NOT FOR CIRCULATION OUTSIDE N.E.H. SUMMER SEMINAR: DRAFT ONLY

 

           From CONSCIOUSNESS AND PERSONS: UNITY AND IDENTITY

     MICHAEL TYE

 

   CHAPTER 1: THE UNITY OF PERCEPTUAL EXPERIENCE AT A TIME

1) It is very widely supposed in both philosophy and psychology that the senses function as largely separate channels of information which generate different sense-specific impressions or experiences.  For example, I see some flowers, a fence, two squirrels, and in seeing them, I undergo visual experiences.  Listening to a nearby bird singing, I hear melodious sounds and in so doing, I am the subject of auditory experiences.  Finding a broken egg on the ground, and smelling its pungent odor, I experience olfactory experiences.  Placing a chocolate in my mouth and tasting its sweetness, I have gustatory experiences.  Running my fingers over the bark of a tree and feeling its roughness, I experience tactual experiences.

 

2) So, according to the received view, if I am using all five of my senses at a given time, I undergo five different simultaneous perceptual experiences at that time, each with its own distinctive sense-specific phenomenal character.  This generates one version of the Problem of the Unity of Conscious Experience.  How is it that if I am undergoing five different simultaneous perceptual experiences, phenomenologically, it is as if I were undergoing one?  How is it that the five experiences are phenomenologically unified?


3) Suppose that at midday a winetaster is tasting a Cabernet Sauvignon.  He sees the red wine in the wineglass beneath his nose, as he brings the wine to his lips.  He smells the rich bouquet of the wine, as he tastes its fruity flavor in his mouth; and in tasting it, he experiences the liquid touching his tongue and the back of his mouth.   Perhaps, as he does this, he flicks a finger against the glass, thereby producing a high-pitched sound.  One way to describe the winetaster’s phenomenal state is to say that he has an experience of a certain colored shape AND further he has an experience of a certain smell AND in addition he has an experience of a taste AND... etc.  But intuitively, this is unsatisfactory.  It misses something out: the unity of these experiences.  There is something it is like for the winetaster overall at midday, as he brings the wine to his lips and smells and tastes it.  There is a unified phenomenology.  How can this be, if, in reality, he is undergoing five separate experiences?  Of course, for each of these experiences, there is something it is like to undergo the experience.  But there is also something it is like to have these experiences together.  And that remains to be accounted for.

 

4) Here is another example.  Holding a ripe apple in my hand, I experience a red surface and I experience a cold surface.  These experiences aren’t experienced in isolation, however.  They are experienced together.    This is part of the phenomenology of my experience overall.  There is a unity in my experience.  In what does this unity consist, given that I am subject to two different token experiences, one visual and one tactual?


5) Phenomenal unity is not simply a matter of the relevant experiences being directed upon a single object in a spatially localized region.  For one thing, in the winetaster case, although the wine he sees is in the glass, the wine he tastes is in his mouth.  For another, the phenomenological unity of experiences to which I am referring can occur even with experiences directed upon widely separated objects.   Standing by the railing of a ship and smelling the sea air, as I look at the ship’s wake in the ocean, I may hear the sound of a tugboat on my left some distance away.   Again, my overall experience is unified.  It forms a seamless phenomenal whole within which smell, sound, and various visual qualities are phenomenologically present.

 

6) The phenomenal unity of simultaneous experiences is also not a matter of their being actual or potential objects of the subject’s attention.  Allowing for the moment that subjects can attend introspectively to their experiences[1], intuitively the unity of simultaneous experiences can exist even without the attention, just as the experiences can.  Walking along a lane filled with leaves, you see many more leaves than you notice.  If your attention is grabbed by one large, yellow and brown, star-shaped leaf, you do not cease to see the other leaves.  They do not vanish from your visual experiences.  They simply recede into the phenomenal background.  Likewise, if some nesting birds nearby take flight and your attention is caught by one shrill sound in particular.  The other sounds are still there -- your auditory experience still encompasses them phenomenally -- even though your attention is focused on one sound in particular.   As you direct your attention appropriately, the phenomenal unity of the visual and auditory experiences is revealed or disclosed -- as you notice the shrill call of the bird while staring at the star-shaped leaf, say -- but intuitively the unity is not created by that attention.  It is there beforehand, unnoticed. 

 

7) This point also undermines the view, popular historically, that unity is imposed on the different sense experiences by thought.  The Kantian suggestion that experiences are unified by the capacity of their subjects to think of them as their own fails to come to grips with the obvious fact that unity is there in the experiences, whatever their subjects can or cannot think.  It is certainly true that without the relevant concepts, a person cannot recognize that there is phenomenal unity and thus, in these circumstances, he or she is ‘blind’ to it.   But the unity itself is not a cognitive matter.   It perhaps is also worth noting that Kant’s proposal fails to do justice anyway to the case of split brain patients who are certainly able to self-ascribe their perceptual experiences, but whose experiences, after the commissurotomy, are disunified (at least in certain special experimental situations).

 

8)  If there really is something it is like to undergo all the sense-specific experiences together at the same time -- if there really is a phenomenological unity--, then there must be an encompassing experience, one that includes the other experiences within itself.  That experience is the bearer of the total, unified phenomenology.

 

9) Note that this experience cannot just be a conjunction of the five modality specific experiences.  The conjunction of two experiences isn’t itself an experience at all.  The maximal experience must be a new experience, one that unifies the other experiences into a single phenomenological whole.  As Bayne and Chalmers put the point in a recent discussion:

At any given time, a subject has a multiplicity of experiences...  These experiences are distinct from each other... But at the same time,.... they seem to be unified, by being aspects of a single encompassing state of consciousness.

 

..... this encompassing state of consciousness .... can be thought of as involving at least a conjunction of many more specific conscious states...  But what is important, on the unity thesis, is that this total state is not just a conjunction of conscious states.  It is also a conscious state in its own right. (forthcoming, p. 000 )

 

10)  That experiences bundle together to form overarching experiences is a view that has counterparts, of course, within each sense.  For there is phenomenal unity not just across senses but within them too.  Thus, Wilfrid Sellars (1968, p. 000) remarks:

           A sense-impression of a complex is a complex of impressions.

Likewise, Sydney Shoemaker (1996, p. 177) says:

...the visual experience of a spatially extended thing is a synthesis of visual experiences of parts of that thing.

 

11)  Seen from this vantage point, the problem of the unity of conscious experience, as it applies to the case of simultaneous perceptual experiences, is, first and foremost, to give an account of the nature of the unifying experience in relation to the other experiences.

 

                    II

12) The problem, stated in this way, is threatened by an infinite regress.  If what it is like to undergo the overall or maximal experience is different from what it is like to undergo the component sense-specific experiences, E1-E5, then there must be a unifying relation between the latter experiences that is itself experienced.   The experience of the unifying relation is not itself a sense-specific experience.  But it is an experience nonetheless; for if there were no experience of the unifying relation, then there would be nothing it is like to have the sense-specific experiences unified.  There are, then, it seems, six experiences: the five sense-specific ones and the experience of unity.   However, the maximal experience isn’t just a conjunction of experiences.  It is a genuinely new unified experience with its own phenomenology.  So, what unites the six experiences together?  It seems that there must be a further unifying relation that binds the experiences.  This relation, however, must itself be experienced.  For the unity is phenomenal.   And now a regress has begun to which there is no end.  

 

13)   There is also a real question as to whether there is a maximal, unifying experience in the first place.  For consider three simultaneous unified experiences, e1, e2, and e3.   If the unity or experienced togetherness of any two experiences requires that there be a unifying experience, then the unity of e1 and e2 requires a further experience E that includes them.  Since E is unified with e3, another experience E’ is now required.   But E’ is also unified with E; so a further experience, E’‘, that includes E and E’ is needed.  And the unity of E’‘ with E and E’ necessitates yet another experience; and so on without end. 

 

14)  Another pressing difficulty is that we are not introspectively aware of our experiences as unified; for we are not aware of our experiences via introspection at all.  This needs a little explanation.


     Suppose that you have just entered a friend’s country house for the first time and you are standing in the living room, looking out at a courtyard filled with flowers.   It seems to you that the room is open, that you can walk straight out into the courtyard.  You try to do so and, alas, you bang hard into a sheet of glass, which extends from ceiling to floor and separates the courtyard from the room.   You bang into the glass because you do not see it.  You are not aware of it; nor are you aware of any of its qualities.  No matter how hard you peer, you cannot discern the glass.  It is transparent to you.  You see right through it to the flowers beyond.  You are aware of the flowers, not by being aware of the glass, but by being aware of the facing surfaces of the flowers.   And in being aware of these surfaces, you are also aware of a myriad of qualities that seem to you to belong to these surfaces.  You may not be able to name or describe these qualities but they look to you to qualify the surfaces.   You experience them as being qualities of the surfaces.  None of the qualities of which you are directly aware in seeing the various surfaces look to you to be qualities of your experience.  You do not experience any of these qualities as qualities of your experience.   For example, if redness is one of the qualities and roundness another, you do not experience your experience as red or round.

     If your friend tells you that there are several ceiling-to-floor sheets of glass in the house and that they all produce a subtle change in the light passing through them so that things seen the other side appear more vividly colored than is usually the case, as you walk gingerly into the next room, you may become aware that there is another partitioning sheet of glass before you by being aware of the qualities that appear to belong to nonglass surfaces before your eyes.  You are not aware of the second sheet of glass any more than you were aware of the first; but you are now aware that there is a sheet of glass in the room by being aware of qualities apparently possessed by nonglass surfaces before you.

     Visual experiences are like such sheets of glass.  Peer as hard as you like via introspection, focus your attention in any way you please, and you will only come across surfaces, volumes, films, and their apparent qualities.  Visual experiences are transparent to their subjects. We are not introspectively aware of our visual experiences any more than we are perceptually aware of transparent sheets of glass.  If we try to focus on our experiences, we ‘see’ right through them to the world outside.  By being aware of the qualities apparently possessed by surfaces, volumes, etc, we become aware that we are undergoing visual experiences.  But we are not aware of the experiences themselves.[2]

     This is true, even if we are hallucinating.  It is just that in this case the qualities apparently possessed by surfaces, volumes, etc before our eyes are not so possessed.  The surfaces, volumes, etc do not exist.

 

15)  Introspection, on the view just presented, is importantly like displaced perception or secondary seeing-that, as Fred Dretske (1995) has observed.  When I see that the gas tank is nearly empty by seeing the gas gauge or when I see that the door has been forced by seeing the marks on the door, I do not see the gas tank or the forcing of the door.  My seeing-that is secondary or displaced.   I am not aware -- I am not conscious -- of either the gas tank or the forcing of the door.  I am aware of something else -- the gas gauge or the marks on the door -- and by being aware of this other thing, I am aware that so-and-so is the case.

     Similarly, in the case of introspection of a visual experience, I am not aware or conscious of the experience itself.  I am aware that I am having a certain sort of experience by being aware of something other than the experience — the surfaces apparently outside and their apparent qualities.[3]

 

16)  What is true above for the case of vision is true for the other senses.  For example, we hear things by hearing the sounds they emit. These sounds are publicly accessible.  They can be recorded.  Similarly, we smell things by smelling the odors they give off.   They too are publicly accessible.   You and I can both smell the foul odor of the rotting garbage.  Odors, like sounds, move through physical space.  We taste things by tasting their tastes.  One and the same taste can be tasted by different people.  Some tastes are bitter, others are sweet.  When we try to introspect our supposed experiences of hearing, smelling, and tasting, the qualities of which we are directly aware are qualities we experience as being qualities of sounds, odors, and tastes.  It seems very natural to suppose that among these qualities are the following: pitch, tone, loudness, pungency, muskiness, sweetness, saltiness, sourness.  But be that as it may, the important point is that when we introspect, the particulars we come across, if any, are sounds, odors, and tastes — the owners, if such exist, of the qualities we find in introspection.  We do not come across, in addition to these things, token experiences of hearing, smelling, and tasting.  Nor do we come across any overarching or maximal token perceptual experience either.

 

17) If we are not aware of our experiences via introspection, we are not aware of them as unified.  The unity relation is not given to us introspectively as a relation connecting experiences.  Why, then, suppose that there is such a relation at all?

.

18) These points serve to remind us that one way to respond to a philosophical problem is to challenge the problem itself instead of accepting it, on its own terms, and trying to solve it. 

 

                    III

19)  Consider the following example (from Parsons 1970) as a way of beginning to get at what seems to me wrong.  Suppose that this statement is true:

     (S1) Jones writes illegibly and Jones writes painstakingly.

It does not follow that

     (S2) Jones writes illegibly and painstakingly,

at least on one natural reading of (S2).  For if Jones writes illegibly but not painstakingly with his left hand and painstakingly but legibly with his right, then (S1) is true but (S2) false.  In this case, there is an event of Jones’ writing illegibly and there is an event of Jones’ writing painstakingly, but these are two distinct events.  There is no event of Jones’ writing both illegibly and painstakingly.  So, (S1) does not entail (S2), but (S2) clearly does entail (S1).  Given the event of Jones’ writing both illegibly and painstakingly, there is, of course, the event of Jones’ writing illegibly.  For the latter is the very same event as the former under a less broad description.  Likewise for the event of Jones’ writing painstakingly.

 

20)   In the case where (S2) is true, there is a kind of unity to Jones’ writing.  Illegibility and painstakingness are combined together in a single instance of writing.  That unity is lacking in the case that (S1) is true and (S2) is false.  But where (S2) is true, there aren’t two different writings, one painstaking and the other illegible, which somehow are unified together to produce a third, overarching writing that includes them.  There is just one writing that may be described in more or less encompassing ways.[4]  To suppose otherwise is to create a new pseudo-problem: the problem of the unity of the event of Jones’ writing illegibly and painstakingly. 

 

21) Here is another example.  Suppose it is lunchtime and I have a sudden and strong desire for a pint of beer with a ham sandwich.  In having this desire, of course, I have a desire for a pint of beer.  It is also true that I have a desire for a ham sandwich.  But patently I don’t have three sudden desires here.  Nor is it the case that having a desire for a beer together with a desire for a ham sandwich just is having a desire for a pint of beer with a ham sandwich.  I might want a beer and also want a ham sandwich while finding the idea of having the two together repellent.

     My sudden desire for a pint of beer with a ham sandwich is a single desire that can be described in multiple ways.  The description ‘desire for a pint of beer’ is incomplete, but unlike the description ‘desire for a pint of beer alone’, it is not inaccurate.

 

22) These remarks apply mutatis mutandis, I want to suggest, to the problem of the unity of conscious experience, as it is usually conceived.  There aren’t five different or separate simultaneous experiences somehow combined together to produce a new unified experience.  Nor are there are multiple simultaneous unified experiences within each sense.  To be sure, if I am the winetaster, the statement

        (S3) I have an experience of a bright red shape and I have an experience of                            a fruity taste and I have an experience of a cassis odor.....

Is true.  And, given that the case is one of phenomenal unity, the following is true too:

        (S4) I have an experience of a bright red shape and a fruity taste and a                                    cassis odor...

Moreover, (S4) entails (S3).  But there is just one experience here, an experience that can be described less fully as my experience of a bright red shape or as my experience of a fruity taste, etc.

 

23) On this view, there really are no such entities as purely visual experiences or purely auditory experiences or purely olfactory experiences, etc in normal, everyday consciousness. Where there is phenomenological unity across sense modalities, sense-specific experiences do not exist.  They are the figments of philosophers’ and psychologists’ imaginations.  And there is no problem, thus, of unifying these experiences.  There are no experiences to be unified.  Likewise, within each sense: there are not many simultaneous visual experiences, for example, combined together to form a complex visual experience.  There is a single multi-modal experience, describable in more or less rich ways.

 

24) ‘Stuff and nonsense’, you may say.  ‘The proposal is empirically false.  Visual experiences are known to arise in the visual cortex, auditory experiences are known to arise in the auditory cortex, and so on.   In the winetaster case, it is surely undeniable that visual experiences are tokened in the winetaster’s visual cortex, as he views the wine he is tasting, experiences of just the same phenomenal type as those that would have been tokened in that cortex, had the situation been the same but his other senses blocked from any information. Of course, these experiences exist!’

25)   By way of reply, let us for a moment indulge in the fiction that there are purely visual experiences in everyday consciousness of the external world.  In these experiences, shape and color are unified.  If, for example, I view a green square, my visual system represents the greenness and the squareness in separate places in the brain; but these qualities aren’t experienced as separate.  They are experienced as qualities of a single thing.  I have an experience that is object-unified, as we might say, even though its physical basis is disunified.[5]  What is the relationship between the experience and its physical basis?  The answer I favor is that the experience, assuming there is one, is constituted by a certain combination of separate and largely independent physical events in the visual cortex, but it is not token identical with that combination.  This answer gives the experience a physical nature; moreover constitution is the relation that bonds macro-events and micro-events, macro-states and micro-states, macro-objects and micro-objects generally.


     Consider, for example, a single cloud in the sky.  The cloud is an aggregate of water droplets.  The ‘is’ in the last sentence is not the ‘is’ of identity.  The cloud in the sky could survive the loss of a few of its constituent water droplets (if, say, a highly localized strong gust of warm air were to cause them to evaporate).  Not so any aggregate of water droplets that contains them.  The loss of those droplets would destroy the original aggregate.  So, the cloud has a modal property the aggregate of water droplets lacks, that of possibly surviving the loss of such-and-such droplets.  It follows by Leibniz’ Law that the cloud is not identical with any aggregate of droplets. In general, ordinary, everyday macro-objects are not identical with aggregates of their parts, since the former differ in their modal properties from the latter.  My car, for example, might have had a different carburettor, but the aggregate of its actual parts could not have failed to contain the actual carburettor.   The car, thus,  is not identical with the sum of its parts.  The relationship rather is one of constitution or composition.

     One need not resort to modal properties to make the above points.  Actual properties will do in some cases.  The clay that constitutes a pot exists before the pot does.  The lump of silver that is melted and formed into a coin exists before the melting process, but the coin does not.    The clay is thus not identical with the pot; the lump of silver not identical with the coin.

     Likewise for macro-events.  Consider the eruption of Mount Vesuvius.  Intuitively, that very eruption might have spewed forth an imperceptibly smaller amount of lava.  Had it done so, the micro-events taking place in the spatial region of Vesuvius’ eruption would have been minimally different and thus the aggregate of those events in that counterfactual situation is not the same as the actual aggregate.  The eruption, therefore, has a modal property that the underlying cluster of micro-events lacks, and the former is not identical with the latter.


     Alternatively, consider all the micro-physical events that compose the emergence of North America (Burge, 1986).  Imagine that these events are embedded within a much larger land mass, so that in the counterfactual situation there is no such thing as North America and its emergence.  Then the aggregate of micro-events has the property of possibly existing without North America; but the event of North America’s emerging does not.  That event is not identical with the aggregate.  The aggregate constitutes the emergence of North America in actual fact, but the relationship is not one of identity.

26) Consider next the following example.   A large chunk of clay is used to make a statue at time t.  The clay constitutes the statue without being identical with it.  Suppose counterfactually that at time t’, where t’ is later than t, an artist cleverly removes much of the clay without remolding it so as to leave behind a small clay pot.  In the counterfactual situation, the clay that remains constitutes a pot at t’.  But in the actual situation it does not.  In actual fact, no clay is removed.  There is, in actual fact, no tiny pot within the statue.  There is only the statue.

     Within the aggregate of lumps of clay composing the statue, there is a smaller aggregate of clay lumps that in a certain counterfactual situation composes a pot.  In actual fact, the smaller aggregate does not compose a pot.  Indeed, it does not by itself  actually compose or constitute any ordinary thing.  Rather that aggregate together with the remaining aggregate form a larger aggregate that composes the statue. 

 

27) I hope that the relevance of all of this is becoming clear.   On my view, at the given time the winetaster is subject to a single experience that represents the color of the wine, the sound of the wine-glass, as it is flicked, the smell of the wine, and so on.  This experience is constituted by a combination of largely independent physical events going on in separate regions of the brain.  Within that combination of events, there is a cluster of events (call it ‘C’) occurring in the winetaster’s visual cortex.   In the extraordinary counterfactual situation in which the winetaster’s nonvisual senses are all blocked, so that no nonvisual information gets in, the winetaster is left with a purely visual experience.    And in that counterfactual situation, C, in the absence of the other pertinent actual physical events, constitutes a visual experience.  But it does not follow from this that in actual fact C constitutes a purely visual experience.  In actual fact, C (wholly) constitutes no experience at all.   There is just one unified experience the winetaster undergoes, and C, in conjunction with the relevant events in the auditory cortex, the olfactory cortex, etc, constitutes that. 

 

28) Perhaps it will be replied that in the example of the statue and the pot, the aggregate of lumps of clay that counterfactually constitutes the pot is in the actual world a purely arbitrary part of the statue, with nothing to mark it out from any number of other arbitrary parts of the statue.  However, the cluster of physical events I have labelled ‘C’, is a nonarbitrary part of the relevant totality of physical events, a token of a physical type with a definite functional role, namely to generate a conscious visual experience with a certain visual unity.

 

29) This begs the question.  I grant that C is a token of a physical type P whose role in a normally functioning brain is to endow the conscious experience of the subject with a visual phenomenology.   But that is certainly compatible with denying that P’s role is to generate in the brain a token experience with an exclusively visual phenomenology.  To suppose that P’s role is the latter is to take for granted the truth of the view I am opposing.

 

30) Furthermore, the fact that C is a nonarbitrary part of the whole combination of physical events constituting the experience is not to the point.   Suppose events E1 and  E2 together actually constitute event F.  Suppose E1 could have occurred without E2  and further that had E1 done so, it would have (wholly) constituted event G.  Still, this is no guarantee that E1 actually constitutes G.  For example, my arm and hand movement relative to Smith and Smith’s arm and hand movement relative to me constitute a certain fight.  Smith’s movement might have occurred without my movement.  Had it done so, it would have constituted an act of aggression on the part of Smith.  But in actual fact that act of aggression does not exist.  In actual fact, Smith’s arm and hand-movement relative to me is a counterpunch; for I hit Smith first.

     Thus, just as Smith’s arm and hand movement might have constituted an act of aggression although in reality it does not, so too the cluster of events, C, might have constituted a purely visual experience but in reality it does not.

 

31) Suppose now that the counterfactual situation I have envisaged for C becomes a reality.  As the winetaster tastes the wine, some extraordinary neural malfunction causes the events other than C that constituted the winetaster’s experience prior to the malfunction to cease.  Before the malfunction, on the proposed view, C does not constitute a purely visual experience.   After the malfunction, it does.   This, it may be charged, is strange.  Why the radical change in what C does?

 

32) The answer, as earlier, is that there is no change.  But before and after the malfunction, C does the same thing: it endows the conscious experience of the subject with a visual phenomenology.  The difference is that before the malfunction, the experience of the winetaster does not have a purely visual phenomenology; after the malfunction, it does.

 

33) Consider next the following example.  Suppose I hear a conversation on my left, as I look at a bed of roses laid out in front of me. Intuitively, my auditory experience — that very experience — could have occurred without my visual experience.  On my account, however, that isn’t possible.  So much the worse, it may be said, for my account!

 

34) Too fast, I reply.   When I try to introspect my auditory experience, I fail.  As noted earlier, what I actually come across are the sounds and the auditory qualities the experience represents.  By being aware of those sounds, I am aware that I am undergoing an auditory experience.  But I am not aware of the token vehicle of that content.  The sounds I experience could have existed without my visual experience of the roses.  Moreover, I could certainly have undergone an experience that represented those sounds (or sounds just like them) without also representing the colors and shapes of roses.  That, I suggest, is all that untutored intuition requires here, once we are clear on what introspection does and does not reveal.  And it is perfectly compatible with these claims that I am the subject of just one experience, an experience that is audio-visual in character.

 

35) But what if my auditory experience goes on longer than my visual experience?  Then my auditory experience has a temporal property my visual experience lacks, and there cannot be a single experience after all.

 

36) I grant that I can experience a sound that continues, in my experience, after I experience anything visually.  But this is all in the content of the experience.  Initially, what I experience is that a sound with a certain pitch and loudness is accompanied by a certain color and shape.  As time goes on, the experienced content changes.  No longer is any shape or color represented.  The sound is represented on its own.  This certainly shows that the represented sound is not the same as the represented shape or color.  But it does not show that there is more than a single experience at a time.

      I do not deny, of course, that difficult questions arise concerning the individuation of experiences through time.  Is the experience I undergo initially -- an experience with an audio-visual content -- the same as the experience I undergo after the color and shape cease?   Is there one experience here with a less rich content through time?  Or is the audio-visual experience replaced by a second purely auditory experience, phenomenally just like the first in its auditory dimension?  Hard questions of individuation through time arise for everyone, however.   They are discussed further in Chapter 4.

 

37) Here is a further worry.  Seeing something entails the presence of a visual experience.  I cannot see X unless X looks some way to me; and for X to look some way to me, it must cause in me a visual experience.  So, to return to the example of the winetaster, since he is seeing the wine in the glass, he must be subject to a visual experience.   However, on the account I am adopting, his experience isn’t really properly classified as visual at all.

 

38) It is indeed true that X cannot look some way to person P unless X produces in P an experience with a visual phenomenology.  But the phenomenology of P’s experience need not be purely or exclusively visual.  It can be partly auditory, olfactory, gustatory, and tactual too.  If a visual experience is understood to be an experience with a visual phenomenology, then the winetaster, as he sees the wine in the glass, is subject to a visual experience.  It’s just that that very experience has a phenomenology that is  auditory, olfactory, gustatory, and tactual as well.

 

39)  Still, it may be complained, on the one-experience view, no explanation is possible for why the beliefs formed directly on the basis of experience are about how things look or how things taste or how they smell, etc rather than about some combination of these.  My response is to deny the datum.  We can know directly that the object looks red and is apparently emitting a loud sound, just as we can know directly that the object looks red and looks round.  We have the ability to attend all at once to qualities, the experience of which (in some cases) requires the use of different senses.  Indeed, it is the exercise of this ability that leads us to think that there is such a thing as synchronic unity in perceptual experience in the first place.  In everyday life, we say things like “The drink looks creamy and it tastes sweet” and “The exhaust smells noxious and it sounds loud.”  The beliefs these remarks are no less cautious or direct than the belief expressed by, e.g., “The card looks square and silvery in color.”  

 

                  IV

40)  If the view of experience I have been defending is correct, then in ordinary, everyday perceptual consciousness, there are no sense-specific experiences to be unified.  So, the problem of unity stated earlier dissolves.   I do not wish to deny, however, that synchronic phenomenal unity is real.

41) The core intuition, lost in the usual way of stating the problem of unity, is that, in normal cases, simultaneously experienced perceptual qualities — the loudness of a sound, the smoothness of a surface, the sweetness of a taste, the pungency of a smell — are experienced together and thus are phenomenologically unified.   These qualities are not qualities of experiences.   They are qualities that, if they are qualities of anything, are qualities of things experienced.    Consider, for example, the case in which I experience a loud noise and a bright flash of light.  The loudness of the noise is unified phenomenally with the brightness of the flash.  Phenomenal unity is a relation between qualities represented in experience, not between qualities of experiences.

 

42) Specifically, phenomenal unity is a matter of the simultaneously experienced perceptual qualities entering into the same phenomenal content.  The perceptual experience a normal perceiver undergoes has an enormously rich, multi-modal representational content -- a content, part of which is nonconceptual, abstract, and appropriately poised.[6]  This part, its phenomenal content, is present not only in veridical cases but also in cases of illusion and hallucination.  It is this content that endows the experience with its phenomenal character (Tye 1995, 2000). 

 

43)  A consequence of the above position is that phenomenal unity goes with the closure of perceptual experience under conjunction with respect to the unified qualities.  Thus, in the case mentioned in (41) in which the loudness of a sound is phenomenally unified for person P with the brightness of a flash of light, the statements

     P has an experience of a loud sound

and

     P has an experience of a bright flash

jointly entail

     P has an experience of a loud sound and a bright flash.

When there is disunity, perceptual experience is not closed in this way.  Cases of simultaneous phenomenal disunification all involve multiple perceptual experiences at a time and multiple phenomenal contents.  Such cases are highly abnormal.  Where they occur, as, for example, with split brain patients, there are simultaneously experienced perceptual qualities entering into different phenomenal contents (each of which is a content experientially represented by the relevant subject at the given time).[7]   


44) Perhaps it will be replied that hallucinations and illusions are not highly abnormal; yet they sometimes involve disunified experiences.  Recall Macbeth’s remark, “Is this a dagger I see before me, the handle towards my hand?  Come, let me clutch thee.  I see thee still, yet I feel thee not.”  Macbeth’s visual experience represented to him a certain colored shape, which he identified as a bloody dagger.  His tactual experience, when he went to touch it, informed him that nothing was there.  Here, it may be said, there are two experiences, one tactual and one visual, that Macbeth underwent at the same time; for a single token experience cannot have an inconsistent content.

     Likewise, in the case of the straight stick in water that looks bent.  Touching the stick, it feels straight.  Touch and vision are thus at odds with one another.[8]  Again, there are two simultaneous experiences.  And again, the subject’s perceptual experiences are disunified.

 

45) These cases, despite how they may initially seem, are not cases of phenomenal disunification.  In the case of Macbeth, his experience by sight and touch was directed upon a single region of space by his hand.  This was part of the phenomenology of his experience overall even though what touch told him led him not to trust his sight.   Similarly, as one looks at the stick in water and touches it with a hand, it certainly seems to one that the stick one is seeing is the same as the stick one is touching.   This, it seems to me, is part of the phenomenology of one’s overall experience.   To be sure, touch tells one something different about the shape of the stick than vision[9]; and they can’t both be right.  But why shouldn’t a single experience have an overall content that is internally inconsistent?  Given the complexity of the processing that underlies the production of experience, it should not be surprising that in some cases a content of this sort is produced.

     There are plenty of examples of experiences with inconsistent contents.  View the  “impossible figure,” shown below.

         INSERT FIGURE 1 HERE

 

Here each set of steps is seen as ascending to the next, even though this is impossible.   Another example is the so-called “waterfall effect,” which involves an illusion of movement (originally of a body of water).  The most dramatic version of this is obtained by staring at a rotating spiral figure.  While rotating, the spiral seems to expand.  But after it is stopped, the spiral seems to contract, while also seeming not to get any smaller.

 

46)   There remains one final objection I want to discuss to the one-experience view.  Consider the experience of a red square next to a green triangle.  That experience, like all other perceptual experiences, has accuracy conditions.   As such, it is a representation.  And intuitively, that representation is complex.  For anyone with the capacity to have an experience of this sort thereby has the capacity to have an experience of a green square next to a red triangle.   The obvious explanation for this connection between capacities is that the two experiences are complex representations sharing the same representational parts.  The experience of a red square next to a green triangle at time t thus has a component representation at t representing a red square and a component representation at t representing a green triangle.   These representations, it may be insisted, have as much a right to be called “experiences” as does the overall representation of which they are parts. 


     Indeed, it may be urged, on the theory developed in Tye 1995 and 2000, it cannot be denied that the component representations are experiences.  For, according to that theory, any perceptual state with a poised, abstract, nonconceptual representational content is a perceptual experience.[10]  And the content, red square, it may be insisted, is just as poised as the overarching content that includes it.  Each content stands ready and available to make a direct difference with respect to what is believed, if attention is appropriately directed.  Furthermore, what is true in this single modality case will be true mutatis mutandis in inter-modal cases too. 

 

47) Consider again the lump of clay that is a statue.  Had that lump of clay been seamlessly embedded within a larger lump of clay which formed a cube, say, it would not have been a statue.  The statue of the earlier example is constituted by a lump of clay that is maximal (that is, not contained within other, larger lumps of clay).   The advocate of the one-experience view can maintain that experiences are, in this way, like statues.  Experiences are maximal PANIC states (states having a poised, abstract, nonconceptual content).  So, even if some proper parts of experiences are representations, they are not themselves experiences.  With this elucidation, the objection dissolves. 

 

48) It is worth adding that it is not entirely clear anyway whether the experience of a red square next to a green triangle really does contain a part that represents a red square and a part that represents a green triangle.  Clearly, the overall representation contains parts that would represent a red square and a green triangle respectively, were they not contained within the overall representation, but it could be insisted that these parts are not  themselves representations, given their actual situation.   In the case of sentences, some parts, namely words, are surely representations.  But it is not so clear that any parts of pictures are.  For some representations, there is arguably a maximality constraint; for others not.

 

49) That completes my discussion of the unity of perceptual experience at a time.  I turn next to the case of bodily experience.[11]

 

 

 

 

 

 

 

 

        NOTES

 


[1].  I don’t believe that this is really possible.  For more, see (14) - (16) below.

[2].This claim is one that the sense-datum theorists would have endorsed, although they would have insisted that the things apparently outside are really immaterial surfaces or sense-data rather than physical surfaces.  After all, sense-datum theorists were at pains to distinguish the act of sensing from the thing sensed.   G.E. Moore is the modern father of transparency.

[3].  There are dissimilarities.  In typical cases of seeing-that, background beliefs play a role in generating the propositional state of awareness.  This is not so in the case of introspection.  For more here, see Tye 2000, Chapter 3.

[4].  Not everyone accepts this claim.  On the view of events elaborated by Jaegwon Kim (19xx) and Alvin Goldman (19xx),  the event of x’s F-ing G-ly is a complex entity consisting of x and the property of F-ing G-ly.  This has the consequence that at the time at which (S2) is true, there are four relevant simultaneous events: Jones’ writing painstakingly and illegibly, Jones’ writing painstakingly, Jones’ writing illegibly and Jones’ writing.  Intuitively, however, events do not individuate in this hyper-fine-grained way.

[5].  I am not suggesting here that object-unity is the same as phenomenal unity.  See paragraph (5) earlier.

[6].  For an elucidation of what it is for a content to be nonconceptual, abstract, and poised, see Tye 1995, 2000.  For further critical discussion and replies, see the web symposium on Tye 2000 at www......

[7].  For a discussion of split brains, see Chapter 5.  In my view, it is a mistake to suppose that the perceptual consciousness of split brain subject is always divided after the commissurotomy.  Indeed, it is a mistake to suppose that even in the highly specialized circumstances in which split brain subjects typically behave anomalously, their consciousness is invariably divided.

[8].  Actually, what really happens is that when one feels the stick, it no longer looks bent: touch corrects vision.  But this does not matter for present purposes, so let us ignore it.

[9].  See ibid.

[10].  I do not deny that perceptual experiences sometimes have conceptual contents too.

[11].  Versions of this chapter formed the basis for talks at Oxford, Yale, the University of Texas at Austin, the University of St. Andrews, the Royal Institute in London, the Southern Society for Philosophy and Psychology, and the Tucson 2002 Consciousness Conference.