensation, which is of one’s stomach lurching upwards, but of the (possible) cause of the sensation. It is possible not to have any very specific internal description of a sensation. English has no word for the sensation-content of “the smell of onions”, which is an external description. If onions were to lose that smell, but we could still smell it in other contexts, we might use “of onions” as an internal description much as we use the term bitter. The locution “a sensation of flying” is an internal description because it is not the sensation occasioned by actual flying; “flying” is used as a kind of metaphor. The fact that this metaphor comes to mind is part of the experience it expresses. Anscombe adds that “sensations” of position are quite unlike this.
- Anscombe points out that the idea that it is by sensation that we determine our body position arises because people think that the “sensation4” of the position itself is not enough and that we need the feelings of various pressures and tinglings to tell us what position we’re in. The reason such feelings are evidence for a particular bodily configuration must be, she supposes, because they have been found to correlate with that position in the past. She thinks Bradbrooke’s example of crossed legs has plausibility because
However, if there is no contact, whatever sensations one has will not even suggest a position. That is, unless the “of” phrase in the description of the sensation is an internal description5.
- It is natural to think of sensations arising when sensitive parts of the body are in contact and
- The contacts in question will naturally arise and would imply that one’s legs were crossed.
- Anscombe now says that the sensation of mutual contact of the relevant parts of the body is of the very kind that misled Braybrooke as to her meaning. It is itself a sensation of position, and deserves to be scare-quoted as “sensation”. Anscombe uses this orthography to call in question the sense in which such “sensations” are sensations are all. She draws our attention to three possible states of affairs; “to give a reflex kick”, “to have one’s legs crossed” and “to have no sensation in one’s cheek”. We can freely prefix to each of these “you know what it feels like ..”; or we might speak of “the sensation of ..” each of them as being synonymous with “what it feels like”. But, she says, this only means what “the thing is familiar to you” would mean in the same context. When asked if I know the sensation of giving a reflex kick, if I imagine anything it’s a kick not a sensation. If I want to imagine the sensation, meaning something other than giving the kick, I may find that I’ve forgotten or never noticed what or where the feeling is6. On investigation, she says, I find, when I give myself a reflex kick and attend to my sensations, a sensation like an electric shock starting behind the knee and running down the leg. I didn’t know this before, though I’d still have said I was familiar with the sensation of giving a reflex kick.
- Not every familiar experience deserves the epithet “sensation of”; for instance, using the expressions “the sensation of being told a fairy tale” or “… of building a summer-house” sounds extravagant or absurd. The reason for the absurdity is that scare-quoted “sensation” should be used for “unitary occurrences in one’s restrictedly bodily history”, which these aren’t, presumably since they take a long time. She thinks a strong reason for calling scare-quoted “sensations” sensations is that it can happen on extremely rare occasions that one wrongly thinks one has given a reflex kick or has one’s leg stretched out. One would then use the expression “it felt just as if …” or “I had the sensation of …” which leads one to think that there is a sensation which is a datum of sensing position and can occasionally occur without the position. The example she gives of an analogous situation that this might be like is “the visual impression of a blue expanse when judging that the sky is blue”. However, she thinks that “I am sitting cross-legged” and “I gave a reflex kick” are not couched in the language of sense-impressions or, properly speaking, to be descriptions of sense-contents in the way that “blue patch”, “tingle”, “pain” etc. are. Consequently, it is supposed that the sensations of giving a reflex kick and so on must be describable in other more general terms not specifically tied to such actions and which will give content to “giving a reflex kick”, etc.
- Taking up another example, there are various equivalent locutions that can be prefixed to “… my leg was bent”; namely “I thought that ...”, “To me it was as if …”, “I would have said that …” and “I had the feeling that …”. However, Anscombe claims that, had my leg been bent, there would most likely have been nothing beyond that fact and my knowledge of it; I would have had the capacity to describe my position straight off without the need of any appearance or sensation to give me the position. The only difference between the two situations – the one in which my leg appeared bent but wasn’t, and the one in which it was bent - may only be as follows; that in the first case my leg is bent and I know it, but in the other case it is not bent but I spontaneously believe that it is. She adds “I may not even think about it at all”. Anscombe claims that “I believed” does not imply that I thought of it or had the idea before my mind even if we take “I believed” to be equivalent to “I had the sensation”. If one now looks for the sensation in question, one may find particular sensations that one hadn’t before known one had, but which one now takes as data under the misapprehension that that is how things must be.
In-Page Footnotes:Footnote 1: Footnote 2:
Footnote 3: Ie. what it feels like, rather than the cause, which is the external description – see next bullet.
- This appears on p. 769 of my one-volume edition ("James (William) - The Principles of Psychology"). The boy’s only sources of feeling were his right eye and left ear (ie., presumably, that he could feel if these items were prodded, not that he used them to determine what he was “feeling”). He couldn’t tell that his limbs were being moved unless they were put under extreme hyperextension, when a dull ache ensued. He couldn’t tell what position he was in if blindfolded and put on a table. Being held upside down produced dizziness, but he couldn’t tell the cause. He had no feeling of muscular fatigue, so that if asked to raise his hand would do so, but not notice later that fatigue had made it drop – thinking it still raised. If his fingers were held shut he would imagine himself opening and shutting his hand when asked to do so. If blindfolded, he can’t tell whether a voluntary movement is large or small or has even taken place. He doesn’t know what extent of movement to expect, nor can he tell if it is prevented. All his normal voluntary movements were made using the cues of eyesight, carefully watching both his limb and the object of the movement.
Footnote 4: Anscombe explains the scare-quotes later.
Footnote 5: Anscombe gives an example of the sensation of arms stretched out in front of one, but I don’t understand it. In this case, there is no contact with anything.
Footnote 6: This isn’t very convincing – don’t we all remember the almost unbearable tickling sensation in the tendon below the kneecap?
Previous Version of this Note:
|Note last updated
||Reading List for this Topic
Authors, Books & Papers Citing this Note
||On Sensations of Position
References & Reading List
||Book - Cited
||Anscombe (G.E.M.) - Intention
||On Sensations of Position
||Paper - Cited
||Analysis, Vol. 22, No. 3 (Jan., 1962), pp. 55-58
||Some Questions for Miss Anscombe about Intention
||Paper - Cited
||Analysis, Vol. 22, No. 3 (Jan., 1962), pp. 49-54
||The Principles of Psychology
||Book - Cited
||James (William) - The Principles of Psychology
Text Colour Conventions
- Black: Printable Text by me; © Theo Todman, 2019
- Blue: Text by me; © Theo Todman, 2019