UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Pain and perception Bassford, Harold A. 1972-12-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


831-UBC_1972_A1 B38.pdf [ 5.65MB ]
JSON: 831-1.0093083.json
JSON-LD: 831-1.0093083-ld.json
RDF/XML (Pretty): 831-1.0093083-rdf.xml
RDF/JSON: 831-1.0093083-rdf.json
Turtle: 831-1.0093083-turtle.txt
N-Triples: 831-1.0093083-rdf-ntriples.txt
Original Record: 831-1.0093083-source.json
Full Text

Full Text

PAIN AND PERCEPTION by HAROLD A. BASSFORD A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of Ph i1ospphy We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August, 1972 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Depa rtment The University of British Columbia Vancouver 8, Canada Da*e Qotr, U 1971-ABSTRACT Among philosophers there is considerable division concerning the propriety and the analysis of talking of a "pain-sense". In this thesis I investigate the relation ship between pain and perception and in particular between pain and tactual perception, with a view to settling the philosophical dispute. The analysis is devoted to answering three central questions. (1) When one has a pain is it correct to say that one is perceiving the pain? (2) When one has a pain is it correct to say that one is perceiving something? (3) How comparable are feeling pain and the various forms of tactual perception? In answering (3) I attend as well to Berkeley's argument which claims that the situations in which we feel pain and those in which we feel heat are sufficiently similar that we should not conceptualize them in the asymmetric fashion we do, externalizing heat as a secondary quality and internalizing pain as a sensation. I first consider a set of arguments by Ryle, Arm strong, Malcolm and others which aim at providing a negative answer to question (1). These are arguments which would show that the word 'feel' in 'feel pain' is not used in its perceptual sense, while it is so used in 'feel heat'. The arguments range over the skil1-denoting aspect of the perceptual use and the necessary lack of such aspect in the pain-ascription use, the interchangeabi1ity of 'feel' and 'have' in pain ascription, but not in perceptual statements, and the incorrigibility of first person pain avowals. I demonstrate that these arguments on the whole do not succeed, and that that part of the arguments which is correct is not adequate to provide a clearly negative answer to question (1). I then proceed to my positive analysis. By paying closer attention to various situations of pain discourse, and by tracing the comparability of pain and tactual sen sation, I develop a model of pain which both does justice to ordinary discourse about pain and which provides answers to my three questions. This can be called an adverbial model, since it would hold that 'pain' in 'feel pain' is a cognate accusative, functioning as an adverb rather than a direct object. References to one's pain and to one's tactual sensation serve as descriptions of one's experience rather than as perceptual claims. Question (1) is there fore answered in the negative. Question (2) is given a qualified positive answer, however, for just as there are perceptual statements paralleling statements of tactual sensation, so too are there perceptual claims which parallel statements of pain sensation. When one feels pain one is not perceiving pain, but one is often perceiving something, such as bodily damage or the pin which causes the damage. The development of the argument shows in what the relationship between pain and tactual perception consists, and thus provides the answer to question (3). In addition I show how the adverbial analysis of pain can account for talk of pain identity and location. Finally, I complete the argument by showing how Berkeley's claim of an indefensible asymmetry has gone wrong, and by showing the kind and extent of perceptual statements which can be made in situations wherein we feel pain. CONTENTS 1. Feeling Heat and Feeling Pain . ......... 1 2. Ryle's Position 15 3. Incorrigibility 34. Pains as Feelings 72 5. The Identity of Pains . 107 6. Externalizing Pain 12Bi bl iography 143 Chapter 1: Feeling Heat and Feeling Pain Contemporary philosophers differ considerably in their analysis of the relation of pain to perception. Some think it obvious that to feel a pain is to be engaged in some sort of sense-perception; some feel this to be a plausible though not obvious thesis; others believe it involves confusion, and work to reject it. For example, one finds Don Locke saying that pains are private in the sense that only one person can perceive them, adding without argument, "I use 'perceive' in the wide but not extravagant sense in which feeling a pain is 1 a species of perceiving." D. M. Armstrong and George Pitcher both recognize grave difficulties, but in their different ways each argue that feeling pain is a species of perceiving. Armstrong analyzes pain as a species of "sense-impression" and argues that "we can give an account 1. Don Locke, "The Privacy of Pains", Analysi s, v. 24 (1963-64), p. 148. -2-of bodily sensation in terms of the concepts involved in 2 perception.... Pitcher holds that "in standard cases, to feel a pain is to be (directly) aware of a perfectly objective state of affairs.1 On the other side of the fence, H. P. Grice, while not specifically claiming that there is no proper way in which we could talk of a pain-sense, denies that "we would be happy to accept (the psychologist's)pain-sense in the way in which sight and smell is a sense.His method is to specify differences between pains and smells, which explain why we externalize the latter, attributing them as qualities of things, but do not do so with the former. Gilbert Ryle, on the other hand, emphasizes the confusion engendered by taking 'feel' in 'feel pain' in a perceptual sense. He claims, "Feeling, in the sense of finding out or discerning, the warmth of things is a kind of perception...; but feeling, in the sense of suffering pains, is not a kind of perception." Any attempt to use them interchangeably would embody "a logical mistake." 2. D. M. Armstrong, Bodily Sensations (Routledge & Kegan Paul, London, 1962), p. 127. 3. George Pitcher, "Pain Perception", The Philosophical  Revi ew, July, 1970 , p. 368. 4. H. P. Grice, "Some Remarks about the Senses", in R. J. Butler, ed., Analytical Philosophy (Basil Blackwell, Oxford, 1962), p. 134. 5. Gilbert Ryle, "Sensation", in Robert J. Swartz, ed., Perceiving, Sensing and Knowing (Doubleday, Garden City, New York, 1965 ) , p. 190. -3-As even this brief survey shows, there are philosophers on all different sides of the question. My goal in this essay is to work out a proper analysis of the relationship between pain and perception, and thus to show which side is correct. My investigation will be aimed at providing answers to three central questions: (1) When one has a pain is it correct to say that one is perceiving the pain? (2) When one has a pain is it correct to say that one is perceiving something? (3) How comparable are feeling pain and the various forms of tactual perception? The importance of separating out the first two questions can be seen by looking at the positions mentioned above. Locke's position is that when we are in pain what we are perceiving is the pain itself. For him, feeling pain would have to be in some sense similar -4-to feeling the heat of a radiator. This leads to a debate over the oddness of taking pains as perceptual objects in the same sense as heat is a perceptual object, and I shall spend considerable time on this consideration. But it is important to see that a negative answer to the question of whether or not pain is an object of perception does not settle the issue. Pitcher, for instance, claims that what we perceive when we are in pain is a damaged portion of our bodies. Thus, one could claim that to feel pain is to perceive something, but not claim that it is the pain itself which is being perceived. It would then undoubtedly be necessary to give an analysis of the pain, but nonetheless the important point here is that questions (1) and (2) can be answered in different ways. Indeed the analysis of pain which I shall develop will lead me to answer a qualified "yes" to the second question, but a "no" to the first. The third question is appropriate because many have recognized a natural comparison between tactual perception and pain, and much of the philosophical -5-discussion of pain has been centered around the appropriateness and extent of the comparison. Indeed a convenient beginning for the argument of the thesis can be found by looking at the attempted comparison of pain and tactual sensation made by Berkeley in his Dial ogues, and at the sorts of replies which have been made to it. Berkeley begins his attack upon the externality of the secondary qualities with a series of arguments designed to show that heat, being the only tactual quality which is secondary, does not "exist without the mind." Two of these arguments turn upon a comparison of heat and pain. His goal is to show that the family of sensations which includes pain should also include heat. His method is to exploit a claimed similarity in the ways in which heat and pain are experi enced. The first argument trades upon the fact that when one feels greater and greater degrees of heat there comes a point at which one instead feels pain. The argument proceeds as follows: 6. George Berkeley, Three Dialogues Between Hylas and  Phi 1onous, in G. J. Warnock, ed., Berkeley: The  Principles of Human Knowledge, With Other Writings (Collins, London, 1962), p. 155. -6-Phil. But is not the most vehement and intense degree of heat a very great pain? Hyl. No one can deny it. Phil. And is any unperceiving thing capable of pain or pleasure? Hyl. No, certainly. Phil . Is your material substance a senseless being, or a being endowed with sense and perception? Hyl. It is senseless without doubt. Phil. It cannot therefore be the subject of pain. Hyl. By no means. Phil. Nor consequently of the greatest heat perceived by sense, since you acknowledge this to be no small pain. Hylas makes what is a natural objection. He takes back his assent to Philonous1 first claim, namely that intense heat is pain, saying instead that the pain is the consequence of the heat: That is, it is painful to touch very hot things, but this is not to say that the pain and the heat can be identified. Philonous, however, has a- reply. Phil. Upon putting your hand near the fire, do you perceive one simple uniform sensation, or two distinct sensations? Hyl. But one simple sensation. Phil. Is not the heat immediately percei ved? Hyl. It is . Phil. And the pain? Hyl. True. Phil. Seeing therefore they are both immediately perceived at the same time, and the fire affects you only with one simple, or uncompounded idea, it follows that this same simple idea is both the intense heat immediately per ceived, and the pain; and consequently, -7-that the intense heat immediately per ceived, is nothing distinct from a particular sort of pain.7 The argument thus is that to feel heat in intense degree and to feel pain comes in some circumstances to having numerically the same feeling. In these circumstances there is no distinction to be made between feeling heat and feeling pain, so in these circumstances there should be no distinction made between heat and pain. Since only sentient beings are subjects of pain, then ceteru s paribus, only sentient beings are subjects of intense heat. An immediate objection to this argument is that o Berkeley is here confusing heat and sensations of heat. What he has misleadingly noted is that heat sensations of great intensity "blend into", perhaps even become in distinguishable from, pain sensations. This fact suggests some sort of conceptual similarity between sensations of heat and pain, but it hardly necessitates a similar analysis of heat and pain. When I say that a given fire is hot I do not just record the having of heat sensations, but rather make an assertion about the fire itself. When the blacksmith says he has a hot fire he may well say so on the basis of how it looks or how many times he has 7. Ibid 8. An argument along similar lines is found in J. F. Thompson, "Berkeley", in D. J. O'Conner, ed., A Critical History of 'Western' Philosophy, p. 243. -8-pumped his bellows, rather than on the basis of how it feels. Nor in this instance is he at all concerned to tell us that we will have certain sensations if we approach the fire. While the claim about a fire's heat may be based upon sensations occasioned by the fire's heat, the claim is not about a certain kind of sensation, nor does it follow from this that the heat of the fire is itself a sensation. Berkeley has been trapped by his own technical terminology. He has Hylas state that heat is a "sensible thing", and then has him accept Philonous' definition of this class of things as "those only which are immediately g perceived by sense." Unfortunately Berkeley does not make it very clear just what he means by "immediately perceive", nor do I here wish to enter into the exigetical debate. What is of interest here is that Berkeley's argument goes through only if 'immediately perceive' applies to two different categories. Berkeley thinks that feeling the heat of the fire involves having a certain sort of sensation. Since these sensations give way to pain, they are felt to be in the same family as is pain. Further, pain, and thus sensations of heat, are "imme diately perceived." Since heat has been defined as something 9. Berkeley, bp.cit., p.152. -9-"immediately perceived" it is easy to slip in this case into the erroneous conclusion that heat is a sensation. The reasoning would have to be somewhat' as follows: "heat is a sensible object, and as such is an object of 'immediate perception'. In the situations in which one typically talks of perceiving heat the objects of imme diate perception are a certain group of sensations, so it follows that heat is a kind of sensation." But where he has gone wrong is in allowing heat and other "sensible objects" to be "immediately perceived", and then providing an analysis of perception wherein only some sorts of sensations can be said to be the proper objects of "imme diate perception". Given this conflict, he should have reworked his technical terms rather than forcing his conclusion that heat is a sensation. But regardless of these confusions, Berkeley does have another and legitimate problem in mind. This is shown by the last argument of the section on heat, which Philonous uses in order to finally convince Hylas of his position. The argument is as follows. Phil. To make the point still clearer; tell me whether, in two cases exactly alike, we ought not to make the same judgment. Hyl. We ought. -10-Phil. When a pin pricks your finger, does it not rend and divide the fibres of your flesh? Hyl. It does. Phil. And when a coal burns your finger, does it any more? Hyl. It does not. Phil. Since, therefore, you neither judge the sensation itself occasioned by the pin nor anything like it to be in the pin, you should not, conformably to what you have now granted, judge the sensation occasioned by the fire, or anything like it, to be in the fire.10 In this argument Berkeley is claiming that the situations in which heat and pain are felt are sufficiently similar for him to challenge the asymmetrical placement of the two concepts in the language. In the case of the coal, I have certain sensations and accordingly ascribe the quality heat to the coal. In the case of the pin I have different sensations, but do not ascribe the quality pain to the pin. Pain is internalized as a bodily sensation while heat is externalized as a secondary quality. Berkeley takes it that the internalization of pain is correct, and accordingly concludes that heat should be internalized as well. While this argument is suggestive, it is by no means conclusive. Should the symmetry of situation turn out to be as complete as he claims, he will have shown 10. Ibid, pp.158-159. -11-need for conceptual revisions, but he will not have demon strated the direction that revision must take. One could as well conclude, "Since you do judge the fire to be hot, so should you judge the fire to be painful." The only support for the direction of revision Berkeley presents is the making of two claims which he apparently takes to be obvious; vis., that no unperceiving thing can be the 11 subject of pain, and that no pain can exist unperceived. Even if this is a correct representation of the everyday concept of pain, it is also certainly true that ordinary language does allow the ascription of heat to unperceiving objects, and does allow its existence even though unperceived. Berkeley's appeal is accordingly not adequate to support his conclusion. This does not diminish the argument's interest, however, for the possibility remains that his conclusion is premature rather than incorrect. It remains to be seen whether the parallelism between feeling heat and pain is complete enough to force some sort of conceptual revision. I believe the tendency of most recent writers on the subject of pain would be to say that there is not enough 11. Berkeley, op.cit., p. 155-157. -12-parallelism. The main direct attack upon Berkeley's arguments is that made by Ryle in his article, "Sensation," and many of Ryle's points have been accepted by other 12 recent writers. Ryle argues that Berkeley confuses two distinct notions of sensation and two categorically different uses of 'feeling'. On the one hand there is the sense of 'feeling' associated with "such things as pains, tickles, feelings of nausea, suffocation, thirst, and the like." One the other hand there is the sense of 'feeling1 which is used when referring to perceiving with the sense of touch. These two different senses have corresponding concepts of sensation. Pain falls into the first class while sensations of temperature fall into the latter. Any attempt to classify them into the same category would involve a logical mistake. To speak, as I did above, of feeling heat and pain is to commit a syllepsis. So far from the feeling of warmth being merely a lower degree of the feeling of pain, the two things are 'felt' only in quite different senses of the word. They are not even species of one genus, as perhaps seeing and tasting are the species of the one genus, perception. They belong to different categories from one another. The attempt to classify felt temperatures with felt pains, and so to show that felt warmth is a state or reaction in ourselves, as pain in some way is, was a logical mistake. 12. Ryle, o p.c i t. All references are to pp. 189-192 -13-It is clear that if it is made out Ryle's claim will apply to Berkeley's arguments. As stated they depend upon the comparability of sensations of pain and sensations of heat, which Ryle would rule out. However, the second argument can be restated without using the word 'sensations'. When I reach out with my hand and thereby stick myself with a pin I feel pain. When I move my hand into the vicinity of a fire I feel heat. I have been taught to say that the pain is in my finger and that the heat is in the fire. Yet the situations are quite similar. Similar situations should be treated similarly, but we seem to have internalized pain and externalized heat. Either a further explanation or some sort of conceptual revision is necessary. Ryle could claim that the force of this argument depends upon taking 'feel' univocally in the phrases 'feel heat' and 'feel pain'. But this is not so. The argument is pointing out the oddity of making such categorically different statements in situations which are for all other intents and purposes quite parallel. Ryle's putative conceptual differences serve to heighten rather than dispell the discomfort caused by Berkeley's dilemma, and it remains desirable to provide further explanation. This does not mean that Ryle's arguments can be dismissed. Regardless of Ryle's success Berkeley leaves us with the task of explaining why heat and pain are so -14-differently placed in our conceptual scheme. But if Ryle is correct, I will have found the answer to my first central question. Ryle will have shown that 'feel' in 'feel pain1 is not used perceptually, with the result that we can say that we do not perceive our pains. He will also have shown that there are major differences between pain and touch, and thus have gone some way toward answering my third question. Accordingly, I shall spend the next two chapters looking ttarefully at the battery of arguments which Ryle and others bring against the position that the use of 'feel' in 'feel pain' is a perceptual use. I shall argue that for the most part the attacks fail, often because they do not fully enough consider the situations of pain discourse. This is not to say, however, that I shall therefore be arguing for the 'perceptual use' position. When I reach the end of Chapter Three I will not have answered any of my three central questions. Rather I shall have cleared away several philosophically popular misconceptions about pain. I shall then go on to provide my positive analysis, and answer my three central questions. I shall also, in the final chapter, answer the challenge posed by Berkeley's second argument. -1 5-Chapter 2_: Ryle ' s Position Ryle bases his case in "Sensation" upon the claim that 'feel' in the sense in which it enters into the language of tactual perception fits into Ryle's "task-achievement" family, while it does not do so in the sense in which it fits into the language of sensation. His argument is as follows: Moreover, feeling, say with one's hand, that the fire is hotter than it was, is finding the answer to a question. The owner of the hand is discriminating some thing, finding out a difference. In some cases he would admit to having made a mistake. He had not been careful enough. But in having a pain there is no finding anything out, no discerning of similarities and differences, no place, even, for mis takes and so no room for carefulness or lack ofcarefulness. Feeling, in the sense of finding out or discerning, the warmth of things is a kind of perception, and a kind at which some people, like bakers and laundry-girls, become better than other people; it is the product of an acquired skill; but feeling in the sense of suffering pains, is not a kind of perception, and there is no question of one victim being better or worse than another at feeling toothaches. In this sense of 'feeling' or 'sensation', pains are not things the feeling of which is the product of an acquired skill.' This argument rests upon (1) the alleged fact that ski11-denoting words are not used when referring to feeling 1. Ryle, "Sensation", p.190. -16-pain but are used when referring to feeling heat, and (2) the implicit claim that this difference is sufficient to constitute a categorical difference in the two cases. Feeling pain is something that one suffers, while feeling heat is something that one does; the one is passive while the other is active. In order to feel heat one does some thing; at least often one actively tries to discover some thing, and if one succeeds one feels it. On the other hand there is no question of trying to and then succeeding in feeling pain. Since there are no attempts to feel pain they cannot be skillful or careful and cannot result in achieving or failing to achieve, and thus in being correct or mistaken. At first blush Ryle seems obviously correct. The notion of detecting does not seem to have a foothold in the language of pain. We can say that John tried to feel the beating of his heart but failed or that he is good at detecting heart beats. But it is strange to say he tried to feel the pain in his heart but failed, or that he is good at detecting his heart-pains. Now, in order to detect something there must be something there to be detected. That is, there appears to be a necessary independence between the thing detected and the act of detecting it. Applying this to the sense -17-of 'feel' that Ryle has elicited, there must be a logical possibility of that which is felt existing independently 2 of the feeling of it. This is evidenced in the case of heat by the non-substitutivity of the categorical and tactual ascription of heat. One could have a feverishly hot forehead without feeling it. However, it is generally held that there is no such independence in the case of pain. It is often pointed out that in all cases in which 'x feels a pain' are appro priate 'x has a pain1 can be substituted without loss of meaning. This substitutivity is usually thought to be transitive, with the supporting claim being that there can be no unfelt pains (if one could have a pain without feeling that pain, then it would be logically possible to 3 find out that one had that pain, by means of feeling it). Also tied in here is the claim that one cannot be mistaken about one's pains. One cannot say "My hand feels sore, but perhaps it isn't really." This cannot be the 2. This is not to rule out.the possibility of there being a certain sort ofentities which are always felt in fact. 3. Ryle explicitly makes both these claims with respect to tickles, which he classifies with pains , in his article "Feelings", Philosophical Quarterly, Vol. I, 1951, p.164. Armstrong uses these distinctions to rule out a "sensible quality" view of pain in Bodily  Sensations, pp. 45-58. The claim that there is a "complete identity in use" or such locutions as 'feeling pain', 'having pain' and 'being in pain' is found in Norman Malcolm's, "The Privacy of Experience", (see especially pp. 132-133) and in Pitcher, o p.c i t. , p.372. -18-perceptual use of 'feels', since, as Armstrong says, "Perception implies the possibility of misperception. "4 Making discriminations implies the possibility of mis-making them but one cannot mismake his self-ascriptions of pain. While this barrage of claims appears formidable, I shall show in the succeeding analysis that they are highly dubious. I shall first show the insufficiency of the task-achievement claim when taken alone. Ryle's argument will then reduce to the above mentioned inter-changeability of 'feel' and 'have' in pain ascriptions and the incorrigibility of first-person pain ascriptions, 0 which claims I shall then challenge. The Achievement of Feeling I shall begin by comparing three examples. Two of them are clearly perceptual; the third is a case of feeling pain. (1) A doctor feels for my faint pulse. After a minute he succeeds, saying,"Ah, now I can feel it." He continues to feel it for some time, paying very careful attention. After a time he feels a very slight flutter. (2) Lying awake at night I gradually come to feel the beating of my heart. I idly feel it for a while, and then find that I cannot "get my mind off it," that it seems to 4. Armstrong, bp.c i t. , p.54. -19-command my attention. This is intensified by the fact that after a while I feel a slight irregularity in the beat. (3) After awakening from a restless sleep I begin to feel a throbbing pain behind my right eye. I lie in the dark for a while feeling it throb. I can't seem to get back to sleep for I can't seem to keep my attention from it. Suddenly I feel the irregularity of the throb and realize that this pain is the same kind as are those which regularly precede my attacks of migraine. I jump up and run to the medicine chest. In case (1) there is an initial task (feeling for the pulse). In case (2) there is no task. Rather there is simply a time at which I am aware of my heartbeat. There could have been a task. If someone asks me to try to feel my heart beating without using my hands, I can lie on the floor, close my eyes and after sufficient con centration (which involves both attending to my heart and ignoring other stimuli), I might succeed in feeling it beating. In this case it is proper to speak of feeling my heartbeat as a success. But in (2) I do not in fact succeed in doing or try to do anything. It just happens. Can one argue that although one does not undertake any task one still scores a success or has an achievement here? While playing soccer I slip and fall in front of -20-the goal. After someone's wild kick the ball bounces off my head and into the goal. I would be credited with a goal here, but I would not be congratulated for scoring it, nor would the announcer say, except with obvious cynicism, "Bassford finally succeeded in scoring a goal ." Nonetheless it might be said that the scoring counts as a success, since any non-rule-excluded case of the ball going into the goal would count as a success. This sort of claim begins to be strained as one gets away from the context of game-playing and competition. Should I accidentally discover Captain Kidd's treasure while digging a pit in which to store mulch for the garden, it would be a joke to say that I succeeded where all others have failed. It is only because finding buried treasure is a task which people often undertake that it is at all appropriate to use achievement words with reference to my finding the treasure. But if one is to speak at all strictly and without joking such words will not be applied in my case. Nor is it appropriate to say that the perception language-game is such that to perceive something is to have a success within that game. Since (2) is a clear case wherein it is inappropriate to make use of the family of success words surrounding the doing of tasks, and is -21-also a clear case of perception, situations like (2) count against taking such a position. To do so is to force as a distinguishing characteristic of a class what is a distinguishing characteristic of a sub-class of that class.. In example (3) there is no initial task performed. Indeed there is a prima facie legitimacy to the claim that there could be no task. One does not know what one would do if told to try to feel a pain in one's arm. And, as was mentioned earlier, there is a general tendency among philos ophers to think that one could not ascribe a pain to some one unless one could also ascribe the feeling of the pain 5 to-him: "one cannot have a pain unless one can feel it." If this is an accurate description of the situation, then, since there is no task undertaking, there is no place for the use of success words. However, as the above paragraph shows, the inappropriateness of success words does not thereby rule out the case as one wherein there is a properly parallel use of 'feel'. The importance of the claim then, is not the question of the possibility of succeeding in a task, but the question concerning the possibility of. misfelt or unfelt pains. 5. I should emphasize that I am not here accepting this sort of claim as correct. Rather I am showing that the task-achievement argument does not go through even if the claim is correct. I explain the possibilities of misfelt and unfelt pains in Chapter Three. -22-It will not do to try to force the fact that although there was no success in example (2) there could have been, while in example (3) there neither was nor, given the assumptions under which the section is being argued, could have been any success. In the first place the description of (2) would have to be quite different for there to be any ascription of success. The situation wherein it was possible to try to feel my heart beating was very different from the actual situation of (2). Further the understanding of (2) is certainly not obviously logically dependent upon the understanding of the situation where I can try to feel my heart beating. If this objec tion is to carry any weight, some such dependence must be shown. The proof of relevance lies with the objector, and not with the person tending toward a perceptual view of feeling pain. There remain two-thirds of the examples to be compared. In each example the doctor or I continue to feel after we come to feel. This may or may not require effort. In (1) it probably does, in (2) and (3) the effort is in trying not to feel. In each case one is attending to something. And if one is able to pay atten tion to something then one can do it carefully or care lessly, etc. -23-It may be thought odd to talk of attending to one's pains. But this is because pains are not usually the objects of attention. The general reaction to feeling pain is to try to get it stopped, or failing that to distract one's attention from it as much as possible. But the fact that the dentist tells one to pay attention to the music etc., rather than to the pain, shows that there is a perfect intelligibility in attending to the pain. There is some strain in the second uses of 'feel' in (2) and (3). This is because most often when we are feeling something we are manipulating it with our hands or other organ of perception. Even though the doctor is not feeling for my pulse after he has found it he does keep his fingers upon my wrist. When I feel the beating of my heart, however, or a pain behind my eye, I do not manipulate anything with an organ of perception. But when the context is specified it is appro priate to use the present progressive of 'feel' in these cases. It would be perfectly natural for the person who has taken drugs to lie upon the couch feeling the beating of his heart. The case of pain suffers here from the same problem mentioned above, namely that pains are not normally liked. Nonetheless I can recall lying upon the couch feeling the throbbing pain in a mildly sprained ankle, in that case because it provided a good excuse for not working -24-upon an essay. The third use of 'feel' in each of the examples is important because it is by means of such examples that one can begin to erode the claim that ascriptions of pain and ascriptions of feeling pain necessarily accompany one another. In each of the cases the subject after feeling the object for a time feels something about the object that he did not immediately notice. In the first two examples there is noticed an irregularity of heartbeat. In the final use there is noticed a certain irregularity in the pain's throbbing. It is not just that I recall that this pattern of throbbing is the same as that which usually precedes migraine. Rather I come to realize that the pain is throbbing in a certain irregular way. This strongly suggests that it is intelligible to say that the pain had this pattern before I felt that it had it. It then makes sense to say that pains can be of a certain sort without the subject of the pain feeling that they are of that sort. In the next chapter I shall show that the suggestion here raised is in fact correct. In summary, then, Ryle's appeal to the task-achievement distinction does not succeed. If his general argument is to succeed it must do so on the basis of the other arguments he presents. That is, it reduces to the -25-supposed different substitution possibilities for 'feel pain1 and 'feel heat1, and to the supposed different corrigibility conditions of the two locutions. 'Feeling and 'Having It is often suggested that one can defeat or at least put at a severe disadvantage the perceptual analysis of feeling pain by appeal to central cases of feeling, without bothering with those peripheral cases wherein one might be tempted to talk about unfelt or misfelt pains. This is thought to be shown by the existence of complete interchangeabi1ity of the categorical ascription of pain with the putative perceptual ascription of pain, but the clear lack of such substitutivity in other cases of perception. Thus Armstrong urges, ...we do speak of feeling a pain or an itch and it may then be argued that this is the perceptual sense of the word 'feel1. But whenever we speak of feeling bodily sensations we can substitute the word 'have' without changing our meaning. This is not the case for the perceptual uses of the word 'feel1. 'I feel the heat of my hand1 does not mean the same as 'I have a hot hand1 for I could have a hot hand without feeling it. But if I have a pain in my hand, it seems I must feel the pain.6 He does not consider this observation conclusive because he does recognize the temptation in some instancesto talk about the unfelt features of pains. Nonetheless he claims 6. Armstrong, Bodily Sensations, p.46. -26-the only reply would have to be some argument showing that "we can have a pain without feeling it." It is worth noting a lack of parallelism in Armstrong's argument. His claim is that 'have' can always be substituted for 'feel' when pain is the object. His counter-argument, however, is that one can have a hot hand without feeling its heat. Here the substitutivity is in the other direction. And it is to this direction of sub-stitutivity in the case of pain that Armstrong turns his attention in the last sentence of the quote. One must assume that as is Malcolm, Armstrong is concerned to claim a "complete identity in use" between 'have x' and 'feel x' 7 where x is taken to be a bodily sensation. There are three ways to attack this position. First, one might accept Armstrong's observations about the present facts of use, but claim this is only contingent. That is, one might claim that there are good reasons for the interchangeabi1ity of the locutions, but that these very reasons show the intelligibility of thinking of feeling pain as involving a perceptual situation. Differences in the speech situation do not automatically rule out comparison, since they can sometimes be appro priately explained. Second, one might turn one's attention 7. Malcolm, "The Privacy of Experience", p. 143. -27-to those instances which cause Armstrong some embarrassment in this present claim, namely those which involve the possibility of misfeeling one's pain and even failing to note the presence of one's pain. Even if one is successful here one will be left with explaining the differences in the more central cases. Third, one can attack the claim that 'feeling a pain' and 'having a pain' do in ordinary cases come to the same thing. In this case one would attempt to bring forth examples to show that these locu tions are in fact not so interchangeable as is commonly thought. To my knowledge Frank Ebersole, in his article "Feeling Eggs and Pains", is the only person who has followed o the third course. Ebersole proceeds by listing several examples to show that sometimes 'I have a pain1 is appro priate while 'I feel a pain1 is not, sometimes the converse, and sometimes they come to the same thing. In addition he suggests that 'I feel a pain1 in those cases in which it alone is appropriate is "tied to the question of sensi bility and numbness." I shall list several of his examples. (1) While hiking with a friend I limp to a halt and sit on a log holding my knee. "What's wrong?" "I have a terrible pain in the knee." (I certainly do not say, "I feel a terrible pain in the knee.") 8. Frank B. Ebersole, "Feeling Eggs and Pains", in his book, Things We Know. Present references are to pages 106-108. -28-(2) A man has suffered paralysis of the legs and is slowly recovering. Every day the doctor touches, probes, moves his legs. He asks, "Do you feel anything?" One day the patient says, "Yes. I feel a deep pain in the ankle." In this example "I feel a pain" is a remark of the speaker's sensibility. When he says, "I feel a pain", the patient asserts he is no longer numb. (3) I have injured a leg and am suffering from an unbearable pain in the knee. I am given a local anesthetic and gradually the pain subsides. My leg becomes completely numb. Later the doctor asks, "How is it now?" "I can feel the pain again." Or "I can feel that pain again." My sensibilities are returning to normal. (4) Now Allister (a researcher of some sort) places my leg in another device so that strong magnetic fields can be created around it. Without touching my leg, he changes the fields and currents. With each change he asks, "What do you feel?" "I feel a tingling sensation," I might say. Or maybe, "I feel a dull pain throughout the whole leg." Here I was asked, "What do you feel". If I had been asked, "Do you notice anything?" I suppose I might naturally have replied, "Yes, I feel a dull pain." I might also have said, "I have a dull pain." (5) Suppose Allister had said, "I'm going to do various things. Tell me anything that happens to your leg. Tell me anything you can about your leg." I should now quite likely say, "I have a dull pain in my leg." In case (1) the 'I feel' locution is said to be totally out of place. In cases (2) and (3) 'feel' is much more appropriate, these situations being tied to information concerning one's sensibility. (4) and (5) present cases wherein there is interchangeabi1ity. But even here there is a hint of non-substitutabi1ity, since -29-there is a suggestion that the form of appropriate reply depends upon the form of the eliciting question. It is to be noted that some sort of attack upon Ebersole's examples needs to be made by the proponent of the perceptual model of feeling pain, as well as by Armstrong, e_t.a_l_. if they wish to defend their claim. It is clear that Ebersole's example (1) is a relevant situation for applying the perceptual model. If 'feel pain' cannot be used here then at the very least the pro-perceptionist will have a major embarrassment to explain. This particular embarrassment will not, however, have to be faced, for Armstrong has a reasonable reply to the examples. In some cases as (1) he can deny that the substitution is untoward. In other cases he can admit that the circumstances surrounding our pain avowals can condition the form that they take.; Thus, when the doctor asks, "Are you in pain now?", it is natural to reply, "Yes, I have a severe pain in my back." But if he asks, "Do you feel pain now", the natural reply is, "Yes, I feel severe pain in my back." The situation makes one locution more natural than the other, but with some clumsiness the other construction can be used and will convey the same meaning to the hearer. Armstrong's statement is that we can always substitute the one locution -30-9 for the other "without changing our meaning." Thus, Ebersole may.be right about what we probably would say, but we would in.no way confuse the hearer if we were to use the other locution. Thus, it simply is not so strange that one might choose the second locution in (1). If my companion asks me why I stopped I might well reply, "I felt a terrible pain in my leg (or a sharp pain in the region of my heart) and had to stop." "Is it better now?" "I still feel it, but it is not so sharp as it was." The natural locutions in the other examples are conditioned by their circumstances, but they as well could be overcome. The doctor would not be overly taken aback if the patient asserts that his probing causes him to have a pain. This also does the job of asserting he is no longer numb. Similarly, (3) "The leg is painful again", (4) "Now I have a dull pain throughout the leg", etc. These appeals to examples of day to day situations do not, then, serve to defeat Armstrong's claim. Consideration of parallel situations to example (1), however, provides a reply to Armstrong along the lines of the first of those possible replies I listed. That is, I shall develop a reply which accounts for the greater inter-9. Armstrong, Bodily Sensations, p.46 -31 -changeability of the perceptual and categorical ascription of pain and which does not entail abandoning the perceptual view. As a beginning of such an explanation, observe that, contrary to Armstrong's assumption, there is often a sub-stitutability of 'have' and 'feel' in tactual perception as well. Say, for example, that I have a heart condition. While on a hike I suddenly stop. I say: "I have to stop, my heart is beating erratically;"..., I feel my heart beating erratically;"..., I have an erratic heart beat." After a while my doctor gets to the point where I am sitting. I tell him my heart has been acting up again. He feels my heartbeat. "It feels perfectly normal now. Go on, but take it easy." He could also have said, "It is perfectly normal now." In these instances and in many others the various phrases are all interchangeable; one will serve as well as any other. Now, the fact that these phrases can often be used interchangeably in the case of tactual perception does not mean that there is no difference between feeling that one's heart is beating and having a beating heart. Nor does a similar possibility show an identity in having and feeling a pain. In order to show a difference it must be shown that there are critical cases which separate the two. That is, one must show that there are no cases in which one could -32-have a pain without feeling it, that there is an unintelligi-bility to the notions of unfelt and misfelt pains. This argument, then, reduces to a consideration of the second kind of attack on Armstrong's position. One might still want to make something of the more pervasive possibility in the case of pain ascription. Most of the time at least there is a substitutivity in the case of pain, but these cases are surely considerably rarer in tactual perception. This can be explained by bringing out the tie between pain and feeling that does not exist between tactual feeling and its objects, and by noting the tendency to often take 'feel' as a hedge in situations of tactual perception. First, as is obvious, the equivalences (as in the above sorts of examples) of 'feel x', 'have x' and 'be x' occur only when x is felt; that is when the discrimination of x is achieved by means of one's sense of feeling. If the doctor listens to my heart with a stethoscope he would hardly report that it feels normal. If my leg is numb (see example (2)) and I am watching the doctor probe my leg, the statement 'you are probing my leg', will not do the same job as 'I can feel you probing my leg'. It is necessary in this instance to specify the sense by means of which I am gaining my information. But if my legs were -33-screened from me, so that the only way I could know what he was doing to them was through feeling, the two could serve the same function. The latter would be more natural given my concern with being able to feel, but the other could do the job. Thus, in some cases the 'feel' locution is exclusively used, since it is the most convenient way of specifying the relevant sense. In many other cases, however, this specification does not have to be made, and in great numbers of these cases one finds the proper inter-changeability of locutions. Now, one arguing the perceptual view of pain would point out that pain is closely tied to the feeling of pain. One makes one's sel f-ascriptions of pain exclusively on the basis of feeling them. There is no need to specify a sense, so it is not unnatural that pain avowals would make fairly free interchange of the categorical and perceptual verbs. Second, in a goodly number of cases of tactual perception, there is a tendency to use 'feel' as a hedge. For example, I pick up a deck of cards and say, "It feels light." "Better count them." "No need, I know there aren't enough, we'll have to find another deck." "Well, why on earth didn't you say so in the first place." In those cases where 'feel ' functions as a hedge, there will be no interchangeable ity with the categorical, for here our purpose is to hedge upon the categorical. -34-The cases wherein one hedges about one's pains are quite small. In cases of very mild pains one might say, "It feels like a throbbing pain all right, but I'm not sure, it feels like a shooting , pain as well." Even this may not be a proper hedge use of 'feel' for I could probably say, "It is like a throbbing pain, but on the other hand it is also like a shooting pain." This difference between pain perception and tactual feeling is undoubtedly important. I shall pay considerable attention to similar issues in Chapter: Four. In the present context, however, it is sufficient to point out that this difference also reduces to the question of unfelt and mis felt pains. A plausible claim, if there are no hedges, is that this is so because there is no sense attachable to the notion of misfeeling one's pain. We hedge when we are unsure, and with respect to pain there is no such possibility. In general, then, the power of the appeal to the synonymity of 'have pain' and 'feel pain' in ordinary use can be seen to rest upon the propriety of the claims that one cannot misfeel one's pain or have unfelt pains. -35-Chapter 3: Incorri gi bi1i ty Given the results of the last chapter, it is appropriate to turn to the question of incorrigibility. Ryle, it will be recalled, claimed that one cannot misfeel one's sensations. Armstrong draws this out in Bodi1y Sensations. He recognizes there a temptation to talk of mistaking one's pains, but feels only relatively minor errors are possible as opposed to the total error one can make with respect to sensible qualities. Further, any tendency to talk of unfelt features of bodily sensations exists because of distraction, and can be handled by the device of introducing counter-factual conditionals; i.e., "if we had paid close attention to our sensations in the past, they would then have had just the Cfelt) character-1 istics that they now feel to have." Before producing examples to bring home the temptation and testing Armstrong's suggested rephrasings, however, it is necessary to consider a more thorough-going contemporary position of incorrigibility. Norman Malcolm holds that first-person ascriptions of pain, given a few specified conditions, cannot be incorrect. To say honestly what my pain is like is to 1. Armstrong, Bodi1y Sensations, p.54. It is to be noted that in his latest work, A Materialist Theory of Mind, he accepts the view that first person pain reports are corrigible. His argument there rests upon a general view about mental states and is not specifically directed to our discourse about pain, as is the case for the present work. I accordingly do not consider those arguments here. -36-provide the criterion for what it is like. This rules out the possibility of mistakes in honest first-person avowals. It also rules out the possibility of not noticing the properties of my pain. If I say that this pain is like the one that precedes my normal headaches and not like the very similar one that precedes my attacks of migraine, and this is followed by an attack of migraine, then there are no grounds for claiming I made a mistake. I had the former kind of pain rather than the latter, and that is that. Malcolm's First Thesis I shall begin by discussing Malcolm's expression of the incorrigibility thesis in his review of Wittgenstein's 2 Philosophical Investigations. There he approvingly interprets Wittgenstein as claiming that "learning the words for sensations is learning 'new pain behavior'." Self-ascriptions of pain replace in the language user the groans, etc. which are the "natural" or "primitive" expressions of pain behavior. This brings to light, "the arresting fact that my sentences about my present sensations have the same logical status as my outcries and facial expressions." Part of this logical status seems to be that both the pain 2. Norman Malcolm, "Wittgenstein's Philosophical Investiga tions," in his Knowledge and Certainty (Prentice-Hall, 1965). Unless otherwise noted all references in this section are to pages 109-111. -37-avowals and the natural pain behaviour are "incorrigible." A man cannot be in error as to whether he is in pain; he cannot say 'My leg hurts' by mis take any more than he can groan by mistake. It is senseless to suppose that he has wrongly identified a tickle as a pain or that he falsely believes that it is in his leg when it is in fact in his shoulder. True, he may be undecided as to whether it is best described as an 'ache' or a 'pain' (one is often hard put to give sat isfactory descriptions of ones feelings); but his very indecision shows us what his sensation is, i.e. something between an ache and a pain. The only real distinction between the pain avowals and natural pain behaviour is that "I don't learn to moan; I do learn the words." One is not allowed to say that the avowal is used to inform another while the natural ex pressions simply occur, or as Malcolm puts it, "escape from one." For, (a) a sentence can be forced from one, can escape one's lips ('My.God, it hurts!'), and (b) a natural expression can be used to in form another, e.g., you moan to let the nurse know that your pain is increasing (you would have suppressed the moan if she hadn't entered the room), yet the moan is genuine. Accordingly, honest pain-avowals are criteria! in deter mining the presence and type of one's pain in precisely the sense as is unfeigned pain behaviour. There are two ways of interpreting the thesis here presented. One may pay close attention to the assimilation of first-person pain reports to one's reaction to pain. In this case one would want to hold that just as one grunts when hit in the stomach so too one learns to say 'I am in -38-pain' when one's head hurts. This position would be that first-person pain avowals should not be taken as statements, should not be thought to be true or false, should not be taken as descriptions of anything. The other interpretation is to allow that first-person pain avowals do have truth values, but to claim that they are not susceptible to disconfirmation, or at least not susceptible if the person understands the concept of pain, and is not making a pain avowal in situations conflicting with the natural expressions of pain. Something along these lines is undoubtedly Malcolm's preferred position. A footnote refers the reader to his "Direct Perception", and there he takes a line similar to this. I shall examine this interpretation presently, but first shall consider the first alternative. The first thesis produces a tremendous gap between first and third-person pain ascription. The first-person locution, 'I am in pain', is a learned form of pain-behaviour, while the third-person locution "Bassford is in pain" is, one supposes, a statement with truth value and a confirmation possibility. If I am in the habit of referring to myself as Bassford, then when I am in pain I shall say, "Bassford is in pain", while another person noting my pain can say as well, "Bassford is in pain." But these locutions will not count as the same statement at all, since in the first-person utterance I am not uttering a statement. -39-One way to overcome the asymmetry is to deny that the third-person case is one of making a statement either. This is to claim that these statements as well are nothing but 3 learned substitutes for natural pain reactions. Here they would not be expressions of pain, but expressions of our reactions to another's pain. One natural reaction to some one's pain is to tend his pain, to run and sooth him, auto matically to look sympathetic. Instead of going "cluck, cluck, cluck" to sooth the person, we say "Don't worry, it will be all right." Instead of (or in addition to) looking with horror, we say, "He's hurt." This has some plausibility. Very often we do not by the use of these phrases utter pro positions, but only soothing or shocked noises. However, this does not do in general. It is clear that our discussion of other's pain is not made up entirely of natural reactions of pity (or glee) at the pres ence of their pain. For one doctor to say to the other, "This patient is in considerable pain, so I had better give him more morphine", is not to substitute a learned pain reaction for a natural one. And when the second doctor expresses the opinion that the patient is only feigning pain in order to get the morphine, the doctors can enter into a heated debate over who is correct. Here we are well 3. This is not Malcolm's claim. He mentions that Wittgenstein assimilates only first person sen sation sentences to expressions of sensation (p. 117). -40-entered into declarative communication, and our sentences properly embody statements. Let us return to the first-person case. Some of the radical asymmetry has disappeared in seeing that in some cases second and third-person pain sentences do function as natural sympathy behaviour. The rest disappears when it is shown the first-person sentences do not always function as expressions of pain. The familiar example of the patient describing his symptoms to his doctor will suffice. In these cases there is clearly a possibility of lying. I want more morphine, now that I have become addicted to it. I grimace and groan, and say to the doctor, "That pain in my back is becoming unbearable again." There is no question of holding back natural reactions, or of the words being dragged out of me. I am simply lying. As such my avowal of pain is false. This is not to say that I am or can be mistaken, but it is to say that I am making a statement. There might be a reply here to the effect that one is not lying, but rather feigning pain behaviour, and that this is not to say anything true or false. To give away this much is to throw the baby out with the bath water. Talk of feigning recognizes the existence of a social situa tion wherein communication can properly be said to take place. Let me elaborate using Malcolm's own examples. One does not learn to moan, but one does learn to use his moans -41-for a purpose. There is a world of difference between my lying on the bed moaning because it hurts so much, and not suppressing a moan so as to inform my nurse of the pain. I could also moan so as to inform her of my pain even though the pain was not quite so bad that I tended to moan because of it. I am moaning in order to communicate something to the nurse, namely the fact that I am in pain. If I should instead sit up and say, "Nurse, the pain is getting worse", I would also be communicating something to the nurse. Indeed I would be stating the fact of which I wish to inform the nurse. As in any other situation of discourse I could here state the facts truly or falsely, accurately or inaccurately. The nurse could tell me I was not telling the truth, and I could try to convince her that I was. It might be that I could not be wrong about my proposition, but it is still one that could be debated, and could be either true or false. The recognition that first-person pain ascription sometimes functions as a pain-reaction should not be allowed to obscure the descriptive functions it can also serve. Malcolm's Second Thesis This does not establish that an honest first-person pain ascription can be false, the negation of which is basically the claim made by the second interpretation above. Malcolm elaborates this claim to incorrigibility in his -42-"Direct Perception." The claim is developed by close attention to the fact that one's avowals of pain often pro vide the best possible evidence for the presence of one's pains. The position is that if a person has correctly used the words ["I have a pain"] on numerous occasions and has never misused them, and if his circumstances are such that it would not be a misuse of the language to say of him that he has [a pain] ... then there cannot be a question of his being in error when he says ["I have a pain"]. There cannot be a question of whether he "takes" something to be [a pain] that is really not one. There can be a question as to whether what he says is true, but it is identical with the question of whether he is fibbing.5 This holds as well for description of the features of one's pains. If a person understands the words 'waxing and waning' and if the other conditions are satisfied, then his assertion that his pain is waxing and waning cannot be mistaken. In this case Malcolm does admit a minimum possi bility of self correction, but the expression must be either because of misspeaking or of a minimal nature: "With regard to anyone's description of his own pain, there is no clear difference in meaning between his assertion that 4. The argument of "Direct Perception" is conducted in terms of after images, but Malcolm says that with appropriate changes it can be applied to pains as well. ("Wittgenstein's Philosophical Investigations", p. 110, footnote). I shall elaborate the argument in these changed terms. According ly, the existence in a quote of ['a pain'] is a substitution for ' an after image'. 5. Malcolm, "Direct Perception", in Knowledge and Certainty, p. 85. -43-he made a_ mi stake and his assertion that he_ expressed himself incorrectly." First-person pain statements which meet these conditions are "1self-confirming' , implying by 6 this that really they have no_ confirmation." Their only entry into confirmation would be to provide confirmation of third-person pain ascriptions. Let us begin by clarifying the assertion that when the above conditions are satisfied "there cannot be a question of his being in error...". Malcolm is not saying that here it would be illegitimate for a person to doubt the assertor's claim, in the sense that under these condi tions there would be no grounds for legitimate doubt, but rather that "it would be entirely senseless for,us to suppose that the assertor was mistaken." This comes to saying that the notion of a mistake just cannot play any role in situations of this sort. My utterance of 'I have a pain1 when I am not misusing language or lying implies that I do have a pain. A second-person's utterance of.'Bassford has a pain1 of course does not imply that I have one. Malcolm would not deny that we are claiming the same thing. He excludes the view that "he is in pain" means only "that his behavior, 6. Ibid., p. 81 -44-words and circumstances are such and such."^ My utterance of 'I am in pain' will express a true proposition if and only if another's correlative utterance of 'Bassford is in pain' expresses a true proposition. The claimed asymmetry is in terms of confirmation, not in terms of meaning. Malcolm is, incidentally, inaccurate within his theory in saying these first-person assertions "have no confirmation." He should say instead that their confirmation involves only the assertation that the assertor is not lying and that he knows the meanings of the relevant terms. Unfortunately this does not completely capture Malcolm's position. For he also wishes to claim that a person accusing me of making a mistake above would not be wrong, but that his remark would be "meaningless." This comes out in Malcolm's case of contradictory testimony about an after-image. His example is of a man who traces an outline of his after-image giving it six points, but says he sees seven points. If he does, not correct this, but rather claims "both that his after-image has seven points and that he has traced it accurately" we can only 7. "Wittgenstein's Philosophical Investigations", p. 119. In his recent book Problems of Mind (Harper, 19 71) Malcolm says that first and third-person psychological statements must exhibit this equivalence in truth value. It is their mutual link to behaviour that prevents my having one concept of pain for myself and another for others (pp. 90-91). -45-"shrug our shoulders at this conflict in his description -and nothing more." If someone's description of a physical reality is self-contradictory, it is sensible to speak of trying to find out whether this one or that one of the contradicting statements is true; and this makes it meaningful to assert "Either this one or that one must be true." When the contradictory statements belong to an after image description the latter assertion is devoid of meaning; for that description is our criterion of what the after-image is like, and consulting it exhausts our inquiry.8 Presumably, the same analysis of a contradiction in a description of a pain would hold. If a man reports that he is in pain, describes the pain as a sharp dull pain, and denies that he means it is a pain which is dull with regular sharp jabs or something like this, but claims instead that it is constantly both dull and sharp, then his assertion is devoid of meaning. This analysis will not do. If a "description of physical reality" is contradictory we do not say we cannot draw any conclusion; we can certainly draw the conclusion that it is false. And we do not know it is false because, as Malcolm claims above, we are unable .to...try to find out which of the two contradictory statements might be true. It is false because of the meanings of the terms involved. We would not know what to make of the description in that we might not know why the individual would make such an 8. "Direct Perception", pp. 82-83. -46-assertion or what could lead him to have been so confused about the meanings of the terms. This is not to say that one would have to be confused about the language in order to make such a claim. A philosopher might well make it because he wanted to enjoy the blank looks, consternation and confusion engendered by the statement. Surely the same thing is the case with the contra dictory statements about pains or after-images. The words are the same in meaning as they are in any other realm of discourse. To say not that the assertion is contradictory, but rather that it is meaningless is very odd indeed. Malcolm would have to contend that the same thing would hold for the third-person ascription. If George says of me, "Bassford has a sharp dull pain", then Malcolm would have to say that what he says is not wrong but meaningless. To claim this is to fly in the face of the fact that words are meaningfully used in situations in which pains are not at issue. George's statement that his knife is both sharp and dull is not meaningless, but rather is false. The only way in which it could be claimed that there is a meaninglessness in the case of pain would be to say that the words used in pain descriptions have different meanings than they do in other realms of discourse. Malcolm would not allow this, but rather make it a condition for saying a person uses the words correctly in his descriptions of his pains and after-images: -47-To be sure, if in his discourse about physical realities he constantly misused the word 'blue1, this would be a sufficient condition for saying that he describes his after-image incorrectly when he says to us, "it is blue". Given the equivalence in truth-values of correlative first and third person pain ascriptions, it follows that the first-person ascription must also be false in this case. Accord ingly, if I had made the statement here made by George it would be correct to say that my utterance was false. My utterance would not play a very useful role in the language-game of pain description, but it would play a_ role. Malcolm is lead to say that the contradictory description is meaningless rather than false because of the version of verificationism to which he is subscribing. If we say the contradictory description is false, and yet allow the existence of the after-image or pain, we can then say that there is another description which is true. But if we do this, then, in Malcolm's view, we have no way of settling the question of what that description would be. This is because, according to Malcolm, the only way of settling this question is by means of the testimony of the subject of the pain or after-image. And if we cannot conceive of what would settle the question, Malcolm would say, we would not really have a question. That this is Malcolm's view is seen in the section quoted previously, wherein Malcolm lays such emphasis on "being able to find out" which alternative is true in the -48-description of physical object. In the cases of contra dictory description of pains, the only way to find out what the correct description would be is, according to Malcolm, no longer open to us. Consulting the statement "exhausts our enquiry", which is to say that there is no sense is talking about the correct description. Now, as I have shown above, Malcolm's position with respect to contradictory descriptions of pain is untenable. Nor should he claim that any statement challenging my description of my pain is meaningless, even when the challenger allows that the adequate confirmation conditions of my statement are met. A person could reply to my description of my pain as follows, "I allow that you know the meanings of the words you are using, and that you are not lying, but your description is wrong." Such a reply is not meaningless, rather it is false. We might say to him that he is speaking nonsense. But this is to suggest to him that he is confused or is trying to confuse us. The trouble with his statement is not that it is meaningless but rather that it is self-defeating. This is because the first part of the statement throws out the grounds on which my statement can be properly doubted, and so is incompatible with the second part of the statement. Such statements,, like talking about the correct description of a pain when all our ways of determining that description are exhausted, do not play a useful role in the day-to-day discussion of pains. -49-We may wonder why people would want to make these kinds of assertions, but we know what to do if people do make them. Rather than saying that they have no sense, we should say that they have little point. Reduced Incorri gi bi1i ty Let us now return to the incorrigibility thesis. Malcolm's verificationist assumption is not his only fault. Though some of his emphasis is correct, he has made several significant omissions as well. Malcolm is correct in noting and emphasizing the fact that first-person psycho logical utterances often provide conclusive evidence for the propositions they enunciate. In most cases if we want to know what kind of pain a person is having we find out by asking him, and are quite correct in being satisfied with his answer. Malcolm is also correct in recognizing that these utterances in conjunction with certain sets of conditions are such that no further evidence could disprove them. But 9 this is not because they have no confirmation. Rather it is because they are made in those conditions which one would wish to check in order to confirm the statements. In principle such statements are always subject to confirmation, though this may only involve seeing that they were uttered in the appropriate circumstances. 9. See "Direct Perception", p. 81. -50-Malcolm should admit that there could be dis-confirmation of first person psychological statements, 1 0 since he allows for misspeaking and for lying. George says his after-image is green. We then notice that his use of 'green' is consistently applied to yellow things and vice versa. This would disconfirm his statement that his after-image was green. Certainly in the case of pain there is a confirmation possibility as well. George says he is in terrible pain in order to get more morphine., The doctor agrees and goes out of the room to get it. But he carefully looks back through the crack in the door, hoping to see George relaxing his act momentarily. Usually any attempts at confirmation or disconfir-mation are unnecessary, because the conditions which would be considered are assumed to be met. The man's language is consistent with our own, there are no reasons for thinking that he is lying, and so on. Here there is no question of confirmation, since there is nothing one needs to do in order to confirm the statement. One who had doubts at this point would indeed be confused about the concepts being employed. But that does not imply that these conditions are met in all cases of first-person psychological utterances. In some cases they clearly are not, and in these cases expressions of doubt and discussions of confirmation or 10 . Ibid., p. 85. -51-disconfirmation are clearly in order. Of crucial importance here is the fact that Malcolm has left out relevant conditions. Malcolm feels the only relevant conditions are understanding the language and truth fulness. I shall argue that one must also add that the utterer must pay sufficient attention to his sensations and must not be in what I shall call a relevantly untrustworthy psychological condition. These conditions are usually not considered, but this is because in central cases they are (properly) taken for granted. Further, since having pain is closely tied to pain-behaviour, any discrepancy between the pain-behaviour and pain-avowals must be explainable. With these additional restrictions the possibility of misfeeling one's pain becomes very real. Revised Reports of After-images I shall begin to elaborate my position by considering Malcolm's discussion of the man who changes his mind about the shape of his after-image. I shall then show that similar considerations apply to the case of feeling pain. A person first reports his after-image is star-shaped. After a moment he reports it is not star-shaped. Normally, says Malcolm, this report would mean the person has a different after-image. If the persons says, however, "No, it isn't star-shaped and my after-image has not changed", we should not take even this as refuting the first report. Instead, -52-we should regard the whole sequence of dis course as uni rite!Ti gi ble. In this sense the subsequent report can discredit the previous report. But it.does not show it to be in accurate . 11 Surely this is just what we would normally think that it did show! The inclusion of the reference to the same after-image would key the listener to the fact that the speaker was revising his report. Such an interpre tation is possible on Malcolm's grounds as well for the speaker could be making known the fact that he mis-spoke. Certainly, he could also be revising his report on the basis of having in the interim paid closer attention to his after-image. People do attend more or less closely to their after-images, pains, etc., and it is not odd for them to change their description on the basis of such closer attention. It might be thought that the identity conditions of after-images and pains are such that this sort of re vision leads to some sort of contradiction, and so is not to be allowed. Malcolm's discussion of after-images in the present section was largely concerned to show the oddity of talking of. observing an after-image. He tries to show that one would not know what it would mean to "look again" or "try to get a better view" of one's after-image. He wants to say that if one does not know what these notions come to in this case, then there is no way to interpret the notion 11. "Direct Perception," p. 80. -53-that an after-image has a different look than it did. There is no way to distinguish the same after-image as looking star-shaped at one moment and not looking star-shaped at another. Malcolm is claiming that the identity conditions of an after-image are tied to the way in which the after-image appears. To first say an after-image is star-shaped and then to say it is not star-shaped, but is the same after-image, is to make a contradictory claim; it is and is not the same after-image. My claim of course was not that the person was holding a contradictory position. His subsequent statement serves as a revision of the correctness of the first. The difficulty with,this, along the above lines, would be the temptation to report this along the lines of "It looked star-shaped, but now it does not." But this is no distinction, since to look star-shaped is, in the case of after-images, to be star-shaped. Accordingly, such a revision is no revision, but rather a contradiction. This shows at best that the temptation must be avoided. The statement with which to correct the first would be better put as, "I thought it looked star-shaped, but I wasn't paying proper attention. I now see that it does not look that way at all." A statement as to how something looks is incorrigible only if one has paid sufficient attention to how it looks. Austin's comment about the corri.gibi 1 i ty of descriptions of looks is relevant here. -54-But certainly someone might say, 'It looks heliotrope,1 and then have doubts either as to whether 'heliotrope' is right for the colour this thing looks, o_r (taking another look) as to whether this thing really looks heliotrope. There is certainly nothing i n  pri nci pi e f i nal , conclusive, irrefutable about anyone's statement that so-and-so looks such-and-such. And even if I say, ' . . . 1ooks . . . to me now ,' I may, on being pressed, or after looking at the thing more attentively, wish to retract my state ment or at least amend it.12 It is true that in the case of after-images we do not literally take another look. But on the other hand we do pay more attention. Assume that I am listening to music and also looking rather absently at a bright light. I look away and my companion asks me what shape my after image is. I reply, "It's star-shaped." But immediately afterwards I say, "No, wait a minute, it looks more like a Christmas tree." At first I reply while attending more to the music than to the after-image. As I attend more to the after-image I have reason to change my description, for it does look differently than I first thought. Again, it is sometimes possible to sufficiently dup licate conditions as to challenge previous statements. I look at a bright light while listening to music. After looking away I give the requested description, but only abstractedly since I am trying to concentrate upon the music. If I am later questioned about the description for some reason, the appropriate thing to do is to look at the light 12. J.L.A. Austin, Sense and Sensibilia (Oxford, 1962), pp. 42-43 -55-again and then attend to my after-image. If I now give a different description, there is a strong tendency to say I misdescribed my first after-image. It is possible for some one to claim that the conditions must be different in the two cases. But I can find situations where the physical conditions can be made as similar as desired, and in such conditions the only obvious remaining difference will be the difference in attention. Mi sfelt Pain Errors of attention are possible in the case of pain as well. They will not often occur, since, as is often pointed out, pains capture our attention; they are very difficult to ignore. But such situations do arise. Not long ago I pulled a muscle in my leg while playing tennis. For a moment I thought I had hurt my foot. I even complained as I began to hop about that I had done something to my foot. It was only after I had lost the point for my side and was not distracted by the game that I realized the pain was really in my ankle. Mistakes are possible with respect to pain quality as well as pain location. I have long had a mistaken belief about the pain of the dentist's drill. I have often thought of the pain caused by the drill hitting a sensitive place as very sharp, stabbing and gruesomely unpleasant. Having had occasion to go to the dentist since beginning the present essay I have come to realize that such pains are actually intense burning pains, -56-though they are indeed gruesomely unpleasant. The pain of the probe and the pain of the drill are not the same, though I have never before attended closely enough to discriminate them. My earlier example of slightly different pains pre ceding migraine and nervous headache is appropriate here. Assume I suffer from a severe form of migraine. Before every attack I have a certain kind of throbbing pain behind my right eye. If, however, I take a migraine pill at other times it makes me quite ill. Unfortunately, nervous tension brings on a similar, but not identical, throbbing pain. Sometimes I make a mistake, but after a while I learn to get it right every time. One might want to oppose this and hold that, unfort unately,.the pain of nervous tension was these times just the same as that preceding migraine. This would be the correct description if I carefully attended to the pain in each case. However, if after my initial bad experiences I began to make more careful discriminations and discovered the above corre lations, it would be proper to say I had not previously paid careful enough attention. Errors of attention are not often made in pain ascriptions, nor are they usually so important when they are made. But in discussion of pain and error they deserve a place of more than usual importance. In Bodily Sensations Armstrong admits the possibility of minor errors, such as I have argued for here, but urges -57-that only very minor errors are admissible. Even if it makes sense to speak of a distinction between the way our sensations feel to be and the way they really are, the gap cannot be a wide one. Our sensations are of necessity, much as they feel to be.13 He uses the difference in possible error to show that there is a workable distinction between bodily sensations and sensible qualities, claiming that "in the case of sensible qualities total error is perfectly possible." This difference in the possibility of radical error might also be thought adequate to maintain a conceptual difference be tween tactual feeling and feeling pain. However, major errors are possible in the case of feeling pain as well. They can occur in instances wherein one is apprehensively expecting a certain sort of sensation, but in which another sensation occurs. One case has been given by charles B. Daniels. Imagine a torture situation; prisoner strapped face down on a table. Information to be extracted. Iron being heated to red hot before eyes of terrified prisoner. The torturer takes up iron, brandishes it before the prisoner's face, walks behind him, and completely unbeknown to the prisoner, applies an ice cube to the prisoner's back. He screams, faints.14 13. Armstrong, Bodily Sensations, p. 56. 14. : Charles B. Daniels, "Colors and Sensations: or How to Define a Pain Ostensively", American Philosophical  Quarterly, v. 4, July, 1967, p. 234, -58-Here the prisoner actually feels a sensation of coldness, but because of the situation thinks he feels a terrible pain. The mistake is not caused because he is not attending carefully,. It comes because of the disturbed state of his mind. A more mundane but quite similar example can be found in a situation faced by most parents. A small child scrapes his knee. The parent attempts to spray on the miracle "ouch-less" antiseptic. The minute the spray touches the knee the child screams and throws himself about. After a minute he stops, for he suddenly realizes that the spray doesn't hurt at all. In fact he hardly felt a thing when the parent began to apply it. As in Daniel's example, the child has built up a certain expectation and the very siight - sensation of coolness which he has is taken to the one of extreme pain. Daniel ' s example also provides a plausible possibility of radically mislocating one's pains. The torturerpasses the iron over the prisoner's back so the prisoner can feel its intense heat. He lifts it away and says, "Now we'll fix that handsome back." He then accidentally touches the iron to the prisoner's foot. The prisoner screams, "My God, my back", and faints. Here we can say that the prisoner's apprehensions were so developed about one area of his body that the sensation in another area of the body was taken to be in the first area. Other errors can be found if one considers cases wherei n • peopl e are hypnotized or victims of self-deception. A case of error due to hypnosis is brought up by Roger Trigg, -59-1 5 in his book Pain and Emotion. Here a patient trained as a hypnotic subject appears to give contradictory pain-avowal s. . A hypnotic analgesia for the left arm was suggested, and the subject was pricked in the left arm four times with sufficient force to puncture the.skin and subcutaneous tissues. After a minute or two, the subject asked the experimenter, 'When are you going to begin?', apparently not having felt any pain. However, from the moment that the pricking began, the subject's right hand had begun to write: 'Ouch, damn it, you're hurting me!' If the writing can be considered an avowal, and the spoken statement considered to imply the statement that he was not feeling pain, then as Trigg states, at least one of his views 16 about feeling pain must be wrong, though both are honest. On the basis of the information provided it is difficult to decide what the correct interpretation is. However, it is the case that the pricking would normally produce pain, and that the written avowal attests that it does. If there were in addition some recognizable pain-behaviour, then we could properly say that the subject's oral statement was a sincere, but erroneous avowal. I should like to consider one more case of self-deception, wherein on one interpretation it seems plausible 15. Roger Trigg, Pain and Emotion (Oxford, 1970), p. 66 (Trigg quotes here from an article by E.A. Kaplan) 16. Ibid. , p. 174. -60-to say that a person mis-locates his pain over a considerable period of time. The example is from Kathryn Pyne Parsons' 17 article, "Mistaking Sensations." She considers there G.E.M. Anscombe's example of a person who says his foot is very sore, but nurses his hand, doesn't object to rough handling of the foot yet points to the foot rather than the hand as the sore part, and so on. Anscombe states that in this case it would be difficult to know what the subject meant. Parsons agrees but suggests that the case is not properly filled out. We may note that the foot is undamaged. We may further know that several of this man's relatives have had a fatal disease which begins with a pain in the hand and that he has developed a pathological fear of dying in the same way. Here, as Parsons concludes, we have good reason to set aside the man's quite honest claim about feeling a pain in his foot. The man's behaviour plus,the psychological ex planation here lead us to say that the man really feels the pain in his hand, but thinks that he feels it in the foot. These cases have provided examples of radical error with respect to one's pains. This defeats Armstrong's con-tention that only minor errors are possible. The conditions 17.. Kathryn Pyne Parsons, "Mistaking Sensations", Philosoph-ical Review, April, 1970. References here are to pp. 208-209, 211. -61-under which first person pain avowals are incorrigible are more restrictive than has often been thought, and provide room for a full-bodied notion of making mistakes. The extra conditions and the according greater corrigibi1ity are usually not noticed, because consideration of feeling pain is usually kept within usual and more central cases of feeling pain where in the relevant conditions are in fact met. Once one looks on the periphery, as is regularly done in cases of perception, these extra conditions become more obvious. We in fact do not usually make errors in feeling pain, but nor do we usually make errors of tactual perception. Nonetheless we can and do make errors in both cases, and in both cases we can specify in what those errors consist. Unfelt Pains It is now appropriate to examine the question of whether one can be said to have pain without feeling it. Several possibilities follow from the above cases of mistake. First consider the example wherein I mistake my migraine pre ceding pain for my nervous headache pain. This sort of occurrence lends itself naturally to talk of unfelt features of my pain as well as to misfelt features. It may take some time to learn to recognize the particular throb which precedes migraine attacks. I may recognize only that I have a pain behind my eye before my migraine attacks and a similar pain associated with nervous headache. After I learn to recognize -62-the particular throb of the pre-migraine pain, it is natural enough to assume that the particular throb was present in pre vious cases as well, but that I failed to notice it. There is nothing relevantly changed in the conditions under which I have my migraine attacks. It would accordingly be inappropriate to think there was something special about the present cases of migraine preceding pain. A difference between hypochondriacs and normal people 18 adds another plausible example. A hypochondriac might well be very interested in his pains and accordingly pay much closer attention to them than do ordinary people, who are only con cerned with the fact that they have them. He accordingly can give much more detailed reports and make much more careful distinctions than does the average man. It is straining to say that this person's pains are much more detailed than those of others in the same circumstances. One should instead admit unfelt features of one's pains and admit that people get better at apprehending these features. Armstrong admits the force of these examples, but argues that all these sorts of cases which involve a lack of necessary careful attention can be handled without admitting "the actual existence of unfelt characteristics of 1 9 sensations alongside the felt characteristics". His 18. This example is suggested by one of Armstrong's. See Bodily Sensations, p. 53. 19. Ibid. , pp. 53-54. -63-device is the introduction of a counter-factual conditional. The newly felt characteristics existed before only in the sense that, if we had paid close attention to our sensations in the past, they would then have had just the (felt) characteristics that they now feel to have.20 In the present case the ordinary man's pains would also have unperceived qualities, but only in the sense that if he should pay close attention, then he would note these qualities as wel 1 . Armstrong's device here is an attempt to make the minimal adjustment necessary to meet the demands of the examples. He is, however, grasping at straws. The question which arises is what relationship is being claimed between the attention and the pain's character. Armstrong wants to say that they are not characteristics which are there but not noticed, but that if we were to attend closely to the pain they would be noticed. This is to suggest that somehow the attention brings about the characteristics. But if one wants to argue this, one could hardly say that the charac teristics were there before, since it is the attention that somehow brings them about. If the claim is to be convincing, something more must be made of the mysterious relationship between the attention and the qualities of the pain. I must admit in passing that there is some plausibility 20. Ibid. -64-to claiming a tie between the intensity of a pain and the attention paid to it. I have noticed during recent trips to the dentist that I am less disturbed by the pain of the drill if I try to pay close attention to the pain and to sort out its various qualities. Should we then say that the attention paid changed the quality of the pain, or should we say the sensation was the same, but that our attitude or reaction to it differed? This question of the relation of the attitudinal to the sensational elements of the pain-experience is an interesting one, but it is 21 not to my purpose here to go into it. What is to my purpose is to note that any possible relationship between the pain's intensity and the attention paid the pain, can hardly be generalized to all the other qualities which the pain may have. In the above paragraph, it was quite natural to talk of my attempting to sort out the various qualities of the pain the dentist was causing. These qualities are not there just because I try to sort them out. Rather I am able to sort them out only because they are there to be sorted out. There is a tendency to think of a pains' characteristics as obvious upon only the slightest inspection. So if we pay close attention and find something new, there is a tendency to say that the quality 21. Several interesting considerations of this problem can be found in Roger Trigg, Pain and Emotion (Oxford, 1970). See especially Chapters 2 and 3. -65-has somehow just arisen, or is conditioned by the attention. But if one begins to attend to his pains, and not just try to get them stopped, one discovers that pains have a fair amount of subtle variations, and that careful attention is sometimes necessary to discover the particulars of a given pain. These examples have all been concerned with unfelt features of felt pains. There remains the necessity of discussing unfelt pains per se. I shall begin by listing three examples. (1) A familiar example is that of the soldier who although showing every sign of intense pain (wounded leg, severe limp, grimace and groan at every step) nonetheless proceeds to save his comrades or whatever, and afterwards stoutly insists that he felt no pain during the entire proceedings. In the face of conflicting criteria we may not know quite what to say. But the behavioural evidence may have been very strong and his attention so riveted to saving his comrades that there is a strong tend ency to override his testimony and say he just was not aware of his pain. (2) Anscombe's example of the hysteric provides another example. This person need not be insistent that he has a pain in his foot while nursing his hand. He may instead give all evidence of having a pain in his hand, while honestly denying that he has any pain at all. Here, in the light of the rest of the story, one says that his hand -66-22 hurts although he does not feel it hurting. (3) Finally, there is a pain reaction of a sleeping man. He straightens his leg and kicks his cat on the foot of the bed, who there upon gives him a vicious scratch. He yelps and jerks his foot away, but does not awaken. We then wake him up. Here 23 we can say that he was in pain but was not aware of it. These cases, however, do not settle the question. In (1) the person was too distracted to feel the pain. In (2) he was in an unsettled mental state. In (3) he was asleep. In each case, pain is ascribed to this person on the basis of his pain-behaviour. Because of this one might want to object that this could not properly be called pain-behaviour unless in some sense of the word the subject were feeling pain. Unless he were in some way feeling it, he could not be reacting to it. A man is having a brain operation. The surgeon probes the wrong place. The man screams and rolls about the table. We are concerned, but the surgeon tells us not to worry, for the man is not aware of a thing, that this is not really hurting him. Later the patient, who had to be conscious during the operation, remarks upon the oddity of hearing himself groaning and watching himself writhe about 22. Henry Beecher (Measurement of Subjective Responses, New York, 1959, p.167) reports a study in which some patients after receiving morphine or phenobarbital , "claimed that the pain was 'quite a bit better1 and yet who continued to be restless, tense, unhappy, bothered greatly by minor ailments (position tubes), and generally uncomfortable. Here it was impossible to believe that the medication had been very successful, despite the relief of pain." Beecher thinks we can therefore distinguish between comfort and pain relief. I suggest we might want also to say that we could gain more information that these people were still in pain, but because of the action of the drugs, were no longer aware of it. -67-when he was not feeling a thing. Here one wants to say that the man was not feeling any pain, but that his body was reacting in a certain way. We are interested in his suffer ing, which is equivalent to "only so far as he is suffering." The apparent pain behaviour here is thought of as response to stimulus, but not as response to the stimulus of pain. His pain behaviour is equivalent, in feel, to the knee jerk ref1 ex. In other cases, however, one can see what is stimu lating the response: the wound in the leg, the malignancy in the hand, the scratch on the foot. In these cases one wants to attribute the response not to some automatic reflex, but to some reaction to felt pain. The man reacts because he feels pain. Because of some alteration in his consciousness he is not aware of this. Normally he would be aware of feel ing the pain, so normally there is no distinction to be made. In extraordinary cases, however, one must do so. This is, I think, a well-founded objection. In all these cases it is plausible to say that the subject felt the pain, but that he was not aware he felt it. This last is valuable, since it serves to break the tie between the existence of pain and the conscious awareness of pain. It does give one important sense to the existence of unfelt pain. On the other hand it suggests a sense in which one 23. A similar example is given by Malcolm in his "The Privac of Experience", p. 133. -68-could not have unfelt pain. This shows the strength of the tie between pain and feeling. I can think of no ex amples which will break that tie. Here at last is a victory for Ryle et_ a]_. There is one sense in which one seemingly cannot intelligibly talk of unfelt pains. The question is whether this dis covery is adequate for the proof of Ryle's general claim. He could argue as follows. In all normal and clear per ceptual uses of 'feel', that which is felt could exist in dependently of its being felt. There is a sense in which pains cannot so exist, so the 'feel' in 'feel pain' is not used perceptually. My feeling is that though this hurts the position Ryle is attacking it does not destroy it. I should like to explore briefly what difficulties this causes . What this basically means is that pains will have an ontological dependency upon perception which other things which are perceived do not have. The fact of ontological dependency itself would not make pains into a terribly unique sort of entity. Smiles exist only when their smiler does. Shadows exist only when that which casts them does. Sim ilarly, pains exist only when they are felt by their owner. What is odd about this case is the fact that pains seem to be ontologically dependent upon feeling which seems to give them a unique status among things which are felt. -69-Th ere are two ways, I think, that one could try to overcome the discomfort this causes. One way is to try to produce other cases wherein the tie also exists, and thus to reduce the uniqueness of the situation. Another would be to admit the uniqueness, but not be concerned with it. That is, one could say that pain plays a fairly unique, role in our lives anyway, and that the presence of enough other similarities to the standard perceptual situation exist (as the last two chapters have shown), not to be overly concerned over this deviation. One possible appeal along the first line is to hallucinatory or perceptual objects. These are perceived in some sense and do seem to have the desired ontological relationship. In so far as they can be said to exist their existence depends upon their perception. For example, assume a man commits a murder. Afterwards he constantly feels the blood on his hands and compulsively tries again and again to wash it off, but succeeds only after he confesses. Here he suffers a persistent tactual hallucination. He can correctly say, "For months and months afterwards I felt the blood on my hands. I tried again and again to wash it off, but still continued to feel it there." This person is using 'feel' correctly. Nonetheless the situation can be appropriately described in other terms. We can say that he thinks he feels blood, but there is none -70-there, or that he feels nonexistent blood on his hands. He has a tactual experience which is the same as one has when one does feel blood on one's hands, but in this case there is no blood to correspond to the experience. This is not the way we talk in the case of pain. We do not suggest that the person who feels pain only thinks he feels pain or feels nonexistent pain. These sorts of locutions are reserved for those fairly rare situations which have been discussed above wherein a person is mistaken about his pains. The example removes some uniqueness from the situation, perhaps, but does not provide as strong a parallel as one would like. There is a stronger parallel possible. One can bring up the other tactual sensations which were discussed in the first chapter. One might then want to argue that just as pains are found not to be liable to much of the argumentation which Ryle ejt aj_ have given, so also sensations of heat and pressure are not liable. If so, one would then have added several other items to the class of things sharing pain's relationship with feeling. These facts, when added to the second mentioned tendency, that of not being particular ly surprised at the uniqueness, suggest that the correctness of the one claim is simply not enough to demonstrate Ryle's general claim. We are thus once again back at the starting gate, at -71 -least so far as having succeeded in giving answers to any of my three original questions goes. This is because the argument to this point has been primarily negative. Ryle was arguing against the position that pains are perceived, and my argument has been directed at showing the inconclusiveness of his arguments. But this does not mean that the opposite position has been shown. I have only allowed the possibility of continuing to argue for it. In the next sections I shall in fact argue that Ryle's answer to my first central question was correct, even though his reasons were not conclusive. I shall show that a further analysis of the concept of pain, which focuses upon the parallels of pain and tactual sensation and upon other locutions substitutable for 'feel pain', provides the needed conclusive evidence. The person who would answer "yes" to my first question has been misled by focussing upon too narrow a part of our pain language. The model which I shall develop will allow me both to do justice to the varied language of everyday discourse about pain and also to answer my initial questions. In the last chapter I shall as well answer Berkeley's attempt to force conceptual revision. -72-Chapter 4: Pains as Feelings In the following I shall be arguing for what can be called an adverbial analysis of pain. In grammatical terms what I shall be arguing is that in.the 1ocution 'feel pain ' , 1 pain'''functions as a cognate accusative. Rather than being a direct object it modifies the verb, and thus acts as an adverb. I shall thus be arguing that 'pain' in 'feel pain' serves to tell what kind of an experience I am having rather than to tell what it is that is the object or cause of my experi ence. I shall proceed by arguing the following points. First, situations in which one describes one's pains can be seen to parallel situations in which one describes one's tactual sensations. Second, descriptions of tactual sensa tions do not represent hedges upon perceptual claims about the world. They rather involve a change of attention from the world to one's own experience. Third, the analysis of pains themselves can be given in terms parallel to that of tactual sensations. Having completed this I shall be able to explain why 'feel' in 'feel pain' does not function per ceptually. This will not, however, force a negative answer to my second basic question. Just as there are perceptual claims paralleling statements of tactual sensation, so too are there perceptual claims which parallel statements of pain sensa--73-tion. I shall thus be able to give a qualified "yes" to the question of whether I am perceiving anything when I am feeling pain. I shall obviously also have gone some way at this point in showing the extent of the parallelism of pain and touch. Let us begin with an experimental situation similar to 2 the one presented by Ebersole. I have volunteered for a series of experiments with Professor Young. He puts up a screen so that I cannot see what he is doing to my hand. He then stimulates the hand in various ways, and I report what is going on. I shall go through several series of examples, with my responses being elicited by slightly different questions from Professor Young. (1) Question: "When you feel something tell me what you feel." (a) He pokes my hand with his fountain pen. "I feel something poking my hand." (b) He jabs me with a pin 1. I shall make the claim of parallel perceptual and sensation statements for tactual perception only. It is often taken for granted that all senses work the same way. While this may in the end turn out to be correct, it cannot be assumed. Several of Wittgenstein's comments in Zettel (Basil Black-well, 1967, p. 84e) are appropriate here. 474. We call seeing, hearing, ... sense perception. There are analogies and connexions between these concepts; these are our justification for taking them together. 475. It can, then, be asked: what kind of connexions and analogies exist between seeing and touching? Between seeing and smelling? Etc. 476. And if we ask this, then the senses at once shift further apart from one another than they seemed to be at first sight. Such an investigation needs to be carried out. It is however beyond my present purposes. 2. See the section Feeling and Having in Chapter Two of the present work. -74-"Ow, damn it." "What did you feel?" "I felt something jab me." (c) He gives me a healthy electric shock. I say pained and rude things. "What was it you felt?" "I felt a very sharp pain run up my arm." Perhaps: "I didn't volun teer to be electrocuted." (2) Question: "Tell me if you notice something." (a) (Let the events here parallel those in (1) ). "Now." "What?11 "You're poking me with something." (b) "Ouch!" "What did you notice?" "Something sharp jabbed me." (c) "IS# etc." "Notice something?" "Don't be absurd," or "'Notice' is not the word for it." "Well, what happened?" "You shocked me with something" or "You gave me a very pain ful shock." In both (1) and (2) I begin by saying I feel something doing something to me. I am primarily concerned with the external stimulus. By (c), however, I am concerned with what goes on within myself. My attention becomes more directed "inward" toward my feeling, rather than "outward" toward that which I am feeling. This might be because of the immediate interest we have in painful feelings. On the other hand, it might be because of using the word 'shock1. Shocks do not exist outside of one's body. It is possible to describe the situation otherwise. (1) (d) Professor Young plays a strong electric field upon my arm. "Ouch, etc." "What did you feel?" Would I say, -75-"I felt a very strong electric field played upon my arm?" Perhaps if I were a physicist or an electrical engineer I would do so. But this would be because they are more familiar with the means of causing shocks. (1) (e) Professor Young rams a needle under my finger nail. I am convulsed with pain and scream. If not tied down I would jump up and hit him. As it is he must somehow cajole me back to calmness. "What did you feel?" "You must be joking." "No, I'm not interested in your behaviour, only in what you say." "I felt an excruciating pain in my hand." If I am calm enough afterward not to be mainly concerned with the pain, I might say, "I felt something rammed under my fingernail." Here the experimental situation and the types of questions asked function so as to elicit statements about the external events and objects I feel . When the experiments get too vigorous I begin to talk instead of what is going on within my body. When I calm down I refer externally again. The difference is whether the intensity of the sensations force me to be concerned with them rather than with what is bringing them about. Let us have Professor Young direct my attention "inward". (3) (a) He pushes his pen against my palm. "I feel something pushing against my palm." "No, this time don't tell me about the objects you feel." "Oh- Well, I feel a pressure on my hand." "That's better." (b) He puts an -76-infra-red bulb near my hand. "I feel a warmth on the hand." (c) He sticks me with a pin. I'm ready for it and do not let it bother me. "I felt a painful jab." (d) He gives me a jolt of electricity. "I felt a painful shock." It can be argued that Professor Young's directions have not succeeded, for I have only become more general. In (a), for example, although I do not say something is pushing on my hand, I still say that there is a pressure being exerted on my hand, and that I feel this pressure. It is possible, however, to elaborate the situation so that I could make use of the same locution, yet not be making the sort of claim argued for here. 3 Let us complicate the (a)'s. (4a) I feel the pen pushing against my hand. "I feel something pushing against my hand." Professor Young takes the barrier away, and I see that there is nothing there. Somehow he has caused me to feel pressure seemingly without touching my hand. (i) "My hand surely does feel as though something were pushing against it." (4b) "I feel pressure on my hand." Again the barrier is removed and I see that there is nothing exerting a pressure. At this point I might repeat my statement, (ii) "I still feel pressure on my hand." Or I might modify it. (iii) "My hand feels as though there were pressure on it." 3. (4a) parallels (la). (4b) parallels (3a). -77-Depending upon how the rest of the situation is filled out, (i) and (iii) can be taken two ways, while (ii) can be taken three ways. I may be expressing a conflict between my sense of sight and my sense of touch. In this case I might be using (i i) to over-ride any claim based on vision, and still insist that there is a pressure being put on my hand. But to try and do this by means of the present locution would be to make my claim very obscurely. This strong claim would best be expressed in more detail. E.G., "I don't care what it looks like. I can feel pressure, and know you are applying it some way." But given the conflict I may be making a much more hesitant assertion about the pressure. Any of the three state ments might be a very guarded claim that there is pressure being applied. That is, the listener might fill out my hedged assertion as, "It looks as though no pressure is being applied, but it feels as though it is, and I tend to think (or wonder whether) there is pressure being applied." While I would allow that with the proper tones of voice and so on this interpretation would be the correct one, I submit that it would not be the only usual or proper way to fill out the situation. In the present case my look around would almost certainly satisfy me that there is no pressure being applied, and my statements would not be taken as expressing any claim, even hesitant to the point of neutrality, -78-about the existence of a pressure. Any such implication could be cancelled by adding a phrase such as, "though of course I realize there is no pressure on it". If these locutions often carry such an implication, the circumstances of the present example would do the same job of cancellation as would this sort of rider clause in other circumstances. Accordingly, these forms of locution need not always express a claim or a hedged claim about perceived events or objects. In the present circumstances, they might well be expressing some tendency to such a claim. That is to say these statements do embody an economical expression of the counterfactual , "If it were not for the fact that I realize no pressure is being exerted on my hand, I would claim that there was." Thus, I might have said to Professor Young, "You really had me fooled. Even now it is hard to believe there is nothing there." The locutions of the present circumstances, however, do more than this. They also embody a reference to my tactual feelings, or, which is the same thing, my tactual sensations. They provide the basis for the counterfactual , which is the fact that I have the same sensation in my hand as I have when pressure is applied to it. Any continuing tendency to want to say that there is a physical pressure being applied has a basis. I can explain any continuing inclination on my part by means of the sensation which I continue to have. -79-Now, one must distinguish between the reference to one's tactual sensations and the hedges used to qualify ones claim about the state of the world. A standard sort of position is that statements like 'My hand feels as though there were pressure upon it' are used only as hedges. However, as the above shows, these statements can also function as reports and descriptions of one's tactual sensation. Further, in proper circumstances the non-hedge uses ('I feel pressure on my hand') can also serve as descriptions of one's sensa tions. Professor Young has at this point succeeded in turning my attention away from the world and toward myself. He could have directed my attention to the same thing in example (3) by specifically asking me to describe my sensations. (5) (a) Professor Young pushes something against my palm. "I feel something pressing against my palm." "No, this time don't tell me about the objects you feel. Tell me about the sensations you have." "Oh. Well, I have a sensa tion of pressure on my palm." "That's better." My reply at (3), "I feel a pressure on my hand", was possible to take as either asserting that there was pressure being exerted on my hand or that I had sensations of pressure. It accordingly became necessary to specify the direction of my attention. The present locution serves to do that without further specifi cation. The other examples can go as follows: (b) He puts an infra-red bulb near my hand. "I feel a sensation of warmth -80-on my hand." (c) He sticks me with a pin. "That was a painful jabbing sensation." (d) He gives me a jolt of electricity. In this case a parallel locution is awkward. I could say, "I felt a painful sensation, as of being shocked." I could also report having a pain and specify its kind, "I felt a bad pain, like one has when one is shocked." At this point the reference to my sensations becomes a straight forward reference to my pains. This suggests a strong similarity between having sensations of pressure and feeling pain. I shall develop this similarity. Tactual Sensation and Tactual Perception Before doing so, however, I wish to spend some time drawing out in more detail the relationship between tactual sensation and tactual perception. This will be useful for its own sake, but will also be helpful in working out the analysis of pain which I shall present. I shall begin by considering further experiments with Professor Young . Let us assume that during the course of his experiments Professor Young comes upon subject X, who reacts strangely. Professor Young sets up his barrier so that X cannot see X's hand, and tells him he is going to do various things to his hand. X is to tell him when he -81-notices something. Professor Young pokes X's hand with his pen. "You have poked my hand with a pen." "Now, don't tell me what it is I am doing to your hand, but rather describe your sensations when I do things to your hand." "I don't quite know what you mean. I can tell you what you are doing to my hand, but you seem to want me to tell you some thing more. I don't know what more there is to tell." Professor Young runs X through other parts of the experiment, by stimulating him in various ways. X correctly tells the professor what is being done to him, but shows no tendency to be able to describe any sensations in these cases. Here there are two possibilities. On the one hand, X may have the relevant sensations, but not have been taught to describe what goes on experiential1y when he discovers things by the sense of touch. On the other hand X may not have a sense of touch which is the same as that of the rest of humanity. Professor Young puts various sets of electrodes into X's brain so that he can stimulate him tactually at will. He now stimulates X's brain in a way which causes most people's right hand to feel as though an object is pushed against it. The barrier remains up. X does not know that it is his brain being stimulated. I shall consider two different replies of X's part. X says, "You are poking my hand with a pen." Professor Young takes the barrier away. After showing all that is going -82-on he convinces X that his hand is not being poked. X says, "But I still want to say that there is a pen poking my hand." "Why?" "Well, it is just like it is when there is a pen pushing against my hand." "What you should be saying is that it feels to you as though there is a pen poking against your hand." "Oh, I think I begin to see what you are after." X is now well on the way to being able to give the sensation descriptions which Professor Young desires. This possibility brings to light an important fact. A person can perceive things tactually, and be able to describe them on the basis of his feeling them, without paying any attention to or without even knowing what it would be to pay attention to his tactual sensations. There is no necessity of making any kind of inference from one's tactual sensations. One can refer to ones sensations, but this is often unnecessary, and perhaps even quite difficult. Though, to be sure, in many cases the difficulty may be that we are in situations wherein our tendency is to want to discover things about our environment. (b) The brain is stimulated, but X just sits there. Professor Young asks what is happening to X's hand. "Nothing." Doesn't it feel to you as though there were something pushing against your hand?" "What do you mean? There is nothing pushing against my hand." Professor Young stops the electrical stimulation and pushes his pen against X's hand. -83-"Now you are pushing your pen against my hand." "Isn't there anything similar between what is going on now and what was going on a moment ago?" "No, nothing." "How can you tell when I am pushing something against you?" "I don't know. It is just that when you do so I know that you are doing so. " The stimulation of the brain should have lead X to make an incorrect claim about a pressure on his hand. Since he did not there is strong evidence that X's knowledge is not founded on tactual perception. Normally we learn that certain things make us feel certain ways. If in the proper circumstances we feel these ways we will make mistaken claims about the world. We also learn to describe our feelings without making claims about the world. If we can make claims about the world in these circumstances without having the relevant feelings, as appears to be the case with X, then we would not say that we are perceiving by means of touch. X's case would be even more clearly deviant if he did have and was able to describe some sensations. If he could, for example, feel pain, a conversation such as the following might ensue upon Professor Young sticking a pin into him. "Ouch!" "Oh, you felt something then did you?" "Yes, when you stick me with a pin it hurts." "Well, is there anything in common between this (touching X's palm with the warm bowl of his pipe) and this (touching X's -84-palm with the flame of his lighter)?" "The latter hurts very much. But other than the pain there is no sensation." At this point one wants to say either that X does feel things, but for some strange reason cannot attend to his feelings, or that he does not perceive the flame, pipe, pin, and so on by the sense of touch. In order for tactual perception to be occurring there must be some sort of feeling 4 being had. Perhaps X, like God but unlike us, can gather knowledge about the world without having any perceptual experience of the world. This claim is in fact too strong, for it is arguable that having tactual sensations is not a necessary condition for talking of perception by touch. X could fulfill enough other characteristics of the sense of touch that we might still want to have the concept cover his case. First, "X-perception" may have the same requirements in terms of physical relationships to objects perceived as exist for normal tactual perception. Before being able to describe the texture of something his hand must have been in contact with it; before talking of an object's heat, he must be in the relation to it which the rest of us would be in when we feel its heat. Second, he may be able to give all those but only those 4. Wittgenstein, Zettel , #472. "Place of feeling in the body: differentiates seeing and hearing from sense of pressure, temperature, taste and pain." -85-descriptions which can be given on the basis of feeling. When Professor Young pushes against X with his pen, X will be able to describe its physical characteristics as well but only as well as would someone skilled at discrimination by touch. He could, that is, describe its size, texture and temperature, but could not describe its colour. Finally, X1 nervous system might be found to be the same as is that of the rest of us. That is to say he might have the same perceptual apparatus. Given these similarities, it might be decided that there is sufficient similarity to extend the concept of tactual perception to cover his case. While example (b) can easily be extended to meet the first two of these conditions it is more difficult to make i meet the last condition, for it is necessary to explain why the appropriate stimulation of the brain did not yield the proper mistaken claim. If we do not allow the claim, then X nervous system will be playing a different role than does that of the rest of humanity. Accordingly the example must modified so as to allow X to make a mistake. But in this case it becomes necessary to explain his mistake without making an appeal to tactual experience. The problem, then, is to keep from reducing the modified example (b) to that of (a), wherein it was found that X merely had to learn,how to refer to his tactual feelings. In example (a) X retracted h claim by saying it was "just like" it was when the pen was -86-pushing against his hand. This sort of retraction leads naturally to talk of tactual sensations. A description of the situation must be provided which does not so lead. Professor Young stimulates X's brain by means of his electrode. "You are pushing something against my hand." Professor Young shows X that there is nothing pushing against his hand. "Why did you claim that I was pushing something against your hand?" "I don't know; I just thought you were. I guess I had better be careful about such things in the future." Professor Young leaves the barrier down, and stimu lates X again, telling X just what he is doing. "I now have a tendency to say you are sticking me again, but I see that you are not. I guess such tendencies have to do with certain sorts of stimulations of my brain." "But don't you have any reason for wanting to make the claim?" "No, I do not. I realize that most people have certain feelings in these cases in addition to tingles, pains, and so on, but I simply do not." This is not completely satisfactory since one wishes to be able to explain X's tendency. A sufficient explanation, however, may come in terms of physiological occurrences (as the example suggests) rather than by any appeal to feelings of any sort. If so, it is proper to affirm that X's nervous system does play the same role as does ours, but still deny that he has sensations of touch. "X-perception" is then seen -87-to have much in common with tactual perception, so that it might be decided that it should fall under that concept. The extension, however, comes because of a fairly complete analogy and not because of an identity. The concept of touch, when applied to X, will not be applied in quite the same sense as when it is applied to the rest of us. This shows that a discussion of tactual sensation is pertinent to an explication of touch. It does not show just what role is played by tactual sensation, nor is it here my purpose to work this question out. Example (a) above demonstrated . that it is not necessary to have any notion of tactual sensations in order to make and report perceptual discrimination by means of touch. On the other hand it also provided an example of the tendency to fall back upon statements of tactual sensation when one is either clearly mistaken or in doubt about one's tactual perception ("My hand certainly feels as though a pin were sticking in it.") Example (b) added to this by showing the difficulty in explain ing certain sorts of mistakes without bringing in tactual experience. References to sensation thus help to explain tendencies to make perceptual claims or to explain mistaken perceptual claims which one has made. I must emphasize that this is not to claim that tactual sensations play the role of incorrigible bases from -88-which we infer perceptual claims. First, as (a) shows,there need be no inference. Second, I should want to argue, just as I did with pains, that distracted attention, psychological disturbances, "mind-sets", and so on, can lead to mistakes with respect to one's tactual sensations. The torturer example (Chap. 3) provides one example. The prisoner had a sensation of cold on his back, occasioned by the application of the ice-cube, which he mistakenly took to be a pain in his back occasioned by the application of the red hot iron. One requires proper conditions in order to make careful scrutiny of one's tactual sensations just as one requires proper conditions in order to make careful observations of the world. It remains to say something about whether sensation statements can play a role in a perceptual situation wherein one is not hedging and in which the question of explaining a mistake does not arise. Normally their introductions would be inappropriate. Assume that the heating system of the building has been functioning improperly, so that the building has been uncomfortably cold. I am standing with my hand near the radiator. In this position I can say that I can feel that the radiator is getting warmer, or if I am a bit unsure of myself, that it feels as though the radiator is getting warmer. In this situation the concern of the group involved is with the warmth of the radiator, and the appropriate statements on my part have to do with the radiator. I could, -89-however, say that I am experiencing sensations of warmth in my hand. Although the statement is true, it would not be appropriate to the situation at hand. Indeed since the situation is such that attention is drawn to the radiator, the statements about my sensations would probably be taken as a peculiar way of saying that the radiator is getting hotter, rather than as an attempt to describe my tactual experience without making any specific references to the world. In order to ensure the appropriateness of my refer ence to my tactual sensations I should have to change the linguistic context in which the utterance is made. This would be difficult in the present circumstances, since everyone is so closely concerned with the functioning of the radiator. Perhaps someone could talk about the chills which are running up his spine, which could lead to a general discussion of the various feelings which we are having. With discussion thus turned in this direction I could then appropriately mention the feelings of warmth which I am having in my hands. Of course, given the context of the example, the attention of the group would probably be drawn once again to the radiator, since the occurrence of the warm feeling in my hand would be immediately connected with the return of heat to the radiator. The discussion of sensations is directed towards the experience itself rather than the cause of that experience. Unless the linguistic situation can be -90-directed to that cause sensation statements are inappropriate. Even then they either produce confusion or serve as inefficient ways of referring to that cause. This last, however, brings up the question of whether or not some form of sensation statement could be used rather than the more economical verb 'feels'. Let me develop this in terms of the above example. I claim that the radiator is getting hotter. Someone asks me how I know, and I reply that I can feel it getting hotter. Might I not also be able to reply that I am having sensations of heat, which are caused by the heat of the radiator? Or, to put the case more strongly, when I make the first claim am I only making the second in a more economical way? Does, that is, 'feel x' mean something like 'have tactual sensations of a certain kind which are caused by x1? This suggestion has an initial plausibility. The statement 'I am feeling the heat of the radiator' will normally be true only if the statement 'I am having sensa tions of heat which are caused by the heat of the radiator' is true. But the former is not just an abbreviation of the latter. Consider a situation in which I make a mistake. Professor Young stimulates my brain by means of his electrodes so that it feels as if he is pushing something against my hand. I make the appropriate mistake and he shows me that I was wrong. To explain my erroneous claim I could say that it certainly felt as though something were pushing -91-against me, which on the present hypothesis would come to saying that I had sensations the same as those which are caused by something pushing against my hand. After Professor Young has explained to me what has happened, I could give a description of this using references to my sensations; e.g., I had sensations of pressure in my hand which were caused by the electrical stimulation of my brain. But the appropriateness of this description produces a major difficulty for the present hypothesis. For if 'feel x1 is an economical substitution for 'have tactual sensations of a certain kind which are caused by x', it would in the present case be proper to say that I felt the electrical stimulation of my brain. This is an entirely inappropriate statement. To insist upon this result is to attempt a conceptual revision and not to explicate the concept of tactual perception. At the very least the causal story will have to be further complicated. Feeling x will have to be a special case of having tactual sensations caused by x. It will be neces sary to specify just what sorts of causal roles are allowable for the object of the feeling. The explication must also be made to account for those diverse components of the sense of touch brought up earlier; i.e., the conditions under which 5 the claims can be made, the apparatus of perception, etc. 5. See pp. 80-81. -92-The final result would be a most complex statement or set of statements. Further, any such analysis would have to provide a different genesis and explication of the notion of tactual sensation than I have here provided. My explication .of statements of tactual sensation has been developed primarily by comparing and contrasting them with statements of tactual perception. But if the notion of tactual perception is to be made out in terms of tactual sensation then the latter notion must be developed independently of the former. I do not know whether such an analysis could be made out. More important for my present purposes, I do not think it would seriously upset my present argument if it were made out. My argument has been to show the different functions that statements of tactual sensation and statements of tactual perception have, while also admitting intimate connexions between the two notions. An independent derivation of the notion of tactual sensation would not disturb those different functions. Accordingly I shall now return to the discussion of pain, show the important conceptual similarities of sensations of pain and tactual sensation, and then analyse the relationship that exists between pain and perception. Feeling Pain At this point the first two parts of this chapter's programme have been completed. The first showed situations -93-wherein references to one's pain paralleled references to one's tactual sensations. The second part showed the legitimacy of talking of tactual sensations when divorced from the percep tual situation. It also showed that they are not therefore irrelevant to that situation. Now I must argue for the appropriate analysis of pain. This is to show that pains are properly discussible as kinds of feeling rather than as the objects of feeling. It must first be demonstrated that in calling pain a sensation comparable to tactual sensations I am proposing an analysis along these lines. This is necessitated by the fact that no one would deny that pain should be called a sensation, but it is still thought necessary to show that 'feel' in 'feel pain1 is not used perceptually. A goodly part of chapters 2 and 3 was dedicated to showing that many of the standard arguments against this perceptual view do not succeed. The one sense in which there can be no unfelt pains embarrassed, but did not overthrow this view. Accordingly, calling pains sensations does not automatically provide an alternative position to the earlier one. However, the analysis of tactual sensations here provided does not allow them to play the role of perceptual object. As was seen in several places (such as interpretation (a) of the experiments with X) it is possible to say, for example, either "My hand feels as though something were -94-pushing against it" or "I have sensations of pressure (with appropriate further specification if necessary) in my hand." The use of the one locution carries the same information as does the use of the other. In the first case one is not referring to the object of any perception, so it would be inappropriate to suggest that in the second case one was. These locutions serve to allow the isolation of one important part of a situation of tactual perception, the feeling component. They allow for the description of one's experience as opposed to the objects or qualities which are the object of that experience. To now make this experience the object of a further but similar experience, which would be the case if tactual sensations were allowed the ontological status of perceptual objects, is to fall into confusion and to defeat the purpose of introducing the terminology in the first place. It also leads to an uncomfortable if not vicious regress. For one could provide a description of the experience of the experience, the experience of the experience of the experience, and so on. It is far better to halt the process at the beginning by recognising that to have a tactual sensation is not to perceive something but to feel a certain way. Given this, it is necessary to see whether the descriptions of pains can come at all to the same thing. The parallels shown between pain and tactual sensation suggest strongly that they do so. The difficulty with immediately -95-accepting this solution stems from the fact that 'pain' is regularly and properly used as the direct object of 'feel1. This suggests a similarity to tactual perception rather than to tactual sensations. I shall overcome this difficulty by showing that there are other locutions equally useable in describing pain and readily substitutable for 'feel pain' which do not carry any such suggestion. Substitutions for 'Feel Pain' As was shown in Chapter Two the possibility of being able to substitute 'have' for 'feel' in ascriptions of pain does not provide the necessary break from perception. The difference is at most a matter of degree, and can do no more than leave pains in a more dependent ontological status than other objects of tactual perception. On the other hand this fact of substitutibi1ity does not overthrow the view of pains as feelings. In appropriate situations it is not incorrect to speak, for example, of feeling sensations of heat in one!s hand. In the experiments with Professor Young there was no linguistic impropriety in going from 'have' to 'feel' when describing sensations. In these cases the uses of 'feel' were cued by the respondent being asked to describe what he felt, but so too are many of the uses of 'feel' with 'pain1. The decision, then, will lie in other possibilities of substitution. -96-Progress can be made by returning to the 'it feels as though' locution which was seen to play a most important role in the description of tactual sensation. It is often observed that pain descriptions are commonly given by reference to the 6 various external and organic causes of pain. In such cases it is possible to substitute the 'feels as though' idiom for the adjectival constructions. For example, a pain in the hand might be described by using the adjective 'stinging' or by saying that ones hand feels as though it had brushed against some nettles. Again, if I wish to describe the pain of a sore throat to my doctor I can tell him that when I swallow it feels as though something is sticking me. In these cases the descriptions of the pain are tied to certain external events because those events provide clear cases wherein one does experience the kind of pain one is wishing to identify. We can teach what a stinging pain is by referring to cases wherein one's pain typically does have the quality in which we are here interested. Similarly the descriptions of one's tactual sensa tions will most easily be given by referring to those situations in which the sort of sensation one wishes to describe is 6. See for example, Bruce Aune, Knowledge, Mind and Nature (Random House, 1967) pp. 130-131; Gilbert Ryle, The  Concept of Mind, p. 193; and D.M. Armstrong, Bodily  Sensations, p. 119. For a criticism of extending this fact to cover all pain descriptions see Roger Trigg, Pain and Emotion (Oxford, 1971), p. 24 ff. -97-experienced. To a large extent our descriptions of pain and our descriptions of tactual sensation bear the same relation to our interaction with our environment. Unfortunately, this straight-forward description by reference to the causes of the particular pains does not capture all of the various pain descriptions. A pricking pain may be explained in terms of the way one feels when one gets a shot in the arm, but a stabbing pain cannot be explained in terms of the way one feels when one is stabbed. Most of us have very little idea of what it would feel like to be stabbed, yet are perfectly conversant with the phrase 'stabbing pain'. There is an analogy to being stabbed here, but there is no causal relationship. One would not explain 'stabbing pain' to someone by telling him it is a feeling such as one has when one is stabbed, nor would one teach anyone the meaning of the term by stabbing him. Rather we would tell him that it is an intense pain of sudden onset and brief duration which travels quickly through a rela tively small portion of the body. The analogy is with the sudden sharp blows which make up the majority of stabs. It is because there is a set of pains which have just this quality about them that we avail ourselves of this descripti 7. Contra Ryle and Armstrong, but pro Trigg (see preceding footnote for page references). -98-not because of any actual experiences of being stabbed. There is accordingly a large number of pains (stabbing pains, shooting pains, "waxing and waning" pains) which are not linked to specific sorts of interaction with the environ ment, and another large number (stinging pains,) pricking pains, "the pains of neuritus and neuralgia") which are so linked. The latter sorts can be conveniently classified by references to the commonly associated causes. The former will also be described in terms drawn from the external 9 world, but the terms will not be drawn in the same way. 8. Armstrong comes close to this position, but seems to fall into confusion. He rejects the analysis of a stabbing pain as "the sort of pain characteristically caused by a stab." But he goes on to say: "Now how can we differentiate a stabbing .pain from other sorts of pain (except with respect to suddenness of onset and intensity) save by using the adjective 'stabbing'? It does not seem possible to do so, and this suggests that the stabbing-quality of the pain is intrinsic to the pain. When we feel a stabbing pain, we feel as if we are being stabbed. (And I think it is irrelevant to point out that often, in actual stabbings, very little is felt at all.) (Bodily Sensations, p. 120)." It is just the suddenness of onset and intensity that does distinguish this sort of pain, and which characterizes its "intrinsic quality". To have a stabbing pain is to hurt in a certain way, the description of which is facilitated by the analogy to stabbing. But this is most certainly not to say that it is to feel as if one were being stabbed. The quality of the pain is intrinsic, all right, since it provides the identifying characteristics, but this does not mean that there is any direct relationship to actual acts of stabbing. We might indeed imagine the experience of being stabbed partly in terms of having a stabbing pain in the relevant part of our body. Here, however, the epistemol ogical priority" belongs to the feeling of pain rather than to the imaginary feeling of being stabbed. 9. This is pointed out by Trigg, op. c i t. , p. 24. As he states, descriptive links may well have to be drawn from the exter nal world in order for communication to take place, but this hardly requires they all be drawn in the same manner. -99-In many of these cases it will instead be done by means of analogies between the qualities of the pains themselves and those of events in the external world. Because of the method of drawing the descriptions, however, the 'feels as if idiom will not provide a translation of the adjectival description. Accordingly, the use of this translation as a means of doing away with unwanted ontological implications and provid ing a parallelism to tactual sensation does not quite succeed. The reason for the divergence can be straight-fowardly explained. Tactual sensations occur most often in perceptual contexts, and so are describable by means of the 'feels as if locution. Some sorts of pains do not have the same links to the world, so are not parallely described. On the other hand some sorts of pains can be so described. Given this the tendency to think of 'feel' as used perceptually when one feels pains of this sub-class loses its force. And since pains of both sub-classes are grouped because of basic qualitative similarities we have a strong case for rejecting this tendency for the more recalcitrant sub-class as well. Still, it is preferable to find a means of transla ting all of the various pain descriptions. The answer to 1 0 this desire is to be found in the use of the verb 'hurts'. 10. There are also the intransitive verbs 'pains' and 'aches'. 'Pains' is not, I believe, an important element in common usage. 'Ache' only works for a certain kind of pain. Accordingly, I shall concentrate here upon the more general verb 'hurt'. -100-If I moan and clutch my stomach an observer can describe what is wrong by saying either that I am having severe stomach pain or that my stomach hurts badly. In both cases he is talking about the way my stomach feels, and not about my 11 perception of something in my stomach. This possibility of substitution has been investigated by Bruce Aune, who recognizes and attempts to overcome its 12 most immediate difficulty. The difficulty is in trying to use 'hurts' as a translation when the pain is described in any detail. We can describe pains as throbbing, stabbing and so on, but cannot say our arms hurt throbbingly or stabbingly. Such adverbs just do not exist. Because of this the suggested translation appears to fail. Aune's way out is by means of the above mentioned fact that pain descriptions involve figurative uses of the relevant adjectives. We describe our feelings of pain by reference to their causes, by analogy with other sorts of things or by reference to the effects our pains have on us. To have a stinging pain is to have the sort of feeling one has when stung by nettles and bees. To have a pulsating pain is to have 11. Though he might also say that I am having stomach cramps. He would then be telling about my pain and about the cause of the pain. There is a hint here that to have a pain of the appropriate sort in the appropriate place is consid erably involved in the notion of "feeling a cramp". If this is correct, it is sufficient to note that this further possibility of reference adds to rather than detracts from the present thesis. 12. Bruce Aune, Knowledge, Mind and Nature, pp. 130-131. -101-a feeling which rises and falls in spurts as does a current. To have a blinding pain is to have one which tends "to make us close our eyes in anguish, thus producing momentary darkness". Given this the difficulty in translation can be expl a i ned. The reason that we cannot advance a plausible translation of 'I have a stinging pain in my foot' into the 'My foot pained me stingingly' idiom is not that the former captures something that could not be captured by the latter, but only that since there is no such word as 'stingingly' in common use, there is no such word to be used figuratively. The ontological point is not, however , affected.by this linguistic accident, for we could obviously introduce such words as 'stingingly' into our current idiom so as to guarantee the above mentioned translation. But once we under stand the force of such words as 'stingingly' in 'I have a stinging pain in my foot', once we understand that they are used to describe the quality of the hurt, we have all the grounds we need for saying that pains are not conceived as peculiar objects but as "qualities": the difficulty of finding a straight forward translation is thus of no signif icance whatever.13 With some slight modification of language it becomes possible to give all pain descriptions in terms which do not suggest that we should take pains to be some peculiar sort of objects, but rather that we should consider them certain sort of feeli ngs. 13. Ibid. , p. 131 . -102-Although I am doubtful about the accuracy of some of his explanatory analogies for the descriptions of pains, I 14 do think that,.i n sum Aune ' s point is-well taken. The only point at which caution must be observed is in the reading of the word 1qualities1. Aune is not suggesting that in having shown that pains are not objects he has thereby shown that they should be thought of as on a par with such things as the secondary qualities. The difficulty with this analogy is that while secondary qualities are not to be counted as objects in their own right, they do count as objects of perception. On the one hand, any individual instance of a secondary quality owes whatever numerical identity it can be said to have to that which it qualifies, as might turn out to be the case for pain under the present analysis. On the other hand secondary qualities are known by means of seeing, touching and so on. This is hardly a proper conclusion of the present argument, since its point is showing that to feel a pain is not to perceive the pain, but to have a certain sort of experience. Pains are experiences which share certain 14. The degree of figurativeness is not always what Aune would suggest. To claim to have a stinging pain is not to literally claim that some object (the pain) is performing some action (the stinging), as the case in claiming that one felt a bee sting one. But while pains do not do something to us similar to what a bee does, they do really sting. To have a stinging pain is to feel the way one does when one is stung. With the translation carried out the metaphor disappears. -103-qualities. If the word 1 qua!ity1 does any more work than this, it is to emphasize thatfeeling a pain is having one's experience "qualified" in a certain way, and thus to stress the tie between a person and his pain. It should be clear that the secondary quality view cannot be derived from the quoted argument. What has been shown is the possibility of translating uses of 'feel pain' wherein 'pain' is a direct object into statements which have intransitive verbs. This suggests that 'pain' in the former cases is a cognate accusative. The translation possibility was demonstrated by showing that the adjectives descriptive of pains can be thought of as giving the quality of the hurt. This does not show that the hurt itself is a quality. Cer tainly the pain is not a quality of the hurting. It i_s_ the hurting, and as such has certain qualities about it. To say that it in turn is a quality of something else requires a further argument. Aune's argument helps to clarify the status of pain, but it does not quite settle the question of parallelism to tactual sensation. So far it has been shown that attribu tions of pain can be handled by the use of the verb 'hurts' with various adverbs specifying the specific qualities of the hurting. In so arguing I have provided a specific verb for the rephrasing of references to sensations of pain. But there are no comparable verbs for references to the various tactual -104-sensations. That is, while I can say that my arm hurts, I cannot say that my arm "pressures" or "warms". If pains and tactual sensations are as similar as I have claimed, some explanation of this difference should be given. The reason is easily discovered by looking at the difference in direction of attention which is natural to situations wherein we feel pain or have tactual sensations. The having of pain is most of the time connected with the malfunctioning of our bodies or with less than desirable interaction with the environment. The sensation of pain is also one which we intensely dislike and which captures our attention, usually quite against our will. Further, in situations wherein we are hurt by something in the world we do not tend to continue the action that lead to being hurt in order to find out about that something. We are usually concerned with the cause of the pain only to the extent of getting away from it. Finally, because of their importance for diagnosis and because of the morbid interest they engen der, these feelings get discussed with regularity. When we have tactual sensations, however, our attention is much more of the time directed toward finding things out about the world, and accordingly not especially directed toward the feelings. Indeed, as the first part of this chapter showed, our attention is usually directed toward our tactual sensations only with some difficulty. -105-Further, we do not have the immediate dislike fortactual sensations that we have for pain. Given this difference of attention and interest in the two situations, it is only to be expected that we should have a more developed descriptive vocabulary for pain than is the case for tactual sensations, including special verbs which have not been necessary for referring to tactual sensations. If for some reason the other sensations should come to play the same kind of role, a similar vocabulary could and would be developed to deal with them. For example, if there were many things in the world which produced the sort of sensation we have when we run our fingernails down a blackboard, then we would undoubtedly have a special name for that sensation, and might well have a special verb for use when we have that sen sation. The parallelism is accordingly made out. With the success of this last section of argument the tendency to argue for the perceptual view of 'feel pain' disappears. One tends this way because of concentrating upon the 'feel pain' locution exclusively. But when one sees that the 'feel pain' locution can be successfully paraphrased into the 'feels as if and 'hurts' locutions one is no longer bewitched by the language. Pain references parallel references to tactual sensation rather than references to tactual perception, and accordingly play a categorically different linguistic role. -1 06-Th e answer to the first central question of the thesis is clearly negative. When we have a pain we do not perceive the pain. This does not, however, mean that when we have a pain we do not perceive anything. As was seen in the experiments with Professor Young, situations in which one can refer to one's tactual sensations are also often situations in which one could instead make perceptual claims about the world. When he poked me with his fountain pen, I could talk about the sensations of pressure caused by the poking or say that he was poking me with the pen. Similarly, when he poked me with the pin, I could refer to the pain thus brought about, or tell him that he was poking me with a pin. Accordingly, though references to pain do not serve the function of making any sort of perceptual claims, it is often the case that when one is in a position to refer to one's pain, one is also in the position to make statements about the causes of one's pain. I shall look at the extent and detailedness possible for these statements in the final chapter. I shall also examine there the last question which remains about the parallelism of pain and touch. This is the question as to why we do not in addition to pain-sensations conceptualize pains or the quality pain as existing externally, as we do in the case of heat. First, however, I wish to undertake some further analysis in order to "cement" the adverbial analysis as being consistent with everyday pain discourse. -107-Chapter 5: The Identity of Pains In order for the adverbial analysis of pains to be rounded off it is appropriate to show that it can satisfac torily account for two of the problems concerning pain which have been of recent concern, namely pain identity and pain location. Can this analysis allow for the day-to-day use of 'the same pain', and for the fact that we locate our pains in various parts of our bodies? The first problem requires sorting out the identity conditions for pains which are to be found in everyday dis course. A.convenient starting place is the position held by Normal Malcolm which rejects the possibility of ascribing 1 numerical identity to pain at all. Malcolm takes his cue from Wittgenstein's remark in the Philosophical Investiga tions, "In so far as it makes sense to say that my pain is the same as his, it is also possible for us both to have the 2 same pain." Malcolm interprets Wittgenstein as here imply ing "that there is not some sense of the expression 'same 3 pain', such that you and I cannot have the same pain." He argues that while one can usefully make a distinction between numerical and qualitative, or as Malcolm says, 1. Malcolm, "The Privacy of Experience", pp. 138-145. 2. Ludwig Wittgenstein, Philosophical Investigations, # 253. Quoted by Malcolm on p. 138. 3. Mai colm, op. ci t. , p. 138. -108-"generic" identity for material objects, no such distinction is applicable to sensations or other "contents of conscious-4 ness." In references to pain we have a use only for generic i denti ty. Malcolm rests his case upon the claims (1) that descriptions of sensations provide the criteria of identity of sensations and (2) that personal pronouns do not have to 5 enter into the descriptions of sensations. If I have a certain kind of back pain, and,you have a certain kind of back pain, both being in the same part of the back and answering to the same description, then we both have the same back pain. There is np further question possible about whether your pain is different from mine, or a different pain than mine. To think there is such a question is "exactly like thinking that because there are two areas 4. Malcolm makes out the distinction by showing the possible ambiguity in 'After dinner Petersen and Hansen smoked the same cigar'. This might mean that there was only one cigar which was passed back and forth (numerical identity) or that each smoked a different cigar, but of the same size, shape, colour, brand, etc. (generic identity). To break the reader free from insistence upon the general possibility of the distinction in all cases he draws attention to styles, colours, opinions, and sudden thought, wherein only qualitative identity plays a role. For exam ple, to say that x and y are the same colour is to say that the colour of x and the colour of y are of the same hue, intensity, etc. To say that x is still the same colour as it was before is only to say that the colour has not faded, or that even though repainted the same colour paint was used. Saying that two people wore the same colour dress does not give rise to the ambiguity as saying that they wore the same dress. 5. Ibid. , p. 143. -109-6 coloured a certain shade there must be two colours." In a footnote he mentions that there is a possible "secondary" sense of 'same headache' used when we are differentiating between different cases or occurrences of the same headache and the same case of the same headache.. But we would object, he says, to talking of different people feeling the same case of headache. We would not know what it is to feel a 7 case of a headache, this phrase not being English. A counter-example suggested by Malcolm's footnote will show that he has overstated his case. Assume that I have been complaining of a headache all morning. My wife goes out to do the shopping and is gone for several hours. Upon her return she finds me lying down with an ice-pack on my.head. She asks, "Is your head hurting again?" I reply, "No, it's the same headache. It never went away." Here the word 'same' indicates numerical and not qualita tive identity, for what I am doing is re-identifying my continuing headache and ruling out the suggestion that my original headache has gone away and another has afflicted me in its place. There is no necessity of the headache remaining qualitatively the same. Headaches can become more or less 6. Ibid. , p. 157. 7. Ibid. -110-intense. They can change from throbbing headaches to dull headaches. Some such qualitative change might well have occurred in the present example, with no change in the Q interchange being required. The description of what I presently feel would not be the same as the description of what I was feeling earlier, so we do not have a case meeting Malcolm's criteria of identity, but we do have a proper and normal ascription of identity. The ascription of identity over a period of time and in the face of qualitative change rests upon the existence of a sense of numerical identity for pains. Malcolm cannot conjure this usage away by substitu ting such phrases as "same case of the same headache" or by referring to it as a "secondary" sense of 'same headache'. Certainly this is not the sense appropriate to the context with which Malcolm is most concerned, but it is one which does regularly and legitimately occur. The introduction of his longer paraphrase does not help his case. Talk of in stances and cases suggests the correlative introduction of kinds. Malcolm's favored contexts are themselves susceptible to paraphrase in terms of kinds of pain. If Malcolm and I both have throbbing pains in our shoulders, it might be said that we have the same pain, or it might 8. Though as a good wife she would undoubtedly go on to ask whether it feels the same or has gotten better. Here the qualitative sense of the word is being used, to which the change in quality is quite relevant. -in -be said that we have the same kind of pains. We thus can equally chide Malcolm for using 'same shoulder-ache' when what he "means" is 'same kind of shoulder-ache'. The different paraphrases help to point out different senses of 'same pain'; they do not do away with one of the senses. Malcolm has gone wrong through his exclusive concen tration upon two different people having the same pain. He is quite correct in his observation that the only sense in which two people can have the same pain is that of qualitative 9 identity. Our concept of pain does not allow pains to move about from one person to another in the way that a cigar can 10 be handed from one person to another. For a fair number of years there has been a headache which moves from one philoso phy graduate student to another, staying only long enough for the thesis to be written and then moving on to the next victim. But this joke only continues because of the aberrant view of pain which it presents. 9. Malcolm is correct to attack Ayer when Ayer says, "Head aches are private; it does not make sense to say that several people are feeling the same headache, "(Malcolm, "Privacy of Experience", p. 139.). Ayer here is erring in the opposite direction from Malcolm and thinking only in terms of numerical identity. Malcolm's difficulty is that he also puts all his eggs in one basket. 10. One can imagine such a model. When the witch-doctor rubs the painful bodily part so as to draw the pain into a rock, the object imagery is being drawn upon. But our normal talk of pain does not square with this strict analogy to material objects. -112-Wittgenstein's remark should be taken as a grammatical remark setting the parameters for the locution 'same pain1 when it is used in interpersonal comparison. Whereas there is not some sense of the expression 'same pain' such that you and I cannot have the same pain, there is a sense of the expression 'same pain' which cannot be legitimately applied in any dis cussion of you and I either having or not having the same pain. The fact that the notion of numerical identity as applied to material objects requires the possibility of using it in the context of interpersonal sharing only shows that pains are not to be thought of in the same terms as material objects. It does not show that the notion of numerical identity cannot be applied to pains. If the statement 'No one else can have my pains' is a remark about the conditions required for indi viduating pains it is not entirely objectionable. Its sin is in the possible suggestion that pains are to be identified similarly to material objects, but are objects falling in a special category. The Adverbial Analysis and Identity The difficulty which now arises is to explain how it is that numerical identity is applicable to pain, but that such usage is inappropriate in the context of interpersonal sharing. The adverbial analysis provides an explanation fairly easily. The key lies in the part of P.F. Strawson's attack on the -113-no-ownership theorists in Individuals which Malcolm quotes as characteristic of the position he wishes to attack. Strawson says that if we think of the requirements of identifying reference in speech to parti cular states of consciousness or private experiences, we see that such particulars cannot be thus identifyingly referred to except as the states or experi ences of some identified person. States or experiences, one might say, owe their identity as particulars to the identity of the person whose states or experiences they are.11 When I identify my headache I am not describing a certain condition, but am referring to the event of my being in that condition. Accordingly, the identification of me as the experiencer is necessary to the making of the identifying reference. Many people can have the same headache, in that their head can all hurt in precisely similar ways. However, to talk of any particular occurrence of that headache is to talk of an event in the psychological life of an individual. It is only with respect to these feelings, with their necessary tie to the experiencer, that the notion of numerical identity has any use. Accordingly, thinking of pains on the adverbial model allows us to understand an application of numberical identity of pain which is necessarily disconnected from the possibility of transferability. 11. Malcolm, "The Privacy of Experience," p. 139. -114-It remains to be seen whether my suggested analysis fits with the various criteria which are used in the day to day ascriptions of numerical identity to pains. Roger Trigg brings forward three considerations, no one alone of which he thinks sufficient: position, temporal continuity and qualita-12 tive similarity. Qualitative identity is clearly not sufficient. One's arm can hurt in precisely the same way on two periods years apart and one's arm can hurt just as does one's leg. In neither case would there be any tendency to speak of these as numerically the same sensations. Nor is qualitative identity a necessary condition, as was shown by the earlier example of my headache. The possibility of qualitative change can be accommodated by the adverbial analysis. An episode of painfulness in a leg, for example, can gradually change from one of a throbbing hurting to one of a burning hurt without perforce being separated into two instances of pain. The case of temporal continuity is more interesting. It is obviously not a sufficient condition for assertions of numerical identity. If I have a dull ache in my stomach which dies down at the time a stabbing pain behind my eye begins, there is no temptation to assert a numerical identity between 13 the two sensations. 12. Trigg, Pain and Emotion, pp. 91-94. 13. Though if there were a regular alternation of the two loca tions and qualities then this might be thought of as a kind of pain of which the present occurrence is an instance. -115-The possibility of its being a necessary condition is not so easily dispatched. Trigg wants to say that it is necessary, though he recognizes that the concept of continuity must be extended in some way in order to account for cases of d i stracti on. It certainly seems natural to talk of having the same headache as I had half an hour ago, even if I have forgotten about it for most of that time. Clearly we are talking of numerical identity. We do not just mean that my present headache feels the same as the one I had half-an-hour ago. We want to say that it i_s the same. 14 Though he does not discuss it,.any such extension must also account for a headache which returns when the A.S.A. tablets which one has taken wear off. Here we do not have another headache which feels the same as the first; we say that the first headache has come back. Aspirin does not cause us to forget or be distracted from the pain; it is an analgetic, bringing about an absence of pain. There might be a tendency in the example of forgetting or being distracted to think that one really was in pain the whole time, but was not fully aware of the pain. Such cases do occur. A familiar similar example is that of a person absently scratching an itch on his wrist while his attention is otherwise engaged, only realizing that his wrist is itching when his attention is called to it. Or again there is the 14. Trigg, op. ci t. , p. 95. -116-football player whose complete concentration is on getting to the goal line, and although he is dragging his injured leg behind him, grimacing with every step, he simply fails to realize until later that his leg was hurting at the time. Even if this sort of account is plausible in some instances, it cannot do for the example of the aspirin tempo rarily alleviating a headache. When the aspirin works I no longer have any pain. It is not just that I fail to notice it. None of the various methods of ascribing pain can be used. The question as to whether I have pain is conclusively answerable in the negative. Even in those cases wherein distraction enters in the present sort of explanation is often implausible. If the distraction leads to my not only not being conscious of pain, but also no longer exhibiting any sort of pain behaviour, then there is no ground for asserting that there is any hurting present. Accordingly, no straight-forward version of temporal continuity will provide a necessary condition for ascribing numerical identity. In order to cover such cases, and yet to keep some sort of temporal continuity necessary, Trigg "extends" the concept by introducing counter-factual,conditionals. If I forget the pain from a blister because of my absorption in a game, my only grounds for talking of 'forgetting it', and for not saying that it had gone, would be that I felt it again the minute my concentration on the game gave -117-way. I could say that I would have felt it but for my absorption, and this gives us ground for thinking of some kind of con tinuity. It is however a logical continuity, rather than that of some shadowy pain which is 'there' in some way, but just happens not to be felt.15 The return of the headache as the effect of the aspirin wears off is similar. To say that it is the same headache is to recognize that the aspirin has worn off, and.thus that the headache would have been continuous over this time if the aspirin had not worked (barring of course, other intervening conditions). Trigg's general claim then, is that a pain which has not been continuously felt can be reidentified only if one can identify intervening conditions whose beginning was coincident with no longer feeling the pain and whose ending was coincident with once more feeling the pain. Tem poral continuity in fact or the counter-factual presumption of temporal continuity is necessary for numerical reidenti-f i c a t i o n . I agree with Trigg that explanations of this sort are usually made when numerical identity is claimed when temporal continuity is not present. I am not certain they must always be made. In the case of my wife coming home after shopping I might well have told her that the headache had gone away, but that it had then come back again. This could have been said 15. Ibid. -118-whether or not there were intervening conditions, such as the efficacious use of medication. But I am not confident that this must be to reidentify the same instance of head ache. I don't think that it would have made any difference to the situation had I said instead that my headache had gone away but that now I had another one. There is a ten dency to use the language of numerical identity in such cases but also a tendency to use that of qualitative iden tity. At this point the sharp distinction between the two sorts of identity does not find a place in our everyday use of the concept. Because of this it hardly seems profitable to push one case as against the other. Trigg's explanation therefore seems to be adequate for those cases where the distinotion could make any sort of difference. The question which now arises is whether or not the adverbial analysis can be made out given the possibility of the ascription of numerical identity in the absence of straight-forward temporal continuity. The difficulty is to account for events retaining numerical identity in the absence of temporal continuity. If I take aspirin for my headache, after which it comes back, then I have the same instance of headache, but it is not clear that I have the same instance of hurting. For I might well say, "My head was hurting so I took some aspirin, but now it has started -119-to hurt again." While it is numerically the same instance of pain, it is not clear that it is numerically the same instance of hurting. To solve this problem, it is necessary to examine the criteria for counting instances of hurting. Consider first the case of a man going out for his evening jog. He jogs for ten minutes, then stops for a moment to catch his breath, then begins to jog again. The implied inter pretation of the above paragraph would have us say that therefore there were two instances of jogging. However, we would count it as such in only very circumscribed cir cumstances, such as wanting to know how many occasions of non-stop jogging the man did during the evening. But normally, we would say that this constituted only one oc casion of jogging. He would be said to have gone jogging twice only if he would have come home, rested a while, maybe had a shower, etc., then got dressed and went out jogging again. Thus, it is possible for an activity to be intermittent and yet counted as numerically the same acti vi ty. Having seen that the uninterruptedness of an activity is not the sine qua non for reidentifying that activity, let us see whether the same thing could be the case for pain. That this could be the case can be seen by once again looking at the example of my wife returning after -1 20-an absence and finding me with a hurting head. Assume once more that she asks me if my head is still hurting. Depend ing upon the circumstances all of the following replies seem possible and perfectly natural: (a) "Yes it is. It hasn't let up at all.", (b) Yes it is. It got a little better for a while, but now it is just like it was. (c) Yes it is. It stopped for a little while, but now it is hurting even more than before. (d) Yes it is. The aspirin I took helped a lot, but now it has become worse than ever. In (a) and (b) the hurting has been continuous, in (c) and (d) it has not. Yet in all the cases it is appropriate to answer the posed question in the affirm ative; in all cases only one occurrence of hurting is counted. Brief breaks in the continuity of the hurting do not need to count against the numerical identity of the hurting before the break and after the break. Undoubtedly there comes a time at which we can no longer identify a present state as part of the same instance of hurting. An affirmative reply to the above question when my wife had been away for several days would be appropriate only if my head had hurt fairly continuously over that period of time. If my head had stopped hurting an hour after she left and only begun to hurt again a short -121 -while before she returned, the reply of that form would be mis leading. While something approaching continuity is required, continuity in the sense of being uninterrupted in time is not. Accordingly, the non-necessity of straight-forward temporal continuity for the making of numerical ascriptions of identity is not inconsistent with the adverbial analysis of pain. This leaves only the consideration of the criterion of location. It is clear that this is not a sufficient condition for the ascription of identity. The pain of burning my hand today is counted separately from the pain of sticking it with a pin yesterday. More interestingly, neither does it seem to 16 be a necessary condition. Most people are familiar with the phenomenon of a pain creeping from one place to another. In "Excedrin Headache Number " the pain starts in the back of the neck and works its way up, ending behind the eyes. By relying upon temporal continuity and qualitative resemblance 17 it is possible to talk of this as one instance of pain. 16. Trigg, op. cit. , p. 92. 17. Trigg's suggestion (p. 92) that this would occur only if the change of location were gradual is accurate in most cases. Assume I have a severe burning pain in my hand, which suddenly stops hurting, with immediately thereupon my knee beginning to feel as though it were being burned, this ceasing and being followed by a burning pain in the neck. It seems more natural to say that three occurrences of exactly similar pain took place, and not that one occurrence of pain took place in three different locations. On the other hand, should this pattern be repeated a few times, it would seem quite possible to talk of one painful episode, the description of which would be more complex than is usually necessary. -122-It is not immediately obvious that the adverbial analysis can account for this possibility. My tendency has been to make identifying references to instances of hurting in terms of the hurting of specific bodily parts; e.g., "My haiidhurts," "My head aches." But this form of locution is not amenable to expressing changes of location. Numerical identity is preserved when one says that the pain in the lower back has gradually moved up to the upper back. However, to say that one's lower back pained one earlier, but that one's upper back hurts now does not convey the same informa tion. It suggests two different occurrences of pain rather than two parts of the same occurrence. The problem this presents is not insurmountable. First, it is a matter of convenience that the term referring to the bodily part be the subject of the sentence. Rather than saying, "My stomach hurts," I can say, "I hurt in my stomach." It is the person and not the bodily part who is the subject of the pain. Second, the importance of analogy to our pain descriptions should be recalled. When I have a stomach I might say, "I feel as though there were a pin sticking in my stomach." In this case I am of course not claiming that-there is a pin in my stomach. Rather I am describing the experience I am having by analogy to the experience of being stuck in the stomach with a pin. My giving of a location for my pain, then does not involve the absurdity of claiming -123-that the pain experience is located in a certain part of my body, but rather provides a further specification of my 18 experience. My experience is similar to that of being stuck with a pin, but more than this, it is similar to that of being stuck with a pin in the stomach. Given this reminder, it is clear that the adverbial model can account for the movement of a pain, though the paraphrase of 'pain' results in loss of linguistic economy. In order to describe the pain creeping up my back, I should have to know the quality of the pain, then find an appropri ate analogy, and finally describe the experience in terms of the analogy. Thus, I could say something like, "I felt as though there were a hot poker in the small of my back, which has very gradually moved up and now seems to be at the upper most part of the back." But, of course, as long as the more economical language does not mislead one into thinking of a pain as a perceptual object, rather than as an experiential state of a person, there is no need to use the clumsier terminology. In accounting for the possibility of pains moving about, the adverbial model has accounted for the location of pain in general. To feel a pain in one's hand is not as such 18. Although conducted in rather different terms, this argu ment owes much to Aune's consideration of pain location (Knowledge, Mind and Nature, pp. 13 2-3). -124-to perceive something going on in one's hand, but rather to feel as though something were going on there. It is not necessary that there be any object or event at the place of the pain or to find at that place the cause of the pain. When our hand hurts we will often want to make a perceptual claim that there is something physically wrong with our hand, but this sort of claim is detachable from our self-ascription of pain, just as are statements of tactual perception and of tactual sensation. Questions of phantom limbs are handled neatly by this analysis. For my phantom limb to hurt is for me to have the relevant sensations. This is clearly not to say that there is something wrong with my non-existent limb. I can have a feeling as of a pin sticking into my phantom limb without having to make any existence claims about that limb. Physi ologists may indeed discover that the cause of the pain is the condition of the nerve endings in the stump, but this 19 gives no reason to locate the pain itself in the stump. I feel as though something were being done to my (non-existent) hand and not as though something were being done to the stump. 19. In fact it appears that this is only part of the story. Ronald Melzack ("The Perception of Pain", Scientific  Ameri can (Feb., 1961), pp.41-49) reports that in the case of phantom limb pain, "All the evidence suggests that the primary focus of physiological disturbance lies in the central nervous system itself." The disturbance to the central nervous system is such that even slight touches near the damaged part, and sometimes on other regions of the body, will trigger phantom limb pain. This evidence is quite consistent with my adverbial model, and also helps lessen the tendency to think of pain location in terms of the location of the pain's cause. -125-The adverbial model thus does justice to the questions of pain identity and pain location. Given this in addition to the considerations of the previous chapter, it can be concluded that it is a viable model for under standing pain discourse. I shall now spend the last chapter completing the remaining parts of the answers to my second and third central questions. -126-Chapter 6: External i zi rig Pain I shall begin by looking once more at Berkeley's dilemma. It will be recalled that the argument was that the situations in which one felt pain (e.g. reaching out and sticking ones finger with a pin) and those in which one felt heat (e.g. putting one's hand on a warm radiator) are very similar. Yet we seem to internalize the pain and externalize the heat, producing an asymmetry which he felt undefensible. Now, in light of the argument of the thesis to this point, it is readily seen that the asymmetry is not so radical as Berkeley would have us believe. When we refer to our pain we are making a statement paralleling the statement we might make about our tactual sensation. But we could also make statements about the world, which would parallel statements about the heat of the radiator. When Professor Young stuck me with his pin, I was able to tell him that he was sticking me with a pin. Accordingly, al though the situations in which we feel heat are indeed parallel to those in which we feel pain, this does not mean that talk of pain and heat come to the.same thing. To try to make it do so is to confuse the two different uses of language which are appropriate in these sorts of situations. -127-Still , Berkeley's argument raises a question which is interesting to investigate. We talk of both sensations of heat and sensations of pain. But when we use the word 'heat' alone, we no longer talk about sensations, but rather are talking about a quality of the external world. On the other hand, when we use 'pain' alone we are still only referring to sensations of a certain sort. It is necessary to explain why there is no externalization of pain as well; to explain why, that is, that we do not ascribe pains as qualities, activities or objects in the external world. Feeling Another's Pain I shall first consider an argument which if success ful would not only provide the explanation, but would also overthrow the adverbial analysis, for it would lead to the conclusion that pains are perceived. This is the argument presented by Don Locke in his article, "The Privacy of Pains.There he argues that it is only because of contingent facts about our nervous systems that we feel only our own pains and not also the pains of others. The contin gency lies in the fact that "our nervous systems are not connected in such a way that whenever you have a toothache I 1. Don Locke, "The Privacy of Pains," Analys i s (March, 1 964), pp. 142-152. -128-can feel it, not in my tooth, but in yours." If we were able to "plug into" one another's nervous systems, he claims, we would come to think of pains as public objects. We might, for example, think of ourselves as being able to "plug into" another's nervous system in order to feel the pain. "My word how unpleasant" we say, and hastily unplug again. Plugging into another in order to feel his pain would be like opening my eyes in order to see what is before me, putting the sauce on my tongue in order to taste it, or putting my nose near the rose in order to smell it. If this, as is logically possible were the case we would naturally come to think of pains as public objects, in just the way that sounds and smells or even tables and chairs are.2 Locke does not spell out in detail why this would be the case, but I think he is mainly relying upon the notion that pains would in these situations become the objects of feeling of innumerable people rather than of just one person, and would thus come to have the characteristics of any other public objects of perception. Let me develop an example. My leg is damaged in an automobile collision. The doctor, a hardened man, plugs into my nervous system, thus coming to "feel my pain". By various probing he can more quickly make an appropriate diagnosis of the injury. Indeed, I may have passed out from the shock of the injury. 2. Ibid. , p. 149. -129-But by "hooking up" below the brain the doctor can feel where the pain is without having to awaken me and ask; he can feel my pain even though I cannot. In general, accordingly, the possibilities of unfelt pain, misfelt pain, and hallucinated pain would become quite straight-forward. Pain itself would come to be identified with certain bodily conditions of which the person in whose body the conditions are located or anyone else may or may not be aware. This individual wi11 , nonetheless , maintain his special concern with and access to his pain, since he is constantly "wired into" his body, while others have to "plug in". Locke's notion is enticing, but the enticement should not be allowed to succeed. I do not wish to debate the possibility of science providing "nerve termi nals" and the subsequent possibility of one person talking of feeling pain in another's arm. My argument is rather that this need not lead to the perceptual object model of pain as Locke assumes. The situation can be understood in terms of the adverbial model, without any objectifi-cation of pain taking place. It is therefore not some contingent peculiarity of our nervous system with respect to our having pain which accounts for pain not being exter nalized. -1 30-Assume that I have "plugged into" your nervous system with the result that the nerve impulses from your right arm are carried both to your brain and to mine. Your arm is painfully stimulated. Now, while it would be natural.enough to say I am feeling a pain in your arm, I could as naturally and equivalently say such things as "I have a pain in your arm," "Your arm hurts me" or, perhaps better in the circumstances than this, "I hurt in your arm." By using any of these locutions I could inform someone what was wrong with me. There need be no implication that we are mutually perceiving some one object or event, or at least that anything we are perceiving mutually is your pain. At best we are both feeling the damage to your arm. But in counting instances of pain there is no need to count only one --both of us are in pain. The.unusual circumstance is not that two people are in pain and feeling the same pain, but that we both locate our pain in numerically the same pi ace,.that we both focus our attention and concern upon the same place. Some complication of the situation's description further reduces the temptation to follow Locke's line. -131-A thumbscrew is applied to your right thumb. Your thumb hurts, all right, but you find you can stand it. I, on the other hand, find the pain virtually unbearable. We are not, then feeling the same pain, but rather feeling different pains in the same place. Given the proper nerve connections, spanking my son might hurt me more than it does him. The doctor gives you a shot of pain-killer. It is of the sort which causes a blockage of signals at a certain point in your nervous system. You sigh with relief. But our nerve paths diverge before the point of blockage. I continue to cry. "Do you feel pain in your hand any longer?" the doctor asks. "No," you reply. "But you must," he says, "because Bassford is still rubbing your hand and crying." "That's Bassford's problem, and not mine. Give him a shot." I am not feeling your pain. You no longer have a pain. Pains are identified and counted in terms of the experiences of individuals first, and only secondarily in terms or where they are located. Locke's interesting science-fiction does not lead us to make pains into public events. We need not say -132-that there is only one pain which you and I and anyone else who cares to "hook in" feels. Rather everyone who "hooks in" will feel pain (and depending upon the rest of their nervous equipment perhaps the same kind of pain) in the same place. It is not because of contin gent facts about our nervous systems that we feel only our own pains or that we fail to externalize pains the 3 way we do heat. Differences of Environmental Interaction The answer as to why we do not is to be found in the different modes of interaction with the environ ment ordinarily involved in feeling pain and feeling heat. The hard work of thinking of the guides to the answer has been done by H. P. Grice in his article, 3. It might be noted that the "plugging in" procedure could give us a check upon people's assertions that they are in pain, and upon the causes of the pain. Joe complains of hurting in his hand. I "plug in". I feel horrible pain. "You'd better get that hand to a doctor. Something is wrong with it." Or I feel no pain. In this case I might want to check the other criteria we have for pain avowals, such as seeing whether Hif;s other pain behavior really is genuine. Or I send him a neurologist to check out the rest of this nervous system, or to a psychia trist. While not absolutely disconfirming a claim to be in pain, this procedure could help locate the source of any pain there may be. -133-"Some Remarks about the Senses." Grice's procedure is to draw comparisons between pains and smells with the differences taken as suggesting why we talk of a sense of smell, but not of a sense of pain. He makes five main poi nts. 1. "Pains are not greatly variegated, except in intensity and location. Smells are." 2. "There is no one standard procedure for getting a pain: one can be cut, bumped, burnt, scraped, and so on. There is a standard procedure for smelling, namely, inhaling." 3. "Almost any type of object can inflict pain upon us, often in more than one way." 4. " ...if a source of pain moves away from a given place, persons arriving in this place after the removal do not get hurt. Smells, on the other hand, do linger in places, and so are 'detachable' from the material objects which are their source." 5. "Though pains do not linger in places, they do linger with individuals after the source of pain has been removed. In this again they are unli ke smelIs." I shall consider these points in turn. Rather than fixing on the comparison with smells, I shall consider them in comparison with heat and cold, this difference being the focus of the present problem. Grice's first point is not well taken. First, there is considerable qualitative variation among pains. The pains of neuritus and neuralgia are different in quality as 4. H. P. Grice, "Some Remarks about the Senses." All references here are to pages 134-135. -134-well as intensity from those produced by the doctor's needle or from that which comes from accidently stepping into scalding water. Second, even if the point were accurate about pains, it would as well cut against externalizing heat, since there are only differences of location and degree amongst the sensations of temperature. The second point, however, does suggest relevant differences. First consider temperature. There are fairly standard ways of coming to feel warmth and cold. If one wishes to find out the temperature of something all one has to do is to see that a bodily part is near to or in contact with that object. By putting oneself into contiguity with an object one can bring about sensations of temperature in the contiguous bodily part. Further, there is a definite relationship between the intensity of the sensation and the spatial proximity of the bodily part to the object. My sensations of warmth become more intense as I move my hand closer to the fire and less intense as I move it away. If there is any degree of intensity in the sensation when the body and the object are close together, the intensity will lessen as the distance apart increases. One might object that the sensations of temperature become less pronounced the longer the contiguity continues. After I have been holding my hand against a radiator for a while the radiator stops feeling as warm to the touch; I stop having sensations of warmth in my hand. In reply, -135-one should note what happens when I now touch this hand with my other hand. Whereas before this would have occasioned no particular temperature sensations in the other hand it now does. The suggestion which extrudes is that something is transmitted from the radiator to the hand, the presence of which can account for the difference. Temperature sensations are tied to a difference in quality between the sensing organ and the thing sensed. Which is to say that sensations of tem perature are normally occasioned by differences of temperature. Do we have such standard ways of coming to feel pain? Sometimes all that is requires is to have physical contact with an object in order to feel pain, such as when we touch an electrically charged fence. But usually the feeling of pain is dependent upon the manner in which we come into contact with the object. It is not sufficient that I come close or touch the knife; I must push against its edge or hit my head with its handle. It is an unfortunate fact that in this world we come to feel pain in many different ways. If we are to allow that we have a sense by which the pain qualities of external objects are uniquely sensed, it is going to be much more complicated in operation than are our other senses. This hardly seems sufficiently useful to justify describing the world in such a manner. It can be objected that there is actually one general description of the way in which we come to feel pain. What -136-is required is not that we inhale, or get into or near to physical contact, but that we produce tissue damage in our 5 bodies. We do not ordinarily feel pain unless the doctor s needle destroys some tissue or the scalding water burns our foot. While this is true, it does not further an attempt to make pain into a quality of external objects. It suggests a link between feeling pain and having tissue damage, but not between the pain and some quality of the object which does the damaging. While this suggests important links between pain and perception, it does not suggest the sorts of links being sought here. I shall return to this point later in the chapter, but at present would like to consider other reasons militating against externalizing pain in the way presently being considered. Grice's third point suggests relevant differences, but is not really accurate. The fact that almost any object can inflict pain upon us is not overly important. Almost any object will serve to provide us with sensations of temperature as well. It is important that objects (such as the knife above) can inflict pain in more than one way, but this fact 5. George Pitcher claims that physiological evidence supports this position in his article "Pain Perception," (Philosophical  Review, July, 1970), p.371. This is supported by the article on "Pain", in the Encyclopaedia Britannia: "Evidence indicates that the adequate stimulus for pain is tissue damage, slight and altogether reversible perhaps, but nevertheless damage." However, this view is disputed by Henry K. Beecher (op. ci t.). who mentions the fact that "light pressure on a sensory nerve can be exquisitely painful", and that this probably does not produce tissue damage (p.10), and also that great tissue injury is often not adequate stimulus for the production of pain, using mainly examples of badly wounded but unsedated (conti nued page 137 ) -137-is by itself more relevant to the above discussion of ways of getting a pain. More important to note here is the fact that the same object, in inflicting pain in different ways, can also inflict different sorts of pain. If we smell a rose several times in a short period it will smell pretty much the same each time. If we feel the warmth of the radiator several times in a short period it will feel pretty much the same each time. But if we are hurt by a knife several times over a short period it may feel quite different each time. We will have one sort of pain when stabbed, another when sliced, another when struck with the hilt, etc. Accordingly, if we external ize pain as a quality then virtually all objects will have to be considered to have a plethora of pain-qualities. If only certain things caused us pain, and then only in certain ways, there might be reason to describe them in terms of pain-qualities. But given the present state of the world, there is no purpose served by so speaking. Grice's fourth point, about the lingering of smells, is not significant. While pains do not linger in places after their source is removed, in the sense that new arrivals in the room will not feel pain, no more do most visual or tactual properties linger. If the stove is removed from the room a person arriving thereafter will not feel the warmth of the stove. He will feel the warmth of the room. And while its smell may linger on, its hardness and blackness (footnote 5 continued) and calm soldiers, who felt no pain (pp.98, 161, 164). One accordingly should say only that we usually have tissue damage when we have pain. -138-w i11 not. Nor is the lingering of pain in individuals after the removal of the source of pain a relevant difference as such. I can continue to smell onions, or to taste onions when the onions are gone. My hand feels warm even after the cause of its warmth is removed. But what is important is.a difference with respect to the lingerings, which is explainable in a way similar to the communicabi1ity of the radiator's heat. The smell of the onions, the taste of the onions, and the warmth of the radiator are all somehow transferred from the original source. If someone else should come into the room they can sniff and recognize that I now smell of onions. The response to a kiss is a complaint that I taste of onions. If my hand is touched the warmth of my hand can be felt (or at least the warmth of the one hand as opposed to the other). But the pain which lingers cannot be so transferred. If the fire has caused me to have a burning pain in the hand, another person will not then get a burning pain by touching my hand. Even if the person should hurt his hand by touching mine, the pain would probably be more like the one such as one gets from blows rather than from burns. In the other cases talk of linger ing suggests the transferability of qualities; in the case of pain it does not. There is accordingly not the same pressure to or reason for externalizing pain. While Grice's points need some developing the develop ment does show that there is good reason not to make pain -139-into a quality or object and also does give some reasons why it is appropriate to do so with heat. With this result, what ever perplexity Berkeley's argument continued to engender at the start of the chapter is resolved. In order to complete the programme of the thesis it is necessary only to look at the sorts of perceptual claims which can be appropriately made when one has pain. I shall end this essay by doing so. Pain and Perception Let me begin with the suggestion which appeared in the previous section of this chapter, that when we have pain we normally have some sort of perceptual awareness of damage to 6 a bodily part. In most cases when one complains that his hand is hurting there is an implication that something is physically wrong with his hand. Given the indisputable fact that pain functions to warn of present or imminent bodily damage, and that we normally feel pain only in the presence of bodily damage, this implication is not untoward. Further, work byphysio 1 ogists has apparently shown that the stimulus normally necessary to evoke pain sensations must be "strong engough to cause at least temporary damage to the affected tissue." 6. This is at least close to the position taken by Pitcher in "Pain Perception". However, I am not quite sure what he means when he says he exposes "direct -real ism", nor am I comfortable with the development of the argument by analogy with glimpses. The comments in this chapter are accordingly independently developed. 7. Ferdinand and Sauerbruch and Hans Wenke, Pain: Its Meaning and  Significance, Edward Fitzgerald, translator (.London, 1963 )., p.30. The word 'normally' must be emphasized here. As pointed out earlier (footnote 5, this chapter), some researchers challenge the universality of the tie between tissue damage and pain. -140-One apparently does not normally feel pain unless one's tissues are damaged or very much threatened with damage (though this should not be taken as suggesting that there has to be any very major damage involved). Finally, if one turns one's attention from the word 'pain' to the word 'hurt' one will find that this word does describe a certain sort of physical condition. When Tiger Tim has his jaw broken by Gentleman Jim he suffers a hurt. This is true even if his jaw does not hurt at the time. Presumably, when he awakens in the locker room he will have a badly aching jaw. One thus has a tendency to say that just as there is a correspondence between having sensations of heat and the presence of heat, so too is there a correspondence between being in pain and the presence of bodily hurt. This suggestion is insightful, but it overly restricts the perceptual information which can be gained from being hurt. To restrict the perception to the disordered bodily state sounds like restricting smells to being properties of gases. I take it that a necessary requirement for smelling something is that some of its molecules lodge in the nose. But this hardly means we should therefore say we smell the molecules and not the substance. It. may also be a necessary requirement for feeling pain that some tissue damage should take place. But this does not mean we should be restricted to saying that what one perceives in such cases is just -141-ti ssue damage. In the experiments with Professor Young in Chapter 4 it was seen that in many situations we could talk either about our pain or about the cause of our pain. When he rammed a needle under my fingernail, I did not, after recovering, think it appropriate to say I felt damage to my tissues but rather that I felt something being rammed under my fingernail. Indeed, to say that I felt tissue damage in my finger would be to respond to the situation most inappropriately. In the present case the only even remotely appropriate answer is to say that I felt something damaging my finger. Nonetheless it is true that in these circumstances we do feel our tissue being damaged. This is parallel to those cases where we say we feel a pressure being applied to us. If we feel something pushing against us it is true that we are feeling a pressure. And if we feel a knife being stuck into us, it is true that we are feeling it damage tissue. But there is no need in either case to restrict ourselves to such general and unspecified claims. In most cases we can be much more detailed in our perceptual claims. I can say, for example, that I felt a pen being pushed against my palm, or that I felt a pin being forced under my fingernail. Very often the perceptual statement which could accompany the pain avowal cannot be very detailed. When we have a dull pain in the lower back often we are only able -142-to make the perceptual claim that something is wrong in our lower back. But other times such statements can be considerably stronger and more detailed. It is quite possible to imagine someone who pays much closer attention to the particular qualities of his pains than we do. A masochist of a scientific bent might learn to discriminate the gauge of a pin or the kind of a whip purely on the basis of the way he hurts when poked with the pin or lashed with the whip. When pain is a topic of conversation we tend because of the nature of the sensations to focus our description and discussion on them. When we are interested in their causes we rely usually upon the other senses. But we need not and do not always rely upon the other senses. Even without them we can make perceptual statements when we are hurt. Getting hurt does not provide a pleasant, convenient, or overly useful way of getting around in the world. But it does occasionally, and some might say all too often, provide a way of doing so. -143-BIBL IOGRAPHY Armstrong, D.M., Bodily Sensations (Routledge & Kegan Paul, London, 1962.) , A Materialist Theory of Mind (Routledge & Kegan Paul , 1 968). Aune, Bruce, Knowledge, Mind and Nature (Random House, 1967). Austin, J.L., Sense and Sensibilia (Oxford, 1962). Beecher, Henry K., Measurement of Subjective Responses (Oxford University Press, 1959). Baier, Kurt, "Pains", Australasian Journal of Philosophy, v. 40 (May, 1 962), pp. 1-23. Berkeley, George, Three Dialogues Between Hylas and Philonous in G.J. Warnock, ed., Berkeley: The Principles of  Human Knowledge, with Other Writings (William Collins Sons and Co. Ltd., London, 1962). Coval , S. , Scepticism and the First Person (Methuen & Co. Ltd., London , 1 966 ). Daniels, Charles B., "Colors and Sensations: or How to Define a Pain Ostensively," American Philosophical Quarterly v. 4, July, 1967, pp. 231-237. Ebersole, Frank B., "Feeling Eggs and Pains", Things We Know:  Fourteen Essays on Problems of Knowledge (University of Oregon Books, Oregon, 1967), pp. 105-125. Grice, H. P., "Some Remarks about the Senses: in R. J. Butler, ed., Analytical Philosophy, (Basil Blackwell, Oxford, 1962), pp. 133-153. Holborow, L. C, "Against Projecting Pains" Analysi s, v. 29 (1968-69), pp. 105-107. Locke, Don, "The Privacy of Experience", in Avrum Stroll, ed., Epistemology: New Essays in the Theory of Knowledge (Harper & Row, New York, 1967), pp. 129-158. Malcolm, Norman, Problems of Mind (Harper, 1971). , "Direct Perception", in his Knowledge and Certainty (Prentice-Hal 1 , 1 965 ). , Wittgenstein's Philosophical Investigations, in his Knowledge and Certainty (Prentice-Hall, 1965). -144-McKenzie, J. C, "The External i zati on of Pains", Analysis, v. 28 (1967-68), 180-193. Melzack, Ronald, "The Perception of Pain", Scientific American (February, 1961), pp. 41-49. Melzack, Ronald and Wall, Patrick D., "Pain Mechanisms: A New Theory", Science (1965, v. 150), pp. 971-978. , "Gate Control Theory of Pain" in A. Soulairac, J. Cahn and J. Charpentier, eds., Pain (London, 1 968), pp. 11-31. Pitcher, George, "McKenzie on Pains", Analysi s, v. 29. , "Pain Perception", The Philosophical Review, July, 1970, pp. 368-393. ~ Ryle, Gilbert, The Concept of Mind (Penguin, 1963). , "Feelings", The Philosophical Quarterly, vol. 1, pp. 193-205. , "Sensation", in Robert J. Swartz, ed., Percei vi ng, Sensing and Knowing (Doubleday, Garden City, New York, 1965), pp. 187-203. Sauerbruch, Ferdinand and Wenke, Hans, Pain: Its Meaning and  Significance, translated from the German by Edward Fitzgerald (London, 1963). Thompson, J.F., "Berkeley", in D.J. O'Connor, A Critical History of Western Philosophy (The Free Press, New York, 1 965 ) , pp. 236-252. Trigg, Roger, Pain and Emotion (Oxford, 1 970). White, A. R., Attention (Basil Blackwell, 1964). Wittgenstein, Ludwig, Philosophical Investigations, translated by G. E. M. Anscombe (Macmillan, 1968), Third Edition. , Zettel , edited by G.E.M. Anscombe arid G.H. Von Wright, translated by G.E.M. Anscombe (Blackwell, 1967). Wolff, Harold G., and Wolf, Stewart, "Pain" Encyclopaedia  Britannica, (Chicago, 1965) v. 17, pp. 30-32. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items