3

ASPECTS OF THE PHILOSOPHY

OF PSYCHOLOGY

Since this chaptercovers a great deal of ground, it may be useful to

offer a preliminary overview of its structure.The first sectionshows
how Wittgenstein's private-languageargument undermines any
attemptto regard psychologicalconceptsas referring to a duplicate
realm of events and entities hidden within the human subject; and
the secondsection shows that acceptanceof this argumentdoes not
entail that psychological conceptsmust be treated as a speciesof
behavioural concepts. The third section demonstratesthat Witt-
genstein's attempt to steer between dualist and behaviourist
misconceptionsof the mental results in a view of psychological
conceptswhich can best be summarizedby characterizingthem as
aspectconcepts.The questionof what might constitutecontinuous
aspectperceptionin this domain is then raised; and a provisional
answer is arrived at in the fourth and final section by developing
an interpretationof the conceptof aspect-blindness in opposition to
one proposedby Stanley Cavell.

MYTHS OF THE INNER

In our everydaydealingswith other peopleour knowledgeof - and
relationship towards - their moods, states of mind, and feelings
seems bound up with a perception of their inner life as being
hidden behind, and hinted at by means of, their outward
behaviour. We speculate, with some interest, about the deep
emotionsconcealedby a friend's placid exterior: we sometimesfind
it impossible to imagine what thoughts are going through a
stranger's mind; and we might be unable to judge whether a
grimace of pain masks an intention to deceive. When philosophers

53
PHILOSOPHY OF PSYCHOLOGY

respond to this picture of 'inner' and 'outer' without careful

examinationof the precise techniquesof its application, or of the
specific grammatical relations between conceptsof the inner and
the outer which this picture dramatizesand conceals,they tend to
construct myths of the inner. Cartesianphilosophical systems,for
example, embody the central tenet that inner states - whose
perception in others is a matter of induction from external
behaviouralevidence- are identified in ourselvesby meansof some
speciesof unmediatedrecognition, the inner eye having no barrier
of flesh and blood to penetrate.Such direct acquaintancewith our
private inner events and processes(i.e. with the referents of our
psychologicalconcepts)is accordinglyseenas the necessaryground
of our capacity to grasp the content and significance of the huge
array of correspondingconcepts:for within the constraintsof such
a myth, attachingmeaningto psychologicalterms could only come
about through a mode of private ostensivedefinition -a ceremony
of inward attention through which eachhuman being confers upon
himself the gift of a languagefor the inner.
The radical incoherenceof this notion of a 'ceremony' is the
guiding principle of Wittgenstein'sexplorationof the philosophyof
psychology; it is therefore crucial that I begin this chapter by
delineatingthe soundnessof that principle. As Wittgensteinputs it
in his later writings (RPP, I , 397), there is a direct and an indirect
way of gaining insight into the impossibility of a private ostensive
definition; and his example of the motor-roller illustrates this (cf.
Zettel, 248). According to a certain design for the constructionof a
steam-roller,the motor is locatedinside the hollow roller whilst the
crank-shaftruns through the middle of the roller and is connected
at both ends by spokesto the roller wall; the cylinder is then fixed
to the inside of the roller. At first glancethis constructionlooks like
a machine but in fact it is a rigid system and the piston cannot
move to and fro in the cylinder. Just as we unwittingly deprived
the motor-roller of all possibility of movement, so the private
linguist deprives his ostensivedefinition of the capacity to perform
its intendedfunction; but this can be seendirectly and indirectly in
both situations.We might seedirectly that the motor-roller cannot
function, since one could roll the cylinder from outside even when
the 'motor' was not running; but we can also trace out the linkages
of the mechanismfrom motor to roller and perceivethereby that it
is a rigid construction and not a machine at all. In this brief

54
PHILOSOPHY OF PSYCHOLOGY

exposition of the private languageargument, we shall begin with

the indirect method of perceiving the impossibility.
A private languageconsistsof words which refer to the speaker's
inner experiences- his sensations,feelings, moods, and the rest;
since these inner experiences are regarded as entities whose
presence or absence can be known only to the person whose
experiencesthey are (i.e. since the terms of the languagerefer to
epistemically private entities), it follows that another person is
incapable of understanding the language which gives them
expression. The point is not that another person is simply
incapable of telling whether the words of this languageare being
used accurately, i.e. are being used only when their referent is
present; it is rather that these words are defined by means of the
entities to which they refer, so that another person's lack of
epistemic accessto those entities entails an inability to grasp the
meaning of the terms which refer to them. The private linguist,
however, claims to be able to understandand utilize the words of
his private languagebecausehe can confer meaning upon them.
He can (he asserts)use a term'S' to refer to (e.g.) a sensationby
associatingthe sign with the sensationand then employing the sign
(perhapsby writing it in a diary) when the same sensationrecurs
in the future.
For the moment, we shall ignore any difficulties concerningthe
private linguist's capacity to grasp the conceptof a 'sensation'(as
the term is standardly used) in isolation from our natural
expressionof sensations,and instead concentrateupon the point
that the ceremony by means of which the sign and the sensation
are associatedis intended to be a kind of definition of the sign'S'.
This basic conferral of meaning takes the form of an ostensive
definition - the private linguist says the sign to himself and at the
same time concentrates his attention on the sensation, as if
inwardly pointing to it. Any definition, however, must be capable
of being used as a standardfor distinguishingbetweencorrect and
incorrect use of the sign it governs; and since the samplesensation
referred to in this putative ostensive definition is in principle
accessible only to the private linguist, he alone is capable of
employing that sample as part of a rule governing future uses of
'S'. In other words, if it can be shown that the private linguist
cannot use the sample as a standardof correctnessin the future,
then the original ceremony cannot intelligibly be described as a

55
PHILOSOPHY OF PSYCHOLOGY

'definition' at all, and the private linguist's claim to be able to

confer meaningon the terms of his languagewill have been shown
to be empty.
Since the sign'S' has been defined by referenceto the type of
sensationinvolved in the original ceremony, then - in order to
mean somethingby'S' when he uses it in the future - the private
linguist must be able to produce an accurate memory of that
particular sensation. The key question for the private linguist
therefore becomes:are any criteria available to him which might
give content to the notion of his being able to producean accurate
memory of the original sign-sensationcorrelation? Note that the
question is posed at the conceptual level; we need to examine
whether it makes sense to imagine the private linguist having an
accuratememory of the samplesensation,rather than whether we
can offer a cast-iron guaranteeof the accuracyof this individual's
memory on a given occasion.It is the intelligibility rather than the
truth of the private linguist's claim to have rememberedthe sample
sensationthat is at issue.
Let us imagine a concrete situation: at some time after the
original ceremony, the private linguist says 'I have "S"'. In order
to check whether he has rememberedcorrectly to which sensation
's' refers, he calls to mind his memory of the sample sensation
upon which he concentratedhis attentionin the original ceremony.
The private linguist feels that this processof appealingfrom one
memory to another is precisely analogousto the case of someone
checking his memory of a train departure-timeby calling to mind
how a page of the timetable looked; and since the latter procedure
is of a sort which is ubiquitous and unproblematicin everydaylife,
the former processmust also be legitimate.
In fact, however, the analogy breaks down becausethe private
linguist is not appealingfrom one memory to another,independent
one. In order to check whether he has correctly rememberedto
which sensation'S'refers (i.e. the meaning he claims to have
conferred on'S'), he calls to mind a memory of his original
sign-sensationcorrelation- perhapsin the form of a mental table
in which the sign'S' is placed next to a memory-sampleof the
relevant sensation;but since this table exists only in his mind, all
he can do is rememberwhich sample goes with'S', i.e. remember
what'S' means.And, of course,it is precisely his memory of what
's' meansthat he is trying to confirm. The processof looking up a

56
PHILOSOPHY OF PSYCHOLOGY

mental table to justify his memory of what'S' means is not

independentof what it is to justify; it is in fact merely a repetition
of the processwhich standsin need of justification, and so cannot
count as a 'justification' of it at all. It is precisely analogousto the
caseof someonebuying a secondcopy of the morning newspaperin
order to confirm the correctnessof its constituentreports.
At this point, the private linguist might simply decide to rely
upon the correctnessof his memory; using his memory twice over
may not strengthenhis case, but this does not in itself entail that
his memory is not to be relied upon at all. After all, in everydaylife
it is often the casethat we have no independentmeansof justifying
our memory-claims,but this does not rendersenselessthe question
of our memory's accuracy.
Here, v ~ reach the crux of the argument.The private linguist's
original ceremony was intended to ensure that'S' was correctly
used if and only if the type of sensation picked out in that
ceremony recurred, and so it follows that the private linguist's
memory of the sample-sensation can only be used as a standardof
correctnessif it is the right memory, i.e. if it is a memory of the
sensationhe originally labelled'S', as opposedto one of another
type of sensationor indeed of any other element in his stream of
consciousness.Accordingly, before he can use his memory of the
sampleof'S' as a standardof correctness,he must be able to show
that this memory is correctly describableas a memory of'S' (rather
than of anything else); but ex hypothesithe only standardavailable
for distinguishing between correct and incorrect use of'S' is the
memory of the sampleitself. It is not that the private linguist must
make use of the same memory a second time to confirm itself;
rather, in order to rememberwhat'S' means,he must be capable
of accurately rememberingthe sample of'S', but in order to be
capableof telling that it is indeed a memory of'S', he must know
what'S' means.Before the private linguist can intelligibly use his
memory of'S' as a standardof correctness,he must first employ it
as a standardof correctnessin order to check upon its suitability
for that role.
No sceptical claim about memory is being invoked here:
Wittgensteinis not claiming that the private linguist has no means
of verifying the accuracyof his memory, but rather that the very
notion of his rememberingthe putative sign-sensationcorrelation
aright lacks any criteria that might renderit intelligible. In order to

57
PHILOSOPHY OF PSYCHOLOGY

identify his memory as a memory of the sample of'S', he must

know what'S' means (i.e. what counts as correct or incorrect
applications of it), despite the fact that it is precisely this
knowledge (of the meaning of'S') that he is trying to remember.
The private linguist makes incoherent demandson his memory,
asking that it perform a task that cannotintelligibly be formulated;
he has unwittingly constructeda rigid set of conceptuallinkages
and thereby deprived himself of the possibility of meaningful
linguistic movement- for if it can make no senseto conceiveof the
private linguist as correctly remembering the sample sensation
used in the original ceremony, and if that ceremony cannot
therefore be said to have established a rule (a standard of
correctness)governing the use of'S', then the ceremonyof private
ostensivedefinition cannot intelligibly be described as a 'rule' or
'definition' at all. The idea that a system of rules for word use
might be grounded on private ostensive definitions is simply
incoherent.
To return to our analogy with the motor-roller: it was also
possible to perceive directly that such a motor-roller could not
function simply by noting that one could roll the cylinder from
outside even when the 'motor' was not running; and the second
strand of Wittgenstein'sattack on private ostensivedefinition can
be seen as involving an equally direct approach. It is best
summarizedin his story about the beetlein the box (PI, 293). The
private linguist's conceptionof his inner world runs parallel to the
following hypothetical situation: supposethat every human being
has a box with something in it which we call a 'beetle', and
supposefurther that no one can look into anyoneelse'sbox, whilst
everyone claims to know what a beetle is only by looking at his
own specimen. Here, we can easily imagine that everyone has
somethingdifferent in his box, or that the contentsof each box are
subject to constantchange;but if the word 'beetle' is to have a use
in these people'slanguage,it would not be used as the name of a
thing. The object in the box can have no place in the language-
game at all, not even as a 'something, I know not what', for it
would be consistentwith our hypothesisthat each box be empty.
In saying, however, that we can divide through by the thing in
the box, we are not denying the existence of the inner; we are
rather rejecting a particularpicture of the grammarof our concepts
of the inner. Although that picture seemsto underwrite the reality

58
PHILOSOPHY OF PSYCHOLOGY

of the inner world by mythologizing psychological events, states,

and processesas entities within that realm, it has the ironic
consequenceof evacuating that world of all substance;for if we
construe the grammar of psychological concepts on the model of
'object and name', the object drops out of consideration as
irrelevant.
Our indirect approach to this issue revealed that private
ostensivedefinition cannotsuffice to confer a techniqueof use upon
the sign'S' because it failed to establish any standard for
distinguishing correct from incorrect uses of'S'; our direct
approachasserts,in effect, that once we assumethat the sign'S'
does have a use in the language, the hypothesis of there being a
private inner entity or processwhich it names stands revealed as
doing no work at all. We should not, however, conclude from the
demonstratedincoherenceof the notion of a private languagethat
human beings are incapableof referring to or speakingabout their
own psychological states and feelings; the point is rather that, in
coming to seethat the grammarof those conceptsin terms of which
we do refer to and speak of our inner life cannot intelligibly be
construedon the model of naming objects, we also ceaseto regard
the correspondingfacets of our heartsand minds as types of object -
i.e. as private entities for which a procedureof naming might be
appropriate.Such a mythologized view of the inner feeds on and
expressesa distorted grasp of psychological concepts: to view
sensations,thoughts, and emotions as inner entities which are
inductively correlatedwith their behaviouralmanifestations-- as if
human inhabitantsof the world each containeda world as well - is
to sever an internal relation without which the technique of
employing the relevant psychologicalexpressionfalls into incoher-
ence. In a crucial sense,the private linguist goes wrong from the
very first momentof his fantasy; he begins by taking it for granted
that sensations can be divorced from their publicly accessible
manifestations, and thus the whole tenor of his thinking is
constrained by a failure to realize that the everyday concept of
sensation(which he is attempting to explicate) is embeddedin a
practice which depends for its coherenceupon the existence of
external expressions and manifestations of the relevant 'inner
event'. The best way to undermine any continued temptation
towards such mythologizedviews of the inner must therefore be to
sketch a picture of the grammar of our psychological concepts

59
PHILOSOPHY OF PSYCHOLOGY

which makesevident the essentialrole played by human behaviour

and its contextsin the applicationand explanationof psychological
terms in general.

CONCEPTSOF THE INNER

The structure of psychological concepts manifests a first-person/
third-personasymmetrywhich the mythological view of the inner
we have been dissectingfails to register accurately; for whilst the
possibility of error exists in third-person applications of psycho-
logical terms, that same possibility is excluded in the caseof first-
personpresent-tenseuses of those terms. If it were correct to view
the inner as a realm of epistemicallyprivate entities to which each
personhas sole access,then first-personpresent-tensepsychological
utterances would have to be viewed as descriptions of those
entities; on such a model, however, it would be at best a contingent
fact that a given speakerdid not err in such descriptions- it would
be logically possible for him to misdescribe the objects of his
inward observation Uust as he might misdescribeobjects in the
external world), and yet the hypothesisthat a speakermight be in
error concerning what he was thinking (as opposed to being
deceitful about it) makes no sense. Our conceptual framework
provides no conceivablegrounds for doubts of this sort (cf. LW,I,
896-8); one might say that the function 'X is in error' has no value
when X = the speaker.
The philosophicaldamagecannotbe repairedby suggestingthat
a speaker'sreports on the entities and events ot his inner world
have certainty conferred upon them because of the uniquely
privileged perspective from which he observes them; for the
conceptual point that error is logically excluded simultaneously
excludes the possibility of regarding those 'reports' as conveying
certain knowledge.The truth is that such utterancesshould not be
regarded as observationsat all - unlike the case of third-person
applicationsof such terms, first-personpsychologicalutterancesdo
not rest upon any evidential basis; they function not as reports
concerningthe speaker'sinner life but rather as manifestationsof
it, and in the absenceof an evidential base the concept of 'error'
lacks any purchase.The vision of human beings as observersof
their own inner world requiresa notion of a contrastbetweeninner
and outer evidence for psychological states in general, the

60
PHILOSOPHY OF PSYCHOLOGY

assumptionbeing that other people must build hypothesesabout a

given person's inner life on the basis of outward behavioural
evidencein default of accessto the inner evidencewith which that
observed individual is acquainted.The grammar of our psycho-
logical concepts reveals that there is no such thing as inner
evidence for the application of such terms; and in removing one
pole of the evidential dichotomy we thereby deprive the other pole
of any significance (LW,I, 77).
Of course, for the advocatesof this distorted vision of human
beings and their relation to their inner world, the collapse of this
evidential dichotomy is tantamountto a collapseof the distinction
betweenconceptsof the inner and conceptsof the outer; if our only
grounds for applying conceptsof the inner must be sought in the
field of human behaviour (in third-person applications of the
relevant terms), has not the private-languageargument simply
effected a behaviouristic reduction of concepts of the inner to
conceptsof the outer? Such a conclusionis too hasty: it fails to see
that a recognition of the lack of observationalgrounds for first-
person present-tensepsychological utterancesdoes not signify a
denial of the role played by the form of such utterancesin the
overall structure of psychological concepts; and it ignores the
possibility that psychological concepts and behavioural concepts
relate to the complex field of human behaviour and utterancesin
very different ways. It is thesegrammaticalpoints which must now
be investigated.
If we begin by noting the different ways in which conceptsof the
inner and conceptsof the outer relate to human behaviour,it is at
once possibleto see the sensein which a grasp of aspectperception
can facilitate our movementswithin this conceptualspace;for, just
as we saw in Chapter I that aspect concepts seem to transcend
immediate perceptualexperienceby invoking non-visual relations
and references (PI, 209b fi), so we might express the difference
between concepts of the inner and concepts of the outer in a
preliminary way by suggestingthat, whilst behavioural concepts
avoid invoking references and relations which go beyond pure
descriptions of perceived bodily movements and patterns of
behaviour, psychological concepts relate to aspects of the human
behaviour we observe (PI, 179a). To describe that behaviour
unhesitatingly in terms of psychological rather than behavioural
conceptsis to manifest the adoption of a certain attitude towards it

61
PHILOSOPHY OF PSYCHOLOGY

- it is to treat that behaviouras directly expressiveof an inner life.

The difference in meaning between psychological conceptsand
bare behaviouralones can be said to rest in part upon the fact that
it is behaviour-in-context that is the bearer of psychological
significance. We can begin to explore the role of 'context' here by
noting that in becoming a fully-fledged member of a human
community a persongains accessto a complex weave of expressive
behaviour, including modes of facial expression and of more
general bodily movementswhich count as manifestationsof inner
states.At least part of the expressivesignificanceof any gestureor
facial arrangement, however, depends upon the field of other
possible expressionsand gestureswhich form part of the culture's
repertoire; it is in contrast to these other points on a given
dimension of variation that the identity of anyone such point or
possibility is constituted and its impact upon (or significance for)
an observer derived. Take the concept of a facial expression: it
could not be acquired by someonewho had seenexamplesof only
one facial expression,who had only ever seen sad faces, for that
concepthas its existenceonly within a play of the features (LW,I,
766); teaching the concept to him requires isolating a particular
dimension of variation by pointing to samples of different points
along that dimension- it involves conveyingone particular way in
which objects can vary, can differ from and resembleone another.
This meansthat the significanceof the dimensionas a whole and of
the specific points along it are intrinsically interdependent:the
general dimension of variation is defined by the various points
along it, and each specific point is given by its difference from the
other points - to say that a face was sad would mean something
very different if our conceptspicked out only two different facial
expressiOns.
It follows then, that part of what makesa joyful smile what it is
(for example) is the fact that it differs from an uncertain smile or
an angry grimace; but the point applies to much more general
patterns of human behaviour. Grief, for instance, is an emotion
manifest in certain sorts of facial expression,bodily gestures,and
utterances.If, however, such grieving behaviour alternatedevery
five secondswith a pattern of joyful behaviour, then it would not
make senseto give either behavioural pattern its normal psycho-
logical designation; and if the person concerned had only two
modes of behaviour open to him (e.g. anger and joy), so that his

62
PHILOSOPHY OF PSYCHOLOGY

facial expression and general behaviour could not embody any

other pattern, then neither available pattern could have its usual
psychological significance. We can thus perceive one sense in
which the psychologicalimport of human behaviourdependsupon
its context - the context here being the field of expressive
possibilities bequeathedto a person by his culture and by those
natural reactionsof his own upon which any culture builds.
Another senseof 'context' that is relevant here is the sensethat
our application of a specific psychologicalconcept to a given facet
of human behaviour is grammatically related to the antecedents
and consequences of that behaviour- in such a way that identical
behaviour, when embeddedin different 'narrative' backgrounds,
carries very different psychological significance. A child who is
crying becauseshe fell over and grazed her knee would thereby be
manifesting pain, but the same behaviour on the part of a child
who has lost her mother would be a manifestationof anxiety. In
theseexamples,it becomesclear that treating human behaviouras
a manifestationof one psychological state rather than another is
equivalent to regardingit as an intelligible human responseto the
circumstancesembodiedin the relevant background.
A similar point can be made with reference to the relation
between the behaviour said to manifest an inner state and the
consequentpattern of behaviour exhibited by the person concerned.
An expressionof thoughtful concentrationon a particular task, if
followed by a series of carelessor mechanicalattempts to perform
that task, would contribute towards depriving the original
attribution of 'thoughtfulness'of any sense.In general, part of the
function of ascribing psychologicalstatesto someoneis to pick out
certain patternsof future behaviourwhich it might be reasonable
to expect from that person -a a function whose point is obvious
given the impact that the behaviourof others may have upon our
own plans and way of life.
However, none of this implies the existence of a necessary
connectionbetweencertain modesof presentand future behaviour;
it is simply that (in contradistinctionto behaviouralconcepts)any
given attribution of a psychological state to a given person's
presentbehaviour makes senseonly if, in general, that behaviour
hangstogetherwith a characteristiccontextof behaviour- in terms
of both antecedentsand consequents- in the absenceof which the
grounds for applying that concept are undercut.

63
PHILOSOPHY OF PSYCHOLOGY

The motivational antecedentsmade manifest in the context of

the behaviour (i.e. the 'occasion'of the behaviour) assumesgreater
importance as the sophistication of the psychological states
attributable to the subject increases.In more primitive contexts -
if, for example, we see someonewrithing in agony in a pool of
blood - the existenceof natural expressionsof the sensationof pain
provides a ground for ascription of pain; but there is no such thing
as the natural expression of someone dissimulating pain (LW,I,
44). In order to justify a claim that a child, on a given occasion,is
crying in order to attract attention rather than as a result of being
in some pain, we cannot simply point to her crying behaviour; we
must rather convinceour interlocutor that the child intends to gain
attention, which involves a reliance on the knowledge that the
pleasureconsequentupon having sympathy directed towards one
provides a plausible motive for the relevant behaviour in this
context.
The same point can be made for other concepts,such as that of
lying, which have no associatednatural expressionin behaviour;in
general, it is part of our understandingof lying that there be an
intelligible motive for so doing, that there be something which
occasionsit. This does not, of course,mean that motivelesslying is
a contradiction in terms: part of the horror in Othello is precisely
due to our perceptionthat Iago's motives for hounding his master
are inadequateor non-existent - but would the evil he thereby
represents be so terrifying if such an absence of motive were
commonplace?If such exceptional cases became the rule, our
concept of lying would lose its point (RPP,I, 780).
At this stage, it becomes relevant to note that the aspect of
human behaviour-in-contextwhich is perhaps most significant m
grounding attributions of sophisticated psychological states is
linguistic behaviour. However, in order to appreciatethe full force
of this remark, we should first acknowledge the way in which
looking at language-useas a complex form of human behaviour
clarifies the relation between linguistic and non-linguistic behavi-
our at a more primitive level. This relation is of course most
evident in the situation of teaching and training children in the
employmentof first-person utterances,but an examinationof such
contexts is not intended to imply that questions of conceptual
genesis are at issue; it is rather meant to highlight a conceptual
connection between language-useand non-linguistic expressive

64
PHILOSOPHY OF PSYCHOLOGY

behaviour which would obtain even if human beings entered the

world with a fully-fledged linguistic capacity. To take the case of
pain and pain-behaviour,for example (PI, 244): a child who has
hurt himself cries and flinches, and an adult can then teach him
first exclamations and later complete sentencesto use in such
circumstancesin the future.
Three crucial points follow from this description of the form
linguistic training typically takes. First, the possibility of teaching
and training in this sphere would seem to depend upon the
existence of more primitive natural expressionsof the relevant
inner state or event in the child's behaviouralrepertoire, since in
their absencethe notion of circumstancesin which it would be
appropriate to use a given linguistic utterance would lack any
content. Second, the description suggests that such linguistic
behaviourshould be treated not as a report basedon observational
grounds but as a replacementfor more primitive non-linguistic
behaviour, and so as performing the function of a manifestationof
the speaker'sinner life. Third, whilst in one sense the linguistic
behavioursubstitutesfor non-verbalbehaviour,in anothersenseit
constitutes an extension of it by giving the child access to more
complex forms of human behaviour.
This last point is of central importance.If the processof teaching
and learning a languageis seen to involve the gradual acquisition
of increasingly complex structures as an extension of the more
primitive ones already mastered,it follows that teaching concepts
such as 'intention' or 'hope' - which (unlike sensations)are not
manifest so overtly in specific forms of primitive non-verbal
behaviouralreaction- can begin with the inculcationof a relatively
primitive fragment of linguistic practice, and thereforerequires only
a correspondinglyprimitive fragment of non-linguistic behaviourto
be a part of the pupil's repertoire. If, for example, a child who is
placed inside a cot with a toy just beyond his reach outside the
railings proceedsto stretch out his hand towards the object and
express annoyance when he fails to grasp it, it would be
appropriateto say that he was trying to reach the toy; and it would
also be an appropriatecontext in which to begin teachinghim how
to expresshis intentions by using the word 'want'.
It may seemthat the abovetype of accountis not appropriatefor
all psychologicalconcepts;for example, it is not clear what sort of
primitive behaviour in a child who has not learnt to speak might

65
PHILOSOPHY OF PSYCHOLOGY

constitute a natural expressionof hope. Difficulties are generated

here, however, primarily because of a tendency to concentrate
upon non-verbal behaviour exclusively, and thus to forget that as
soon as certain (admittedly primitive) forms of linguistic behaviour
are mastered,the pupil thereby gains accessto the sort of logical
spacenecessaryfor the manifestationof certain sophisticatedinner
states.'Hope' is in fact a good exampleof what is meant here (PI,
174a). If a child crawls towards the door, then turns and asks
'Daddy home?' before turning back to the door, it would make
senseto say that she was hoping that her father would return, and
we could then perhaps begin to teach her the relevant primitive
fragment of our linguistic practiceswith the concept.And of course
as her commandof past and future tenses develops, so the child
will acquire the type of linguistic behaviour necessaryfor the
attribution of more complex modes of hope to be rendered
intelligible.
The preceding description of language acquisition can thus be
seen as embodying two general conceptualmorals. First, it shows
that the denial of the mythological vision of the inner (i.e. the
refusal to treat first-person psychological utterances as reports
about observed inner entities) does not leave us bereft of an
account of the process of learning to employ such utterances:
despite their not being employed on the basis of evidence, their
correct use can be seen as a matter of employing them in
appropriatecircumstances(e.g. in the case of 'I'm in pain', only
when the speakeris in pain), and thus of the pupil assimilatingthe
relevant forms of words into his behavioural repertoire. Second,
this description highlights the fact that the grounds for ascribing
psychological states and occurrencesto a given person lie in the
complex field of behaviour-in-contextby emphasizing that the
process of teaching the relevant first-person utterance (i.e. of
grafting it on to more primitive behaviouralforms) is controlled by
a grasp of which patternsof non-verbal behaviour can intelligibly
be treated as expressiveof the psychologicalstate involved. Once
the linguistic version of this behavioural pattern is correctly
assimilatedby the pupil, he is now capableof exhibiting modes of
behaviourwhich might make the attribution of more sophisticated
psychologicalstatesintelligible.
In other words, the point about the earlier description of the
child's primitive hoping behaviour is that the precise meaning of

66
PHILOSOPHY OF PSYCHOLOGY

her utteranceis an essentialelementwithin it; it alone permits us

to suggestthat the child has something specific in mind, and to
relate that specific somethingto the other aspectsof her behaviour
in that context in a way which gives point to our application of the
concept of hope. There are, however, additional and equally
important respectsin which the preciseform of words uttered by a
speakershould be seen as internally related to the precise type of
inner state thereby made manifest. This becomesevident in the
expressiveaspectof languagein general.
To begin with, a central part of understandingmany linguistic
utterances is the capacity to see them as expressive of the
particular human being who utters them. Take as an exampleour
hearing someonesay 'I am afraid': we will almost never have to
ask the speakerwhether he was making fun of his fear, discovering
it in himself, reluctantly confessingit or expressingit like a scream,
becauseour awarenessof the tone of voice and context of the
utterance is enough to provide this information. Just as human
behaviour-in-context is expressive of mind, so the linguistic
behaviour of human beings can give expressionto the speaker's
inner state when perceived in context. And just as human
behaviour in general - with its flexibility and variety of facial
expression and bodily movement - has the logical multiplicity
necessaryfor the application of a complex field of psychological
concepts, so such elements of verbal behaviour as intonation,
inflection, and the speed and rhythms of the utteranceprovide a
similar multiplicity within a more narrow focus for the application
of those concepts.
There remains, however,a further respectin which language-use
must be viewed as expressiveof mind. A human being in a stateof
deep despair may come across Marlowe's line in Dr Faustus
('perpetual cloud descends') and acknowledge it as a uniquely
appropriate articulation of the state of mind in which he finds
himself; it may even be a means of allowing him to understand
more dearly the precise nature of his own mood and feelings by
permitting him to put that facet of his inner world to words for the
first time. It is a commonplaceof artistic criticism that the process
of a widening educationin poetry and literature is simultaneouslya
process of clarifying and extending our comprehensionof the
possible varieties of human suffering, exaltation, and experience;
and it is undeniable that certain aspects of our appreciation of

67
PHILOSOPHY OF PSYCHOLOGY

certain types of aestheticobject rest upon a capacity to respondto

(and find assimilable) certain complex modes of manifesting
psychologicalstates and attitudes.
Such aesthetic understanding- that specific field of human
relationshipsto language,the world and other people - is simply
one exampleof a more generalphenomenon,one which is evident,
amongst other ways, in the case of primary and secondary
meaningsof terms. As we saw in Chapter2, three related qualities
characterizethe primary-secondarymeaningrelationship:it is only
if a word has a primary sensefor the speakerthat he can use it in
the secondaryone; the secondarysense is not 'metaphorical',in
that no alternative word could expresswhat the speakerwants to
say; and explanationsof secondarymeaning can only make use of
examplesof the word's primary employment.As I also pointed out
in that earlier chapter,a large proportion of the examplesof words
with primary and secondarysenses that Wittgenstein offers fall
acrossthe boundarybetweenthe inner and the outer realms in the
following way: primary senses tend to refer to the public,
intersubjectiveworld whilst secondarysensestend to relate to the
inner world of the subject. This is true of 'calculating' and
'calculating in the head'; 'unreality' and the 'feeling of unreality';
'fitting' and the feeling that a name 'fits' its bearer; 'word-meaning'
and the 'experienceof word-meaning'; knowing where a city lies
and feeling as if we knew; and so on. The generalway of stating the
parallel primary-secondary/outer-inner relationship which we em-
ployed in that earlier context was that it only makes senseto say
that someonehas had certain experiences(feelings, inclinations) if
he manifests linguistic behaviour of a sort which presupposeshis
mastery of a public, intersubjectively checkable technique (PI,
208).
Thus, it becameclear that a large rangeof inner states,attitudes
and experiencesare made manifest in a person'sbehaviourby his
finding a certain form of words appropriatein a given context -
indeed, the relevant concepts can only intelligibly be applied to
anotherpersonon the basis of his uttering the appropriateform of
words. In effect, this highlights the role of linguistic behaviouras a
complex field which permits fine-grained distinctions between a
person's psychological states in the absence of apppropriate
distinctions being manifest in non-verbal behaviour (e.g. the
difference between a dull ache and a throbbing pain). Further-

68
PHILOSOPHY OF PSYCHOLOGY

more, the point that a given speaker's capacity to employ the

relevant terms in their secondary sense is dependentupon his
having mastered the technique of their primary employment
suggestsa way in which that speaker'slinguistic training might
plausibly lead to his assimilating such secondarysensesof words
and going on to use them in appropriatecontexts.
Putting these two observationstogether, we can reinforce the
conclusion that, contrary to certain mythical views of the inner, a
human being's inner states are not merely inductively and
externally related to the repertoireof behaviour through which he
makes them manifest. Rather, the gradual acquisition of increas-
ingly complex patterns of linguistic behaviour should be seen as
being at the same time the provision of expanding access to a
sophisticatedfield of verbal reaction which can assist in constituting
as well as refining the form and structureof his self-expressionand
thereby of his inner life.
Of course,in suggestingthat clarity follows from regardingfirst-
person psychological utterancesas being on the same level (as it
were) as non-linguistic elementsof a person'sbehaviour,we are in
effect simply highlighting the asymmetrybetweenfirst-person and
third-person uses of psychological expressions from a slightly
different perspective.For whilst third-personusesof such terms are
grounded on observation of the subject's behaviour, to say that
first-person uses of the same terms should be treated as linguistic
behaviourwhich is an expressionof the subject'spsychologicalstate
is equivalentto saying that it should not be viewed as a description
based on self-observation.The subject is certainly not basing his
utteranceon observationof his own actions or behaviour;but since
that utteranceis not thereby being treated as a bare noise, this
view of first-person psychological expressions avoids anything
which might be sensibly regarded as behaviourist reductionism.
Since the private-languageargument showed that such linguistic
expressionsare not based on observationof epistemically private
entities, the conclusion would seem to follow that such utterances
are not the result of any sort of observation.
This conclusion is itself essentially a restatement of the
grammaticalremark with which this section of the chapterbegan:
namely that the conceptsof doubt and error (i.e. of the subject's
making a mistake) have no application to first-personpresent-tense
psychological utterances.This can easily be made explicit. If a

69
PHILOSOPHY OF PSYCHOLOGY

person's relation to his own behaviour did parallel that of his

relation to another person's behaviour with respect to the
ascription of psychologicalstates,i.e. if he had to make both types
of ascriptionby observingthe relevantperson'sbehaviour,then the
types of doubts and errors which can occur in the caseof ascribing
such statesto others could also occur in the caseof ascribing them
to himself. Yet it is precisely this type of error which is excludedin
the case of first-person psychological utterances, and which is
another of the logical distinguishing marks of psychological
concepts.If, for example,all usesof the form of words 'I was lying'
were the result of inferencesmadefrom the speaker'sobservationof
his own behaviour, then that utterancecould never be used as a
corifession (RPP,I , 703), i.e. it would lose its characteristicrole, one
in which the issue of whether the speaker has made a mistake
cannot intelligibly be raised.
A further way of expressingthis asymmetricalimmunity from
error would be to say that, in a situation where we are attempting
to discover someone'spsychologicalstate, the truth of that person's
own expressionof his inner state is guaranteedif he is sincere,
whereasthe sincerity of our ascription of a given state to him is not
enough to guaranteeits truth (PI, 224f). The special role of first-
personutterancesin the grammarof psychologicalconceptsis thus
that special applications can be made of (special consequences
drawn from) confessions whose truth is guaranteed by their
truthfulness. The fact that doubts can be raised concerningthird-
person ascriptions which make no sense in the case of the
corresponding first-person form means, for example, that if a
person'sfuture behaviourfails to hang togetherwith the nature of
the psychological state to which he earlier gave linguistic
expression we can exclude the hypothesis that the subject
misidentified his inner state from considerationas an explanation
of the discrepancy.
We might thereforesummarizethe conclusionsof this section as
follows: psychological concepts retain their distinctiveness from
behavioural ones despite the absence of epistemically private
entities as referentsof those concepts.In treating humanbehaviour
as expressiveof an inner life, we are applying to it conceptswhich
relate essentiallyto behaviour-in-contextand which embodya first-
person/third-personasymmetry: and these are elements which
logically distinguish psychological from behavioural concepts.

70
PHILOSOPHY OF PSYCHOLOGY

What must now be mademore explicit is the sensein which we can

legitimately treat psychologicalconceptsas aspectconcepts.

ASPECTSAND THE INNER

In the first section of this chapter, we saw that psychological
concepts cannot be dismissed as candidatesfor the category of
aspect concepts on the ground that they refer not to facets of
human behaviourbut to a realm of entities and processeswhich is
entirely distinct from that of the behavioural: to regard these
conceptsin that way is to fall victim to an ultimately incoherent
mythology of the inner. The secondsection of the chapter blocked
off the other obvious way of denying that psychological concepts
relate to facets of human behaviour,viz. that of claiming that such
conceptsrelatepureLy to behaviour,that they are in fact behavioural
concepts.To make such a claim is to fall victim to the obverseor
undersideof the first mythology we identified, by assumingthat a
denial of that particular fiction of the inner is equivalent to the
denial of the inner itself. I had already noted an initial similarity
betweenaspectconceptsas they were describedin the previous two
chaptersand the psychologicalconceptswe have been examining
here: namely that concepts of the inner invoke references and
relations which go beyond the bare behaviourexhibited by human
beings in just the way that aspect concepts applied to visual
symbols invoke references and relations which are not purely
visual. Having now advocated a perspective on psychological
conceptsin which linguistic expressionsare treatedas sophisticated
refinements and extensionsof non-verbal behaviour, we are in a
position to see that: there are further respectsin which conceptsof
the inner resembleour paradigmaticaspectconcepts.
We saw earlier, for example, that the capacity to perceive a
particular aspect of a given object (of a cipher, or a painting)
presupposeda graspof a generaldimensionof variation in terms of
which the object could be located by comparisonand contrastwith
other objects (PI, 20gb, 21Ob), and that this dimension was
articulatedin the relevant conceptualtechnique.It has beenone of
my central aims in this chapter to show that precisely the same
points can be made about facets of human expressivebehaviour,
and to reveal thereby the role of our linguistic practice in the
provision of this particular frame of reference. The expressive

71
PHILOSOPHY OF PSYCHOLOGY

significance of (e.g.) a facial expression is partly constituted by

contrast with the other facial expressionswhich our system of
psychologicalconceptspicks out as such. In other words, what we
mean by a specific psychological term is partly constituted by its
relations of contrast with other such concept-termsin the system;
and the usefulness (as well as the possibility) of applying that
system of concepts to human behaviour rests upon the fact that
humanlinguistic and non-linguistic behaviour-in-contextexhibits a
correspondingvariability and multiplicity - indeed it is this logical
multiplicity in human behaviour that is often projected upon a
hypothetical inner world of subtly differentiated entities and
processesby those attachedto mythical views of the inner. We thus
have no reason to regard psychological concepts as significantly
different in nature from the pictorial and linguistic aspects we
examinedearlier: they do not pick out a separaterealm of entities
hidden behind human behaviour, but rather constitute a system
which applies to the behaviour itself, but in ways which differ
significantly from purely material (i.e. behavioural)concepts.They
take up aspectsof such behaviour.
If this assertion is correct, then we should be able to find
analoguesof the key conceptualcategoriesWittgenstein identifies
in other areas of aspect perception in the realm of psychological
concepts. As is often the case in this investigation, examples of
aspect-dawningare by far the easiestto come by; and, again as one
might expect, the examples which predominate are of aspect-
dawning without a changeof aspect- analoguesto living pictures
rather than to duck-rabbits. There are many occasionsin which
one is profoundly struck by the particular shadeof consciousness
manifest in someone'sexpressionor behaviour; on such occasions,
it is not just that we see that the personis fearful or joyful - we see
the fear in his stance,the joy in her face. Similar experiencesmight
be cited in relation to languageas well as to facial expressionor
behaviour; for in certain contexts, we can experiencethe expressive
meaning of a form of words, hear the emotion in an utterance.
For the purposes of this book, however, the most important
consequence of regardingpsychologicalconceptsas aspectconcepts
is the implication that continuous aspect perception might be the
appropriateway of categorizingour ordinary attitude towards the
behaviour of others. As in earlier chapters, the central way of
elucidatingthe natureof that attitude must be in contrastwith that

72
PHILOSOPHY OF PSYCHOLOGY

of aspect-blindness:we must ask ourselves what the difference

might be betweenseeingin terms of the relevant aspectsand merely
knowing that such aspectsare there to be seen. Since, however, the
notion of aspect-blindnesswas not specifically defined by Wittgen-
stein with respectto the sphereof other minds, we can only apply
it within that sphereby imagining what a consistentextensionof it
from the analogoussituation with respect to pictures and words
might be.
Aspect-blindnessin relation to pictures involved the need to
interpret what the picture might be intended to representfrom a
direct perceptionof its arrangementof colours and shapes,i.e. from
its propertiesas a material object in its own right; and the forms of
verbal and non-verbalbehaviourwhich characterizesuch blindness
were seento be manifestationsof a certain generalattitude towards
pictures -a mode of treating them which revealed an orientation
towards them as material objects rather than as representative
symbols or meaningful objects. This was my interpretation of
Wittgenstein'sremark that the aspect-blindregard pictures as we
do blueprints - they could not immediately see the pictured scene
or object in the picture.
If this is a correct readingof how the aforementionedrelation of
knowing is to be understood,then the analogousapplicationof the
conceptof aspect-blindnessin the sphereof psychologicalconcepts
must be one in which the aspect-blindneed to infer or hypothesize
the psychological significance of a piece of behaviour from an
immediate perception of its constituent behavioural elements:
rather than seeing the friendliness of a glance, they infer from the
shape,colour, and movementof the perceivedglance that it must
have been friendly (RPP,I, 1102). The aspect-blindthus manifest
an orientation towards human behaviour in which it is treated as
behaviour rather than as human behaviour- they do not treat it as
behaviour expressiveof mind.
It follows by contrast that continuous aspect perception is
characterizedby modesof verbal and non-verbal behaviourwhich
reveal that one is treating the other's behaviouras human behaviour
- as behaviour which is expressiveof mind. The most prominent
manifestation of this attitude - as always - has to do with the
immediacy or readiness-to-handof the relevant forms of descrip-
tion: someonewho spontaneouslyand unhesitatinglydescribesthe
behaviourof others in terms of psychologicalconcepts,who would

73
PHILOSOPHY OF PSYCHOLOGY

find it unnaturaland difficult to describewhat he seesin terms of

purely behavioural concepts,who does not regard his description
in terms of psychological concepts as one option among many -
such a person thereby makes manifest the fact that he regards the
behaviour of others as human behaviour, as precisely the sort of
behaviour it is.
This emphasison the immediacy of forms of description does
not, however, exhaust our means of characterizing continuous
aspectperceptionin the domain of psychologicalconcepts.Further
points suggest themselves,once again upon the model of points
employedin other domains to characterizethe analogousattitude.
We might begin by mentioning an element of the general Gestalt
built up in psychologicalconceptswhich has been left in abeyance
thus far, viz. the relation that those ascribing the given psycho-
logical state have towards its manifestationin anotherperson. We
noted earlier that if someone's'smiling' expressionwere merely one
of three which he or she was able to adopt, it would not be what we
call a smile; and part of the reasonfor that is not just the truncated
field of expressivecontrastsin which it is embeddedbut also our
own consequentinability to respondtowards it as we do towards a
smiling face (cf. LW,II, Ill). The responseof others to a certain
psychologicalstatein anotherpartly characterizesthat inner state:
pain is something we respond to in others by sympathizing or
failing to do so, love and hatred in another person are not
phenomenato which we are indifferent, being certain that someone
else is glad to see me involves feeling secure in my own pleasure
(LW,II, 58). The same point holds for concepts not so closely
linked to the emotions; a certain society might consist of people
who regardedlying behaviourby others as a sign of madness,and
who therefore respondedto its manifestationby locking up those
who were caught acting in that way. Would this be our conceptof
lying?
In other words, a crucial aspectof understandinga psychological
conceptinvolves graspingthe role of the relevant techniquein the
form of life concerned.It picks out a mode of human behaviour
towards which we are very likely to have a certain attitude, since
the behaviourof others affects us and since the mode of behaviour
concernedis accessibleto us in our own life - it is an option we can
adopt or shun. As in the examplesof continuousaspectperception
examined in earlier chapters, therefore, this reaction to human

74
PHILOSOPHY OF PSYCHOLOGY

behaviour as the specific human behaviour it is manifests a

particular orientation or attitude towards what one is perceiving:
unless one were taking its status as a particular form of human
behaviour for granted, one could hardly be so immediately and
unhesitatingly sensitive to its more fine-grained features or to its
general value in human intercourse.
A related point - one which is also dependent upon the
particular grammar of psychological concepts - is even more
pertinent. If the significance of a given segment of human
behaviour as a manifestationof the inner is dependentupon the
characteristicantecedentsand consequentswhich hang together
with a given psychological state, as well as upon the field of
contrasting modes of behaviour available within the relevant
culture, then a familiarity with the culture itself must be crucial for
the capacity to apply the relevant conceptsin an intelligible way.
This sort of information is not likely to be given to us merely by
asking the inhabitants of that culture, since, although they learn
the relevant concepts against the background of these available
contrastsand patterns, they do not thereby learn to describe that
background: a familiarity generated through experience would
therefore seem to be required.
Of course,a similar point holds at the level of individuals as well
as of individual cultures. The multi-faceted array of facial
expressions,gestures,and behaviour syndromesacquired by one
individual will not be identical to that of any other, nor to that of a
hypothetical set of Platonic ideal types stored in their shared
culture - and minor deviations from expected paradigms might
therefore carry great psychological significance. Only extended
acquaintancewith the individual concerned can generate the
familiarity with his own repertoireof expressivebehaviourwhich is
essentialfor psychologicalconceptsto be applied to him with any
subtlety or fine shading.
Thesepoints are not, of course,intendedas empirical hypotheses
concerning the amount of experience necessary to master the
application of psychologicalconcepts.They are rather intended to
highlight the fact that our typical masteryof those conceptsis not
so much a question of learning universal rules as a matter of
knowing one's way around within a cultural or personalworld -a
world in which specific co-ordinates are defined by relations of
similarity and contrast with other co-ordinates,and make sense

75
PHILOSOPHY OF PSYCHOLOGY

only againsta backgroundor frame of referencewhich is taken for

granted. Another way of putting this would be to say that a
capacity to respond to the specificity of the given cultural or
personal world presupposesor manifests an unhesitating recog-
nition of that world as the world oj a person: unless one regardedhis
or her behaviouras behaviourexpressiveof mind, one could hardly
respond to it as giving expression to a particular mind in a
particular way. We saw earlier that such knowing one's way
around in a particular realm characterizes continuous aspect
perception; now we can also see that this familiarity is related to
one's familiarity with the relevant conceptual technique in the
specific and varied contexts of its application.
We can now perhaps appreciate more fully the aptness of
Wittgenstein's'physiognomy' metaphorin capturing our senseof
the readiness-to-handof words. The familiar physiognomy of a
friend's face, and the familiar physiognomyof a word's face: both
forms of expressionmake manifest our sensitivity to the unique
identity of that which we are observing,to our sensethat they each
make up a world that we have assimilatedand in which we are at
home. Thus, our knowing someone well comes out in the
unhesitatingway we take for grantedthe genuinenessor insincerity
of his expressivebehaviour: our familiarity with the fine shadesof
his behavioursensitizesus to the preciseshadeof consciousnessto
be seenin his face and gestures,and so we come to know our way
around with him - his habits, tastes, enthusiasms,and typical
modes of expressing or concealing his thoughts and feelings. A
crucial part of this understandingis thus a matter of being able to
make the right connections- to see a particular expressionas a
manifestationof a precise state of mind, to link that inner state
with something in the situation to which that person is likely to
have responded,to foreseejust how his future behaviourwill reveal
or repress that response and what the consequencesof such
revelation or repressionmight be.
As we saw in Chapter 2, a precisely analogous capacity for
moving around within tie-ups is part of linguistic understanding-
the full significance of a given utterance may depend upon its
precise nature as a form of words (what sort of person from what
sort of backgroundwould chooseto expresshimself that way?) and
upon the precisechoice of words within the phrase,a choice which
takes accountof the field of force of a given word and the tie-ups of

76
PHILOSOPHY OF PSYCHOLOGY

similarity and contrast with related words. In all these ways,

linguistic understandingcan be seen to involve a sensitivity to the
specificity of the utterance, an awarenessof the specific physi-
ognomy it turns to our gaze. Indeed, we might regard this
linguistic application of the word 'physiognomy'to be not so much
a metaphoricalas a secondacyuse of the word - for what other word
has the implications that this one amassesfrom its primary use in
relation to facial expressions, implications which we need to
expressthe similarities in conceptualstructure and attitude which
we can now see in the domains of languageand of people?
With this analogy betweenwords and people in mind, we should
note in conclusion a further feature of our typical attitude to
human behaviourthat has emergedin this chapter. My delineation
of psychologicalconceptsand their grammarrevealed the applica-
tion of such conceptsto involve referenceto certain characteristic
surroundingsof human behaviour - not just to a given type of
behavioural pattern but also a relation of that pattern to the
occasion of the behaviour and to certain types of consequent
reactionsand behaviour. Theseconcepts,in effect, draw togethera
certain Gestalt of behaviour (including future actions), utterance,
and occasion;but even though the relation betweenthe elementsin
the general pattern helps to constitute the meaningof the relevant
psychological expression, their interconnectionsare not necessary
ones. Human behaviour exhibits little of the identity of circum-
stances, of reactions to given circumstances, or of the exact
similarity of future behaviour manifest by those in a given state
which might give point to a conceptualstructure in which such
elements were necessarily yoked together. On the other hand,
human behaviour does involve the sort of rough and approximate
regularitieswhich render useful the application of a set of concepts
which treat it as falling into a variety of loose patterns that re-
appear.
If, however, our psychologicalconceptsare such that a looseness
of fit between the elements of certain behavioural Gestalts is
postulated,then our capacityto apply them must involve an ability
to perceive human life as 'the same occurring again - but with
variations', i.e. a capacity to respond to new combinations of
behaviour,utterance,and circumstanceas a variation on one of the
loose patterns.This point applies not only to the variety of human
circumstancesencounteredby adults in their day-to-day existence;

77
PHILOSOPHY OF PSYCHOLOGY

it applies also to the learning processwhereby a child's grasp of

very simple, rudimentary psychological states is transformed
(refined, extended) through time and experienceinto the grasp of
an adult upon larger types of variety and complexity. He, too, must
be capableof treating the latter type of pattern as a more complex
variant of the former; and since in this respect he resembles
someonecapableof seeing bare schematicdrawings as pictures of
rabbits, i.e. since both such capacitiesinvolve drawing relations of
comparisonwith other examplesof things falling under a certain
conceptualdimension or system (whether that of rabbit pictures
and rabbits or of psychological states), we can provide a further
ground for regarding the perception of psychological states as a
type of continuousaspect perception.
The above preliminary points should be sufficient to allay any
fears that the notion of aspectperceptionis being extendedillicitly
to the psychologicalrealm; what we must now do is to explore the
philosophical significance of that extension by examining in more
detail the consequencesthat aspect-blindnessmight have in this
domain.

BLIND TO THE INNER

Imaginationis called for faced with the other, when I have to
take the facts in, realize the significanceof what is going on,
make the behaviourreal for myself, make a connection.'Take
the facts in' meanssomethinglike 'seehis behaviourin a certain
way', for example,seehis blink as a wince, and connectthe
wince with somethingin the world that there is to be winced at
(perhapsa remark which you yourselfwould not wince at), or, if
it is not that, then connectthe wince with somethingin him, a
thought, or a nerve. 'Seeingsomethingas something'... is the
principal topic of the chief sectionof what appearsas Part II of
the Investigations.
(CR, 354)
The above quotationfrom StanleyCavell's The Claim of Reasonl can
stand as an appropriateway of summarizingthe main conclusion
to be derived from the previous sections of this chapter; it might

S. Cavell, The Claim ofReason(Oxford: Oxford University Press,1979)-

hereafterknown as CR.

78
PHILOSOPHY OF PSYCHOLOGY

also be useful to follow the avenuesof exploration Cavell takes as

leading off from this basic connection between psychological
concepts and aspect perception, since it seems at least arguable
both that he managesto raise the correct issuesand also that the
errors he makes in delineating them can help to refine our
understanding of Wittgenstein's detailed treatment of aspect
perception.
Cavell begins to explore this terrain with the following
statement:
The idea.. .is that humanexpressions,the humanfigure, to be
grasped,must be read. To know anothermind is to interpreta
physiognomy,and the messageof this region of the Investigations
is that this is not a matterof 'mere knowing'. I have to read the
physiognomy,and see the creatureaccordingto my reading, and
treat it accordingto my seeing.The humanbody is the best
picture of the humansoul - not, I feel like adding, primarily
becauseit representsthe soul but becauseit expressesit. The
body is the field of expressionof the soul.
(CR, 356)
The best way of characterizingthis formulation, I would suggest,
is to describe it as holding error and insight in a tension which
threatens to collapse them together at any moment. The error
involved is stated more explicitly elsewhere when Cavell says:
"'Seeing something as something" is what Wittgenstein calls
"interpretation'" (CR, 354); for it is one of the fundamentalaims of
Wittgenstein'streatmentof aspectperceptionto show that aspect-
dawning and continuous aspectperception are a matter of seeing
rather than of interpretation (cf. PI, 212d). For Wittgenstein, the
notion of interpretationcarries connotationsof making inferences,
forming hypothesesor drawing conclusions- as if when someone
sees a friendly glance in another'seye, what really happensis the
direct perceptionof shapes,colours, and movementwhich are then
interpretedto mean that the glanceis a friendly one (RPP,I, 1102).
A crucial motivation for stressing the aptness of the concept of
seeingin thesecontextsis precisely to underline the sensein which
the friendliness of the glance is as directly, as immediately
perceivedas the colour of the eyes might be thought to be.
In employing the notion of interpretation to characterizeour
relation to other people Cavell is therefore using a conceptwhich,

79
PHILOSOPHY OF PSYCHOLOGY

as defined by Wittgenstein in this context, can only help to draw

the reader away from one of the key insights which emergesfrom
Wittgenstein'streatmentof this issue; and the misleadingconnota-
tions ramify beyond this initial stage. As we noted in the previous
chapters,part of the force of employing the conceptof seeingwith
respectto aspectperceptionis that it characterizesthe perceiver's
general orientation towards the object of his perception- seeing
behaviouras expressiveof mind involves treating that behaviourin
a certain way, taking its statusas the field of expressionof a mind
for granted. As we just saw, Cavell preserves these ideas of
'attitude' and 'mode of treatment'in the way he characterizeshis
notion of interpreting a physiognomy; but, becausehe links them
together with 'interpretation',he makes it easy for any reader of
Wittgenstein to misinterpret the nature of the relation between
those three ideas. Given the connotation of making inferences
which is carried by 'interpretation'in these stretchesof Wittgen-
stein's writings, it would be natural to think that Cavell is positing
a three-stageinferential processat the basis of our relationship to
other people: first we interpret the physiognomy,then we must let
that interpretation mould our perceptionof the person, and then
we must treat that personin a way which accordswith the way we
see him. In reality, Wittgenstein is positing internal relations
betweenthese concepts:for him, to say that we see the behaviour
as expressiveof mind is to say that we treat it in an appropriate
way; describing oneself as seeing emotion in another's face is a
manifestationof a certain generalattitude towards the person; and
accordingly, someone who needs to interpret the perceived
physiognomy cannot intelligibly be said to have the attitude
towards that behaviour (the capacity to treat it appropriately)
which is grammatically bound up with calling the relation one of
seemg.
What makes Cavell's use of the term 'interpretation' seem so
perverse is that it is clear in other parts of his text that he is
concerned to emphasizethe very insights that his terminological
predilection encourageshis readers to cancel. He says in the
passagequoted earlier, for example, that interpreting a physi-
ognomy is not a matter of 'mere knowing'; and in anotherpassage
he states: 'Wittgenstein'sexpression"The human body is the best
picture of the humansoul" is an attemptto...expressthe idea that
the soul is there to be seen,that my relation to the other'ssoul is as

80
PHILOSOPHY OF PSYCHOLOGY

immediate as to an object of sight' (CR, 368). The accuracy of

these insights does not, however, excuse or justify Cavell's use of
'interpretation'; it rather highlights its inexplicability, and does
nothing to lessenthe importanceof recognizingjust how much of a
potential for misinterpretationis createdby his decision to employ
the term in this way.
A second issue Cavell takes as leading ofr from the observation
that our relation to the inner states of others involves aspect
perceptioncan best be approachedby registeringthosefacets of the
grammaticalstructureof psychologicalconceptswhich lead people
to say that we can never be certain of someoneelse's inner state.
Although it is important to note that this conclusion lacks the
generalscopeoften claimed for it - there are countlesssituationsin
which the criteria for the application of the relevant concept are
fulfilled and intelligible grounds for doubt are clearly lacking: we
know that a tennis-playerin the Wimbledon final intends to win,
that a woman who picks up an umbrella before leaving the house
believes that it might rain - it is also important to acknowledge
that there are many situations in which grounds for claiming to
know with certainty that a personis manifestinga given inner state
are not available. The crucial question then becomes:how are we
to accountfor this lack of certainty?
The answer cannot simply be that the evidenceprovided by an
individual's behaviour and its surroundingsis merely an indirect
symptom of the inner reality to which our psychologicalconcepts
refer. Our examinationof first-person utterancesshowed that they
were not grounded upon evidence uniquely accessible to the
subject; and if no such direct, inner evidence actually exists, it
could make no sense to talk of behavioural evidence as being
indirect merely because it is external - just as the dichotomy
betweeninner evidenceand outer evidencebreaksdown when one
pole of the contrast is removed, so the direct/indirect dichotomy
must collapseif it makes no senseto talk of inner evidence.In the
only situations where it makes sense to talk of 'evidence for the
inner' at all, that evidencetakes the form of behaviour (verbal and
non-verbal) in a certain context. It must thereforebe the casethat
the uncertaintywe feel about ascribing psychologicalstateson the
basis of such evidencelies, not in any putative contrastwith other
types of evidence which can provide certainty, but rather in the
nature of the evidenceitself.

81
PHILOSOPHY OF PSYCHOLOGY

We might say that the relevant grammar embodies indeter-

minacy: that is, that the conceptual structures involved are
founded upon modes of evidence which are not capable of
providing certainty in every case. Even when we have a person's
utterancesto go on, in certain situationswe are unable to point to
grounds which might justify our being certain that the speakeris
being sincere or truthful - the criteria of truthfulness (the fine
shades of present behaviour which lead us to trust or doubt
someone's sincerity) cannot in themselves guarantee veridical
knowledge in all cases. The characteristic of our psychological
concepts which is relevant here is that they unify disparate
elements - occasion, utterance, and behaviour - into a Gestalt
which fits together only loosely. There is no necessaryconnection
betweenthe elementssuch that the presenceof one guaranteesthe
presenceof the others; and yet the conceptcan only be said to be
correctly applied if the person involved exhibits reactions and
behaviour which, in his circumstances, can be seen as an
instantiation of this general pattern or as a variation of it. In
principle, moreover, one key element of that pattern is never
available when one looks at a person'sfacial expressionor present
behaviourand attemptsto assesshis stateof mind, viz. the pattern
of his future behaviour and reactions which it is part of the
function of psychological conceptsto predict. Even in the case of
first-person utterances,questions of their truthfulness will in the
end often be settled by looking at how the individual's future
actions hang together with his professedinner states.
Of course,long experiencewith a particularculture and a specific
individual can lead to increased sensitivity concerning the fine
shadesof that person's behaviour and the relation of those fine
shadesto the more coarse-grainedreactions and behaviour with
which they are drawn together by our psychologicalconcept. But
this observation about the connection between experience and
understandingin this area (cf. PI, 227h) carries as its obverse a
reminder of the role played by imponderableevidence in ascribing
psychological states (PI, 228b-d). The point here is that even in
those situations in which no evidence is available by means of
which we might convince others of the validity of our judgements,
we feel absolutely certain that, e.g. this personwhom we know so
well is genuinely unhappyrather than merely seekingattention. If
such certainty cannot be cashedin terms of intersubjectivelyvalid

82
PHILOSOPHY OF PSYCHOLOGY

grounds for a knowledge claim, how are we to understandit?

Wittgensteinsuggeststhat we regard it as a matter of taking up a
position in relation to the person concerned(cf. MS 174, p.IIS);
our certainty that he is unhappy is manifest in our attitude of
sympathy towards him, our attempts to discover the source of his
unhappinessand alleviate it - in short, in a variety of modes of
behaviour which reveal our taking the genuineness of his
expressionsof unhappinessfor granted.
We can thus seeonce more the importanceof that elementin the
general Gestalt built up in psychological conceptswhich concerns
the relation those ascribing the given psychological state have
towards its manifestationin anotherperson. It is this facet of such
conceptsand their use which Cavell treats as central in the third
idea he draws from Wittgenstein'sreferencesto aspectperception
in the psychologicaldomain:
Wittgenstein'sexpression'The humanbody is the best picture of
the humansoul' is an attemptto ...expressthe idea that the soul
is there to be seen,that my relation to the other'ssoul is as
immediateas to an object of sight, or would be as immediateif,
so to speak,the relation could be effected. But Wittgenstein's
mythology shifts the location of the thing which blocks this
vision. The block to my vision of the other is not the other'sbody
but my incapacityor unwillingnessto interpret or to judge it
accurately,to draw the right connections.The suggestionis: I
suffer a kind of blindness,but I avoid the issue by projecting this
darknessupon the other.
(CR, 368)
This diagnosisof the problem of other minds is, of course,based
upon a particular reading of the conceptof aspect-blindnessas we
first meet it in relation to the duck-rabbit:
We may say that the rabbit-aspectis hidden from us when we
fail to seeit. But what hides it is then obviously not the picture
(that revealsit), but our (prior) way of taking it, namely in its
duck-aspect.What hides one aspectis anotheraspect,something
at the samelevel. So we might say: what hides the mind is not
the body but the mind itself - his his, or mine his, and
contrariwise.
(CR, 369)

83
PHILOSOPHY OF PSYCHOLOGY

And from this point Cavell enters into a stimulating and

sophisticateddiscussion of the ways in which individuals might
understandtheir separationfrom (and closenessto) other peopleas
the result of particular stances of the mind rather than of
metaphysicallimitations. In effect, by exploring the sensein which
people are responsiblefor everything that comes betweenthem, he
suggeststhat the relation to other minds might be more fruitfully
conceptualizedin terms of acknowledgementand denial rather
than knowledge and ignorance.
The interest of Cavell's discussionmust be left for the readerto
explore, since the discussionitself defies summary; what is most
pertinentfor the purposesof this chapteris the questionof how far
Cavell's use of the concept of aspect-blindnessmatchesor refines
that of Wittgenstein. Cavell himself points out (CR, xv) that
beyond a certain stage in his writings on acknowledgement,he no
longer regardshis citations of the Investigationsas interpretationsof
it; in order to draw this chapterto a close, it would seemuseful to
attempt to identify the point of transition.
The crucial facet of the concept of aspect-blindnesswhich
Cavell's use of the notion does highlight is the sensein which the
certaintiesand uncertaintiesinvolved in relationswith other people
should be seen as grammatically distinct from knowledgeclaims.
The example employed earlier - our trust in the genuinenessof a
friend's expressionsof unhappiness- reveals a relation between
knowledgeand certainty which mirrors that manifest in the role of
Welthild propositions: our certainty is so fundamentaland open to
so little serious questioning that it functions as a hinge around
which our more specific dealings with the person concernedwill
revolve; and the revelation of its being erroneous(i.e. of our trust
being misplaced) would entirely disrupt the world of our personal
interrelationshipwhich it helped found.
Two points flow from the role of imponderable evidence in
groundingsuch certainties:first, that getting someoneelse to share
our certainty is more akin to inducting them into a particular
world-view (or rather, into a particular interpersonalworld) than
to presentingevidence to justify a hypothesis; and secondly, that
doubts sufficient to disrupt the certainty can arise from evidence
which patently has no intersubjectiveground or strength. In all
theserespects,the caseof Othello and Desdemonacan standas an
exemplificationof the issuesat stake; it is no accidentthat Cavell's

84
PHILOSOPHY OF PSYCHOLOGY

examinationof the conceptof acknowledgementin relation to other

minds should culminate in a reading of that tragedy. And in this
way, our recognition that conceptsof the inner are aspectconcepts
can help us to see that continuousaspectperception- conceivedof
as involving the perceiver'sunhesitatingtreatmentof a person as
the particular person he is - raises issues which may be correctly
graspableonly againstthe backgroundof such discussionsas those
in On Certainty and other late texts.
The same notion of an attitude which is reflected in a familiar
orientation within the world, and of the contrast between this
subject-world relation and a relation of cognitive confrontationor
knowledge claims is, however, also central to the Heideggerian
contrastbetweenpresence-at-hand and readiness-to-hand as modes
of relation to the world. Cavell's treatmentof aspectperceptionin
the sphere of other minds thus further suggests that a full
understanding of aspect perception will involve plotting the
relation between two texts which seem at first sight to inhabit
entirely incommensurable philosophical traditions - namely,
Section xi of the Investigations, and the first Division of Being and
Time. This interpretative task is one that I will begin in the next
chapter.
The difficulty which arises at this point for Cavell's interpretation
of aspect-blindnessis that he needs to read that notion in a way
which conflicts with Wittgenstein if he wants to reach the
conclusion at which he aims. We have already seen in rough
outline the form which aspect-blindnessin the psychological
domain must take if it is to be modelled upon that disability as
defined in other contexts. Just as the person who is blind to
pictorial aspects regards paintings as we do blueprints, so the
person blind to psychologicalaspectsregardshuman behaviouras
behaviour rather than as the field of expressionof a heart and
mind: he has to infer from the physical properties of a face the
inner statewhich is therebyrevealed.The problem for Cavell is not
that aspect-blindnesswould have serious consequencesfor its
sufferers; it is rather that according to Wittgenstein, those
consequencestake a form different from the one Cavell suggests.
To elucidate his own senseof these consequences,Wittgenstein
hints at an analogy which might be used here in place of the
reference to blueprints and pictures: the aspect-blind regard a
human being's behaviouras we would the behaviourof a robot, of

85
PHILOSOPHY OF PSYCHOLOGY

a constructionwhose behaviouris mechanicaland thus describable

by means of geometrical concepts(RPP,I, 324). The implications
of this analogyare not easy to fathom, but once they are elucidated
they can be seen to follow directions traced out in the previous
section of this chapter. Whatever else one thinks of the idea of
'geometrical'conceptsin the domain of descriptionsof behaviour,
this much at least is clear: they could not be conceptswhose use
relates to fine shades of behaviour and whose structure yokes
elementsinto a loose, flexible Gestalt, becausethen they would lack
the objectivity of application and the sharpnessof boundaries
which are characteristicof conceptsin the field of geometry. The
primary implication of the analogywould then seemto be that the
aspect-blindcannot see (or regard) human behaviour in terms of
the fine shades, the variety and the· flexibility which our
psychological concepts pick out and presuppose. They are
incapable of applying our psychological concepts directly and
unhesitatingly to behaviour and must instead infer its freight of
human significance from those physical features of it that can in
principle be described geometrically - and in the process of
inferencemuch of that freight is lost becauseit is not capturablein
geometricalterms in the first place. In this sense,the aspect-blind
would be blind to an aspectof the humanityof human behaviour,to
part of what makes it behaviour expressiveof mind.
A further elementof significancein the robot metaphorlies in its
purging of individuality as well as of humanity. Human behaviour
does not consist in machine-tooled,preciserepetitions of a limited
repertoire of movements that is invariant between cultures or
persons,but rather in irregularities and variations of texture which
iriflect culturally-relative paradigms of expressive behaviour in
specific ways and which together produce a weave of behaviour
with a particular physiognomy- an individual style or character.It
follows that to regard someoneas if he were a robot is to make use
of a model which must deprive that person's actions of the fine
shadesand the sheer specificity which go to make up our senseof
him as a particular human being. The capacity to register such
individuality is manifest in the capacity to see a given piece of
behaviour as significant in the context of one interpersonalworld
but not in another, as a variation of one paradigm rather than
another; and such capacities presupposean unhesitating recog-
nition of the individual as a person- as someonewhose behaviour

86
PHILOSOPHY OF PSYCHOLOGY

is expressive of mind in the first place. As was hinted in the

previous section, therefore, lacking the capacity to regard persons
as personscondemnsthe aspect-blindto a failure to distinguish the
individuality of the person with which they are dealing.
On this reading of the consequencesof aspect-blindnessin
relation to concepts of the inner, Cavell's difficulty becomes
obvious; for it is not easy to see how such a disability can lead us
into the issues of acknowledgementand denial in quite the way
Cavell proposes. He argued, for example, that when the rabbit
aspectof the duck-rabbit is hidden from us, 'what hides it is then
obviously not the picture (that reveals it), but our (prior) way of
taking it, namely in its duck-aspect. What hides one aspect is
another aspect, something at the same level' (CR, 369). This
implies that it is some compound of the perceiver's and the
perceived's states of mind which blocks vision, which causes
aspect-blindness;and the implication that aspect-blindnessis thus
a function or reflection of how two personsrelate to one anotheras
personsis confirmed by the tenor of later passages:
If somethingseparatesus, comesbetweenus, that can only be a
particularaspector stanceof the mind itself, a particularway in
which we relate, or are related (by birth, by law, by force, in
love) to one another- our positions,our attitudes,with reference
to one another.
(CR, 369)
It is difficult to imagine that a failed love-affair might lead to
someone treating her ex-lover as if he were a robot; and this
difficulty arises from Cavell's inclination to treat aspect-blindness
as something 'on the same level' as the aspect(s) to which one is
blind. Of course,we are not meant to think of the aspect-blindas
lacking the capacity to apply the relevant aspect concepts
altogether:just as the original aspect-blindpersoncould see that a
schematicdrawing was meant to representa cube, so the person
blind to psychological aspectsis not incapable of drawing some
inferencesabout the state of mind expressedin the behaviour he
directly perceives.The defect of aspect-blindnessis not so much an
inability or unwillingness to draw the right conclusion (cf. CR, 368)
- as if the aspect-blind simply relate in the wrong way to
(mistreat?) another person - but rather the need to draw
conclusionsat all. The defect is thus not on the same level as the

87
PHILOSOPHY OF PSYCHOLOGY

aspects to which one is blind: it is not a defective personal

relationship that is involved in aspect-blindness,nor a complete
inability to view the other as a person, but rather an inability
direct!J to perceive the other as a person.

I have argued, then, that Cavell's desire to extend his reading of

Wittgenstein on aspect perceptioninto the realm of interpersonal
acknowledgementand denial tends to contradict the primary
thrust of his recognition that Wittgensteinuses aspect-blindnessto
highlight the sensein which continuousaspectperceptionis not a
matter of knowing. Since I have already registeredmy intention to
treat this latter theme as fundamentalto Wittgenstein'spurposes,
the conclusionfollows that the point at which Cavell's discussionof
aspectperceptionstops being an explication of Wittgensteinis the
point at which his interest in interpersonalacknowledgementand
denial submerges his intermittent awareness that it is the
immediacy rather than the distortions of such relations that is at
stake in the Wittgensteiniantext.
I should, however, make it clear that the problems I have
unearthedin Cavell's treatmentof this issue should not be seenas
a condemnationof the insights with which his accountis studded.
In particular, the suggestionthat the notion of continuous aspect
perception may be analogous with, or otherwise related to,
Heidegger's concept of readiness-to-handis one which must be
pursuedin the next chapter. The interpretativeanchor in such a
project must be a sense of the nature of continuous aspect
perception; and that sense is itself primarily determined by our
picture of its contrast, aspect-blindness-a picture which I have
tried to build up over the last three chapters.It therefore seems
right that my final paragraphsshould summarisethe composite
picture at which we have arrived by combining the results of the
earlier chapters with those revealed by our examination of
psychologicalconcepts.
If, as Cavell suggests,the attitude of attachmentto our words
mirrors our attachment to ourselves and to other persons (a
suggestion reinforced by noting that linguistic behaviour is a
sophisticateddomain of human behaviour in general), then those
who lack the attachmentto words will have a counterpartin those
lacking the analogous attachment to other persons. Since this
former attitude of attachment can be partly characterizedas a

88
PHILOSOPHY OF PSYCHOLOGY

familiarity with - an unhesitatingcapacity to move around within

- language, and its lack as a form of linguistic stumbling (as if
one's linguistic joints were in splints), we are led to conclude that
an aspect-blindperson'sbehaviourin relation to other personsis to
be imagined as equally a matter of hesitation, stumbling, stiffness
of joints. When we add to this stiffness the point madein Chapter2
that the aspect-blindlack the second nature bequeathedto us by
language (that range of spontaneouslinguistic reactions which
forms the substratumof certain language-games,experiences,or
feelings), then it becomesclear that the behaviour of the aspect-
blind will be just as robotic in nature as they take the behaviourof
others to be. In other words, just as the aspect-blind regard
anotherperson'sbehaviouras we would that of a mechanicalman,
so their own behaviourmust be regardedas similarly mechanical-
as lacking in the variability and flexibility (the fine shades)which
determine the individual physiognomy of human behaviour and
provide the necessarylogical multiplicity for the application of
sophisticated psychological concepts. As we saw earlier, for
example, their inability to comprehendthe secondarymeaningsof
terms is alone sufficient to deprive their behaviour of features
which might ground the ascription of many feelings and statesof
mind to them.
Thus, by a reflexivity which is perhapsnot surprisingwhen one
remembersthe degree to which self-consciousnessand conscious-
nessof other selveshave been linked in the history of philosophical
treatmentsof theseconcepts,the aspect-blindperson'sblindnessto
the humanity in others parallels a dehumanizing blankness in
himself; and it would seem to follow that philosopherswho are
prone to characterizethe ordinary human relation to languageand
other people in terms which resembleWittgenstein'scharacteriza-
tion of aspect-blindnesslay themselvesopen to the Kierkegaardian
charge of denying their own humanity.
One might wonder whether such an extreme charge can be
levelled at anything other than straw men. In fact, however, some
philosophersdo characterizethe humanrelation to languageand to
other personsin that way; and to prove it, I shall begin the next
chapterwith an accountof the theorizing of someonewho is at the
forefront of contemporaryanalytical philosophy and is vulnerable
to precisely this charge. I shall, however, end the chapterby going
on to show the ways in which Heidegger'ssystemof thought (like

89
PHILOSOPHY OF PSYCHOLOGY

Wittgenstein's) is predicated upon an awarenessof the need to

avoid that type' of error, and so might reveal further ranges of
significance in the human attitude I have been calling continuous
aspectperception.

90