Deep Thoughts: Consciousness

Wittgenstein, Consciousness, and the Mind
Pär Sundström

In traditional accounts of mental phenomena, consciousness often played a central role. For example, Locke, at one point in the Essay, considers whether it might be the case that we always think but are not always conscious of it. He dismisses this idea on the ground that, to think is to be conscious that one thinks: «'tis altogether as intelligible to say, that a body is extended without parts, as that any thing thinks without being conscious of it , or perceiving, that it does so. ... thinking consists in being conscious that one thinks» (book 2, chap. 1, §19).

In more modern accounts of mental phenomena, however, the role of consciousness is often downplayed. Instead, the mind is often accounted for in terms of «outer» or «objective» phenomena like mechanisms, dispositions, abilities and even environmental features. Different versions of this «outward turn» in the theorising about the mental can be found in the psychoanalytic literature, in behaviourism in philosophy and psychology, and in cognitive science. Foot note 1

In this paper, I examine one instance of this trend. In a series of passages, the later Wittgenstein criticised the idea that each mental phenomenon must involve some «inner state or process», such as an experience, sensation, feeling, or inner imagery. His treatments of this idea concerns a variety of specific mental phenomena, including searching for a red flower (BB, 3), Foot note 2 comparing from memory (BB , 85ff.), understanding how to continue a series of numbers (PI, §§151ff.), and reading (PI,§§156ff.). These passages are among the most well-received of the teachings of the later Wittgenstein. Sympathetic readers have cherished them, and critical readers have usually turned their attention elsewhere. According to many sympathisers, the passages show that one may exemplify any of the phenomena at issue without being in any particular type of conscious state.

I shall argue that the passages do not warrant this conclusion. Moreover, the conclusion is arguably false. For at least some of the phenomena in question, a strong case can be made for holding that in order to exemplify them, it is necessary that one is in a particular type of conscious state.

This is not to embrace the view--which might have been Locke's--that all mental phenomena are conscious. But it is to suggest that Locke and other thinkers of the past may have been more right about the importance of consciousness for mental phenomena than we are currently inclined to think. Even if the conscious mind does not exhaust the mind, it may be a rather large and important part of it. Foot note 3

The remaining agenda is as follows. In section I, I present the relevant passages from Wittgenstein. In section II, I document some conclusions that have been drawn on the basis of these passages. In section III, I assess what the passages show and do not show.

I Wittgenstein's passages

In The Brown Book, Wittgenstein considers a scenario where an agent A shows a colour sample to another agent B upon which B goes and fetches an object that has the same colour as the sample, using his memory of the sample. In this case, we may say that B compares the colour of the object he sees with the colour of the sample he has just seen. But what makes it the case that B compares the two objects? What, for example, distinguishes the case where B compares the two objects from the case where he just by chance picks up an object that has the same colour as an object he has recently seen?

According to Wittgenstein, we are tempted to think that an essential component of such comparing is «a specific experience of comparing and recognising» (BB, 86); unless the agent B has such an experience, his performance is not one of comparing.

However natural this thought is, Wittgenstein believes it is wrongheaded. To convince us of this, he urges us to examine «closely» what really unites performances that we are prepared to label `comparing'. If we perform this close inspection, we find no one type of experience characteristic of comparing. Instead, we find that there is a great number of states of mind, all « more or less characteristic of the act of comparing». They include:

memory images, feelings of tension and relaxation, satisfaction and dissatisfaction, the various feelings of strain in and around our eyes accompanying prolonged gazing at the same object, and all possible combinations of these and many other experiences (BB, 86).

Individual acts of comparing are thus, according to Wittgenstein, a diverse lot. They resemble and differ from one another in various ways, but there is no one feature common to all of them.

We find that what connects all the cases of comparing is a vast number of overlapping similarities, and as soon as we see this, we feel no longer compelled to say that there must be some one feature common to them all (BB, 87).

Dialectics reminiscent of this passage show up in many other places. I have already mentioned the discussions of searching for a red flower ,understanding how to continue a series of numbers and reading . In yet other passages Wittgenstein discusses what is involved in intending a picture to be of so-and-so (BB, 32f.), believing what one says (BB, 144ff.), counting a number of objects (BB, 149f.), recognising (BB, 165f.), pointing to an object's shape (PI, §§33ff.), intending (PI, §591), saying `it'll stop soon' meaning the pain (PI, §§666ff.), writing a letter to so-and-so (Z, §7), looking for a photograph in a drawer (Z, §8), saying `come here!' meaning a person A (Z, §21f.), and lying (Z, §189f.). In each case, he considers the relation between the phenomenon under discussion and some inner state or process--such as an experience, sensation, feeling, or inner imagery--or class of such. And in each case, he suggests that no state or process out of the range considered is essential to the phenomenon.

A note about the relation between these passages: As has been observed (see, e.g. Malcolm, 9), it seems natural to understand a subset of the passages as different inquiries into one more general phenomenon, namely the phenomenon of meaning something by, say, a gesture or expression. This subset may include (all or some of) the passages about intending a picture to be a portrait of so-and-so ,believing what one says ,counting a number of objects ,pointing to an object's shape ,saying `it'll stop soon' meaning the pain , writing a letter to so-and-so and saying `come here!' meaning a person A. It is less natural, however, to understand all the passages as different inquiries into the general phenomenon of meaning something by a gesture or expression. For example, it is unnatural to construe the passages concerning comparing, searching and recognizing as inquiries into this general phenomenon.

I mention this to caution against quick generalisations over the class of mental phenomena investigated in Wittgenstein's passages. Even if one should become convinced that something is true about the general phenomenon of meaning something by an expression or gesture, the insight may not readily generalise to all of the mental phenomena under discussion. I will return, below, the importance of recognizing the diversity of the phenomena under discussion. Foot note 4

II Suggested lessons of the passages

What do the passages that we have just looked at show? According to Norman Malcolm, they have dramatic implications for how we should understand mental phenomena. Malcolm thinks we should conclude, on the basis of the passages, that one may exemplify any mental phenomenon regardless of «what goes on in one's mind or thoughts»:

We are tempted to think that your meaning the color must have been something that went on in your mind. But it might be that what went on in your mind or thoughts had nothing to do with what you meant. ...Instead of looking inside ourselves we should be looking around us, at the context in which our words and pointing are located. We should be searching horizontally instead of vertically. This temptation to look in the wrong direction besets us whenever we are perplexed about the concepts of mind. Wittgenstein's admonition applies to all of them (1970, 15-6).

Malcolm's suggestion is perhaps unusually radical. But with a few reservations, it is representative of what many readers take Wittgenstein to have shown in the passages under consideration. Foot note 5 The reservations are three: First, other commentators are usually more guarded than Malcolm in their claims about the range of phenomena to which Wittgenstein's results apply. They usually don't draw conclusions about mental phenomena generally, but only about some subset of them. Exactly which subset they have in mind is not always explicit, but my impression is that many readers take Wittgenstein to have shown something significant about at least the range of phenomena considered in the passages listed in section I above. Foot note 6 Second, not all commentators formulate Wittgenstein's result in terms of what goes on in all of «mind or thought». Many formulate them instead, more cautiously, in terms of what goes on in «consciousness», or what goes on that can be «introspected». Foot note 7 And third, not all commentators take Wittgenstein to have shown that exemplifying a given phenomenon may have « nothing to do» with what goes on in consciousness. Mostly, commentators have concluded that some relation fails to hold between exemplifying a given phenomenon and being in a particular type of conscious state. In some places, it is concluded that there is no type of conscious state the having of which amounts to exemplifying the phenomenon in question. Foot note 8 In other places, it is concluded that there is no type of conscious state the having of which is necessary to exemplifying the phenomenon. Foot note 9 And in yet other places, the conclusion is that there is no type of conscious state that a subject is always in when he or she exemplifies the phenomenon. Foot note 10

To summarise this, each of the following conclusions have been drawn, by more than one commentator, on the basis of Wittgenstein's passages:

(CI)
There is no phenomenon P (out of the range discussed in the passages listed in section I) and type of conscious state C, such that: being in C is to exemplify P.
(CN)
There is no phenomenon P (out of the range discussed in the passages listed in section I) and type of conscious state C, such that it is necessary that: if one exemplifies P then one is in C.
(CA)
There is no phenomenon P (out of the range discussed in the passages listed in section I) and type of conscious state C, such that it is always the case that: if one exemplifies P then one is in C.

On a natural understanding of these claims, they can be ordered with respect to strength, (CA) being the strongest and (CI) the weakest. That is, (CA) implies (CN) but not vice versa, and (CA) and (CN) both individually imply (CI) , but not vice versa.

To simplify my discussion, I shall set the weakest of the claims, (CI) , to one side. I'm not quite sure what expositors of Wittgenstein have meant by it, and a discussion of this would take us too far afield.

Setting (CI) aside, I shall in the following cast doubt on (CN) and (CA) . I think it is doubtful that these claims are true. And I think it is clear that they are not justified by the passages in question. I turn to discuss these issues now. Since casting doubt on the (weaker) claim (CN) amounts to casting doubt on the (stronger) claim (CA) , I will target the former.

III What the passages show and do not show

To assess what Wittgenstein's passages show and do not show, it is crucial to distinguish two different interpretations of the temptation of thought that Wittgenstein identifies and tries to exorcise.

On the first interpretation, the temptation is to say, for any phenomenon P under consideration, that to exemplify P one must be in any one out of a narrow range of states of mind. I shall not try to make absolutely precise what this narrow range of states of mind is, but we shall think of it as including sensations ,mental images ,and feelings , while not including awareness that something is the case. Thus, on this interpretation, the temptation is to say things like: to exemplify P one must have a particular type of sensation, or feeling, or image ; but the temptation is not to say that: to exemplify P one must be aware that something is the case. I will sometimes refer to this as the «sensationalist» interpretation of the temptation.

On the second interpretation, the temptation is to say, for any phenomenon P under discussion, that to exemplify P one must be in any one out of a broader range of states of mind. On this interpretation, the temptation is (in part) to say that: to exemplify P one must be aware that something is the case. Foot note 11 ,Foot note 12

With this distinction in place, I shall in the remainder of the paper try to make the following claims plausible:

(1) If Wittgenstein's temptation is understood narrowly or sensationalistically, it is right to reject it in the case of each phenomenon under discussion (comparing, reading, understanding, and so forth). I can compare from memory, or exemplify any of the other phenomena under discussion, without having any particular type of feeling, sensation or mental image.

(2) However, rejecting the temptation as narrowly or sensationalistically conceived does not by itself give us any reason to conclude that there is no type of conscious state that one must be in to exemplify these phenomena. This is because there are arguably conscious states other than sensations, feelings, and imagings; and consequently, there may well be some type of conscious state I need to be in, in order to compare, even though there is no sensation, feeling or image I need to have in order to do so.

(3) We would be justified in thinking that no conscious state is necessary for exemplifying any of the phenomena in question if we could reject a certain version of the temptation as broadly conceived.

(4) However, it is doubtful that this temptation should be rejected in the case of each phenomenon. Certainly, Wittgenstein's passages don't provide sufficient reason for such a sweeping rejection.

While I shall say something in favour of each of these claims, it should be obvious that I have a stake only in (2) and (4), and I will consequently devote most of the space to these. I happen to believe that (1) and (3) are true as well; but should they be false, that would not harm my argument. I will attend to the claims in turn.

(1) We should reject Wittgenstein's temptation as narrowly, or sensationalistically, conceived.

I will not spend much energy defending this claim. The claim is, in effect, that there is no particular type of sensation, image or feeling that I need to have in order to compare, or exemplify any other of the phenomena at issue. And this is clearly right. It may not have been obvious or widely appreciated before Wittgenstein pointed it out to us. Thus, philosophers like Hume and Locke probably thought otherwise. But by now, philosophers have largely absorbed this particular lesson of Wittgenstein's.

(2) But to reject the temptation as narrowly or sensationalistically conceived does not by itself give us any reason to think that there is no type of conscious state that one must be in to exemplify any of the phenomena at issue (because there are arguably conscious states other than sensations, feelings, and imagings).

This claim requires much more defence. To defend it, I introduce the following claim:

(AN) It is necessary that: if one compares then one is focally action aware that one is comparing.

Later on, I shall defend this claim. Foot note 13 But for present purposes, much less is required. First, I shall explain what it is to be «focally action aware» that one is doing so-and-so. Second, I shall try to make plausible that such focal action awareness is a type of conscious state . If this is so, then there are conscious states other than sensations, feelings, and imagings. And then, there may well be some type of conscious state--for example, the state of being focally action aware that I am comparing--that I need to be in, in order to compare, even though there is no type of sensation, feeling or image that I need to have to do so. And that suffices to establish (2).

What, then, is it to be focally action aware that one is doing so-and-so? It will take some work and care to pinpoint this phenomenon. But as a first characterisation, I am typically focally action aware that I am doing so-and-so when I'm in a position to report, without special prompting , that I am doing so-and-so. By `special prompting', I have in mind here the act, which can be performed by myself or somebody else, of bringing to my attention the possibility that I am doing so-and-so. Such prompting might take the form of the question `Are you (or: am I) doing so-and-so?'.

To illustrate, suppose I am comparing. And suppose further that, as I do so, there is a set of propositions that I'm in a position to sincerely express, without special prompting, in reply to the question `What are you now doing?' and another set of propositions that I'm in a position to express in reply to this question only given prompting. Let us suppose that the first set of propositions includes I am comparing and I am helping so-and-so find something that agrees with this sample , and that the second set includes I am knitting my brows ,I am exerting pressure on the soles of my shoes ,I am trying to make up to so-and-so for letting him down the other day , and I am suppressing an urge to do something more fun . In such a case, I am typically focally action aware that I am comparing and that I am helping so-and-so find something that agrees with the sample. In contrast, I am not focally action aware that I am knitting my brows or that I am trying to make up to so-and-so for letting him down the other day--although I may in some sense be aware that I am doing these things.

I intend this characterisation to fix the reference of `focal action awareness' rather than defining the term. Thus, focal action awareness is not defined as some kind of behavioural disposition. It is, rather, ostended as that kind of awareness that is characteristically present in the kind of circumstances described and exemplified above.

A few words need to be said about the relation between focal action awareness and attention . Focal action awareness does not require attention. Or at least, being focally action aware that one is doing so-and-so doesn't require that one attends to the fact that one is doing so-and-so . Just think about a typical case where you are comparing, say, the colours of two objects and are in position to report, without special prompting, that you are doing so. In the typical such case, what you are attending to is surely the colours of the two objects . You are not attending to the fact that you are now comparing them. (And, as is often pointed out, if you start to attend to the fact that you are comparing them, this will distract you from what you are doing and make it harder for you to tell whether the objects match or not.) But while you are attending to the colours of the two objects, you still have a certain kind of awareness of the fact that you are comparing them. If you are asked what you are doing, you can effortlessly report, without special prompting, that you are comparing. And this is plausibly because you have this awareness--«focal action awareness»--of what you are doing.

That you are focally action aware that you are doing so-and-so does not mean, then , that the fact that you are doing so-and-so is the focus of your attention . It means, rather, that it is in the focus of your action awareness , that is, your awareness of what you are doing . When you compare two objects, you are typically aware, in some sense or other, of doing a number of things. And, even if you attend only to objects outside of you and not to any of the things you are doing, it is often or always possible to distinguish what you are more focally aware of doing from what you are less focally aware of doing. If you are asked what you are doing, there are some true answers you can provide without special prompting, and other true answers you can give only given special prompting. The former answers typically express what you are more focally aware of doing; the latter answers what you are less focally aware of doing. Foot note 14

Now, some may doubt that focal action awareness is present in the kind of cases I have described. Thus, it is sometimes suggested--often on phenomenological grounds--that, when all your attention is directed «outward», you are in fact not in any way aware that you are comparing even if you can report that you are doing so without special prompting. Your ability to effortlessly report on what you are doing may foster an illusion that you have such awareness. But closer phenomenological inspection will tell you that, in reality, you become aware that you are comparing only when you are asked what you are doing. Foot note 15

However, I don't see any reason--whether phenomenological or of some other kind--to think that, in this kind of case, I become aware that I am comparing only when I'm asked what I am doing. Let's distinguish two kinds of cases in which you are comparing the colours of two objects and are asked, in the midst of this performance, what you are now doing. In one kind of case, this question causes you to shift some of your attention away from the colours of the objects and towards the fact that you are comparing them. In the second kind of case, the question doesn't cause any such shift in attention. (Think, for example, about situations where you are, as we say, very «focussed on your task». In those situations, you may report that you are comparing when asked what you are doing, but while reporting this you remain just as attentive to the colours of the object. In such cases, your report will often be somewhat absent-minded). I trust that both cases are familiar. Now, I don't find, in either of these cases, any evidence that I become aware that I am comparing only when I'm asked what I am doing. Consider the latter case first. In this case, there are indeed some changes in my awareness as the question is asked. First, I become aware that a question is being asked, and later, I am aware that I am answering this question. But I fail to detect any changes in my awareness in addition to those. In particular, I find no evidence that I suddenly come to realise that I'm comparing only when the question is posed. Rather, it seems to me that I'm reporting something I was aware of all along. This phenomenological impression is to some degree confirmed by the very fact that I can effortlessly report what I'm doing while staying fully «focussed on the task». It is relatively easy to report something that one is already aware of. If I became aware that I'm comparing only when asked what I am doing, this would be cognitively more disruptive, and it would be at least slightly more surprising that I could stay fully focussed on my task while having this realisation and then reporting it. Consider now the former kind of case, where the question what I'm doing causes a shift in my awareness. In this case, there is a change in my awareness in addition to my being aware of a question and my answer to it, because there is a change of attention. When I'm asked what I'm doing, I come to attend (or attend more) to the fact that I am comparing. But, again, I see no reason to think that I only then become aware of the fact that I'm comparing. As far as I can tell from any phenomenological inspection, the truth is rather that I come to attend (or attend more) to a fact that I was already aware of but didn't pay (much) attention to.

Now, focal action awareness that one is doing so-and-so is, I submit, a type of conscious state. Admittedly, this suggestion brings us into a disputed, and rather murky, territory. As has often been remarked, we possess many concepts of consciousness. Foot note 16 And it is true that there is at least one sense of consciousness such that some philosophers deny that there are, in that sense, conscious states other than sensations, images, and feelings. Thus, Michael Tye has argued that sensations, imagings, and feelings are the only states that are «phenomenally» conscious; that is, they are the only states that are ever like something to be in. Foot note 17 However, this view is controversial. Many--perhaps even most--philosophers of mind hold that states other than these--for example thoughts and desires--can also be like something to have and thus be phenomenally conscious. Foot note 18

The issue is elusive. All parties of the dispute agree that there is something it is like for me when I, say, think a certain thought. Tye insists that what it is like is always solely a matter of what sensations, images, and feelings accompany the thoughts. But others insist that this is not so; that the thought itself can be like something to have. It is unclear, at least to me, what could settle this dispute.

In the present context, however, there is something more to say. As we are discussing what Wittgenstein's passages show and do not show, it is of interest to consider Wittgenstein's favoured criteria for a state's being conscious. And it seems that focal action awareness that one is doing so-and-so satisfies these criteria, at least, for being a conscious state.

When Wittgenstein tried to determine whether some psychological phenomenon was a «state of consciousness»--or even, sometimes, whether it was a mental state--he characteristically considered whether the phenomenon displayed a certain kind of duration, which he called «genuine duration». For example, in Remarks on the Philosophy of Psychology , he says:

I want to talk about a «state of consciousness», and to use this expression to refer to the seeing of a certain picture, the hearing of a tone, a sensation of pain or of taste, etc. I want to say that believing, understanding, knowing, intending, and others, are not states of consciousness. If for the moment I call these latter «dispositions», then an important difference between dispositions and states of consciousness consists in the fact that a disposition is not interrupted by a break in consciousness or a shift in attention. (And that is of course not a causal remark.) Really one hardly ever says that one has believed or understood something «uninterruptedly» since yesterday ( R2 , §45).

Think of this language-game: Determine how long an impression lasts by means of a stop-watch. The duration of knowledge, ability, understanding, could not be determined in this way ( R2 , §51.)

The general differentiation of all states of consciousness from dispositions seems to me to be that one cannot ascertain by spot-check whether they are still going on ( R2 , §57). Foot note 19

It seems that focal action awareness that one is doing so-and-so satisfies the conditions for being a conscious state that one can extract from passages such as these: Foot note 20 To begin with, focal action awareness that one is doing so-and-so is interrupted by a break of consciousness. For example, if I am focally action aware that I am doing something and then suddenly fall into dreamless sleep, my focal action awareness is interrupted. Further, it makes sense to say that one has been focally action aware of something uninterruptedly for a certain period. For example, I can sensibly say that for the past half-hour, I have without interruption been focally action aware that I am working on this paper. Again, it is possible to determine by means of a stopwatch for how long someone is focally action aware of something. Thus, the kind of assessment I just made--that for half an hour I have been focally action aware that I am working on this paper--could have been made with greater accuracy by using a stopwatch. (To be sure, it will not always be possible to make this kind of assessment very precise even with a stopwatch, since the beginnings and end points of focal action awareness are not always sharply delimited. But in this regard, focal action awareness does not differ from sensations of pain or taste, which are cited by Wittgenstein as examples of states of consciousness.) Finally, one can ascertain by spot-check whether one is still focally action aware of something. For example, ten minutes ago I was focally action aware that I was working on this paper, and I can now determine that I am still focally action aware that I am doing that. In contrast, an hour ago I was focally action aware that I was brushing my teeth, and I can now determine that I am no longer focally action aware of doing that.

True, there is one condition for being a conscious state which Wittgenstein mentions and which is not satisfied by focal action awareness: focal action awareness is not interrupted by any shift in attention (for this condition, see the first of three passages quoted above ). I may shift my attention, say, from substantive to formal aspects of the paper, and yet remain focally action aware that I am working on it. But then, it seems that this condition should at least not be a necessary condition for being a state of consciousness, by Wittgenstein's own lights. Wittgenstein thinks that hearing a tone and a sensation of pain are conscious states (first passage quoted above), and neither are those are interrupted by any change of attention. If I hear a tone or feel a pain, I may shift my attention in various ways and still hear the tone or feel the pain.

So, focal action awareness that one is doing so-and-so is, arguably, a kind of conscious state. That is at least what many--perhaps most--contemporary philosophers of mind would think or allow. And it is also the verdict one seems to reach if one considers Wittgenstein's criteria for a state's being conscious. But if focal action awareness is a conscious state, then there are conscious states other than sensations, feelings, and imagings. And hence, there may well be a type of conscious state--for example, a certain state of focal action awareness--that I need to be in, in order to compare, even though there is no sensation, feeling or image that I need to have to do so. And that suffices to establish (2).

(3) We would be justified in thinking that no conscious state is necessary for exemplifying any of the phenomena in question if we could reject a certain version of the temptation as broadly conceived.

To repeat: On the broad interpretation, the temptation is to say, for any phenomenon P under discussion, that to exemplify P one must be in any one out of a broader range of states of mind. On this interpretation, as opposed to the former interpretation, the temptation is (in part) to say that: to exemplify P one must be aware that something is the case.

Wittgenstein's discussion of reading--or one strand of this discussion at any rate--affords us an example of a temptation thus broadly conceived. Part of the formulation of the temptation in this case is: «A man surely knows whether he is reading or only pretending to read!» ( PI , §159). Judging by this particular formulation, it seems that the temptation under discussion is to say that some sort of awareness or knowledge that one is reading is necessary for reading. Foot note 21

If we could show this temptation to be false, then I think we could reasonably conclude that there is no type of conscious state that one must be in, in order to read; and analogous conclusions would be warranted for the other phenomena under discussion: for example, if awareness that one is comparing is not necessary for comparing, then it is reasonable to conclude that no conscious state is necessary for comparing.

Why would these conclusions be reasonable? After all, awareness that one is doing so-and-so is at most one type of conscious state. How can it be reasonable to conclude that no conscious state necessarily accompanies doing so-and-so on the basis of ruling out that that type of conscious state necessarily accompanies doing so-and-so?

The answer is that there is, arguably, no more plausible candidate , among the states of consciousness, for being a necessary accompaniment of comparing than the (focal action) awareness that one is comparing. Contrast the focal action awareness that one is comparing with, say, the feelings of tension and strain that Wittgenstein considers in The Brown Book. Clearly, focal action awareness that one is comparing is a more plausible candidate for being a necessary accompaniment of comparing than are feelings of strain and tension. Arguably however, there is no more plausible candidate, among the states of consciousness, for being a necessary accompaniment of comparing than the focal action awareness that one is comparing. That's why it would be reasonable to conclude that no conscious state is necessary for comparing, if we could rule out that focal action awareness that one is comparing is not necessary for doing so. Foot note 22

As I said above, I think this is correct, but it is not essential to my argument. Essential, on the other hand, is the following point:

(4) It is, however, doubtful that this version of the broad temptation should be rejected in the case of each phenomenon under consideration. Certainly, Wittgenstein's passages don't provide sufficient reason for such a general rejection.

To support this, I return to the claim (AN), which says that in order to compare, one must be focally action aware that one is comparing. If (AN) is true, then a certain version of Wittgenstein's temptation (broadly conceived) is true for at least one of the phenomena under discussion, namely comparing from memory . Above, I explained what it is to be focally aware that one is comparing, but I did nothing to defend the claim (AN) . I will defend it now.

As a preliminary to defending (AN) , I want to make two points. The points are both rather obvious, but they have nevertheless been blurred in some commentaries on Wittgenstein. Taking notice of them will clear the way for a sober assessment of (AN) .

First point: Comparing is not an ability or capacity. To be sure, comparing presupposes at least one ability, namely the ability to compare. But it is not that ability, or any other ability. Comparing is rather the exercise of the ability to compare. Now, it is evident that I may be able to compare without being focally action aware that I am comparing (and also without being aware that I am able compare). But this must not be confused with the possibility of comparing without being focally action aware that one is comparing. Foot note 23

Second point: The question of what is necessary for falling under a concept (say, the concept comparing ) must be distinguished from the question of what is necessary for understanding the concept, as well as from the question of what is necessary for coming to understand the concept. To understand, or come to understand, the concept the winner of the race one doesn't have to be the first one to cross the line. In fact, one doesn't have to be in the race at all. But to fall under this concept--that is, to win the race--one must be the first one to cross the line. Now, it's clear that one can understand the concept of comparing without being focally action aware that one is comparing (and also without being aware that one understands the concept). And it seems plausible, if perhaps less obvious, that one can come to understand this concept without being or ever having been focally action aware that one is comparing (and also without being aware that one is coming to understand the concept). But none of this implies that one can compare without being focally action aware that one is comparing. Foot note 24

With these potential confusions out of the way, I shall try to bring out the case for holding (AN) . My strategy will be to contrast comparing with some other things one can do.

In the case of many things one can do, it is indeed clear that one can do them without being focally action aware that one is doing them. Breaking a window and touching wet paint are cases in point. I can obviously break a window or touch wet paint without being focally action aware that I am doing any of this, and indeed, without being focally action aware of anything; I can do these things when I am in dreamless sleep. The same is true of performances often associated with comparing, such as picking something that in fact agrees in some respect with a given sample.

In the case of some other things one can do, it may be necessary that one is focally action aware of something , and there may even be some necessary restrictions on what one is focally action aware of when one does them. Cases in point may be displaying generosity and proceeding brutally. But it is again clear that I can do either of these things without being focally action aware that I am doing precisely it . Focal action awareness that one is displaying generosity is clearly not necessary for displaying generosity, for example.

In contrast, it is less clear that I can compare something with something else without being focally action aware that I am doing so. Imagine a case: A person S is standing by a bookcase. His or her eyes are running over the shelves. After a while, he or she grabs a book. But suppose it is not true that S is in a position to report, without special prompting, that he or she is comparing. We may suppose that S might without special prompting report on doing other things (like I'm waiting for so-and-so or I'm trying to figure out what so-and-so's reading habits are ). Or we may suppose that there is nothing that S can report on doing without special prompting. In either case, we have reason to think that S is not focally action aware that he or she is comparing. And it seems that we thereby have strong reason to withdraw or contest any judgment to the effect that S is comparing. This is true if we conceive of S from a third-person perspective, but also if we do so from a first-person perspective: If Iin this kind of situation am not focally action aware that I am comparing, it seems that Ithereby have strong reason to contest any judgment to the effect that I am comparing. Even if there are features of the situation that indicate that I am comparing, the mere fact that I am not focally action aware of doing so gives me reason to think I am not. The contrast with the case of (say) picking something that in fact matches a sample is striking: even if I lack focal action awareness that I am picking something that matches a given sample, I can very easily be convinced by other evidence that I am doing so.

I take this to show that (AN) has, at the very least, some initial plausibility: if the absence of focal action awareness of comparing gives us reason, both from the first-person and third-person perspective, to infer the absence of comparing, then we have reason to believe that there is some intimate connection between the two.

Are there any convincing reasons to reject (AN) , despite its initial plausibility? I'm not aware of any. An exhaustive discussion of this issue would make this paper too long. I shall confine discussion to two challenges that can be derived from Wittgenstein's passages. I hope that my responses to these will contribute to make (AN) seem plausible.

The first challenge I adapt from Wittgenstein's discussion of reading, in the Investigations , §§156ff. Consider a hypothetical drug which had no impact on people's capacities for comparing, but which made them unable to be focally action aware of what they were doing. If such a drug is possible, then it is obviously possible to compare without being focally action aware that one is comparing.

However, the question is precisely whether such a drug is possible. After all, the connection between comparing and focal action awareness that one is comparing does not seem entirely accidental. Comparing is an activity that requires some degree of sustained attentional involvement. It seems to be--in every sense--impossible that somebody just happens , by chance, to compare a set of objects with a given sample. (Again, there is a striking contrast with the case of picking an object that in fact matches some sample: somebody may perfectly well happen, by chance, to do that.) Now, it may be that the sustained attentional involvement required for comparing in turn requires that one be focally action aware that one is comparing. Perhaps such awareness is needed in order to keep sufficient cognitive resources enlisted for the task at hand. I'm not insisting that this is true, but the idea does not seem ludicrous. And if it is right, then any drug that impaired an agent's capacity to stay focally action aware of what he or she is doing would necessarily impair the agent's capacity for comparing.

To be sure, we may sometimes feel that we are able to imagine or conceive of someone who--for example under the influence of a drug--compares without being focally action aware of doing so. But this may reflect that we are at those moments not too clear about what it is to compare and/or be focally action aware. We know from other cases that we are prone to make such mistakes: Someone who is sufficiently ignorant of geometry may find it conceivable that the square of the hypotenuse of a Euclidean right-angled triangle is not equal to the sum of the squares of its other two sides. Nevertheless, it is necessary that the square of the hypotenuse of a Euclidean right-angled triangle is equal to the sum of the squares of its other two sides. Foot note 25 Similarly in this case: If it seems to us conceivable that someone compares without being (focally action) aware of doing so, this may just reflect that we are unclear about what it is to compare and/or be (focally action) aware; if we had a sufficiently clear understanding of this, it would perhaps be apparent that it's necessary that anyone who compares is focally action aware that he or she is comparing.

The second challenge I want to consider is this: Every now and then, we forget what we are doing in the midst of an activity. After a while we may ask ourselves, `What am I doing again?' and only then does awareness return: `Oh yeah, I am comparing'. This may seem to show that it is not only conceivable that someone compares without being focally action aware that he or she is comparing, but that this happens in fact, and so is obviously possible. Foot note 26

In reply to this challenge, I think it is unproved that one in fact ever continues to compare while being thus forgetful. There is of course a sense in which one may compare the whole day long even if, during the day, one takes both a lunch break, a nap and a dinner break. That is the sense of `comparing' in which comparing does not have what Wittgenstein called `genuine duration'. But there is another sense in which comparing does have genuine duration. This is the sense in which I stop comparing when I take a break, and resume comparing after the break. Foot note 27 It should be obvious that we have been concerned with comparing--as well as searching, pointing, reading, and so forth--in this latter sense all along. And it is unproved that I ever continue to compare in this sense while I am forgetful of what I am doing.

To be sure, a maximally reflected verdict on (AN) would have to consider further objections. But I hope the above discussion has contributed to make (AN) seem plausible.

Now, (AN) is a claim only about comparing . What about the other phenomena under discussion, searching, pointing, understanding, and so forth: Can it be argued, along similar lines, that to exemplify any of these, one must be focally action aware that one is doing so?

At this point, generalisations cannot be made across the board. In the case of some of these other phenomena, this claim may be defensible. For example, it may be defensible to hold that in order to search for a red flower I must be focally action aware that I am searching for a red flower. But this kind of claim is clearly not plausible for all the phenomena at issue. For example, it is clearly false that I understand how to continue a series only if I'm focally action aware that I understand this. Similarly, I think it is clearly false that I intend to do so-and-so only if I'm focally action aware that I intend to do so-and-so.

Thus, we should recognise that Wittgenstein's temptation as broadly conceived has very different degrees of plausibility depending on which phenomenon we are talking about. While the temptation is clearly indefensible in the case of, say, understanding and intending, a certain version of it may well be defensible for, say, comparing and searching.

IV Conclusion

I have distinguished two interpretations of the kind of temptation Wittgenstein tried to exorcise in his investigations into comparing, searching, understanding, reading, and other psychological phenomena. I have argued that (1) on a narrow or sensationalistic interpretation of the temptation, it is right to reject it in each case. But (2) the rejection of these narrow temptations does not give us reason to draw the sort of anti-Lockean conclusions about consciousness and the mind that many have taken the passages to establish. Specifically, it does not warrant the conclusion that no type of conscious state is necessary for, say, comparing or searching or reading. (3) We would, on the other hand, have reason to draw this conclusion if we could show that a certain version of the temptation as broadly conceived should be rejected. But then (4) it is far from clear that this temptation should be rejected in the case of each phenomenon under discussion. Certainly, Wittgenstein's passages do not provide sufficient reason to do so.

Wittgenstein cautioned us, in the Investigations : «When we do philosophy, we should like to hypostatize feelings where there are none» (§598). This is no doubt a valuable warning. If we think that each psychological phenomenon essentially involves some feeling, then we hypostatise feelings where there are none, and thus distort our understanding of the mind. Fortunately, philosophers have largely absorbed this lesson. But 50 years after the Investigations , it seems we have reason to issue a complementary caution: When we do philosophy, we are sometimes led to deny the existence of consciousness where there seem, after all, to be some. For example, we are sometimes led to conclude, on the basis of the claim that one can exemplify a given mental phenomenon without having any particular type of sensation or image or feeling , that one can exemplify that phenomenon without being in any particular type of conscious state . And then we end up denying consciousness where there (essentially, necessarily, or universally) is some, and thus distort our understanding of the mind. This lesson may be just as consequential for our understanding of the mind as the one that Wittgenstein taught. And I think it remains to be fully absorbed. Foot note 28

References
Baker, Gordon and Hacker, Peter (1986) «The Grammar of Psychology: Wittgenstein's Bemerkungen Über die Philosophie der Psychologie », in Ludwig Wittgenstein: Critical Assessments , vol. 2, Stuart Shanker (ed.), London: Croom Helm, pp. 352-72. Originally published in Language and Communication ,2, 1982, pp. 222-44.
Block, Ned (1995) «On a confusion about a function of consciousness», Beharioral and Brain Sciences ,18 , pp. 227-87.
Budd, Malcolm (1991) Wittgenstein's Philosophy of Psychology , London: Routledge.
Chalmers, David (1996) The Conscious Mind: In Search of a Fundamental Theory , Oxford: Oxford UP.
Descartes, René (1984) The Philosophical Writings of Descartes: Volume II , John Cottingham et al. (transl.) by Cottingham et al Cambridge: Cambridge UP.
Flanagan, Owen (1992) Consciousness Reconsidered , Cambridge, Mass.: MIT Press.
Güzeldere, Güven (1997) «The many faces of consciousness: A field guide» in The Nature of Consciousness: Philosophical Debates , Ned Block et al (eds.), Cambridge, Mass.: MIT Press, pp. 1-67.
Kripke, Saul (1982) Wittgenstein on Rules and Private Language: An Elementary Exposition , Cambridge, Mass.: Harvard UP.
Locke, John (1975) An Essay Concerning Human Understanding, Peter H. Nidditch (ed.), Oxford, Clarendon.
Malcolm, Norman (1970) «Wittgenstein on the nature of mind», in Studies in the Theory of Knowledge , Nicholas Rescher (ed.), Oxford: Blackwell, pp. 9-29.
McGinn, Colin (1984) Wittgenstein on Meaning: An Interpretation and Evaluation , Oxford: Blackwell.
Nagel, Thomas (1974) «What is it like to be a bat?», Philosophical Review ,83 , pp. 435-50.
Pitcher, George (1964) The Philosophy of Wittgenstein , Englewood Cliffs, N.J.: Prentice-Hall.
Searle, John (1992) The Rediscovery of the Mind , Cambridge, Mass: MIT Press.
Strawson, Galen (1994) Mental Reality , Cambridge, Mass: MIT Press.
Tye, Michael (2003) «Qualia», in The Stanford Encyclopedia of Philosophy , Edward N. Zalta (ed.), URL =
Wakefield, Jerome, and Dreyfus, Hubert (1991) «Intentionality and the phenomenology of action», in John Searle and his Critics , Ernest Lepore and Robert van Gulick (eds.), Oxford: Blackwell, pp. 259-70.
Wittgenstein, Ludwig (1965) The Blue and Brown Books [BB], New York: Harper & Row.
Wittgenstein, Ludwig (1997) Philosophical Investigations/Philosophische Untersuchungen [PI], second ed., G.E.M. Anscombe (transl.), Oxford: Blackwell.
Wittgenstein, Ludwig (1980) Remarks on the Philosophy of Psychology, vol. 1-2 [R1-R2], G.E.M Anscombe and G.H. von Wright (eds.), Oxford: Blackwell.
Wittgenstein, Ludwig (1981) Zettel [Z], second edition, G.E.M Anscombe and G.H. von Wright (eds.), Anscombe (transl.), Oxford: Blackwell.
Pär Sundström
Department of Philosophy and Linguistics
Umeå University
Sweden


[Foot Note 1]
It is true that there has been, in philosophy and also to some degree in cognitive science and the neurosciences, a renewed interest in consciousness lately. But this interest often coexists--at least in philosophy--with the view that consciousness is inessential for much or most of our mental life. That combination of attitudes is expressed by, e.g., Chalmers (1996) . Despite being deeply fascinated by consciousness, Chalmers maintains that large domains of the mental can be accounted for without any invocation of it.
[Foot Note 2]
I use the following abbreviations. BB :The Blue and Brown Books .PI :Philosophical Investigations .Z:Zettel .R1-R2: Remarks on the Philosophy of Psychology, vol. 1-2.
[Foot Note 3]
I am not alone in suggesting this. Notably, John Searle (1992) has argued vigorously that the role of consciousness in the mind has been underestimated in recent philosophy of mind and cognitive science. Galen Strawson (1994, chapters 6 and 11) also argues this point. I view my undertaking in this paper as supplementing these arguments. While Searle and Strawson defend the central role of consciousness in the mind mainly against claims arising out of cognitive science, I defend the same general view against claims arising out of the Wittgensteinian tradition.
[Foot Note 4]
See section 3, claim (4)
[Foot Note 5]
In the following, I am trying to summarise assessments such as those arrived at by Budd, pp. 21-6, McGinn , pp. 93-117, Pitcher , chapter 11, Kripke, pp. 40-51, and Baker and Hacker, pp. 358-9.
[Foot Note 6]
The impression is based on the following. While many commentators (for example, Pitcher, McGinn and Kripke) have had their focus on the phenomena of meaning and understanding , at least some (for example, Budd, and Baker and Hacker) have explicitly taken Wittgenstein to demonstrate something about a broader range of phenomena. And no commentator that I am aware of has discussed the possibility that Wittgenstein's results may apply to only some of the phenomena considered in the relevant passages.
[Foot Note 7]
See Kripke, pp. 49f., McGinn, pp. 104ff., Baker and Hacker , p. 359, and also Budd, p. 25.
[Foot Note 8]
See Kripke, pp. 41f., McGinn, p. 96, and Pitcher, pp. 259f.
[Foot Note 9]
See McGinn, pp. 96-7, Pitcher, pp. 259f., Baker and Hacker, p. 359, and Budd, p. 26.
[Foot Note 10]
See Pitcher , p. 260, and Kripke, pp. 43ff.
[Foot Note 11]
Distinctions that seem similar to the one I am drawing here have been recognised as important to the assessment of the passages under discussion. Examples are Budd's distinction between an «extrinsic» and an «intrinsic» conception of states of consciousness ( Budd, pp. 22-3), and McGinn's distinction between «quotidian» and «queer» conscious contents ( McGinn, pp. 7-9 ). ( See also Kripke, pp. 51ff., and PI ,§§187-97.)
Despite drawing these distinctions, however, neither Budd nor McGinn pays any serious attention to the second--broad--interpretation of the temptation above. This may suggest that their distinctions do not, after all, coincide with mine. ( If they have drawn just the distinction that I draw, then I believe they have failed to appreciate the consequences of this distinction for the assessment of Wittgenstein's passages.)
[Foot Note 12]
Exegetical note: The two interpretations do not seem to me to do equal justice to each of the passages under consideration. I believe some passages suggest a narrow interpretation of the temptation that they discuss, while other passages suggest a broad interpretation. (And in yet other passages, there are aspects that suggest a narrow interpretation and other aspects that suggest a broad interpretation.) But whether this is right or not will not matter to my discussion.
[Foot Note 13]
See the argument for claim (4) below.
[Foot Note 14]
A parallel distinction between the focus of attention and the focus of some mode-specific awareness can be made in other cases. Thus, suppose I attend to something that I see. It may still be possible to distinguish more or less focal elements of my audition. If you were to ask me what I hear, I might be able to answer that I hear a tune without special prompting, but answer that I hear the faint noise of distant traffic only given special prompting.
[Foot Note 15]
This claim is made by Wakefield and Dreyfus, 268.
[Foot Note 16]
See, e.g., Güzeldere , pp. 8-9, and Chalmers, pp. 25-31.
[Foot Note 17]
Tye, section 2. Familiarly, the specification of `consciousness' in terms of `what it is like' became widespread in philosophy through Nagel (1974) .
[Foot Note 18]
Galen Strawson is the most emphatic and articulate defender of this view; see sections 1.3-1.4. But the view is also embraced by, e.g., Block (p. 230), Chalmers (pp. 9-10), and Flanagan (p. 64).
[Foot Note 19]
See also PI , §148, and p. 59, Insert (a); Z, §§72, 75-8, 81-5, 472; R1 , §836; R2 , §63.
[Foot Note 20]
It is not clear to me whether the conditions stated in the passages should be understood as necessary or sufficient or both. I only claim that whether understood as necessary or sufficient or both, focal action awareness seems to satisfy them.
[Foot Note 21]
Other passages that invite a broad interpretation of the temptation under treatment are the discussions of writing a letter to so-and-so, and of looking for a photograph in a drawer, in Z, §§7 and 8 respectively.
[Foot Note 22]
To draw this kind of conclusion, one must be careful to really identify the most plausible candidates, among the states of consciousness, for being necessary accompaniments of whatever phenomena one investigates. An illustrative failure to do so is provided by one passage from Malcolm . In the relevant passage (p. 10), Malcolm initially follows Wittgenstein in rejecting the idea that a person who points to an object's colour must attend to, and in that sense be aware of, the object's colour . With this rejection, I have no quarrel. But on the basis of it, Malcolm proceeds to the conclusion quoted above, that pointing to an object's colour may have «nothing to do» with what goes on in mind or thoughts. And this conclusion is not justified. It is not justified because attending to, or being aware of, an object's colour is not the most plausible candidate, among the states and processes of mind and thought, for being necessary to pointing to an object's colour. A more plausible candidate is being aware that one is pointing to the object's colour. (And clearly I can be aware that I am pointing to an object's colour without attending to, or being aware of the object's colour .)
[Foot Note 23]
The distinction between abilities and exercises of abilities is often blurred in commentaries on Wittgenstein. For example, Budd claims that Wittgenstein, in the Investigations, by `reading' understands «the ability to follow certain kinds of rule» (p. 27; emphasis added). But what Wittgenstein says is: «reading is here the activity of rendering out loud what is written or printed» ( PI , §156; emphasis added.) Later on, Budd infers that the rules of chess need not be present in the mind of someone who decides to play chess, on the ground that those rules need not be present in the mind of someone who can play chess (p. 37).
It is worth making a note about understanding in this context. The word `understanding' can be applied to (i) something one can be said to do on a given occasion--such as understanding (or meaning) something by an expression or gesture--as well as to (ii) something one cannot be said to do on a given occasion--such as understanding (or knowing) the meaning of an expression. (By saying that understanding in the former sense is something one can be said to do on an occasion, I don't mean to suggest that it is necessarily an action . The main point is that it is something that takes place at a time . Thus, I can understand--or mean--one thing by `bank' on one occasion, and another thing by `bank' on a different occasion. But one may object to saying that these are actions on the ground that it seems nonsensical to order somebody to understand or mean an expression in a certain way. Cf. Z, §51.) Now, understanding in sense (ii) may well be an ability; but understanding in sense (i) is not. To understand (or mean) something by an expression on a given occasion is the exercise of an ability. Nevertheless, the distinction between these two kinds of understanding has been blurred in commentaries on Wittgenstein as well.
[Foot Note 24]
These distinctions have been blurred in commentaries on Wittgenstein as well. For example, Baker and Hacker say this:
Since there is no such thing as private samples ... psychological concepts cannot be explained by introspection. All introspective reports can provide is a variety of experiences or phenomena accompanying understanding, believing, fearing, wanting. These accompaniments are neither necessary nor sufficient for the application of such psychological predicates (p. 359.)
This passage says, first, that a concept like fearing cannot be explained by introspection, and second, that nothing reportable on the basis of introspection is necessary or sufficient for the application of predicates like «fearing». But these claims, uttered in one breath, are very different. The latter is a claim about when somebody falls under the concept fearing ; the former is a claim about what it takes to acquire it.
[Foot Note 25]
This is, of course, what Arnauld pointed out in the fourth set of objections to Descartes' Meditations .
[Foot Note 26]
This objection echoes the short discussion of looking for a photograph in a drawer in Z, §8.
[Foot Note 27]
Familiarly, Wittgenstein himself suggested that many psychological predicates display this ambiguity, and that our understanding of psychology would be enhanced if we had different words for the different senses. See PI ,§577.
[Foot Note 28]
Work on this paper was funded by The Bank of Sweden Tercentenary Foundation and The Swedish Foundation for International Cooperation in Research and Education (STINT). For helpful discussions and comments, I thank Eddie Cushman, Mats Furberg, Ingvar Johansson, Sten Lindström, Peter Nilsson, Fred Stoutland, Barry Stroud and Bertil Strömberg.