ARTICLES
In this article, I identify three ways in which Wittgenstein opposed an idea of epistemic asymmetry between the first person and the second-or-third person. Examining the questions of 1) absence of doubt about my own experience and uncertainty about the experiences of others, 2) ineffability of subjective experience and 3) immediacy of my knowledge of my own experience contrasted with my merely inferential knowledge about the experiences of others, I see Wittgenstein’s remarks about “inner and outer” as a many-faceted denial of the claim that people’s minds are in some deep way unknowable to others. These considerations also serve to clarify Wittgenstein’s relation to behaviorism.
Wittgenstein is undoubtedly an important philosopher to consider when tracing the history of the so-called problem of other minds. There was a boom of writings on the topic from philosophers of the analytic tradition after the middle of the 20th century, probably largely inspired by Wittgenstein’s remarks on solipsism and related matters. However, when Wittgenstein’s contribution to the problem of other minds is explicitly assessed, the central attention is usually given to rather narrowly limited points in his writings. He is applauded for seeing a conceptual problem in the place of the traditionally conceived skeptical problem of other minds (Avramides 2001). Sometimes his talk of “criteria” is seen as an attempt to answer a skeptic of other minds directly (Hyslop 1995). Michel ter Hark (1990, 1991) is an exception because he gives a detailed reading of Wittgenstein’s late philosophy of mind from the viewpoint of the other minds problem, and I generally agree with his account. However, I see Wittgenstein fighting against the idea that people’s minds are unknowable to others in several different contexts at once.
I suggest that Wittgenstein’s late writings about the theme of “inner and outer” are a holistic attempt to deny a significant epistemic asymmetry between first person and second-or-third person. By epistemic asymmetry I mean the assumption that each of us has a privileged epistemic access to his own mind (and to his own mind only), making our first-personal knowledge of our own minds better in quality, more immediate, or otherwise superior to any interpersonal knowledge.
I examine several contexts where Wittgenstein first identified a tendency to think that our knowledge of mental goings-on in other people is inferior to our first-personal knowledge, and then proceeded to criticize such a tendency. These contexts can be treated as variants of the problem of other minds. I also suggest how Wittgenstein’s late writings about psychology help to see his relation to behaviorism in the right light. Here I have drawn much inspiration from ter Hark (1991).
I will identify three ways in which there seems to be an asymmetry between a subject’s first-personal knowledge of his own mind and his knowledge of the minds of others. In the main body of this paper, I examine Wittgenstein’s handling of each of these topics in turn.
First, there is incorrigibility about my own experience, contrasted with some inevitable amount of uncertainty about the experiences of others. In feeling pain or seeing a patch of red, there is no room for doubt or mistake in my own case. In contrast, it is always at least in principle possible that another person only seems to undergo a certain experience, while actually he does not. I claim that Wittgenstein did not deny this; he admitted that at least some types of statements about our subjective experiences are incorrigible, but he attributed this incorrigibility to the logical role of those statements as avowals, not to a privileged access. He pointed out that certainty has different standards when we talk about the experiences of others, thus implying that a comparison between first-personal and third-personal knowledge here is misguided.
Second, there seems to be something ineffable about the fundamental qualities of private experiences. A characterization or description of an experience cannot fully disclose it to another person; the subject always knows more about his private experience than he can communicate. I claim that for Wittgenstein, such a view was based on an unrealistic view of what the human activity of “describing one’s experiences” is meant to accomplish, or needs to accomplish.
Third, each of us knows his own experience immediately, while we get to know the experiences of others through inferences based on their behavior. When one has an experience, the thing itself is present for the subject, but in observing someone else having an experience, what is present for observation is a piece of behavior which only suggests the presence of an “inner” experience. I claim that here Wittgenstein opposes the implication that seeing the experiences of others manifested in their bodily behavior is a second-best thing compared to some other, better way of being in touch with them.
My exposition of Wittgenstein’s treatment of the first form of epistemic asymmetry has two parts. First, I examine his account of subjective experience statements and their authority. Second, I show how he characterizes our attributions of experience to others and what he has to say about certainty and uncertainty in that context.
Wittgenstein repeatedly states that doubt about one’s own sense-experience is unintelligible. His favorite example is the impossibility of being wrong about whether one is in pain.
The truth is: it makes sense to say about other people that they doubt whether I am in pain; but not to say it about myself. (PI §246)
[Referring back to PI §283 (“Couldn’t I imagine having frightful pains and turning to stone while they lasted?”)]: Suppose I were in error and it was no longer pain? – But I can’t be in error here; it means nothing to doubt whether I am in pain! (PI §288)
The expressions used here are striking: it “makes no sense to say” that I doubt whether I am in pain; it “means nothing” to doubt my pains; “there is no such thing” as my being mistaken about them (LW II: 34-36, see also LW II: 30-31, 92). It is not merely the case that I never go wrong in attributing experiences to myself because I am in a better position to observe those experiences than anyone else. Rather, the claim is a stronger one: doubt is somehow “logically excluded” in this case. My subjective experience is an area where questions about my being right or wrong about the features of my experience are just not relevant; they are never raised and they need not be raised.
Even just in principle, why is doubt not possible? It can be thought that for Wittgenstein, it is a matter of what he calls “grammar”. This means that first-personal immunity to error is a norm upheld in our language. We are just not prepared to count anything as a mistake in the context of such first-personal statements, and we would consider first-personal expressions of doubt baffling:
[I]f anyone said ‘I do not know if what I have got is a pain or something else’, we should think something like, he does not know what the English word ‘pain’ means; and we should explain it to him. […] If he now said, for example: ‘Oh, I know what “pain” means; what I don’t know is whether this, that I have now, is pain’ – we should merely shake our heads and be forced to regard his words as a queer reaction which we have no idea what to do with. (PI §288)
But it is clearly not an arbitrary norm, adopted just because we have decided to trust people on these things. Wittgenstein’s private language argument gives reason to say that when someone applies a concept to her present subjective experience, there is no way, even for the subject herself, to make a distinction between a correct and an incorrect application of that concept in the absence of public and interpersonal criteria. The application of a concept like “pain” to one’s subjective experience is rather like a primitive reaction, which is not aptly described as “recognizing one’s sensation as one of pain”, because there are no criteria for telling the difference between a correct recognition and a misrecognition, outside of what the subject feels appropriate to say in that situation. This, in effect, collapses the distinction between appearance and reality in the case of present first-personal experience. Most importantly, Wittgenstein recognizes that the immunity to error of my first-personal verbal expressions of pain is not the result of my superior epistemic access to my pains, but rather a fundamental feature of how first-personal experience-talk works. Such talk is essentially subjective, in the sense that it does not rely on naming one’s sensations according to some objectively assessable criteria. Subjective criteria, on the other hand, are an oxymoron because “rules” that are only followed privately are not rules (as argued in Wittgenstein’s discussions about rule-following). This implies that while discussing sense-experiences, I can rightfully apply concepts like “pain” to my subjective experience without relying on any kind of criteria (PI §289). Thus, the situation is as Wittgenstein sums it up in PI §258:
[W]hatever is going to seem right to me is right. And that only means that here we can’t talk about ‘right’.
This is the reason why error is excluded in principle in these cases: the necessary context required for making a difference between success and error is not there. Thus, there remains the fundamental fact that first-personal statements of one’s experience are the undisputed starting point for the language-game of talking about subjective experiences (PI §290). Introducing objective criteria can make these statements revisable in some contexts, but they are always logically primary.
Subjective experience is an area where “how things are” and “how things seem to me to be” are collapsed together. That simple point guarantees that whenever a person gives an honest report of their experience, we don’t have to (and we as a matter of fact do not) take into account the possibility that such a report might contain a mistake. But it is essential that the report must be honest. There is a huge family of cases where we see another person saying and doing things that suggest, for example, being in pain, but we entertain doubt as to whether she really is in pain. In those cases, we do not suspect the person to be in error about her pain, but we suspect her of being insincere. And this threatens to introduce an unbridgeable gulf between my first-personal case, where the possibility of error about my pain is peculiarly absent, and the case of other people, where there always seems to be the possibility of pain-behavior occurring without the person feeling pain: the possibilities of pretending or play-acting.
The problem is not only about the pains and other sensations of fellow human beings; it is a problem about a whole range of mental phenomena. As well as play-acted sensations, there are faked emotions, insincerely stated beliefs, hidden intentions and concealed desires. Regarding all these, our epistemic relation to the minds of others seems to be marred with incurable uncertainty: there is always the possibility of the appearance being different than the reality, and we are never able to check what the reality in the other person’s realm of consciousness is like. This incurable uncertainty about whether the appearance given by others matches the reality of their inner experience is in the strongest possible contrast with the utter lack of appearance-reality distinction in first-personal experience. Is this not a fundamental kind of asymmetry between knowing oneself and knowing others?
Wittgenstein’s reply to this problem is twofold. First, he argues that the incurable uncertainty has no real place in people’s lives in practice; it exists merely in philosophical reflection, where it gives rise to the idea of skepticism about other minds. Second, he argues that insofar as there is some fundamental element of uncertainty in our attributions of mental states to others, this uncertainty should be viewed as an essential part of our relations to one another, and not as an epistemic shortcoming.
The basic point made by Wittgenstein about pretense is that pretending is a “complicated pattern” (e.g. LW II: 55) that has to be learned like any other sophisticated skill. There are natural expressions of sensations and other experiences that are in place long before anything like the ability to pretend makes an appearance. Apparently Wittgenstein is also here suggesting what he probably would call a “grammatical” point: mastering the concept of pretense, that is, being able to pretend and take some displays of behavior in others as cases of pretense, necessarily requires being able to take some displays of behavior in others as genuine expressions. Because pretense requires such a concept-mastery acquired through a complex interaction between human beings, the cases where pretense is even imaginable are actually rather limited. We never normally take into account the possibility that a newborn child might pretend, and for a good reason: not because we know newborn children to be honest, but because the prerequisites for anything to be called an act of pretense (or honesty) are not fulfilled in the case of the newborn (e.g. LW II: 39-40; PI II, xi: 194).
The idea that there is something blocking us from ever being certain of what goes on in others is connected with a false philosophical idea of “essentially inner events that no one else, in principle, could witness and which I am unable to reveal or describe to another person” (Moran 2001: 91). The truth is that “in countless cases” (LW II: 94) we are perfectly certain about the mental processes in someone else. This is clearly true whenever we recognize something as a universal natural expression of an experience occurring in normal circumstances. When seeing someone being burned by a flame and screaming, there is just no point in thinking: “of course there are always two possible cases; one of pain-behavior with pain-experience and one of pain-behavior without pain” (see LPE: 287). Doubting the authenticity of this situation would not have the normal consequences of doubt. It would rather be like trying to doubt that the future is connected with the present (see PI II, xi: 190).
So, in many occasions we are justifiably certain in our attributions of experiences to others. But the second part of Wittgenstein’s reply to these worries is to point out that certainty and uncertainty regarding the experiences of others are of a special type, which gets misrepresented if it is contrasted with, for example, mathematical certainty:
In mathematics a particular kind of evidence that can be clearly presented leaves no doubt open. That is not the way it is when we know that someone was glad.
There can’t be a long dispute in a court of law about whether a calculation has this or that result; but there certainly can be about whether someone was irritated or not.
But does it follow that one can know the one and not know the other? More likely what follows is that in the one case one almost always knows the decision, in the other, one frequently doesn’t. (LW II: 85)
Rather than saying that knowledge of other minds is inferior to knowledge of some other things, Wittgenstein characterizes knowledge of other minds as being of a different kind. Mathematical certainty is generally achieved through a definite procedure that is not controversial. But there is no definite procedure for assessing people’s reports and expressions of their sincerity; we cannot lay down anything resembling a proof here. In Wittgenstein’s terms, the “language-game” played by experience-ascriptions is altogether different from those played by mathematical concepts; it does not include a determinate set of rules.
I can be as certain of someone else’s sensations as of any fact. But this does not make the propositions ‘He is much depressed’, ’25 × 25 = 625’ and ‘I am sixty years old’ into similar instruments. The explanation suggests itself that the certainty is of a different kind. – This seems to point to a psychological difference. But the difference is logical. […]
The kind of certainty is the kind of language-game. (PI II, xi: 190-191)
What is essential to the practice of judging the sincerity of the experience-reports of others is that it is based on evidence that is often extremely complicated and difficult to characterize; and that in it nothing plays the role of conclusive evidence which everyone is forced to accept on pain of irrationality. Wittgenstein briefly remarks that it is partly based on “imponderable” (unwägbare) evidence (PI II, xi: 194; LW II: 95) that includes “subtleties of glance, of gesture [and] of tone” (PI II, xi: 194). Moreover, I might be “quite incapable of describing the difference” between such subtleties that for me make the difference between a genuine and a pretended expression in the other (PI II, xi: 194). If two people disagree on how to assess this subtle evidence, there is no universally valid procedure to solve such disagreements:
I am sure, sure, that he is not pretending; but some third person is not. Can I always convince him? And if not is there some mistake in his reasoning or observations? (PI II, xi: 193. See also Z §§554-556.)
These are clearly rhetorical questions that are meant to be answered in the negative. Anyway, Wittgenstein also remarks that there is such a thing as a better and worse judgment about the experiences of others; knowledge of people (Menschenkenntnis) is a skill that can be learned. But what one in this case learns is
…not a technique; one learns correct judgments. There are also rules, but they do not form a system, and only experienced people can apply them right. Unlike calculating-rules.
What is most difficult here is to put this indefiniteness, correctly and unfalsified, into words. (PI II, xi: 193)
A report or confession of one’s experience is the authoritative account of his experience to others, and when questions are raised about the sincerity of the report, these cannot be settled by any simple and stable set of criteria. Such questions are in a way open-ended. But it is a thoroughgoing motive in Wittgenstein’s discussions that this should not be seen as a philosophically significant flaw in our knowledge of other minds; it is just a constitutive difference. There is no good reason to compare our knowledge of other minds unfavorably to other areas of knowledge in this respect, and in particular not to the subject’s knowledge of her own mental states. Here Wittgenstein battles against epistemic asymmetry by not taking my first-personal accounts of my experience as the paradigm case of certain knowledge, and on the other hand by respecting our knowledge of the mental states of others as its own type of knowledge, with its own peculiar characteristics.
The second form of epistemic asymmetry concerns the thought that it seems impossible to put the essential qualities of my first-personal sensations, feelings and experiences into words. There appears to be something “ineffable” about subjective experience. It is a familiar fact that sometimes, in the face of highly unusual and novel experiences, words fail to capture them. This is a relevant issue in dream research, for example (Revonsuo 2010: 84).
But it can be argued that it is not only unusual experiences but conscious experiences in general that have an ineffable element in them. It is one thing to point out that, as Hume in his Treatise of Human Nature noticed (T 1.1.1.9; SBN 5), spoken or written words cannot produce the taste experience of pineapple in someone who is not already familiar with it; however, this does not mean that we could not still describe the taste of pineapple in words. But it is a further thing to argue that even the words we legitimately use in such descriptions are not about the intrinsic qualities of the sensation as such, but describe them in a roundabout way, via metaphor and comparison with publicly accessible entities. David Chalmers writes:
We have no independent language for describing phenomenal qualities. […] Although greenness is a distinct sort of sensation with a rich intrinsic character, there is very little that one can say about it other than that it is green. In talking about phenomenal qualities, we generally have to specify the qualities in question in terms of associated external properties, or in terms of associated causal roles.
[…] When we learn the term ‘green sensation’, it is effectively by ostension – we learn to apply it to the sort of experience caused by grass, trees and so on. (Chalmers 1996: 22)
I assume it to be uncontroversial that Wittgenstein would agree with Chalmers at least in this sense: sensations or feelings cannot be a basis for an independent language, separated from a public communicative context which gives experience-words their normative properties. I will not go into any detail with this, but rather assume it as a given that this is Wittgenstein’s position: he indeed recognizes that we can name and describe subjective experiences only with the help of publicly identifiable objects. What he denies, I argue, is that this poses any special problem for communicating them. Instead he recognizes the variety of situations in which and purposes with which people describe their subjective experiences to each other, and denies that there is any goal or purpose that communication about subjective experiences is constitutionally unable to achieve.
As Lagerspetz (2012) has observed, the tendency to think that there is some impossibility in principle of describing subjective experiences stems from failing to appreciate descriptions as actions in a communicative context. Wittgenstein emphasizes the uses in which descriptions of experiences are put in human life. As shown in e.g. PI §244, Wittgenstein saw first-personal experience-talk as importantly connected with primitive expressions. For reasons of space, I will not here go deeper into the possible interpretations and limitations of Wittgenstein’s expressivism and his non-cognitive thesis of avowals (for discussion, see Rodriguez 2012; Hacker 2005; Bar-On & Long 2001; Macarthur 2010; Robjant 2012). I merely note the general point, made abundantly clear by the PI II section ix, that Wittgenstein saw utterances of first-personal subjective experience as capable of serving both expressive and descriptive roles, with mixed and intermediate cases.
The point that avowals sometimes are expressive in nature is enough to alleviate the problem of ineffability of subjective experience to some extent. As Macarthur (2010) explains, the later Wittgenstein is opposed to the assumption that language, in general, always serves some one simply definable function. In the particular case of first-personal experience-talk, he is opposed to the idea that all such talk is in the business of describing some “inner” event. Rather, in many contexts what is primarily at issue is not the descriptive accuracy of what the subject says, but its status as an avowal; an expressive utterance of the subject, which invites the hearers to attend to her. Moran (2001) elaborates, albeit primarily in the context of beliefs, the way in which avowing and describing one’s state of mind are importantly connected. So even if it were true that we cannot display the phenomenal qualities of our experiences by verbal descriptions in any simple way, that does not mean that we cannot communicate them at all, because much of first-personal experience-talk is not descriptive anyway.
In PI II xi, Wittgenstein illustrates the many facets of such talk:
Are the words ‘I am afraid’ a description of a state of mind?
I say ‘I am afraid’; someone else asks me: ‘What was that? A cry of fear; or do you want to tell me how you feel; or is it a reflection on your present state?’ -- Could I always give him a clear answer? Could I never give him one?
We can imagine all sorts of things here, for example:
‘No, no! I am afraid!’
‘I am afraid. I am sorry to have to confess it.’
‘I am still a bit afraid, but no longer as much as before.’
‘At bottom I am still afraid, though I won’t confess it to myself.’
‘I torment myself with all sorts of fears.’
‘Now, just when I should be fearless, I am afraid!’
To each of these sentences a special tone of voice is appropriate, and a different context. (PI II, ix: 160)
Some of these cases are clearly descriptive ones. There is such a thing as describing one’s subjective experience, and such a description can be successful or unsuccessful in communication. But most importantly, Wittgenstein points out that “[w]hat we call ‘descriptions’ are instruments for particular uses” (PI §291, emphasis in the original; see also e.g. PI II, xi: 170-171). Therefore, whether a description is successful or unsuccessful is dependent on the purpose for which the description was put forward, and on whether it produces a desired kind of understanding between the speaker and the hearer. It is clear that in different contexts, a good description of a subjective experience will amount to different things. At a doctor’s office, a description of one’s pains serves its purpose if it makes the necessary distinctions about the location of the pain, it’s quality, intensity, frequency, duration and so on, enabling the clinician to form a hypothesis about its cause. Such an account would be a very bad description of a similar pain in a tragic poem, for example; and similarly, what counts as a brilliant poetic simile of excruciating pain might be completely useless for a doctor.
Pains can be described as dull, splitting, burning, sharp, stabbing, and so on. It can be argued that such descriptions are possible only by borrowing our vocabulary from the category of public objects, and that such a derived way of talking can never fully communicate the first-personal, experienced character of pain. I think Wittgenstein has an implicit, if not an explicit, counterargument to the latter claim. I will present it here following Lagerspetz (2012).
Subjective experiences seem ineffable only if we hold a confused view of what counts as a sufficient description. Hume noticed that even the best verbal description of the taste of pineapple, for example, is unable to produce the taste-experience of pineapple in someone who is not already familiar with the taste from his own experience. But descriptions should not even be expected to do such a thing. The way to produce a taste experience of pineapple in someone unacquainted with it is to offer her a suitable sample of pineapple. Descriptions of tastes can serve a variety of different purposes, but straightforwardly producing novel taste experiences is not among those purposes. A person can describe the taste of pineapple to someone else so that the other can guess whether she will like pineapple or not; descriptions of different foodstuffs can be used to make a systematic list of them with a number of categories, for practical purposes; or two people tasting pineapple may compare their taste-descriptions to see how their taste-vocabularies differ. In all those contexts what counts as a sufficient description will be relative to the interests of the speaker and hearer. It is not reasonable to expect descriptions to do the same job that samples do, because description is an altogether different instrument. So there is no reason to claim that the inability of descriptions to produce novel subjective experiences in us is a flaw in our ability to describe experiences. (See Lagerspetz 2012: 291-294.)
However, it is still possible to protest that even if it is not the point of descriptions to reproduce the described experience in another person, no description of a subjective experience can ever be exhaustive. The felt qualities of experiences like pains and tastes are richer than our vocabulary for describing them, and therefore, the subject of experiences always inevitably knows more about them than she can verbally or otherwise communicate. But this protest might be based on a philosophical prejudice about what counts as an “exhaustive” description. In actual practice, descriptions are said to be inadequate, adequate, vivid or comprehensive in relation to the context of describing and the goals of the persons involved. A description of one’s pain to a doctor is exhaustive insofar as it gives her all the information she needs; an abundant pain-description, explaining all the nuances of the experience, its minute-by-minute development, its exact location and alterations in its location, might from the doctor’s perspective be a worse pain-description than a more compact one, because the former will contain lots of redundant information. Descriptions are generally given as answers to actual or potential questions, and their adequacy can only be assessed in light of the questions they are meant to answer. (For one particularly good illustration of this, see PI §368.) It does not make any sense to talk about an absolutely exhaustive description of one’s experience, abstracted from the communicative context; because it is not clear even for the subject herself what such a description should look like. It is not the case that in principle, it is possible to give a complete description of a subjective experience in the abstract sense, but as a matter of fact we are unable to do that. Rather, there is nothing that would count as such a complete description in the abstract sense, even in principle (Lagerspetz 2012: 290).
Wittgenstein has no reason to see any fatal philosophical problem about our ability to describe our subjective experiences to each other. But this does not mean that he held this kind of communication to be always unproblematic. On the contrary, he notes that a “human being can be a complete enigma to another (PI II, xi: 190). Another person may remain such an enigma even when he “does his utmost to make himself understood” (LW II: 28). Wittgenstein sees the breakdowns which sometimes do happen in human communication as breakdowns of understanding, that is, failures to relate properly to other people. This can happen when the speaker and hearer do not share a context of common interests, goals and motivations of action, or a background of at least some relevant experiences familiar to both. Understanding, in this case, can be said to consist of the ability to have an exchange of questions and answers which both parties can spontaneously develop and enrich. When this does not happen and the discussion terminates, there will be a feeling that something about the other person remained “hidden”. In normal circumstances, the basic interpersonal attitude which Wittgenstein called our “attitude towards a soul” (PI II, iv: 152) will surely still remain intact. But it will be coupled with uneasiness; uncertainty about what the other is aiming at with her descriptions of her experiences and about how she is using her self-expressive words.
For Wittgenstein, failures to understand what goes on in the minds of others are essentially of this type. They are not the result of some deep ineffability of subjective experiences, but of inabilities to share contexts of action (or what Wittgenstein called “forms of life”) with others. This is also the proper context for Wittgenstein’s remark: “If a lion could talk, we could not understand him” (PI II xi: 190; LW I §190). If a lion could talk, there is no reason why it would not be able to talk about its lion-like subjective experiences as well as humans can talk about theirs. The difficulty would rather be the vast difference between a lion’s life and human life, maybe vast enough to make it impossible for us to find the right questions to ask about lion-like experiences, so that the lion could understand what we want to know of it.
The third and final form of epistemic asymmetry that I will discuss concerns the relation between our perceptions of human behavior and our beliefs about mental events that are the causes of behavior. Supposedly, our beliefs about the minds of others are formed on the basis of their behavior. The subject’s own beliefs about her mental life, by contrast, are (normally) not grounded in observations of her own behavior. First-personal experiences are “just felt”; our awareness of them is not grounded in anything further than just the experiences themselves.
When I have a pain, I have direct access to the pain itself, whereas in the case of others I have direct access only to bodily movements like grimaces, gestures, sounds and speech; in short, to pieces of pain-behavior of others. This view has some familiar philosophical consequences. First, it seems that even in ideal circumstances, the best we can do is to become assured that other people probably have pains, sensations and other experiences; although this probability will be overwhelmingly high, it will not amount to certainty. Second, it would be strictly wrong to say that pains, sensations and other experiences are ever perceived by anyone other than the subject herself; it is only the behavior which suggests the presence of these things that is being perceived. This can be taken to be an existentially flavored philosophical problem in its own right: we cannot meet the minds of others first-hand. Above (in section 3.2) I covered Wittgenstein’s treatment of the first point. I will not here go deeper into Wittgenstein’s notion of certainty in On Certainty and elsewhere. Instead I will proceed to the theme of perceiving the minds of others, linking this topic to Wittgenstein’s discussion of aspect-seeing.
It would be unintuitive to deny that our evidence of mental events of others is, in a real sense, constituted by their bodily behavior. What Wittgenstein is doing is subtly criticizing the status which we give to that behavioral evidence. The essential point of this criticism is briefly stated in PI §246:
Other people cannot be said to learn of my sensations only from my behavior, – for I cannot be said to learn of them. I have them. (PI §246, all emphases in the original)
The questions of our knowledge of other minds and our knowledge of our own minds are, as ter Hark (1990: 141) puts it, two sides of the same coin for Wittgenstein. For him, the motivation to say that behavioral evidence for the experiences of others is indirect and insufficient stems from the fact that we compare it with the first-personal case, where we supposedly have another, better kind of evidence. In the first-personal case, we seem to have direct, introspective evidence of our experiences.
Wittgenstein breaks this asymmetry by insisting that it is wrong to construe our relation to (at least some types of) our own mental states in terms of introspective evidence. It is not that I know of my pains and sensations because I introspectively see or feel the sensation or feeling; rather, I simply have the pain or sensation. The relation is even more intimate than the alleged direct introspective access. If I knew of my pains by consulting introspective evidence, then there would be no reason why this introspection could not sometimes go wrong, resulting in me being mistaken about my own pains, which Wittgenstein held to be nonsensical (see above, section 3.1). Rather, Wittgenstein writes explicitly:
It is not as if he had only indirect, while I have internal direct evidence for my mental state. Rather, he has evidence for it, (but) I do not. (LW II: 67)
We only construe behavioral evidence for the mental states of others as indirect evidence because we have an idea of some superior type of evidence, compared to which the usual behavioral evidence is a second-best thing. But what we have in the first person is not a good point of comparison, because there the relation between us and our mental states is not evidential (e.g. LW II: 92). Further, it is not clear if any other idea of a superior type of evidence is coherent. In the Brown Book, Wittgenstein remarks that “people have often talked of a direct transmission of feeling which would obviate the external medium of communication” (BBB: 185), and proceeds to question whether it makes sense to postulate such a direct medium of communication in contrast of the usual, “indirect” one. Such a medium would be something like telepathy, or what C.D. Broad (1925: 328-330) called “telegnosis”: a cognitive situation where the perceiver would be involved solely with a mental event belonging to another mind. It would be a topic of a whole separate discussion to see whether any such situations can be coherently described. One could conjecture that Wittgenstein wanted to answer in the negative. Therefore, it is wrong to say that we learn of the sensations of others only from their behavior, but right to say that we learn of them (simply) from their behavior. It is just the “only” which is inappropriate.
Much of Wittgenstein’s writings on perception and psychology revolve around aspect-seeing (see PI II, xi; LW I §§165-180, 735-785; LW II: 12-18; Z §§208-226). There is a natural link between aspect-seeing and perceiving other minds, which shows itself when Wittgenstein talks about the possibility of seeing other human beings as machines or automata:
But can’t I imagine that the people around me are automata, lack consciousness, even though they behave in the same way as usual? -- If I imagine it now – alone in my room – I see people with fixed looks (as in a trance) going about their business – the idea is perhaps a little uncanny. But just try to keep hold of this idea in the midst of your ordinary intercourse with others, in the street, say! Say to yourself, for example: “The children over there are mere automata; all their liveliness is mere automatism.” And you will either find these words becoming quite meaningless; or you will produce in yourself some kind of uncanny feeling, or something of the sort.
Seeing a living human being as an automaton is analogous to seeing one figure as a limiting case or variant of another, the cross-pieces of a window as a swastika, for example. (PI §420)
This indicates that seeing living human beings as lacking minds is seeing them under a certain aspect. It also indicates that it is a very unusual aspect, one that can be summoned only briefly and only in favorable circumstances. Wittgenstein admits that at least in some sense, people can be imagined to be mindless machines, but points out that such imagining never has more than trivial psychological consequences. Our certainty about the minds of others is not threatened in practice; we just cannot see people that way.
In PI II, xi, Wittgenstein explores the range of the concept of “seeing”. He is interested in the conceptual issues around the phenomenon of seeing a picture according to an interpretation, or under a certain aspect, in which case the perceiver in some sense “sees” things in the picture which are not strictly speaking “in” the picture. One of his examples is the following:
I see that an animal in a picture is transfixed by an arrow. It has struck it in the throat and sticks out at the back of the neck. Let the picture be a silhouette. – Do you see the arrow – or do you merely know that these two bits are supposed to represent part of an arrow? (PI II, xi: 173, emphases in the original)
Wittgenstein then goes on to say that “it must be possible to give both remarks a conceptual justification” and that the question concerns the sense in which this can be said to be a case of seeing (PI II, xi: 173-174). Eventually, he writes:
‘To me it is an animal pierced by an arrow.’ That is what I treat it as; this is my attitude to the figure. This is one meaning in calling it a case of ‘seeing’. (PI II, xi: 175, emphasis in the original)
The term “attitude” immediately reminds us of Wittgenstein’s brief remark in PI II iv, which seems to sum up his view of the nature of human beings’ relations to one another:
My attitude towards him is an attitude towards a soul. I am not of the opinion that he has a soul. (PI II, iv: 152, emphasis in the original)
Wittgenstein often discusses seeing feelings and emotions manifested in a human face, and these discussions are entangled with discussions of aspect-seeing, clearly indicating a connection (Luckhardt 1983: 333). Equally clearly he construes the cases of feelings and emotions in others as cases of seeing. In Zettel he makes this very explicit (Z §§220-226). When seeing an aspect in a picture, we see both the picture and the aspect, but we do not see them as two separate things. Similarly with seeing an expressive human face:
What do psychologists record? -- What do they observe? Isn’t it the behaviour of human beings, in particular their utterances? But these are not about behaviour.
‘I noticed that he was out of humour.’ Is this a report about his behaviour or his state of mind? […] Both; not side-by-side, however, but about the one via the other. (PI II, v: 153, emphases in the original. See also RPP I §§287-292.)
The point is that there is a conceptual distinction to be made between the sense in which we see the physical features of a face and the sense in which we see a feeling manifested in them; but both are cases of seeing. The latter is not inferred from the former (Z §225). Rather it is displayed in the former, leading Wittgenstein to say that “the human body is the best picture of the human soul” (PI II, iv: 152).
Of course, part of the interest of the phenomenon of aspect-seeing is the possibility of aspect-blindness. It can happen, for a variety of reasons, that someone is unable to see a feeling manifested in the behavior of the other person. Here, as much as with the theme of ineffability, it should be noted that Wittgenstein is not claiming interpersonal relations to be always epistemically unproblematic. What he does claim is that the minds of others are not in any peculiar way inaccessible to us. I might not always be able to see the feeling manifested in the facial expression of the other person; but when I do see it, I literally see it in the face, rather than somewhere behind it.
Wittgenstein’s way of studying mental phenomena quite fundamentally involves the thought that there is no privileged first-person perspective that is helpful in understanding what mental phenomena are (e.g. RPP II §§31-35, 531; PI §314, 413; PI II, xi: 174). To that extent, his approach has a methodologically behaviorist tone. But logical behaviorism is in no way attributable to him; his expressive analysis of first-personal experience-talk rules that out, as Luckhardt (1983), Fogelin (1976: 174-176) and others have observed. In his (1991), ter Hark offers a reading which shows how Wittgenstein’s “attitude towards a soul” and his remarks about the open-ended nature of experience-attributions to others (see 3.2 above) also preclude his account of third-personal experience-talk from being interpreted as logically behavioristic. Anyway, there is still a sense of ambivalence in Wittgenstein’s relation to behaviorism. I think his essential critique of it can be put in terms of inner and outer: Wittgenstein is opposed to behaviorism insofar as it construes the behavior we observe in others as “outer” events, which forces one to deny the existence of the things we usually call “inner” processes (see PI §308).
What Wittgenstein implies is that the behaviorist is right to insist on the publicity and observability of mental phenomena, but that the notion of “behavior” which underlies or is suggested by behaviorism is misguided. It is an impoverished concept of “mere behavior” (see LPE: 278-279). The bodily movements of others are not mere behavior to us, comparable to the “behavior” of gases or planets, because the former have a uniquely human aspect for us. They are joy-behaviors, pain-behaviors and sorrow-behaviors, and we know what joys, pains and sorrows are by living as members of a community where such things occur.
Introspectionist psychology attempts to study the “inner” events of human consciousness, which are reachable by introspection and then reported through speech or some other medium; but in any case through some “outer” event. Wittgenstein is opposed to this dichotomy. But he is also opposed to behaviorism insofar as it remains trapped in this dichotomy and only denounces one half of it, claiming that those which are usually called “inner” events are in reality nothing but “outer” events (PI §§304-314; LPE: 278-284). The right thing to do is to see the talk of “inner” experiences and their “outer” manifestations as tools for making conceptual distinctions in the continuous fabric of human life. That fabric is whole and open to view as it is, and most of the time it does not invite the quite specialized distinction of inner and outer.
Tero Vaaja is a researcher/doctoral student in philosophy at the University of Jyväskylä, Finland. In his PhD project, conducted at the Department of Social Sciences and Philosophy, he investigates the variations of the problem of other minds, tracking the historical context in which the problem has emerged, its connections with notions like intersubjectivity, consciousness and solipsism, as well as its role in several contemporary discussions. His work has been kindly supported by the Finnish Cultural Foundation and the Faculty of Social Sciences of the University of Jyväskylä.