ArticlePDF Available

Metarepresentation and human capacities (2002)

December 2002
Pragmatics and Cognition 11(1):93-140

DOI:10.1075/pc.11.1.05bej

Authors:

University of Seville

Both metarepresentation and cultural learning have an identical origin. The imitation of new and complex motor patterns (also articulatory-phonetic patterns) is a crucial skill not only because it enables cultural transmission but also because its high requisites give rise to the exclusively human mind. The premotor plan at the base of such imitation requires the ability to fictionalize bodily postures, which implies a second line of awareness. Only by means of this second line can the human being deal with situations different from his own real and current situation. This permanent double cognitive architecture explains finger-pointing, laughing, and also all capacities considered metarepresentational. Evocation and metabeliefs (these latter being the starting point of syntax) result in language formation. There are two operating modes of the double line, which starting respectively with motor learning and finger-pointing, give rise to pretence and metabelief, and whose distinct nature can also account for the difference between fictionally triggered emotions and morality.

Content uploaded by Teresa Bejarano

Content may be subject to copyright.

Metarepresentation and human capacities*

Pragmatics & Cognition 11:1 (2003), 93–140.

issn 0929–0907 /e-issn 1569–9943©John Benjamins Publishing Company

<TARGET"bej"DOCINFOAUTHOR"TeresaBejarano"TITLE"Metarepresentationandhumancapacities"SUBJECT"P&C,Volume11:1"KEYWORDS "cultural learning; ﬁctionally triggered emotions; laughter; metabelief; morality; motor imitation; origin of syntax; pointing; pretence; ‘theory of mind’"SIZE HEIGHT "220"WIDTH "150"VOFFSET "4">

<LINK"bej-n*">

Teresa Bejarano-Fernández

University of Sevilla

Both metarepresentation and cultural learning have an identical origin. The

imitation of new and complex motor patterns (also articulatory-phonetic

patterns) is a crucial skill not only because it enables cultural transmission

but also because its high requisites give rise to the exclusively human mind.

The premotor plan at the base of such imitation requires the ability to ﬁc-

tionalize bodily postures, which implies a second line of awareness. Only by

means of this second line can the human being deal with situations diﬀerent

from his own real and current situation. This permanent double cognitive

architecture explains ﬁnger-pointing, laughing, and also all capacities con-

sidered metarepresentational. Evocation and metabeliefs (these latter being

the starting point of syntax) result in language formation. There are two

operating modes of the double line, which starting respectively with motor

learning and ﬁnger-pointing, give rise to pretence and metabelief, and whose

distinct nature can also account for the diﬀerence between ﬁctionally trig-

gered emotions and morality.

Keywords: cultural learning; ﬁctionally triggered emotions; laughter; meta-

belief; morality; motor imitation; origin of syntax; pointing; pretence; ‘theo-

ry of mind’

1. Introduction

For the last twenty years metarepresentations, that is, mental states dealingwith

other mental states (for example, beliefs about beliefs) have been extensively

studied. The number of surveys on the ‘theory of mind’ (or a subject’s under-

standing of his own mind and another’s mind) is ever-increasing. Against this

background, some authors have claimed to have found the link between

metarepresentation and other human skills. Leslie (1988) suggested that

<LINK"bej-r49">

children’s symbolic games and the human ability to perceive others’ false beliefs

94 Teresa Bejarano-Fernández

were connected. In Cosmides and Tooby (2000) we can read a long list of such

<LINK"bej-r16">

phenomena — representation of goals, suppositional reasoning, representation

of a personal past, ﬁction, ﬁctionally triggered emotion, morality. As Sperber

<LINK"bej-r78">

(2000:7) says: “Understanding the character and the role of this meta-

representational capacity might change our view of what it is to be human”.

It is in the context of these fascinating hypotheses that I wish to make some

suggestions on both these issues. Firstly, I intend to deal with the topic of

whether or not there is a relationship between motor imitation and meta-

representational capabilities. Secondly, I intend to deal with how we should

consider pretence and metabelief (or ‘false belief’), whether as successive stages

of one single capability or rather as examples of distinct capabilities which at all

ages could explain the whole list of human metarepresentational capabilities.

As regards the ﬁrst question, I am going to suggest that motor imitation of

new and complex patterns — but not any other kind of motor imitation — is

crucial because its high requisites give rise to the exclusively human mind. This

imitation or motor learning requires, as I shall later explain, what I will call,

from now on, a double line of awareness. The notion of the double line is

fundamental in my approach. This double line, which provides the subject with

the mental tool to deal with situations diﬀerent from his real situation, is, in my

view, the cornerstone of motor learning and all the metarepresentational

abilities.

By linking metarepresentation with this type of imitation I am aiming at a

conclusion of much broader scope: Imitation of new and complex motor

patterns, which, I must underline, also include articulatory-phonetic patterns,

enables cultural transmission from experts to apprentices. Thus, culture and

metarepresentational capacities would share a common origin.

I shall argue that the requirements of imitation of new and complex motor

patterns are, at the same time, similar to and diﬀerent from the requirements of

ﬁnger-pointing. How should one describe this similarity and diﬀerence? This

leads us to the second nuclear question of this paper.

What is the relationship between pretence and ‘false belief’? In my view,

there is barely any time-lag between the ages at which they appear: the design of

the ‘Maxi Test’ (Wimmer and Perner 1983) is excessively demanding (‘Maxi

<LINK"bej-r91">

Test’ = ‘Sally-Ann Task’ = ‘Changed Location Task’). Pretence and false belief

are not successive stages of one single capacity. As far as the functioning of the

second line of awareness is concerned, I distinguish between two diﬀerent

modes, or, more precisely, two diﬀerent relationships between this second line

and the ﬁrst or primary awareness. I shall call these two modes, respectively, the

Metarepresentation and human capacities 95

parallel double line and the colliding double line. Pretence is an exampleof the

parallel mode of the double line, whose starting point could be seen in the

imitation of new and complex motor patterns. False belief, by contrast, is an

example of the colliding mode of the double line, with ﬁnger-pointing as its

starting point.

The capacities listed by Cosmides and Tooby are, in my approach, attribut-

ed to either of the two modes. On the one hand, evocation which, as I shall

argue, derives from the imitation of new and complex motor patterns, would be

sometimes implemented as pretence: this pleasant activity evolutionarily arose

to exercise and strengthen the capacity to evoke. On the other hand, the

colliding double line also counts on a pleasure similarly designed by evolution:

laughter, which would strengthen the ability to discover false alien beliefs and

expectancies. By considering those beliefs which are not compatible with mine,

the capacity involved in ﬁnger-pointing extends its original scope. This list of

paired capabilities, a member of which belongs to the parallel double line while

the other belongs to the colliding double line, can be enlarged with one more

binomial which comprises two of the capacities mentioned in Cosmides and

Tooby’s list: ﬁctionally triggered emotion / morality. Thus, in this paper I will

deal with the following list of paired capabilities: imitation of new and complex

motor patterns / ﬁnger-pointing; evocation / false belief; pretence / laughter;

and ﬁctionally triggered emotion / morality.

Although in a somewhat more complex way, semantics and syntax are

linked with the double line too. From the starting point of the imitation of new

and complex motor patterns, in which the articulatory-phonetic patterns must

be included, I aim at explaining self-controlled evocation or the power to create

and deal with symbols. In my view, without this power of evocation, language

could not have come into being. However, since all linguistic meaning, either

that of nouns, verbs or adjectives, is inextricably bound to syntax, the power to

evoke images is only an ingredient of semantics. Syntax, the crucial ingredient

of language, can also be traced back to the double line. This connection,

however, is not direct. It is, I will propose, bridged by metabeliefs. I share the

widely accepted opinion that the primeval syntax is that of the categories of

thema and rhema. This syntax is a consequence of the speaker perception that

alien beliefs, as far as they are incorrect, insuﬃcient or not updated, require

modiﬁcation. The perception of such an incomplete belief, in other words, this

metabelief is — I submit — the cognitive content of the thema, and renders it

subject to a syntactical complement. At that very moment, the ‘Holophrastic era’

or one-word-utterance era could be surpassed. Only then could the long and

96 Teresa Bejarano-Fernández

complex process whereby particular syntaxes and semantics came into being be

triggered oﬀ.

These suggestions are integrated and presented as follows:

The metarepresentational capacity — the double line of awareness —

would have appeared in response to the demands of the imitation of new and

complex motor patterns or motor learning (3.2).

The most widely accepted list of metarepresentational capabilities, of

considerable size nowadays, should, in my view, be even longer. In addition to

the already mentioned motor learning, I suggest adding ﬁnger-pointing (3.3,

5.2) and laughter (5.1, 5.2) to the list.

There are two distinct operating modes of the double line which, instead of

being successive stages, are triggered oﬀ almost simultaneously. On the one

hand, the parallel double line gives support to motor learning, evocation (4),

pretence (5.1) and ﬁctionally triggered emotions (7). The colliding double line,

on the other, gives support to ﬁnger-pointing, laughter, false beliefs (6) and

morality (7). Language can only arise by means of a combination of capacities

belonging to the parallel double line with others belonging to the colliding

double line: articulatory-phonetic learning (3.2), evocation or prelinguistic

semantics (4), syntax and syntactically moulded semantics (6), all this being

implemented in the context of a type of communication which appeared with

ﬁnger-pointing.

The reader will have probably noticed that section 2 — on the kinaesthetic-

visual matching of chimpanzees — could have been omitted without seriously

‘aﬀecting any of my proposals. Why is it then that I have included it? This

research ﬁeld is open to many explanatory approaches and sometimes it is hard

to choose between them. The needier and more humble we regard ourselves,

the better it is for our search of new constraints and exigencies for assessing

theories. With this purpose in mind, Section 2 and parts of Section 3 try to

show that the double line of awareness is a mechanism both peculiar to human

beings and, at the same time, presaged or preluded by some abilities already

seen in chimpanzees.

Why have I coined the new term ‘double line of awareness’? For a similar

purpose, Leslie coined the term ‘decoupling’. This term, like that of ‘meta-

representational operator’, aims at describing this capacity of ours to attend to

something diﬀerent from our own, real and current situation. However, I have

preferred not to use these two terms because in both cases the mechanism so

labelled is interpreted as an occasional operator without consistency in time. By

contrast, I defend the existence of a permanent mechanism. In my view, in the

Metarepresentation and human capacities 97

human cognitive architecture there is room for two discrete systems — two

lines of awareness — each of them possessing its own sensorimotor, cognitive

and emotional apparatus.

What about the term ‘theory of mind’? This term, as ‘metarepresentation’,

refers mostly to the ﬁnal product and not to the cause of the process. Moreover,

it is almost exclusively used to deal with the perception of mental states.

Although some authors, such as Gopnik and Meltzoﬀ (1997), have suggested

<LINK"bej-r35">

that the perception of mental states would include the perception of alien

kinaesthetic states, many other researchers concerned with the ‘theory of mind’

have overlooked this. In addition, I want to detach myself from a legacy of the

‘theory of mind’. This legacy, shared by most of the authors with this approach

(but see Baron-Cohen 1999:272), is the statement that the set of exclusively

<LINK"bej-r5">

human skills we are considering would have been evolutionarily achieved

because of the advantages of a mentalistic, not behaviouristic, prediction of

someone else’s conduct — remember all that has been said about Machiavellian

intelligence and the evolutionary ‘cognitive arms race’. We cannot deny these

(sometimes successful) predictions in modern human beings, but in the origin

the forces of evolution were yielding other even more crucial skills than this

Machiavellian intelligence. In my view, these other skills were cultural learning

and language.

I cannot hope to provide even the shadow of a fully convincing proof of my

proposal in this paper. The desire to ﬁnd even a faint trail leading to a compre-

hensive and parsimonious theory of all these human skills is well worth all our

eﬀorts. Although a parsimonious explanation is not necessarily true, parsimony

is particularly desirable in this matter: we must not only comply with the well-

known methodological principles — see also Chater (1999) — but also explain

<LINK"bej-r13">

how evolution gave rise to the exclusively human abilities in such a short time.

2. Kinaesthetic-visual matching: A great feat, though an insuﬃcient feat

2.1 A great feat

Chimpanzees are able to imitate, using even parts of their own body which they

cannot see, simple movements (nonautovisible vs. autovisible movements).

Custance, Whiten and Bard (1995) have proved it. Using a number of training

<LINK"bej-r18">

actions, researchers encourage the subject to mimic whatever act the experi-

menter performs. Subjects are then tested on novel actions. Chimpanzees were

98 Teresa Bejarano-Fernández

as able in copying actions they could not see performed with their own body as

in copying those they could see. If, for example, a chimpanzee sees how an

experimenter taps his own head, the chimpanzee is also likely to tap his own.

There is no obvious correspondence between a chimpanzee’s bodily experience

of his own head and his solely visual perception of the experimenter’s head.

However, his imitative tapping of his head clearly shows that the correspon-

dence has been made. This power to establish such correspondences explains

why chimpanzees can recognize their own mirror images. Animals can pass this

mirror test only if they are in some way able to imagine the external appearance

of their own movements and their own body, a body which in most cases they

only in part can see. Chimpanzees (probably, all great apes) have this faculty.

Are apes the only animals in possession of this faculty? Pigeons have been

shown to imitate a model who operates a treadle either with their beak (which,

although they may be able to “see” it, clearly looks very diﬀerent from the

model’s beak) or their foot (Zentall, Sutton and Sherburne 1996). From these

<LINK"bej-r92">

observations Whiten (2000a:147) concludes: “At least within the limits of well-

<LINK"bej-r89">

rehearsed actions like this, kinaesthetic-visual matching may be a feature of

avian and mammalian imitation”. Maybe, but this is not the only plausible

explanation. Perhaps pigeons are using some other clues. Since pigeons are

unable to recognise their own mirror image, we must be cautious, the more so

if we take into account Meltzoﬀ’s surprising ﬁndings in new-born babies which

have been later given a new, more parsimonious interpretation (see infra).

Certainly, what I want to propose is that only humans can achieve the imitation

of new and complex motor patterns by means of a premotor plan: my sugges-

tion is not concerned with the exact number of species capable of kinaesthetic-

visual matching. However, given that so far there is no conclusive evidence on

the issue, I think we should not easily accept that kinaesthetic-visual matching

is also available to other animals, not only apes. The rest of section 2.1 relies on

this presumption.

Arboreal and bipedal locomotion has much to do with the acquisition of

these remarkable abilities. Since arboreal locomotion obliges the monkey to see

its own hands, it can perceive the noticeable similarities between its own hands

and those of others. This does not happen in other animals. A cat playfully

chasing its own tail cannot be compared to a monkey contemplating its own

hand. The cat’s partial vision of its tail or limbs is insuﬃcient to trigger in its

brain the revelation of similarities with another’s corresponding limbs. Mon-

keys and apes not only perceive their hands much better than a cat perceives its

paws, but the distinctive shape of their hands also facilitates the mental drawing

Metarepresentation and human capacities 99

of the correspondence between the monkey’s and a conspeciﬁc’s hands. In the

chimpanzee, the complete homology between its own body and that of a

conspeciﬁc is fulﬁlled. However, the turning point in this process is, I insist, the

addition of a distinctive visual image to the kinaesthesia of one’s own hand. The

mental drawing of a correspondence between one’s own hand and another’s

hand is achieved when this same process is reversely applied, i.e., when the

visual perception of another’s hand is givena kinaesthetic interpretation by the

subject contemplating it.

In the premotor cortex of monkeys, di Pellegrino et al. (1992) and

<LINK"bej-r22">

Rizzolatti et al. (1996) have found what they called ‘mirror neurons’. These

<LINK"bej-r73">

neurons are a subset of the grasp-related neurons of the bilateral premotor

cortex. They ﬁre when the monkey makes meaningful hand movements, such

as placing or taking away objects from a table, grasping food from another

experimenter or manipulating objects, but also when it observes these move-

ments made by the experimenter.

The making of tools and the implementation of techniques has always been

regarded as the evolutionarily decisive mission of the hand. However important

these achievements are, the truly decisive role of hands in evolution was, in my

opinion, to give rise to the mental drawing of the homology I am discussing.

Perhaps no useful imitative learning could have ever taken place without the

bridgehead eﬀect of hands to overcome the gap between the mere visual

perception of others’ bodies and the animal’s own bodily sensations.

The above mentioned abilities of chimpanzees — kinaesthetic-visual matching

— belong to Piaget’s fourth stage of development of imitative skills (Piaget 1959).

<LINK"bej-r66">

Before that, in the third-stage, children are able to imitate hand movements of

other individuals, but they are unable to imitate nonautovisible movements. The

starting point is, also in children, the imitation of hand movements.

Meltzoﬀ (see, e.g., Meltzoﬀ and Moore 1983) however, claims to have

<LINK"bej-r57">

found in newly born babies traces of imitative behaviour carried out with parts

of their body which they cannot directly see. If Meltzoﬀ is correct, then the

drawing of the homology between one’s own body and that of another’s would

not be, contrary to Piaget, a feat of learning, but an innate gift. Currently this

nativism is under attack. Some critics have seen in these allegedly innate

responses of new-born babies the same epidemic and automatic mechanism

whereby the bored members of an audience contagiously yawn or babies tend

to begin to cry one after another in a neonatal ward. Anisfeld (1991), after

<LINK"bej-r2">

reviewing the laboratory results of Meltzoﬀ, has recorded in new-born babies

only one clearly imitative behaviour of nonautovisible movements: babies

100 Teresa Bejarano-Fernández

consistently stick out their tongues in reply to the model’s doing so. But Jones

<LINK"bej-r45">

(1996) has oﬀered a convincing explanation of this behaviour. According to

Jones, new-born babies can voluntarily control only mouth movements, any

other reaction being spasmodic. For a baby, an adult face nearing his is such an

interesting spectacle (the more so if a tongue is sticking out of that face!) that he

wants to come into contact with or come nearer to the scene. But in order to

achieve his goal the baby can voluntarily put in motion only his mouth muscles.

Therefore, the baby will either open his mouth or occasionally stick out his tongue.

Consequently, Jones concludes that these behaviours are not true imitation.1

2.2 An insuﬃcient feat

However complex a capacity kinaesthetic-visual matching is, it has its limits.

These limits are, using Piaget’s terminology, those of the fourth developmental

stage of imitation. In short, fourth-stage children cannot yet imitate new and

complex motor patterns.

This boundary — new and complex motor patterns — is crucial. Obvious-

ly, only the imitation of a new pattern allows cultural transmission. We must

take into account that a new pattern must always be complex.2Besides, contrary

to the terminology of Meltzoﬀ (1988) and Tomasello, Kruger and Ratner

<LINK"bej-r57"><LINK"bej-r84">

(1993:497), I am not saying that there is imitation of a new pattern if there is

mere implementation of an already-known pattern onto a new object; the

novelty must reside in the motor pattern. At this point, we arrive at the above

mentioned description of the crucial boundary: the imitation of new and

complex motor patterns.

Now, I would like to deal with a topic of increasing concern to prima-

tologists. Laboratory and ﬁeld research have proved that the imitative skills of

primates had been overestimated (Visalberghi 1993). Apes do not even repro-

<LINK"bej-r86">

duce short motor sequences, however much encouraged to do so. More recently

it has even been conﬁrmed that no useful conduct learnt by apes derives strictly

from motor imitation, but from what is now called ‘emulation’ (Tomasello,

<LINK"bej-r84">

Kruger and Ratner 1993), or ‘program-level imitation’ (Byrne and Russon 1998),

<LINK"bej-r9">

a kind of learning in which only the goals or results,3not the motor patterns of

the model, are relevant for the imitator. Washoe, the female chimpanzee reared

by the Gardners almost as a human baby, certainly obtained some advantages

from her motor imitative skills in her artiﬁcial world, but none of her abilities

could ever have a selective value in the wild. Chimpanzees’ motor imitative

skills must be interpreted as a by-product of their kinaesthetic-visual matching.

Metarepresentation and human capacities 101

Recent discoveries in primatology dramatically devaluate the role formerly

assigned to motor imitation and clearly suggest a “deﬂationary” revision of the

scope and importance formerly ascribed to it.

I must make one more comment before Section 3: what about the dialects

learnt by singing birds or the probably more complex but similar skills learnt by

parrots in labs? Sterelny (1999) mentions the singing imitation of birds as an

<LINK"bej-r79">

argument against what he considers an overestimation of the imitative skills.

But is singing imitation of birds motor imitation? Marler (1991) has shown that

<LINK"bej-r53">

the white-crowned sparrows (Zonotrichia leucophrys) learn in their ﬁrst months

of life, before even being able to sing, the dialect of neighbouring adults by

simply adding the particular notes and cadences they hear to an innate, basic,

auditory pattern. This innate pattern seems to be much more similar to adult

singing than a child’s babbling is to spoken words. However, in my view, the

crucial diﬀerence between bird singing and human post-babbling speaking lies

in the way they both store in their brains the sounds they learn. The bird storage

device is auditory, whereas the human’s is motor. When Marler made the

young birds deaf before they were about to start singing (young birds do not

sing), they were neither able to sing according to the dialect they had been

hearing for weeks nor, curiously enough, to use their innate, basic pattern,

something that birds reared in isolation could easily accomplish. By contrast,

when children, after the babbling stage, begin truly linguistic learning, they

seem to store the spoken sounds they are exposed to in an articulatory manner,

i.e., as a premotor plan to be eventually fulﬁlled (infra, 3.2). The enormous

storage required by language could not be obtained in the auditory way of birds.

When I speak of the highly demanding requisites of imitative learning, I am

referring to that kind of learning which, in my view, needs a premotor plan

prior to the ﬁrst bodily imitation.

3. Motor imitation of complex and new patterns

3.1 The New World of human motor imitation

I agree that in apes motor imitation is a minor and almost useless faculty. I can

only see one true advantage of their kinaesthetic-visual matching, the advantage

of making the observer know in which direction another individual is looking.

However, in humans imitation is not only applied to simple movements as in

apes, but to new and complex motor patterns. However small a change this may

102 Teresa Bejarano-Fernández

seem, it is a giant leap for mankind. At this new level, imitation becomes the

fundamental ingredient of cultural transmission and learning and, consequent-

ly, its adaptive value is enormous.

The adaptive value of cultural learning is indisputable. The bone of conten-

tion is how far that learning relies on or even requires motor imitation as such.

Byrne and Russon (1998) have strongly rejected the idea. The allegedly impressive

<LINK"bej-r9">

goals that apes achieve by means of program-level imitation, as compared with the

poor results they obtain by means of motor imitation, have led these authors to

apply the same rule to humans. In my opinion this approach is misleading.

According to Byrne and Russon motor imitation plays no fundamental role

in humans either. Consequently, they reject the idea that the articulatory-

phonetic skills of language can be learnt by means of motor imitation. In their

words “children imitate spoken words at the program-level, copying the organiza-

tion of phonemes that make up words, not the physical sounds” (Byrne and

<LINK"bej-r9">

Rusonn 1998:683). Let us clarify their argument: since phonemes are not real

models but abstract speech entities neither to be spoken nor heard, they can in no

way be subject to motor imitation.

In my view Byrne and Russon are correct in their reasoning but their

premises are erroneous. Why do they take for granted that in articulatory-

phonetic learning the model to be copied is the particular way in which an

individual utters a sound, that is, with a speciﬁc timbre, intonation and pitch?

The model that any language oﬀers to a learner consists of a set of utterances

produced by the social group of speakers he belongs to. Out of this set the

learner obtains the phonemes to be copied. In other words: the model is already

phonemic not phonetic. Thus, there is no contradiction in assuming that,

although the learner does not try to reproduce the individual and particular

features of sounds, he learns the articulatory patterns of his language by motor

imitation. And the same goes for non-linguistic patterns. In this case, the model

is likewise a social model, i.e., a model certainly taken from particular occur-

rences, individuals and circumstances, but abstracted from them too. The

“deﬂationary” approach overlooks the social context of both language and

human technical learning, reducing motor imitation to a high-ﬁdelity copying

of a particular individual on a particular occasion, that is, to clearly absurd,

slavish mimicry.4

Whiten (2000b:495) claims that no imitation is exact. “All imitation has a

<LINK"bej-r89">

schematic character, varying in ﬁdelity according to the features it assimilates.

Fidelity is a continuum”. Whiten is right, but in my opinion this continuum is

mostly caused by the fact that models can be more or less concrete, i.e., they can be

Metarepresentation and human capacities 103

abstracted from one occurrence or from many occurrences. Thus, if the model has

been abstracted from many occurrences, then imitation can be completely faithful

to the model without, however, coinciding with a particular occurrence.

In my opinion, motor imitation in apes is a useless side-eﬀect with no

adaptive value, but in humans it becomes a fundamental tool of great eﬃciency.

As in the case of the proverbial ugly duckling which turned into a swan, motor

imitation only ﬁrst exhibited its strength when, in human beings, it began to be

applied to learning new and complex patterns. That moment marked the dawn

of a new era of changes and novelties.

However diﬀerent the imitative skills of apes and humans are, there are

obviously evolutionary connections between them. Let us consider the above

mentioned ‘mirror neurons’ of monkeys. Curiously enough, mirror neurons are

located in an extremely revealing area of the monkey’s cortex. In human brains

that area in the left hemisphere corresponds to the rostral part of Broca’s area,

i.e., of the premotor cortical area controlling speech. I have already said that we

must understand language learning as a particular imitation of new and

complex motor patterns. Some authors consider that this peculiar location of

mirror neurons strongly suggests that they could have contributed to the

correct emission and reception of primitive bodily or manual messages, the

forerunners of language. Arbib and Erdi (2000:532) suggest a slightly diﬀerent

<LINK"bej-r3">

account: “That specialization may correspond to verbs or verb phrases and

seems separate from the functions of naming”. In my opinion, there is an easier

explanation. Language is learned by motor imitation; consequently, its relation

to mirror neurons, rather than a problem, is what is to be expected. The new

world of the imitation of new and complex motor patterns is an extension and

further development of previous imitative skills of minor complexity (imitation

of simple movements). And these minor imitative skills have their starting

point, I suggest, in the mental drawing of the homology between one’s own

hand and another’s hand. Thus, it makes sense that the area originally dealing

with this function maintained in the course of evolution its pre-eminent role in

motor imitation (see infra, note 17 for more detail).

Note that I am not suggesting that these ‘mirror neurons’ mediate imita-

tion. Being able to imitate simple hand movements is of little use. If a hungry

animal sees others grasping fruits on a tree, he is likely to grasp the fruit, but his

doing so does not mean that he is actually imitating others’ hand movements.

The complete kinaesthetic-visual matching of chimpanzees certainly has an

adaptive advantage which, in a certain sense, is related to imitation. This

matching, in my view, was selected because it allows the observer to know in

104 Teresa Bejarano-Fernández

which direction another individual is looking (I shall deal with this topic later).

By contrast, ‘mirror neurons’ were probably selected because they allow the

animal to distinguish between his own hand and another’s visibly close hand, a

distinction that arboreal locomotion would have required: these neurons would

then originally have had no relation whatsoever with imitation. What I am

suggesting is that once the ‘mirror neurons’ and the kinaesthetic-visual match-

ing were available, their potential could be later transferred (as necessary but

still insuﬃcient elements) to another function.

3.2 Why is the imitation of complex and new motor patterns

such a diﬃcult and demanding task?

Why do chimpanzees never go beyond the mere imitation of simple move-

ments? Why does an 8 month-old child in Piaget’s fourth stage of development

in imitative skills take up to ten months more to reach the peak of imitation of

complex and new motor patterns, the only kind of imitation which really pays

dividends? This intriguing question can only have one reasonable answer:

imitation of complex and new motor patterns is a highly demanding and

diﬃcult task. But now a new question arises: why is this imitation such a

diﬃcult task?

In order to answer this question, I will use Piaget’s fundamental concept of

latent imitation. According to Piaget, in the sixth developmental stage of

imitation there appear two new features of dramatical importance. On the one

hand, the ﬁrst reproduction of a new and complex motorpattern is carried out

without the hesitations or tâtonnements still seen in the previous developmental

stage. On the other, that ﬁrst reproduction often takes place hours after the

child’s having seen the model. On these grounds Piaget suggests that, while

observing the model’s complex pattern, sixth stage children elaborate a

premotor plan or latent imitation.

How could this premotor plan or latent imitation be interpreted in the light

of the recent proposals on mental models for motor control? Grush (1994:209)

<LINK"bej-r37">

following control/robotics literature, distinguishes two kinds of mental models:

he calls them the upper model (most probably peculiar to humans) and the

lower model. “The lower model is an inverse model: input the desired end state

and it will determine the commands necessary to achieve the goal”. This is

called an inverse model “because it basically works in reverse, i.e., it contrasts

with what the actual controlled system — the musculoskeletal system — does,

which is to take a sequence of eﬀector commands as input and produce an end-

Metarepresentation and human capacities 105

goal state as output”. The upper model is “a forward model, that takes as input

(perhaps via eﬀerent copy) the same signals that would normally give rise to action

and produces as output a ‘mock’ version of the aﬀerent signals the peripheral

system would have produced had it been given that same input”. It is clear that

Piaget’s latent imitation involves a forward model, i.e., one of those models which

Grush — correctly, in my view — considers “steps to higher cognition”.

Let us now look at the requisites of latent imitation. These requisites will make

clear the diﬃculties experienced in acquiring the imitation of complex and new

motor patterns. There cannot be imitation of complex and new motor patterns

without what I call a double line of awareness. Let me explain this concept.

While contemplating another’s complex and new motor pattern, a child,

doing nothing himself, can elaborate a premotor plan of the pattern.5How is

this accomplished? In order to correctly choose the motor mechanisms required

by step n, the imitator-to-be must proceed from the bodily results of the

movement produced in the previous step n-1. But, since the imitator-to-be

makes no movement in this process, and, more precisely, has not made the

movement n-1, his own posture cannot serve as an indication as to how step n

must be performed. Thus, for the elaboration of the premotor plan the subject

must mentally simulate ﬁctitious bodily results.6

Cosmides and Tooby (2000: 87), when dealing with the ‘Representations of

<LINK"bej-r16">

Simulations of the Physical World’, have, in brackets, a similar comment:

“Hegarty has also suggested that there is a kinaesthetic sketchpad for simulating

body motions; personal communication”.

The subject must mentally simulate ﬁctitious bodily results. How is this

simulation possible? I suggest that there must be two lines of sensory-motor

awareness simultaneously active in the imitator-to-be. The primary awareness

of the subject is obviously concerned with his real, immediate and current

environmental and bodily situation, where direct needs and real threats or

dangers can be overlooked only at great risk to himself. To understand how

latent imitation works I postulate a second line of awareness, only present in

humans, which permits the subject to simultaneously deal with another (not

‘real, immediate and current’) scenario.7I have called this cognitive mechanism

double line of awareness.

In my view, this requisite, the possession of a double line of awareness, is

the link between imitation and metarepresentation. Whiten has aimed at

linking metarepresentation and imitation (even of simple movements) in a way

which is not, in my opinion, convincing. Whiten and Ham (1992:271): “To

<LINK"bej-r89">

imitate in the visual mode involves B copying an action pattern of A’s that was

106 Teresa Bejarano-Fernández

originally organized from A’s point of view… (This pattern) has then to be re-

represented, i.e., metarepresented, so as to be performed from B’s point of

view”. In my view — and I agree with Sterelny (1998) — this “translation

<LINK"bej-r79">

between the perspective-of-action of another and of oneself” (Whiten

<LINK"bej-r89">

2000a:144) — is far from being a metarepresentation. By contrast, the learning

of new and complex motor patterns, although not a ‘belief about a belief’, is a

very demanding task which, in my view, can be achieved only by means of the

cognitive architecture supporting metarepresentation.

More parsimoniously, Whiten (2000b: 501) declares only that “knowing

<LINK"bej-r89">

you are imitating is a kind of metarepresentation”. How should we interpret

this “knowing”? Is Whiten referring to a simultaneous occurrence of two

awarenesses on the subject? His remark would be in line with my proposal. But

I insist, however, that this duality would have only appeared to permit the

imitation of new and complex motor patterns. This only, this restriction, is, in

my opinion, also missing in Donald’s (1991) view and in his concept of

<LINK"bej-r24">

mimesis. This is the point I am going to deal with now.

3.3 Does the imitation of simple movements require no double line of

awareness?

I have spoken of the kinaesthetic-visual matching achieved by apes. Does this

matching require no double line of awareness? Monkeys’ mirror neurons, and

the easy copying by chimpanzees of simple nonautovisible movements strongly

suggest that there must also be a double line of awareness at work. But is this so?

Chimpanzees are able to kinaesthetically interpret a conspeciﬁc’s simple

movements. In this respect, they certainly possess the faculty of attending to

both kinaesthesias: their own posture in repose and another’s movement. On the

other hand, they have never been observed reproducing new and complex motor

patterns. I have already suggested that this incapacity to imitate a complex and new

motor pattern derives from the chimpanzee’s inability to ﬁctionalize a bodily

posture diﬀerent from its current one (more precisely, to ﬁctionalize a bodily

posture resulting from the ﬁrst motor steps to be copied). However, when a

chimpanzee kinaesthetically interprets another’s simple movement, is he not

ﬁctionalizing that movement while still in repose? Is this a contradiction?

The answer, in my opinion, is no. This observing chimpanzee would be

focusing on its own carrying out of the action in the immediate future, i.e. on

its own immediate expectation. Animal awareness must certainly provide room

for the immediately succeeding movement. By contrast, in my view, animal

Metarepresentation and human capacities 107

awareness is unable to manage ﬁctional situations, i.e., situations diﬀerent from

both the current situation of the animal and the animal’s immediate expectation.

This is why the kinaesthesia inferred by means of kinaesthetic-visual

matching by a chimpanzee is inescapably bound to the real bodily posture of

the observing animal. In other words, the ‘expectation’ about its own possible

movement extracted by a chimpanzee on seeing another’s movement cannot be

later used as a basis for new expectations. This dramatically inhibits the

sequential linking of expectations of any premotor plan, and consequently no

learning of complex motor patterns arises in a chimpanzee. Chimpanzees can

certainly copy one or other individual step of a motor sequence, but since no

pattern can be reduced to a fragmented succession of single movements without

mutual connection, or, in Luria’s words, without ‘kinaesthetic melody’, the

imitative skills of chimpanzees are very limited. Premotor complex plans need

to work on a ﬁctional basis beyond the reach of chimpanzees’ mental processes.

Chimpanzees’ expectation — expectation about their own (possible)

immediate movement — is inescapably bound to their current bodily posture

and, consequently, is not sequential. This is not the only diﬀerence between a

chimpanzee’s expectation and a true second line of awareness. There are two

crucial diﬀerences.8Let us now examine the second one. A chimpanzee can, at

most, only foresee its own imitation of a conspeciﬁc’s movement. The kinaes-

thetic-visual matching allows the observing animal to attend to its own bodily

posture and to an inferred kinaesthesia. However, as I have tried to illustrate, these

two situations are not taken into account by the observing animal as simultaneous

stands. Unable to bear in mind the two situations as simultaneous, the observing

chimpanzee is also unable to ascribe to the conspeciﬁc a perceptual ﬁeld in which

he, i.e., the observing animal, is included. This fact explains why untrained

chimpanzees are unable to indicate an object or direction with their arms or

ﬁngers, i.e., why they cannot point in the way we can. Let us examine this issue.

Simple as these gestures seem to us, chimpanzees have not discovered the

extremely useful results to be obtained from them. Why? The question is not

superﬂuous. A number of reasons make this incapacity remarkable. Firstly,

chimpanzees’ communicative behaviour includes a number of requests.

Secondly, arm- or ﬁnger-pointing serves to clearly indicate the precise goal of

a request. Finger-pointing is not only highly useful in human societies, with

their enormous number of available objects, but it would also be useful to

chimpanzees in the wild and certainly much more useful than their vague

asking gestures. Thirdly, chimpanzees are able both to produce this simple

motor pattern and to kinaesthetically interpret it.

108 Teresa Bejarano-Fernández

Then, why are they unable to produce and interpret arm- or ﬁnger-pointing

when untrained?9The answer is that a mere kinaesthetic interpretation of arm-

or ﬁnger-pointing does not suﬃce to correctly understand these gestures. As

mere bodily movements, arm- or ﬁnger-pointing is absurd or meaningless. It

neither serves to grasp, release or push an object. For a correct interpretation it

must be understood as a communicative act. The observer must imagine

himself in a kinaesthetic and contextual situation other than his current one.

Chimpanzees can do this. However, the interpretation of communicative acts

is more demanding. Let us remember Grice’s non-natural meaning (Grice 1982;

<LINK"bej-r36">

Sperber 2000: ‘Fifth scenario’). To understand arm- or ﬁnger-pointing a

<LINK"bej-r78">

chimpanzee observing these gestures would have to ascribe to the inferred

kinaesthesia of the pointing animal a visual ﬁeld encompassing it -i.e., the

observing animal- as distal object.10

A current debate may be clariﬁed if we adopt this view. Some authors, e.g.,

Gallup (1998), have suggested that because chimpanzees recognise their own

<LINK"bej-r29">

mirror image they are self-aware and can also attribute self-awareness to a

conspeciﬁc. In other words, Gallup considers the kinaesthetic-visual matching

of chimpanzees to be powerful enough to allow them toattribute self-awareness

to others. By contrast, Povinelli et al. (2000) maintain that the alleged kinaes-

<LINK"bej-r67">

thetic-visual matching of chimpanzees is not necessary to account for their

ability to assess what a conspeciﬁc can and cannot see: Their assessment can be

fully achieved by means of more primitive strategies. I agree with Gallup that it

is the kinaesthetic-visual matching11 of chimpanzees that allows them to assess

a conspeciﬁc’s visual direction. This must have been the adaptive advantage

whereby evolution, from monkeys’ ‘mirror neurons’, eventually gave rise to

chimpanzees’ kinaesthetic-visual matching. Besides a visual direction, chimpan-

zees could also ascribe some feelings (de Waal 1998) and goals (Whiten and

<LINK"bej-r87"><LINK"bej-r89">

Ham 1992) to their kinaesthetic interpretation of a conspeciﬁc’s body. Howev-

er, this ability has a signiﬁcant limitation, and Povinelli is right in limiting the

power of the kinaesthetic-visual matching of chimpanzees: chimpanzees are

unable to use this matching in order to determine if a conspeciﬁc can see them

or not. To accomplish this task, chimpanzees use only rather primitive mecha-

nisms.12 This might be the reason why, as has been frequently suggested,

blushing and embarrassment appeared at such a late date in evolution.13 As I

have already said, a chimpanzee can understand another’s kinaesthesia, but

cannot understand it as a kinaesthesia simultaneous with its current one.

The central question of this debate is, of course, how far and to what extent

chimpanzees are capable of seeing others as having a perspective on the world —

Metarepresentation and human capacities 109

seeing others as what Tomasello (2000) calls “intentional agents”, and not only

<LINK"bej-r84">

animate beings. Whereas nowadays there is some agreement that chimpanzees

are unable to deal with ‘false beliefs’, similar agreement on chimpanzees’ ability

to see others as having a perspective on the world has not been reached. In my

opinion a chimpanzee can interpret others as intentional agents but not

simultaneously as other individuals. A chimpanzee, in my view, lacks the

second line of awareness; consequently, when he understands the kinaesthetic

situation and visual direction of a conspeciﬁc, he can use only the ﬁrst line of

awareness (more precisely, his own immediate expectations) to do so. My

opinion does not simply mirror the general agreement that a chimpanzee is

unable to ascribe to conspeciﬁcs a false belief. I go further in restricting

chimpanzees’ mental capabilities. Chimpanzee A cannot even ascribe to

chimpanzee B, observing A, a visual perception of A’s body. Povinelli and

<LINK"bej-r67">

Prince (1998:76) mention this idea: “It occurred to us that perhaps the apes had

diﬃculty understanding themselves as objects of the visual perspective of others”

(emphasis added). But they, ﬁnally (see, supra, note 11) reject the idea or, more

precisely, reject the idea that it is there rather than at some lower point where

chimpanzees’ limits are to be found. See also Bruner and Kalmar (1998:314).

<LINK"bej-r8">

I suggest that no animal can imitate new and complex motor patterns if it

lacks the power to conceive a premotor plan of these patterns, a power that

requires a double line of awareness to trigger it oﬀ. It is for this reason that it is

so diﬃcult for beings capable of imitating simple, even nonautovisible, move-

ments to reach the imitation of new and complex motor patterns. As probably

no other animal, except for humans, can implement this kind of imitation, the

double line of awareness could be peculiar to humans.

The extraordinary growth of the premotor cortex supports the above

mentioned suggestion. Neither in a chimpanzee brain nor in a Neanderthal

brain (bigger in volume than that of current humans) does the premotor cortex

reach a proportion comparable to that of the human brain.

We can also stress the particular human hemispheric specialisation. The

contralateral hemispheric functioning is not, obviously, a distinctive feature of

humans. Besides, not only humans but a number of animals clearly exhibit a

preference for the use of a particular half of their bodies when accomplishing

certain motor patterns. However, only in humans is a new hemispheric

specialisation added to this contralateral functioning. Only in humans is the left

hemisphere (to right-handed people) responsible for the control of all cultural

movements, i.e., learnt by means of imitation of complex and new motor

patterns. ‘Analyticity’, ‘sequencing’ or ‘generativity’ would derive from that

110 Teresa Bejarano-Fernández

basic function. Thus, in (right-handed) people who are for some reason obliged

to use their less skilful hand (the left) to carry out a cultural motor pattern, it is

always their left hemisphere which controls the action, although it is not the

contralateral hemisphere with regard to that hand. Brain control of language

also reveals this hemispherical dichotomy. In right-handed individuals the

articulatory-phonetic skills, which we learn by means of imitation of complex

and new motor patterns, depend on the left hemisphere,14 whereas intonation

which we do not need to learn, is controlled by the right hemisphere. Could this

particular human hemispheric specialisation be related to the proposed double

line of awareness? The left hemisphere seems in charge of the imitation of new

and complex motor patterns. This imitation — I have suggested — requires the

second line of awareness, i.e., a ﬁctionalized sensorimotor control.

To suggest that the left hemisphere is related with the second line of

awareness is certainly rash.15 I have mentioned it because it permits me to insist

on an idea I have dealt with in the introduction. There, in defence of my

coining the term ‘double line of awareness’, I held that no former concepts do

justice to the fact that there must be two discrete and almost complete systems

of awareness. This is something I deﬁnitely maintain. In this paper I shall argue

that each of these two systems not only possesses a complete sensorimotor

framework (left and right body halves included), but also cognitive and

emotional frameworks. All these capacities are, in my view, duplicated in

humans. In this respect, we must take into account that the duplication of

already existing mechanisms (cf. Hodos and Butler 2001: 123) can imply only a

<LINK"bej-r42">

simple mutation, one powerful enough though to qualitatively change cognitive

functioning. This is certainly speculation, but matches very well with what we

are looking for. Now, leaving rash speculations aside, let me return to my

proposal. I suggest that curiosity, this need to know others’ stories, our gossip-

ing, our “addiction to narratives”, together with the remarkably stronger (as

compared with other primates) receptiveness (see note 1) that human babies

show to gestures and voices of adult individuals, are ways whereby the perma-

nent system of the second line tries to avoid situations analogous to those seen

in sensory deprivation experiments.

4. Premotor plan, double line and evocation

Let us now consider Piaget’s account of the power to evoke absent scenes. Does

this power for ﬁction coincide with what I have called the second line of

Metarepresentation and human capacities 111

awareness? In other words, what is the relation between these two proposals?

Piaget’s account of mental evocation reversed the causal line of former and

seemingly more natural explanations. He suggested that deferred imitation

(sometimes performed hours after seeing the model) of a new and complex

motor pattern is the cause, not the consequence, of the mental image evoked.

The link between the carrying out of this pattern and the model would, in

Piaget’s view, trigger the visual evocation of absent scenes. Before examining the

objections to Piaget’s proposal, I shall focus on some of its numerous merits.

Evocation, despite being concerned with external scenes, is under the direct

control of the subject, environmental factors playing no immediate role in the

process. Thus, it makes sense that the evocative power must be rooted in motor

abilities, i.e., in abilities that are directly under control of the subject. In

addition, the idea that the evocation is an ability of high-level requirements

gathers ever increasing support nowadays. Let us see these arguments.

Some authors (Duncan 1995; Duncan et al. 1996) link goal evocation with

<LINK"bej-r26"><LINK"bej-r26">

factor g(general) of human intelligence. Likewise, it has often been suggested

that the evocation of goals subject to a larger number of requirements gives rise

to cognitively more demanding tasks (see also Gattis et al. 1998). In this respect,

<LINK"bej-r30">

it seems clear nowadays that goal evocation is not necessary to explain goal-

guided animal behaviour: this can be explained by means of more parsimonious

mechanisms: ‘innate consummatory patterns’ (Lorenz’s 1966), learning and

<LINK"bej-r50">

‘satisfaction gradient’ (cf. Dickinson and Balleine 1991:294; Papineau

<LINK"bej-r4"><LINK"bej-r63">

2001:169–171).

Likewise, the scope of animal memory is currently being devaluated. Most

researchers now reduce animal memory to the status of a plain behavioural

guide obtained from one single episode or a number of them, a mechanism that

does not necessarily imply the evocation of these episodes.

Finally, the Rapid Eye Movements of sleeping animals are not in contradic-

tion with the presumed status of evocation. REM mostly correspond to dreams.

But dreaming, contrary to evocation, requires no double line of awareness

whatsoever. In dreaming, the usual primary line of awareness is switched oﬀ

and replaced by another, still primary, line, which must not be confused with a

second line. Accordingly, we cannot consider dreams as true evocations. A

mental recalling of past images cannot be interpreted as genuine evocation if, as

happens in dreams, those scenes are not taken as absent or past.

However, despite all these favourable signs, Piagetian explanation of

evocation is clearly insuﬃcient. Why does the link between two motor patterns

— the model pattern and the deferred reproduction of it — necessarily lead to

112 Teresa Bejarano-Fernández

a visual evocation of the absent model? Why can real and current imitation

trigger the visual image of an absent scene?

I think that the fundamental concept of premotor plan may shed some light

on this issue. Deferred imitation cannot take place without a previous premotor

plan and this, as I have suggested, cannot occur without a double awareness.

Consequently, the double awareness does not appear as an unnecessary or

unjustiﬁed embellishment of deferred imitation, but a sine qua non of the

imitation of new and complex motor patterns.

Piaget wisely warned against interpreting deferred imitation as a conse-

quence of a previous visual image of an absent model. However, by postulating

that deferred imitation inevitably requires — or is a consequence of — a double

line of awareness I am not falling into this trap. On the contrary, in line with

Piaget’s approach, I suggest that deferred imitation is the cause (not the

consequence) of visual evocation. We must bear in mind that the double line is

not at the beginning a visual evocation previous to a deferred imitation but the

sequence of ﬁctional postures that the premotor plan demands.

What happens, then, when the premotor plan turns into real motor

execution? This renders the ﬁctional postures totally superﬂuous. Nevertheless,

however superﬂuous the double line then becomes, it is already installed.

Consequently, now it can be given another function, and can be used, for

example, to carry out the visual evocation of the behaviour from which the

motor pattern was copied.

The double line of awareness performs three diﬀerent and in some way

progressive functions. In the drawing of the premotor plan, the second line

channels the ﬁctional bodily awareness. Later, in deferred imitation, it is the

second line which gives rise to the visual evocation of the behaviour. I have

already described these two functions. What is the third function?

Some motor patterns provide the subject with visual, or, more generally,

exteroceptive, sensations similar to those perceived in the model. Autovisible

movements, e.g., hand movements, are of this kind, but they are not the only

ones. Articulatory-phonetic patterns16 have the same consequences.17 In a

deferred imitation of a previously heard utterance the subject not only pro-

nounces the sounds but also hears them. This fact makes any evocation of the

exteroceptive sensations on the part of the imitating subject unnecessary. Then,

what is the role played by the second line here? What is the content of the

evocation? Let me put it this way. In the above mentioned modiﬁcation, i.e.,

from the ﬁrst function to the second, the double line was no longer required to

channel ﬁctitious postures, and began to be used to visually evoke behaviour

Metarepresentation and human capacities 113

which at that very moment cannot be seen. Now, in the new modiﬁcation, i.e.,

from the second function to the third, the double line does not have to deal

with this kind of evocation, and consequently, it is available for further use,

such as the evocation of the environments in which both the model and the

learner (or latent imitator) performed their actions. In short, only imitation of

the simplest kind, such as the imitation of autoperceptible movements could

trigger the most diﬃcult kind of evocation.18 Note that the ﬁrst linguistic

learning of deaf people involves the other kind of autoperceptible movements:

hand movements.

The dramatic novelty of the third function is that environment has no

direct relation with the motor pattern observed. It is important to note that the

visual bodily results of a movement (no matter if it is a nonautovisible move-

ment and, therefore, the imitator cannot perceive but only evoke it) have a

direct correspondence with it. By contrast, in this third function of the double

line, the relation between the evoked scenarios and the imitated motor pattern

can be a conventional, not natural, association. The arbitrary nature of languages,

i.e., the diﬀerences between one language and another, derives from this fact.

However, the meaning of words does not automatically arise from this

mechanism. Even names, the words most closely identiﬁed with concrete

objects, are given their meanings by means of syntax and as sentence compo-

nents. The word “tree”, but not the tree itself, requires a verb and is normally

matched with one adjective or another.19 On focusing on a word, the mind not

only activates the scenarios in which that word has been learnt. The most usual

syntagmatic connections of the word, i.e., the words and syntactic types most

frequently associated with it, will eventually be available or preactivated.

Let me summarise what I have said in this section. The double line of

awareness, whose original use20 consisted in making the imitation of complex

and new motor patterns possible, gives rise to cultural learning (articulatory-

phonetic and of manual techniques) and to evocation, with or without lan-

guage, of possible scenes. In addition, the imitation of complex and new motor

patterns would introduce the need to keep a blueprint of the sequence already

performed. When it comes to imitating a new and complex motor pattern, the

actions of the imitator cannot only be guided by his taking into account the

goal. It is the model learned, not the goal, which directs his imitative move-

ments.21 Bodily stands already accomplished must be paid due attention to until

the whole pattern is completed. Thus, the working memory could amplify its

scope and complexity.

114 Teresa Bejarano-Fernández

Now, I will concentrate on the two operating modes of the double line of

awareness. For brevity, I will call these two modes the parallel double line and

colliding double line. In order to properly understand the diﬀerence between

them, I will now deal with symbolic games (pretence) and laughter.

5. Parallel double line and colliding double line

5.1 Pretence and laughter as evolutionarily designed exercises

of the double line

In a social group any member unable to use a double line of awareness would

have been, against others with that faculty, most probably doomed. In my

opinion, the advantages of the double line of awareness are such that evolution

may have been rewarding its progressive implementation with pleasant conse-

quences from childhood onwards. Pleasure may have lured children into

practising the double line as happens in symbolic play (pretending), laughing

and joking.

A number of researchers (see a recent review in Bjorklund and Pellegrini

<LINK"bej-r7">

2000) have long pointed out some of the adaptive and concrete rewards that

pretending can produce in the course of human development. My proposal is

more general. I suggest that symbolic games (pretence) and laughter must be

interpreted as particular ways of encouraging the advanced implementation of

a double line of awareness. Pretence and laughter are, in fact, mainstreams of

the double line, and, more importantly, of its two basic modes of functioning

respectively.

In pretend play, children ride on a broomstick or on an invisible horse,

pretend to eat from an empty spoon, or use a banana or even their hand as a

telephone. It is the motor pattern, but not the prop, that is indispensable. In all

these games children imitate, out of a real and suitable context, a motor pattern

extracted from a model. This is “the ability to pantomime acts, that is, to imitate

them in a situation in which direct external cueing is absent”, (Suddendorf

<LINK"bej-r82">

1999:245); see also Luria and, in the 19th century, Hughlings Jackson. Since the

bodily pattern is actually carried out, the double line of awareness does not need

to ﬁctionalize postures and can then be used to evoke the model scene or the

common features observed in similar scenes. In pretence, the implementation

of the second function — out of the three progressive functions already seen in

Section 4 — would be rewarded with a phylogenetically designed gratiﬁcation.

Metarepresentation and human capacities 115

(Sequential imagination is easier when “it is prompted by physical action” —

Rieser, Garing and Young 1994: 1277). The child knows perfectly well that he is,

<LINK"bej-r71">

for example, riding on his mother’s broomstick in the hall. However, his second

line of awareness is simultaneously concerned with his horse and a prairie. “The

playing child monitors events occurring in reality and maintains the duality

between the two modes of thought (…) Thoughts about reality run parallel to

thoughts about pretence” Golomb and Kuersten (1996: 215).

<LINK"bej-r34">

I consider that children’s practising of evocation in pretend play is a

mechanism designed by evolution in order to prepare them for more demand-

ing skills in which the double line also plays a fundamental role: ‘evocation

without a bodily display’ of possible or ﬁctitious scenes, and ‘displaced speech’.

Thus, pretend play, although it is in part ontogenetically previous to these

abilities, would also be an evolutionary consequence of them: recall the very

similar situation of babbling.

Laughter, I suggest, has the same evolutionary explanation as pretend play.

Reddy (1991) has investigated the jokes played by very young children —

<LINK"bej-r70">

“tricks” in Newton et al. (2000). At the end of their ﬁrst year children can

<LINK"bej-r60">

already playfully give and take objects by turns. In giving the object the child is

well aware that the playmate is expecting it from him. When he playfully

decides not to give the object, his double line of awareness is simultaneously

concerned with his hidden intention and the other’s misled expectation. This

intentional clash causes the laughter to appear. See also Stern (2001:146):

<LINK"bej-r80">

“Infants ﬁddle with the very timing of interactions and expectations so as to

create pleasurable violations”.

It is not surprising that two-year old children are mostly unable to under-

stand any other kind of teasing or jokes. This give-and-take game provides

children with abundant and very easy clues for teasing. Firstly, prior to the

teasing, the object has changed hands many times before in a regular sequence

that must trigger the attribution of expectations to the playmate. Secondly, the

joke directly involves the child’s own voluntary action, that is, something under

his immediate control and consideration. In addition, the child obviously

perceives that he is the centre of his playmate’s visual ﬁeld: this power of

perception, the most primary result of the colliding double line (see supra, 3.3),

was already at the disposal of the child. Thus, the attribution of expectations to

the playmate is an attribution to a second-person (in contrast to the Maxi Test).

Having noted that this sort of jokes are not very demanding, we must not

forget that the child is already in possession of a theory of alien mind once he

understands false expectations or beliefs of others. The sensorimotor origins of

116 Teresa Bejarano-Fernández

the theory of mind are obvious. I agree with Gopnik and Meltzoﬀ (1997: 130,

<LINK"bej-r35">

133): “Rather than having a concept of psychological agents, young infants

seem to have a concept of person, which combines mind and body (…)

Kinaesthetic sensations have a great deal in common with classic mental states

(…), seem to be a primitive kind of mental states”. However, I disagree with their

attribution of kinaesthetic-visual matching to new-born babies (see supra, 2.1).

Since the ﬁrst line of awareness (the child’s intention not to give the object)

and the second line of awareness (his estimation of his playmate’s mistaken

expectation) refer to the same act and time, they collide making the child joker

laugh.22 Unlike pretend play, where the two lines of awareness run parallel, in

teasing others the two lines are in conﬂict.

In my view, the diﬀerence between pretence and laughter is not secondary

or irrelevant. Pretence and laughter serve as preparatory exercises for two

diﬀerent abilities involving the double line. A child joker, using his second line

of awareness in conﬂict with his primary one, is preparing himself to later pay

due attention to others’ beliefs and, in this way, to understand and produce

predicative and interrogative utterances (infra, 6).

5.2 The diﬀerence between the colliding and the parallel double line

We must now remember the two human capacities mentioned in Section 3: the

premotor plan and ﬁnger-pointing. Both require, in my view, the second line of

awareness. However, there is a signiﬁcant diﬀerence between them, and this

diﬀerence is similar to the one observed between pretending and laughing. Let

us consider this point.

The premotor plan is considered by the subject as one of his possible

conducts in the future. Certainly, this possible future conduct must be a

complex sequence, not only a simple immediate expectation. This is why it

requires the double line (supra, 3.3). However, let me now stress a diﬀerent

aspect: nothing in the premotor plan is in contradiction with the subject’s

considering such a sequence as one of his future conducts. Let us now consider

the ﬁctitious bodily sequence which must be evoked. In the subject, is it linked

to the model or rather to the imitation carried out by the subject himself?

Neither the one nor the other. The subject does not provide the mental contents

of his parallel double line with a concrete location, and consequently he can

consider them as future possibilities of his. This feature of the parallel double line

is, by the way, indispensable for the phonemic or ‘social model’ I deal with in 3.1.

Metarepresentation and human capacities 117

Finger-pointing is very diﬀerent. Let us remember howpointing comes into

being. The observer A must not only infer the kinaesthesia of the subject B, who

is ﬁnger-pointing, but A must also attribute to B the capacity to see him (A).

This cannot be accounted for by the simple and primitive mechanism set in

motion when an animal detects another’s eyes ﬁxed on him. The observer A

must deal, thanks to his second line of awareness, with a visual ﬁeld in which he

is part of the scenery. Thus, the observer can by no means take this particular

visual ﬁeld for one of his possible, future visual ﬁelds. This particular content

(contrary to what we have seen with respect to the contents of the parallel

second line) is attributed determinately and precisely only to an alien self. This

attribution is crucial to the collision between such a content and the primary

awareness of the subject.

There is another way of expressing the diﬀerence between these two modes.

Bear in mind that, in my view, the double cognitive architecture was only

needed when the subject started to attend to a cumulative sequence of move-

ments and postures diﬀerent from his own current movements and postures.

Thus, the primary function could not have been to attend to a punctual

kinaesthesia diﬀerent from one’s own, current kinaesthesia: the kinaesthetic-

visual matching would have been suﬃcient. But, once the second line is

available, things are diﬀerent. Now, i.e., in human beings, the accurate imita-

tion of an alien kinaesthesia would not need to be interpreted as the subject’s

own expectation. Rather, it would be implemented by the second line and it

could eventually be in inherent conﬂict with the subject. Thus, we can conclude

that the parallel double line, but not the colliding double line, is linked to

cumulative sequentiality, whereas the colliding double line, but not the parallel

double line, requires a precise and determinate attribution.

The same diﬀerence which we have seen between the premotor plan and

ﬁnger-pointing can be found between pretending and teasing. Nothing in the

ﬁctitious, pretended during a child’s play, scene is in contradiction with the

child’s considering such a scene as one of his future conducts. Contents are not

ascribed to a particular location, nor are they attributed to in a precise way. The

play scene and the previous model scenes, the child and the adults he is imitating,

the child in his current situation and in the future — all these elements, indistinct

from one another and at the same time, belong to the content of the second line of

awareness. Therefore, in a strict sense, none of these alternative possibilities may be

identiﬁed with that content. By contrast, the teasing child knows what he (the

child) is about to do and, while thinking so, he by no means can conceive as a

future possibility of his any other belief which contradicts his current knowledge.

118 Teresa Bejarano-Fernández

Let me now compare my proposal with some others. Since the discovery of the

similarities between pretence and the perception of alien beliefs — two abilities, on

the other hand, clearly very diﬀerent — many authors have suggested two stages,

or two successive types of mind corresponding respectively to these capacities. The

widely admitted lapse (over two years) between the ages at which these abilities

appear has been used as an argument by these authors (e.g., Perner 1991; Barresi

<LINK"bej-r65"><LINK"bej-r6">

and Moore 1996; Suddendorf 1999). They all accept this time-lag as a premise.

<LINK"bej-r82">

By contrast, I suggest that the parallel double line and the colliding double

line appear and develop simultaneously. In children, not only ﬁnger-pointing

and imitation of new and complex motor patterns appear almost coincidentally.

The perception of false belief (implied, as has been already seen, in precocious

teasing) appears quite simultaneously with pretence. I think that the

phylogenetic process must have been very similar: the two operating modes of

the double line must have appeared simultaneously, perhaps with a slight

priority as far the imitation of new motor patterns is concerned. As soon as the

double line was achieved and ready (in its role of parallel double line) to permit

the imitation of new and complex patterns, it was also possible to understand

oneself as the object of the visual perspective of others (i.e., the colliding double

line also came into being). The crucial step — or the crucial mutation, if you like

— was the implementation of the double line and not of a particular mode of it.

How then did the idea that there is a considerable time-lag between the

beginning of pretence and the beginning of the perception of false beliefs

receive such wide acceptance? In my previous comments on children’s teasing,

I have already mentioned in passing how the diﬃculty of false belief had been

exaggerated, i.e., how it was ascribed to an excessively late age, but I shall deal

with this issue later (see infra, 6). For the moment, let me state that the Maxi

Test, as elaborated by Wimmer and Perner (1983), is excessively diﬃcult. I

<LINK"bej-r91">

suggest that the main reason for this diﬃculty is the need to keep sequential and

cumulative track of Maxi’s mental states, and, more concretely, to achieve this

when, for a while, both Maxi has been absent and the spectator has been trying

to keep track of the mental states of another character (see infra, note 35).

6. The perception of beliefs diﬀerent from one’s own, current beliefs

and the origin of syntax

Predicative (that is, declarative or descriptive) communication was often

deﬁned as that sort of utterances where language is supposed to conform to the

Metarepresentation and human capacities 119

world. This classic approach, nevertheless, overlooks signiﬁcant aspects of

predicative communication. Problems do not only arise when considering the

controversial case of lies: lying has always been regarded as a blameworthy

conduct and it must be statistically infrequent if it is to be successful. Sincere

and true declarative utterances do not ﬁt well with the classic approach either.

Firstly, consider the case of negative predicative utterances. Where do negative

predicates arise from? How does the feature which must be denied come to a

speaker’s mind when trying to conform his utterance to the world? Recall the

debate on Russell’s ‘negative facts’; Bergson is also an essential source. Negative

predicates cannot be traced back to perceptual data: even Gibson agrees on this

point (see also Wegerif and Mercer 2000:189). Secondly, the classic approach to

<LINK"bej-r88">

aﬃrmative declaratives is subject to an equally serious objection: how does the

speaker choose the feature to be predicated? Any object eventually acting as a

sentence subject has innumerable known features which could be chosen for

descriptive, i.e., predicative, purposes. However, each predication involves the

selection of one and only one of that huge class of features. Any theory of

declarative language is incomplete if it does not account for the mechanism

whereby predicates are chosen. In my view, the classic deﬁnition of declarative

speech as a linguistic product aiming at mirroring the world is no longer of any

use. We should better approach the topic by giving the speaker’s perception of

others’ beliefs a fundamental role in his production of declarative utterances.23

For example: “When I tell you that Liverpool won the football match, I am

doing so in order to give you new information that I believe you do not have,

and that you might be interested in or want (…) Since many pragmatics rules

involve tailoring one’s speech to what the listener needs to know, pragmatics can

be seen as intrinsically linked to a theory of mind” (Baron-Cohen 1999:262, 266).

<LINK"bej-r5">

The recollection of one’s own, previous belief would serve likewise. However,

it is not clear that this is easier to achieve than the former. We should take into

account that a recollection of one’s own false belief would be as diﬃcult

(perhaps, even more diﬃcult: see supra, Section 5, on the teasing child and also

infra, in the present section) as the knowledge of others’ beliefs. Children pass the

Deceptive Box- or Smarties-Task (own past false belief ), elaborated by Gopnik and

<LINK"bej-r35">

Astington (1988), and the Maxi Test (alien false belief ) at the same age.

Obviously, sometimes, unlike in the Deceptive Box Task, one’s own

previous beliefs can have been true (true in the past but already no longer true).

The best way of referring to these beliefs is by calling them outdated beliefs.

What can be said about the recollection of one’s own outdated beliefs? The

perceptive updating constantly and automatically erases past perceptions to deal

120 Teresa Bejarano-Fernández

with new ones. As I said in section 4, past episodes may serve as a means for the

animal to guide its behaviour without necessarily being evoked. Millikan’s

<LINK"bej-r59">

(1998) ‘perceptual reidentiﬁcation’ would certainly provide the current

perception with clues and instructions derived from past perceptions. However,

this does not imply the simultaneous awareness of two diﬀerent episodes — the

past one and the current one. I maintain that the recollection of one’s own true

belief concerning a past experience is achieved by means of the second line of

awareness. (See Cosmides and Tooby 2000: 93: “Decoupling and Episodic

<LINK"bej-r16">

Memory”.) But my argument is more precise here: these beliefs, as well as

another’s false beliefs and one’s own past, false beliefs, need the colliding double

line on the part of the subject. Let us remember the conclusion of 5.2: Evoca-

tions carried out by means of the parallel double line are neither linked to a

precise episode nor anchored in a particular place. Obviously, one’s own

outdated beliefs do not comply with these speciﬁc and delimiting criteria.

Now, let me proceed with my argument. The well-known link between

declarative communication and alien false beliefs can, in my view, lead to some

interesting conclusions. I suggest that alien beliefs are the content of the thema

(or subject in the cognitive-communicative syntax — the ﬁrst, basic and

empirically universal syntax, in my view). The thema has been deﬁned as the

object or event whose knowledge the speaker ascribes to the hearer. In my

opinion this is a correct but still imprecise deﬁnition. We should also under-

stand that knowledge ascribed to the hearer by the speaker is seen by the latter

as insuﬃcient, not updated or partly erroneous. And this metacognitive (or

second-level) mental state, this alien belief, which demands correction, is, in my

view, the thema. The image of the object as it really is (or, more precisely, as it

is known by the speaker) includes all the features it exhibits at that moment

and, therefore it does not need to be collocated with the predicate. Thus, the

object as it is known by the speaker cannot be the thema. The thema or subject

is what actually needs to be brought together with the rhema or predicate. This

incompleteness of the thema contrasts with the self-suﬃciency of holophrastic

expressions and makes the thema the starting point of syntax.24

Let me put this in other words. In replies, the thema, if any, echoes to some

degree the question. Likewise, in usual predications (i.e., not prompted by a

previous question) the thema is also an echo, not, obviously, of a previously

proﬀered question, but of the mental content or belief attributed by the speaker

to the listener.

Metarepresentation and human capacities 121

In direct and indirect quotations meanings are being metarepresented. As

Récanati (2000b:354) puts it: “It is an act of simulation in the scope of a serious

<LINK"bej-r69">

assertion”. Likewise, the syntax of simple predications involves, in my view, a

metarepresentation: The thema is, for the speaker, the false or insuﬃcient

representation which he attributes to his listener. Thus, the same cause which,

in referred speech, makes syntax more complex would also be causing the

primeval syntax.

In my view, metabelief does not start with direct or indirect quotations, but

earlier. It really starts when an utterance is heard by someone who disagrees

with that utterance. In such a hearer the conversion of that utterance into a

metabelief has already taken place, although the conversion may not yet be

expressed verbally. Obviously, hearers do not tend to utter: “The person who is

now talking to me thinks (/You think) that p”.25 However, the hearer has

attributed the utterance to somebody else and has restricted the truthfulness of

that utterance to the mental world of that speaker.

Now, in order to formulate my proposal on the thema, let us try to answer

this question: what happens to the incorrect, insuﬃcient or outdated belief

which that hearer (and would-be speaker) attributes to the former speaker (and

would-be hearer)? More generally, what happens to the belief that the utterer of

a predicative statement attributes to his hearer? It will become the cognitive

content which the speaker will use as the thema of his predication. In other

words, having perceived the alien belief, the speaker can already foresee that the

hearer will not provide the term designating the particular object with the same

cognitive content that he, i.e., the speaker knows. Consequently, if the speaker

used it with the cognitive content that he knows, he would commit an error

similar to that error — also a communicative error, according to Mackenzie

<LINK"bej-r52">

(1985:340) — which in arguments is called begging the question. We must here

remember the pragmaticits’ stress on the double mission of communication: to

attend both to the reality being described and to the communicative context.

On the other hand, the rhema would be the element capable of modifying

the thema and (with the exception of lies) getting it closer to the view that the

speaker has at that very moment. Aﬃrmative predication would then add some

features to a belief seen as insuﬃcient by the speaker at that particular moment.

Negative predication, by contrast, would remove some features incorrectly

included in that belief. The most basic predicate of children, i.e.,the starting point

of their predicative communication, is usually and simply “no”; in other words, it

is restricted to a mere rejection of a belief diﬀerent from his own at that moment.

122 Teresa Bejarano-Fernández

I must make some remarks on others’ beliefs involved in linguistic produc-

tion. These beliefs need not necessarily be false, outdated or incomplete beliefs

from the point of view of the speaker (i.e., the “false beliefs” of the ‘theory of

mind’). They can be conceived both as less or (by the interrogative speaker) as

more complete and updated than his own beliefs. Interrogative communication

is clearly distinct from all kinds of animal exploratory behaviour. The human

questioner must necessarily have conceived other mind diﬀerent from his own.

But this conception of an alien mind cannot be explained by forcibly reducing

it to “knowledge of others’ false beliefs”.

Likewise, others’ beliefs involved in linguistic production would not

necessarily have to be inferred. I wish to emphasise that, although in the Maxi

Test (Wimmer and Perner 1983) they must be inferred (from the accumulative

<LINK"bej-r91">

sequence of the two ﬁrst scenes), in real life they can be linguistically received.

(cf. Jerison (1988:78) “Our brain has to distinguish the reality generated by

<LINK"bej-r44">

one’s own sensory information from the reality generated by verbal information

from another individual”. See also Wilson (2000:441): “It would be interesting

<LINK"bej-r90">

to check whether the use of overt linguistic devices would facilitate comprehen-

sion. For example, is it easier to attribute a false belief when expressed or

implied by an utterance (“The ball is in the cupboard”) than by inferring it on

the basis of non-communicative behaviour?” (Bruner and Kalmar 1998: 318.)26

<LINK"bej-r8">

This linguistic reception as well as the easy inference that we have seen in

jokes observed by Reddy allow us to solve a dilemma correctly detected by

Risjord. As this author pointed out, “Gricean analysis (of indicative, predicative,

utterances), in which the speaker must regard the hearer as the kind of thing

that has beliefs” (Risjord 1996:467), and the age at which this kind of sentences

<LINK"bej-r72">

appear in children are in contradiction with the disappointing results obtained

in the Maxi Test27 by children capable of true declarative (or ‘predicative’)

communication. In the face of this incoherence, Risjord rejects Grice’s doctrine

on predicative communication. For me, on the other hand, the contradiction is

simply solved by paying due attention to the easy accesses (Reddy’s “teasing and

mucking about in the ﬁrst year”, and linguistic comprehension) to the false

beliefs of others. We must bear in mind that a child’s ﬁrst predications occur in

reply to some adult linguistic intervention. Thus, considering a child’s facility

for linguistic comprehension, it is not overadventurous to presume that

children perceive auditively alien false beliefs much earlier than they are able to

pass the Maxi Test, perhaps even before they are able to recollect their own false

(or merely outdated) beliefs.

Metarepresentation and human capacities 123

I ﬁnd no serious objection to considering linguistic reception as the

ontogenetic cause (together with laughter) of the perception of alien false

beliefs. In this case, there would be a bidirectional causal relation: from lan-

guage to false belief and from false belief to language.In ontogenesis, there is no

risk of a vicious circle.

Let us now deal with the evolutionary genesis. It is unnecessary to mention

laughter here: the pleasure of laughter would have evolutionarily arisen as a

consequence of the huge usefulness obtained by the perception of alien beliefs

with the aid of language. My question, therefore, is this: could linguistic

reception also be the cause of the evolutionary genesis of the perception of alien

false beliefs? I think so. It is of course true that, in my view, no proto-human

creature, so to speak, could have ever uttered a truly declarative or interrogative

sentence before he was capable of conceiving minds alien to his own. However,

the linguistic reception capable of prompting the perception of alien false beliefs

need not be reduced to the reception of predicative or interrogative sentences.

Vocatives28 or perfectly speciﬁed requests could have triggered oﬀ the whole

process.29 In some cases, hearers of such cries or concrete requests could have

come to imagine the false beliefs of an utterer who lacks the knowledge they

have, e.g., that the individual being addressed is absent or that the thing being

requested is missing.30 Among creatures already capable of meaningful articula-

tory-phonetic patterns (I prefer not use the term word: see supra, Section 4),

this hypothetical situation could have been the link between a presyntactic,

appellative era (the ‘Holophrastic era’) and the human syntactic language.31

As I said earlier, “no” is the most precocious rhema of children. Something

similar could have been the original set-up. From the still pre-predicative

rejection of a conative message (from the rejection of a heard command or

demand) the speaker could have easily come to rejecting (using now a declara-

tive “no” functioning as rhema) the belief involved in the message he heard.

Whereas the ﬁrst kind of rejection would not imply a metabelief, the second

certainly would.

The perception of false alien beliefs would not only give rise to the need of

predicative communication, it would also — and this is my proposal in this

section — provide the tools for fulﬁlling that need. The cognitive content

which, with regard to a particular reality, a speaker ascribes to his hearer is only

the ﬁrst half of a whole, which lacking its corresponding second half, is incom-

plete. This “ﬁrst half”, this incomplete, non-holophrastic meaning would be

genetically crucial: it would have been the ﬁrst genuine word ever uttered, that

is, the ﬁrst ‘meaning for syntax’.

124 Teresa Bejarano-Fernández

So far, I have dealt mostly with the communicative function of syntax. Let

us now see what happens when the erroneous, insuﬃcient or merely outdated

thought which is corrected or updated by means of a mechanism similar to

predication is the past belief of the speaker himself. I think that this process

would be at the root of some exclusively human skills. Let us examine the

human ability to interpret the footprints of predators. Vervet monkeys are

unable to understand the menace of their predators’ footprints (Cheney and

<LINK"bej-r14">

Seyfarth 1992: 146). This incapacity could be easily explained if the monkeys are

actually conﬁned to mental associations within the range of an inverse model.

In the inverse model, the neutral stimuli being conditioned precede the

‘unconditioned’ stimulus. Thus, when the subject faces them again, they give

rise to an immediate concrete expectation, an expectation which, as I have

already said, does not need evocation. By contrast, the explanation of footprints

requires awareness of an outdated perception as such. Describing a current

footprint in terms of its corresponding outdated perception would yield, then,

a cognitive gain which relies on a mechanism similar to predication (current

footprint = outdated perception + subsequent departure of the predator). Thus,

the mechanism of predication (probably developed to permit interpersonal

communication) redeems the uselessness of the subject’s own past belief.

The Buytendijk Task oﬀers us another similar example.32 In Buytendijk’s

experiment, animals (normally chimpanzees) are faced with a row of containers

and they see the bait being placed in the ﬁrst one, which is later covered. Later

on, without the animal knowing, the bait is placed in the second container. Of

course, the animal will rush to the ﬁrst container in search of the bait. After its

failure to ﬁnd it, the animal is shown that the bait is in the second container.

Put again to the test, the animal looks for the bait in the second container. Once

again, the animal fails and is shown the bait in the third one. The experiment is

carried out a number of times, always placing the bait in the next container in

the row. No animal was able to discover the apparently obvious rule: “search

this time for the bait in the container next to the one where you found it last

time”. Animals are apparently unable to single out the current container of the

bait in terms of his past beliefs. “Present container = previous container + 1”:

this formula suggests the origin of our numerical sense. As in the case of the

footprints, the awareness of past beliefs as such and the addition of the neces-

sary element to correct such beliefs would also be a sine qua non.

What happened ﬁrst? This type of ﬁndings or, rather, the interpersonal

declarative communication? It is hard to say.33 What I am prepared to say is

that there is a close relationship (see Dascal 1983) between pragmatics, the

<LINK"bej-r19">

Metarepresentation and human capacities 125

communicative use of language, and cognition. Over the last few years, this

issue has been the object, in one way or other, of my intellectual activity.

7. Morality, ﬁctionally triggered emotions and human work

Beyond the above mentioned capacities, there are other human skills which

would be produced by the double line working parallel or in collision. Let us

remember what Cosmides and Tooby call ‘ﬁctionally triggered emotions’. We

could also label these emotional responses Radfordian or Waltonian emotions.

Perhaps Augustinian or Pushkinian (“pleasurable tears”).34 They operate, in my

view, through the parallel second line.35 Consequently (supra, 5.2), these

contents are not attributed in a precise and exclusive manner. In the spectator’s

mind, they belong both to the model and to the latent imitation that he, the

spectator, carries out, and, therefore, they cannot be attributed in a precise

manner to either of them. Since they are not fully alien, the spectator’s emo-

tions are not true compassion. Since these emotions triggered by the ﬁctitious

characters do not completely belong to the spectator either, they can be also a

pleasure to him, even though they mirror human suﬀering.

Let us now consider “morality”. This should be interpreted as an imple-

mentation of the colliding double line. Thanks to the colliding second line, the

subject can see himself as part of others’ view of the situation and, accordingly,

as a potential agent in it, e.g., as a help mate of others. Note that in these cases

the use of the parallel double line — which would imply a passive response, just

as in dealing with ﬁction — would not be moral.

Morality, as happens with all the abilities derived from the double line,

would then be an exclusively human capability. An animal, certainly, could

perceive, in a ‘mindblind way’, others’ eﬀorts and attempts: these are, as many

authors have already suggested, originally subject to perception as self-propelled

motions. Those animals (probably only apes) able to accomplish a kinaesthetic-

visual matching could perhaps perceive another individual’s feeling when seeing

it in a location and posture already known by the observing animal (de Wall

1998). But, as I have already argued, they would not perceive that feeling as a

truly alien feeling, instead they would perceived it as a possible, immediate

expectation for himself. Consequently, although it is of little importance here,

they could not perceive it as integrated in a cumulative sequence. In human

beings, however, things are quite diﬀerent: they are diﬀerent as a consequence

of the double mental line or, more precisely, of the colliding double line. After

126 Teresa Bejarano-Fernández

perceiving a mental state — visual ﬁeld, expectation or belief — radically

incompatible with one’s own one, a human being is able to conceive a mind

diﬀerent from his own. Beyond this point, the needs and urges of another

individual can be perceived as internal phenomena of an alien mind and,

therefore, can demand a certain response in the observer (as long as his atten-

tion remains concentrated on that reality) — a response which, on the other

hand, may be harmful or disturbing for the observer. When I say “harmful” I

mean harmful all the way, not only causing damage which could be later

compensated, as happens in so-called “reciprocal altruism”.

By deﬁnition, a subject will see as false all those alien beliefs which are

diﬀerent from his own and will restrict them to the limited world of an alien

mind, keeping them from operating in the external reality in which he is living.

By contrast, the needs of other individuals, however diﬀerent from one’s own

needs they may be, are never false. Consequently, for the observer, they are not

only an ingredient of an alien mind but also of the external reality or environ-

ment of the observing individual. It is perhaps this diﬀerence as compared to

the perception of alien beliefs that leads Nichols (2001:450) to say that only

<LINK"bej-r61">

“minimal mindreading” is required to perceive alien distress.

Let us now recall that the knowledge acquired by an animal from its

environment guides its behaviour: this function is at the very origin of the ﬁrst

sensorial and perceptive capacities. Certainly, the perception of alien mental

states is very sophisticated knowledge, beyond the reach of animal perception.

Nevertheless, the perception of others’ needs is also knowledge of the environ-

ment. In this respect, Taylor (1999) is right when he stresses the perceptive

<LINK"bej-r83">

character of the process whereby someone else’s distress is perceived. The

novelty of this sophisticated perception lies only in the fact that now, for the

ﬁrst time ever, certain knowledge about the environment erects inside the mind

of the observing individual a new judgment centre on what is convenient and

what is harmful, that is, a judgment centre diﬀerent from the own judgment

centre. In other words, for the ﬁrst time inevolution, the two functions ascribed

to perception — to inform about the environment, on the one hand, and to

serve the perceiving animal, on the other — can be clearly separated to the

point of being mutually incompatible. Consequently, the observing individual

can opt to reject the knowledge available on such a peculiar reality; in other

words, he can opt to hide part of what he now knows as true.

The diﬀerence between ﬁctionally triggered emotions and morality

mirrors the general diﬀerence (supra, 5.2) between the parallel and the

colliding double line. In each case, this diﬀerence is displayed in its own

Metarepresentation and human capacities 127

particular way. Nevertheless, these two operating modes and all their variants

always comply with an identical rule: the second line of awareness is, at the

same time, supra- and sub-ordinate to the ﬁrst.

It is subordinate because it is the awareness of one’s own, real and current

situation which takes the lion’s share of the individual’s resources and guides

his behaviour. The goal of the ﬁrst line of awareness, which consists of turned-

on innate consummatory patterns, sets immediately in motion all the resources

of the individual, and activates any external subgoals which, having been re-

encountered by the subject, are linked to that innate pattern by conditioning or

learning. Animals, which only have a primary line of awareness, do not have

external ﬁnal goals (supra, 3). On the contrary, the (either selﬁsh or truly

altruistic) goals of the second line of awareness — either external, evoked goals

or others’ goals which may have a high cost36 for the observer — cannot set the

individual in motion until he of his own accord makes a lengthy enough mental

eﬀort. The individual must connect those goals with his ﬁrst line innate goals,

the only ones which can trigger oﬀ his action (I reject, e.g., Searle’s “gap”

(2001:ch. 3). This mental eﬀort marks the diﬀerence between a truly human

work and that of beavers, for example. The exclusively human work involves the

simultaneous evocation of external goal with attention to instrumental tools

provided by the situation.37 The external, evoked goal and its pleasant, expected

consequences — “appraisal and aﬀect mutually regulate each other” (Mascolo

<LINK"bej-r54">

and Fischer 1998:351, 377) — must be channeled through the second line of

awareness.38 Contrary to what happens in animals with their goals (which are

never external), now, to prompt the action there is only one means available:

evocative eﬀort. There is surely a good reason for the frequent metaphors which

link second line goals and the satisfaction of innate consummatory patterns. In

these cases, language (concretely, the rich semantic and syntactic connections

implied in these words) is probably contributing to the mental eﬀort peculiar to

human work.

On the other hand, the second line of awareness is supraordinate to the

ﬁrst. It is supraordinate because only this skill to evoke and conceive imagined

scenes and alien beliefs or expectations gives rise to the notion of mental states,

which is clearly a more outstanding feat than the mere possession of mental

states. In other words, our metacognitive abilities are far more relevant for the

study of exclusively human cognition than the classic topic of the relationships

between body and mind. Because of this supraordination I disagree with some

simulationist statements. Simulationism certainly stresses the role of imitation in

the making of the ‘theory (forged by the subject) of mind (his own mind or not)’.

128 Teresa Bejarano-Fernández

This stress39 is akin to my own approach. However, simulationism — e.g.,

Goldman (1993) — does not usually do justice to the fundamental role of alien

<LINK"bej-r33">

beliefs. In my view, only alien beliefs or, more precisely, beliefs diﬀerent from

one’s own current beliefs, give rise to the perception of mental states as such.40

We must understand that the very notion of mental state will be beyond the

reach of the subject in so far as he is unable to abandon the so-called ‘ﬁrst-

person point of view’.41 Only the second line of awareness, the awareness of

beliefs in conﬂict with one’s own, or of ﬁctitious or non current scenes can

make the subject acquainted with this notion.

8. Conclusion

In the late 80s, an intellectual adventure which has sometimes been called In

search of a metarepresentational link (Nielsen and Dissanayake 2000), appeared.

<LINK"bej-r62">

In some respects, in this paper I contribute to this debate by trying to:

– Trace our inquiry back to the very origin of metarepresentational capaci-

ties. I suggest that the starting point — the appearance of a double line of

awareness — was originally required by the imitation of new and complex

motor patterns or motor learning (the bridge between chimpanzee’s

kinaesthetic-visual matching and human culture).

– Show the variety of skills derived from the double line. Cosmides and

<LINK"bej-r16">

Tooby (2000) linked Leslie’s ‘decoupling’ (or ‘metarepresentational

operator’) to a long list of cognitive processes. In addition to that list I have

linked to the double line some other items: it is not only the already

mentioned motor learning, but pointing, laughter, and the origin of syntax

as well.

– Distinguish the two general classes of skills derived from the double line of

awareness: those arising from the parallel double line and those arising

from the colliding double line. In my view, these two modes must not be

interpreted as successive stages.

Is this enough to suggest that complex imitation and the double line of aware-

ness is the crucial factor in the evolutionary step from apes to humans? Obvi-

ously not. Let us recall the basic, classic objection to the importance of imita-

tion: in imitation the imitator never surpasses his model. Of course, the

imitation of new and complex motor patterns and the double line, by giving rise

to the skill of evocation and symbols, would be necessary to accomplish a truly

Metarepresentation and human capacities 129

creative problem-solving ability: Nowadays, there is almost universal agreement

that the experience with external symbol systems provides the necessary

representational scaﬀolding for creative problem-solving (see Clark and

<LINK"bej-r15">

Thornton 1997; Karmiloﬀ-Smith and Inhelder 1970). However, it is doubtful

<LINK"bej-r46">

that symbolic scaﬀolding (even with the helpful addition of the incredibly rich

semantic and syntactic connections of linguistic symbols) is suﬃcient; in other

words, that it is the enabling condition, for the ﬂash-of-insight style

reconception or ‘redescription’, for truly creative problem-solving. Therefore,

unless a more direct relation between the double line and creative problem-

solving can be proved, we will have to admit that not all exclusively human

psychic processes have their immediate origin in the double line.42

Notes

*I am particularly thankful to Prof. Dascal. Both his sharp criticism and his generous

<DEST"bej-n*">

suggestions have considerably improved this work.

1. Obviously, my objections to nativism of the kinaesthetic-visual matching must not be

interpreted in the sense that I have overlooked the early susceptibility of infants (Rochat

<LINK"bej-r75">

2001), where adult gestures and voices are concerned. This susceptibility is, undoubtedly, a

necessary requirement for the imitative capacity.

2. It is true that phylogenetically, a new simple pattern can by means of a rupture arise from

an old and more complex unitary pattern. Likewise, gross synergisms, that are in place and

functional at birth, diﬀerentiate over time such that the infant can isolate component

movements. But this is not relevant when we are dealing with imitation. An innate motor

sequence only has constituent parts at a subpersonal level. Thus, these parts, as long as they

have not been previously individualised in the subject, are not imitable by that subject.

3. Goals vs. mere results: Whiten and Ham (1992) vs. Tomasello et al. (1993).

<LINK"bej-r89"><LINK"bej-r84">

4. Social models are also more stable. Heyes (1993:999) argued that “imitation is insigniﬁ-

<LINK"bej-r41">

cant in relation to culture, because imitation does not insulate information from modiﬁca-

tion through individual learning in the retention period between acquisition and retransmis-

sion”. In my view, the social model is perfectly retransmitted except when useful innovations

are found. Thus, Heyes’ claim is simply the classic and basic objection against the decisive

role of imitation — imitation does not allow the imitator to surpass his model.

5. In latent imitation, in a premotor plan, there is no diﬀerence between autoperceptible and

nonautoperceptible movements.

6. According to Jeannerod (1994), motor preparation, which is an entirely non-conscious

<LINK"bej-r43">

process, must be distinguished from motor imagery, whose content can be accessed

consciously. In my view, motor imagery would originally arise exactly in this latent imitation.

Jeannerod proposes that it originally arises in cases where execution is blocked or hindered.

130 Teresa Bejarano-Fernández

My objection is as follows: could it originally have been of any use to store the motor imagery

in these cases? Just as the child’s perceptive updating disposes of outdated beliefs, beliefs that

at a merely biological level are useless, as shown by the Deceptive Box Task (Gopnik and

<LINK"bej-r35">

Astington 1988), so too motor preparation rejected by circumstances would also be deleted.

Only after motor imagery has emerged to make imitation of new and complex motor

patterns possible, can motor imagery be used with other functions. In other words, unlike

Jeannerod, I think that motor imagery is originally an interpersonal process. Remember that

“motor image cannot be entirely segregated from visual image” (Jeannerod 1994:189) and

<LINK"bej-r43">

that children are unable to achieve a conscious motor image of their own movements until

they are conscious of the movements of others.

7. Glenberg (1997), commenting on the evocative recall of past episodes, says that in these

<LINK"bej-r31">

cases the subject’s awareness of his real situation is suppressed. I think that suppression is not

a correct way of describing the procedure involved. Suppression only takes place in dreams,

that is, when all or most voluntary motor activities of the subject are switched oﬀ and the

subject is in a relatively safe resting place. Under diﬀerent circumstances any total switching

oﬀ of the ﬁrst line of awareness may be suicidal. Except for dreams, no other consideration

of past, ﬁctitious or possible scenarios involves the true suppression of the ﬁrst line of

awareness, but a double line of awareness.

8. The ﬁrst diﬀerence corresponds to the parallel double line, whereas the second one goes

with the colliding double line. See Section 5.2 for details.

9. On captive chimpanzees, Povinelli, Bering and Giambrone (2000: 530) propose: “Given

<LINK"bej-r67">

that humans automatically interpret such gestures mentalistically, it is hard to imagine the

circumstances under which captive chimpanzees would not develop a gesture that structural-

ly resembles pointing — regardless of how they understand it”. In my view, these lines

(Vygotskian ﬂavour in a non-Vygotskian cake, I would say) oﬀer a lucid account.

10. Let us consider how all children enjoy hide-and-seek. Could this game be a strategy to

encourage the double line of awareness exclusive to humans? In watching the movements of

the seeker, hidden players ascribe to him a kinaesthesia, an expectation and a perceptual ﬁeld

diﬀerent from their own. What matters here is that in the game the hidden players imagine

themselves as objects of another subject’s internal states, something which requires two

simultaneous lines of awareness. In teasing games (see infra, Section 5.1) we will also ﬁnd a

similar collision of others’ (presumed) expectation with the true knowledge of the subject.

11. Povinelli et al. (2000:519) argue that a low-level model is also adequate for gaze-

<LINK"bej-r67">

following. “The chimpanzees (unlike 3-year-old children) failed to distinguish between at-

target and above-target: they simply chose the container on the same side of the apparatus to

which the experimenter’s head was turned”. According to Povinelli et al., this shows that “the

chimpanzees do not imbue it (the gaze-direction of the experimenter) with referential

signiﬁcance”. I am not persuaded: A weak cue is better than no cue. What chimpanzees, but

not children, lack is the ability to interpret another’s gaze as a communicative signal.

12. Baron-Cohen (1995) conjectures a specialized innate mechanism that detects the gaze-

<LINK"bej-r5">

direction of another organism. In my view, such a function is performed by the kinaesthetic-

visual matching. By contrast, the mechanism I am referring to must be more primitive and

available to a larger number of animals. It would only allow an animal to perceive that

Metarepresentation and human capacities 131

another’s eyes are ﬁxed on him and to expect some alien reaction, most likely, an attack or

a ﬂight.

13. Why is blushing a speciﬁc display of shame? Why has shame been phylogenetically

ritualised for social communication? See some suggestions in Parker (1998:129n).

<LINK"bej-r64">

14. Hauser and Andersson (1994:3946): “A monkey’s right ear leads the orienting response

<LINK"bej-r40">

only if the call is from species-typical repertoire”. This specialization could be the origin of

the linguistic specialisation of the left human hemisphere. Certainly, without complex

imitation, the right ear’s advantage would be of little help. Nevertheless, it might be

interesting to investigate whether that specialization appears in adults as a consequence of,

perhaps, the situation of the right ear in the womb (better situated for hearing than the left

one).

15. However, in the lively debate of the 60s and 70s on split-brain patients, some authors

proposed that even the normal human brain would possess dual consciousness, one in each

hemisphere.

16. All animal cries are obviously autoaudible, but here there is no learned motor (supra,

Section 2) model to copy.

17. The articulatory-phonetic patterns are autoaudible, just as the movements of the hand

are autovisible. This similarity ﬁts with the above mentioned homology between the area of

‘mirror neurons’ in a monkey’s brain and Broca’s area.

18. In internalised language, which a child takes time to acquire (see Vygotsky), the linguistic

sounds cease to be autoperceptible. Is the auditive image evoked in this case? I believe that

such an evocation would be self-defeating. All the evocative eﬀort should be concentrated on

the meaning. Besides, as it is a social and not an individual model, that is, phonemic and not

phonetic, it cannot really be heard. Thus, if in internalised language auditive evocations were

to occur (in a sick brain, for example), they could only be of voices with their individual

pitch.

19. All linguistic theories agree on this, not only the old distributionalism or the recent

Latent Semantic Analysis. The syntactical aspect of word meanings was a favourite discussion

topic of some commentators of Pulvermuller (1999). Let me insist on this point. Dor

<LINK"bej-r68"><LINK"bej-r25">

(2000:344): “We do not have access to word meanings via introspection. What we do have

access to are the outputs of word meanings — which enable us to make introspective

judgements about the proper usage of words”. In my view, the usual syntagmatic connec-

tions of the word not only provide these outputs with meaning, but are, strictly speaking, the

meaning or, at least, that part of the meaning which makes it completely linguistic. See also

McGilvray (1998). Finally, note that this conception renders our easy syntactical production

<LINK"bej-r56">

more explainable (cf., e.g., Harris 1998). Not surprisingly, Kintsch, the creator of Latent

<LINK"bej-r39">

Semantic Analysis — a statistical approach to syntagmatic associations — is also the creator

of the concept of ‘long-term working memory’ (pre-activation of elements in order to make

them more easily available).

20. “A cognitive mechanism may be used for many purposes, but only one function is likely

to be critical for driving costly evolutionary change at a given time”. Byrne (2000:556)

<LINK"bej-r9">

reminds us of this evolution theorem.

132 Teresa Bejarano-Fernández

21. Although the ﬂexible, merely goal-guided, behaviour of superior animals meant an

evolutionary step in the past, it was only by means of the partial rejection of that ﬂexibility

(rejection which was required by the imitation of complex and new motor patterns) that the

impressive human ﬂexibility came into being. Byrne’s objections to motor imitation are not

only the result, as I have said, of his overlooking the social and abstract model, but they are

probably a reaction against this apparently bizarre step back in the process too.

22. When an adult laughs at a funny story, the second line of awareness is the mistaken

estimation which he, previous to the ‘punch line’, had made of the character’s belief. Of

course, these expectations of the adult who laughs are not his primary or biological expecta-

tions before his immediate situation (just seeing how erroneous these were could by no

means make him laugh), but the expectations he attributes to the character. These — which,

in my opinion, are channelled through the second line of awareness — are not subject to the

automatic updating and, consequently, remain alive in our attention for a short period of

time. This lapse of time is enough to permit them to collide with perceptions which

contradict them. In short, whereas in the case of the child playing a joke on another, it is the

detected false expectation of the victim which collides with reality, in the case of the adult

hearer of a joke or funny story, it is the expectation falsely attributed to a character which

does so. About stories, see infra, Section 7, and, more concretely, note 35.

23. Or, more precisely, in his non-underemployed predicative sentences. As an example of

underemployed speech, consider Malinowskian phatic speech. Likewise, bearing in mind that

the speaker always aims at oﬀering a certain image of himself or, more precisely, at modify-

ing or elaborating it, not only by referring to himself, but, more frequently, by not overlook-

ing the diﬀerent aspects of these utterances (Amossy 1999), the real communicative purpose

<LINK"bej-r1">

will sometimes be only this. It is clear that I accept that a considerable portion of predicative

utterances do not comply with their most genuine function — which could not be imple-

mented without the predicative mode.

24. A central element in this issue is not yet empirically, neurophysiologically researchable,

namely, if the perceptual systematicity whereby the scene ‘A is hitting B’ is distinguished

from the scene ‘B is hitting A’ can be achieved without the support of an argumental

structure similar to syntax. My opinion, contrary to Fodor or Bickerton (Calvin and

<LINK"bej-r10">

Bickerton 2000: Chap. 10), is that perceptions may have that systematicity without needing

anything similar to syntax or linguistic semantics. In other words: the importance of the

belief ascribed to the hearer can be assumed either in a theoretically soft way or, as I propose,

in a theoretically hard way. In the soft approach, that belief certainly conditions the speaker,

but only in the sense of prompting him to select, out of his box of beliefs (beliefs already

conﬁgured syntactically), the appropriate one for that communicative situation.

25. We have here a true assertion about a false belief. This truth, obviously, is what makes

false beliefs so interesting. See Récanati (2000a:58): The more the explicitness, the less the

<LINK"bej-r69">

relativization.

26. Jerison, a zoologist, and Wilson, a linguist, are only peripherally concerned with the

‘theory of mind’ research. Sperber (2000: 3) comments on this tension: “Much recent work

<LINK"bej-r78">

on metarepresentations is pursued within the framework of cognitive science and of

Metarepresentation and human capacities 133

philosophy of cognition. As such it gives great importance to mental representations of

mental representations. [By contrast,] in rhetoric, semiotic or hermeneutic studies, work was

mostly about the public representation of public representations”.

27. Quite signiﬁcantly, results are less disappointing in the task of explaining post hoc

another’s behaviour after being so requested by the researcher (Robinson and Mitchell 1995).

<LINK"bej-r74">

This request of an adult, this why-question, is the most helpful of clues for a child (McCabe

<LINK"bej-r55">

and Peterson 1997). Turnbull and Carpendale (1999: 347) describe another example in which

<LINK"bej-r85">

adult speech makes it easier for children to accomplish this task. See also Haden et al. (2001).

<LINK"bej-r38">

28. Or, more precisely, cries which perfectly determine the identity of the addressee. I

assume that there is neither vocatives without other noun cases, nor names without other

complementary sentence elements.

29. Vocatives or, alternatively, requests. By contrast, animal cries conﬂate “mum!” and

“care-for-me” or, more precisely, precede this diﬀerentiation. Note that the animal cry is

unable to transmit to the hearer a false belief of the utterer (e.g., about where the mother can

be found). There must be semantic learning which provides that diﬀerentiation. This

primitive semantics would certainly be holophrastic and very diﬀerent from that of true

language, but, I insist, would be suﬃcient to perfectly determine the identity of the individual

or object being referred to.

30. Sabbagh and Callanan (1998:494), exploring “how parent-child interactions might

<LINK"bej-r76">

contribute to children’s understanding of the concept of metarepresentation, i.e., to

understanding of thoughts as representations of reality”, looked “not only at parents’ use of

mental terms, but also at contradictions”. Likewise, I am looking for ‘contradictions’ in a

pre-syntactic, pre-predicative historical stage.

31. With respect to language acquisition, it is evident that there cannot be a close relation

between ontogeny and phylogeny. The linguistic environment of children accounts for their

learning success. But let us put a new question: What about their linguistic failures? These

progressively decreasing failures reveal the intrinsic diﬃculty of some language functions

(the diﬃculty of predicative and interrogative communication). These diﬃculties could be

extrapolated to phylogeny. This possibility is, at least, less disputable than the extrapolation

of the successes.

32. Luria (1981) repeatedly insisted on the importance of this experiment, although I ﬁnd his

<LINK"bej-r51">

comment too vague.

33. If what happens in laughter can be taken as a reliable clue, then the sequence of laughter

modes, as they appear in children, will support the primacy of false alien beliefs over false

own past beliefs. See supra, note 22: false own past beliefs only become funny when the

subject attends to a story. Besides, this clue suggests that own past beliefs would be rescued

more easily when the subject is ‘the monitor’, not ‘the doer’ (cf. studies on collaborative

learning).

34. Apart from the problem known as ‘Redford’s paradox’, Goldie (1999) gives us a lucid

<LINK"bej-r32">

classiﬁcation of the ways in which we discern characters’ emotions.

35. Certainly, this interpretation of ﬁctions must not overlook how often authors, in order

to provide their characters with vigour and substance (or, more exactly, with mind), prompt

134 Teresa Bejarano-Fernández

readers and spectators to perceive their character’s false beliefs. How must we explain this?

The parallel second line, when successively tracking the doings of characters, one after the

other (note that this is exactly the Maxi Test), could even be able to encompass a new

(embedded) double line, either in the parallel or in the colliding mode. cf. the “third or still

higher-order mental states”, as suggested by Dennett (1983) or Perner and Wimmer (1985).

<LINK"bej-r65">

36. Conﬂicts are necessarily included in truly moral decisions (see, e.g., Deigh

<LINK"bej-r20">

1996:216–217). Nil-cost aid, aid which is provided without the mentioned mental eﬀort,

cannot be regarded as within the realm of truly moral decisions.

37. Millar and Tesser (1992: 270): “Individuals may focus on either aﬀective or instrumental

<LINK"bej-r58">

aspects when holding favourable attitudes or evaluative reactions and when performing a

behaviour. Attitudes are expected to be poor predictors of behaviour when attention in the

attitudinal and behavioural contexts was not focused on the same type of considerations”. In

other words, human work needs, in each of its stages, this double attention, if it is to be fully

performed and completed. See also Shapiro and Hudson (1997:28).

38. Ford and Maher (1998:200–201) do not accept that “the process of selectively energizing

<LINK"bej-r28">

thoughts and perceptions relevant to a person’s current goals” be “deﬁned as thinking about

thinking”. In my view, the evocation of selﬁsh goals cannot be mistaken with the colliding

second line. Do these authors refer only to this confusion of evocation and metabelief? If they

do, I agree with them. But I cannot agree if their approach is interpreted in a wider sense.

The goals evoked in speciﬁcally human work are a result of the second line.

39. Against this stress the observation is often made — see, e.g., Stich and Nichols (1995) —

<LINK"bej-r81">

that mental states ascribed to others are not completely identical to those which the subject

would have if he were really in the shoes of the subject he observes. This fact derives, in my

opinion, from the shortage of resources of the second line, when compared with those of the

primary line.

40. This topic, not the role of imitation, is the central issue of the debate between the main

doctrines about how humans achieve their ‘theory of mind’, between simulationism and the

theory-theory. Consider Gopnik and Meltzoﬀ (1997) (see supra, 5.1) and bear in mind that

<LINK"bej-r35">

Gopnik is not a simulationist but a champion of the theory-theory. Simulationists not only

maintain that the subject guesses what will be going on in others’ minds by putting himself

in the other person’s shoes, but also that the ﬁrst-person point of view provides easy and

privileged access to the concept of mental state as such. I, together with many others, think

that this easy access must be understood as providing the subject only with direct knowledge

of the content of his own and current mental states: see Sperber (2000: 120). By contrast, the

<LINK"bej-r78">

self-attribution of these mental states — current beliefs, for example- as such mental states

cannot be interpreted as a necessary, or even frequent operation, and certainly it would be a

much more complex cognitive feat than the mere attribution of mental states to other

individuals.

41. Properly speaking, it would be inappropriate to call it the ‘ﬁrst-person point of view’. It

is incorrect to talk about a ﬁrst-person view until the subject is able to attend to alien

thoughts. This critique of simulationism is related to what social psychologists call ‘funda-

mental attribution error’: We see ourown acts as determined by circumstances, while seeing

Metarepresentation and human capacities 135

the acts of others as stemming from traits of character. Both the concept oftraits of character

and the concept of belief arise similarly in an interpersonal way.

42. Should we playfully explore the hypothesis that “unless” suggests? In creative problem-

solving, the subject has to deal with a situation of total impasse: There is nothing in the

primary line of the subject which makes him question this view. But by means of a second

line he could imagine that an unknown alien mind could have the solution to the problem

(see an incidental commentary in Currie 1995: 162). This conﬁdence in the existence of a

<LINK"bej-r17">

solution — a conﬁdence which, I suggest, only the second line of awareness can provide —

could shake the well-established and perversely “ﬁxed” ﬁrst interpretation of the data. Then,

a radically new phenomenon arises: one’s own current, complete and updated belief is seen

as a mental state (Kuhn 1990; Carpendale and Chandler 1996; see supra, note 40). In other

<LINK"bej-r48"><LINK"bej-r11">

words, it is then that the metacognitive ascent takes place at an intrapersonal level, or, more

exactly — in order to include the ‘currentness’ — at a strictly intrapersonal level. We observe

that conﬁdence in a teacher or an experimenter is certainly decisive when subjects look for

solutions, e.g., to very diﬃcult series: “Type-2 learning is only observed in subjects who have

been explicitly informed that such an abstract structure might exist” (Dominey 1997:71).

<LINK"bej-r23">

Although these are not genuine processes of discovery, the observation is thought-provoking.

I have just read Clark (2001:133) — “ Our mental images seem to be more interpretively

<LINK"bej-r15">

ﬁxed (than perceptions). (In order to achieve a reinterpretation) we must perceptually re-

encounter visual forms” — and I cannot help extending this already long note. I think that

perceptual reencountering is not the usual perceptual reidentiﬁcation. Instead, it must be a

humble attitude which puts one’s own perceptual belief — one’s own ﬁrst line — in

quarantine.

References

Amossy, R. 1999. “L’ethos au carrefour des disciplines: rhétorique, pragmatique, sociologie

<DEST"bej-r1">

des champs”. In R. Amossy (ed), Images de soi dans le discours. Lausanne: Delachaux et

Niestlé, 127–154

Anisfeld, M. 1991. “Neonatal imitation: A review”. Developmental Review 11: 60–97.

<DEST"bej-r2">

Arbib, M. and Érdi, P. 2000. “Précis of ‘Neural Organization: Structure, function and

<DEST"bej-r3">

dynamics’”. Behavioral and Brain Sciences 23: 443–467.

Balleine, B. and Dickinson, A. 1991. “Instrumental performance following reinforcer

<DEST"bej-r4">

devaluation depends upon incentive learning”. The Quarterly Journal of Experimental

Psychology 43B: 279–296.

Baron-Cohen, S. 1999. “The evolution of a theory of mind”. In M.C. Corballis and S. E.G.

<DEST"bej-r5">

Lea (eds), The descent of mind. Oxford: Oxford University Press, 261–277.

Barresi, J and Moore, C. 1996. “Intentional relations and social understanding”. Behavioral

<DEST"bej-r6">

and Brain Sciences 19: 107–154.

Bjorklund, D.F. and Pellegrini, A.D. 2000. “Child development and evolutionary psycholo-

<DEST"bej-r7">

gy”. Child Development 71: 1687–1708.

Bruner, J. and Kalmar, D.A. 1998. “Narrative and metanarrative in the construction of self”.

<DEST"bej-r8">

In Ferrari and Sternberg (eds), 308–331.

136 Teresa Bejarano-Fernández

Byrne, R.W. 2000. “Evolution of primate cognition”. Cognitive Science 24: 543–570.

<DEST"bej-r9">

Byrne, R.W. and Russon, A. E. 1998. “Learning by imitation: A hierarchical approach”.

Behavioral and Brain Sciences 21: 667–721.

Calvin, W. H. and Bickerton, D. 2000. Lingua ex Machina. Cambridge, MA: The MIT Press.

<DEST"bej-r10">

Carpendale, J.I.M. and Chandler, M.J. 1996. “On the distinction between false belief

<DEST"bej-r11">

understanding and subscribing to an interpretive theory of mind”. Child Development

67: 1686–1706.

Carpenter, M., Nagell, K. and Tomasello, M. 1998. “Social cognition, joint attention and

<DEST"bej-r12">

communicative competence from 9 to 15 months of age”. Monographs of the Society for

Research in Child Development 63:1–143.

Chater, N. 1999. “The search for simplicity: A fundamental cognitive principle?”. The

<DEST"bej-r13">

Quarterly Journal of Experimental Psychology 52A: 273–302.

Cheney, D. and Seyfarth, R. 1992. “Précis of ‘How monkeys see the world’”. Behavioral and

<DEST"bej-r14">

Brain Sciences 15: 135–182.

Clark, A. and Thornton, C. 1997. “Trading spaces: Computation, representation, and the

<DEST"bej-r15">

limits of uninformed learning”. Behavioral and Brain Sciences 20: 57–90.

Clark, A. 2001. “Reasons, robots and the extended mind”. Mind and Language 16: 121–145.

Cosmides, L. and Tooby, J. 2000. “Consider the source: The evolution of adaptations for

<DEST"bej-r16">

decoupling and metarepresentations”. In Sperber (ed), 53–115.

Currie, G. 1995. “Imagination and simulation”. In M. Davies and T. Stone (eds), Mental

<DEST"bej-r17">

Simulation. Oxford: Blackwell, 151–169.

Custance, D. M., Whiten, A. and Bard, K. A. 1995. “Can young chimpanzees imitate arbitrary

<DEST"bej-r18">

actions? Hayes and Hayes (1952) revisited”. Behaviour 132: 839–858.

Dascal, M. 1983. Pragmatics and the Philosophy of Mind, vol.1: Thought in Language.

<DEST"bej-r19">

Amsterdam: John Benjamins.

Deigh, J. 1996. “Empathy and universalizability”. In L. May, M. Friedman and A. Clark (eds),

<DEST"bej-r20">

Mind and morals. MA: The MIT Press, 199–219.

Dennet, D. 1983. “Intentional systems in cognitive ethology”. Behavioral and Brain Sciences

<DEST"bej-r21">

6:343–390.

Di Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V., and Rizzolatti, G. 1992. “Understanding

<DEST"bej-r22">

motor events: A neurophysiological study”. Experimental Brain Research 91: 176–180.

Dominey, P. F. 1997. “Reducing problem complexity by analogical transfer”. Behavioral and

<DEST"bej-r23">

Brain Sciences 20: 71–72.

Donald, M. (1991). Origins of Human Mind. Three Stages in the Evolution of Culture and

<DEST"bej-r24">

Cognition. Cambridge, MA: Harvard University Press.

Dor, D. 2000. “From the autonomy of syntax to the autonomy of linguistic semantics”.

<DEST"bej-r25">

Pragmatics & Cognition 8 (2): 325–356.

Duncan, J. 1995. “Selection and control”. In A.D. Baddeley and L. Weiskrantz (eds),

<DEST"bej-r26">

Attention: Selection, Awareness and Control. Oxford: Clarendon Press, 53–71.

Duncan, J., Emslie, H., Williams, P., Johnson, R. and Freer, C. 1996. “Intelligence and the

frontal lobe: The organization of goal-directed behavior”. Cognitive Psychology 30 (3):

257–303.

Ferrari, M. and Sternberg, R. J. (eds).1998. Self-Awareness: Its Nature and Development.New

<DEST"bej-r27">

York: Guilford Press.

Metarepresentation and human capacities 137

Ford, M.E. and Maher, M.A. 1998. “Self-awareness and social intelligence”. In Ferrari and

<DEST"bej-r28">

Sternberg (eds), 191–218.

Gallup, G.G. Jr. 1998. “Self-awareness and the evolution of social intelligence”. Behavioural

<DEST"bej-r29">

Processes 42: 239–247.

Gattis, M., Bekkering, H. and Wohlschläger, A. 1998. “When actions are carved at the

<DEST"bej-r30">

joints”. Behavioral and Brain Sciences 21: 691–692.

Glenberg, A.M. 1997. “What memory is for”. Behavioral and Brain Sciences 20: 1–56.

<DEST"bej-r31">

Goldie, P. 1999. “How we think of others’ emotions”. Mind and Language 14: 394–423.

<DEST"bej-r32">

Goldman, A. I. 1993. “The psychology of folk-psychology”. Behavioral and Brain Sciences 16:

<DEST"bej-r33">

15–28.

Golomb, C. and Kuersten, R. 1996. “On the transition from pretence play to reality: What

<DEST"bej-r34">

are the rules of the game?”. British Journal of Developmental Psychology 14: 203–217.

Gopnik, A, and Astington, J.W. 1988. “Children’s understanding of representational change

<DEST"bej-r35">

and its relation to the understanding of false beliefs and the appearance-reality distinc-

tion”. Child Development 59: 26–37.

Gopnik, A. and Meltzoﬀ, A.N. 1997. Words, Thoughts and Theories. Cambridge, MA: The

MIT Press.

Grice, H.P. 1982. “Meaning revisited”. In N. Smith (ed), Mutual knowledge, London:

<DEST"bej-r36">

Academic Press, 223–243.

Grush, R. 1994. “Motor models as steps to higher cognition”. Behavioral and Brain Sciences

<DEST"bej-r37">

17: 209–210.

Haden, C.A., Ornstein, P. A., Eckerman, C. O. and Didow, S.M. 2001. “Mother-child

<DEST"bej-r38">

conversational interactions as events unfold: Linkages to subsequent remembering”.

Child Development 72: 1016–1031.

Harris, C. L. 1998. “Psycholinguistic studies of entrenchment”. In J-P. Koenig (ed), Discourse

<DEST"bej-r39">

and Cognition. Stanford: CSLI, 55–70.

Hauser, M.D. and Andersson, K. 1994. “Left hemisphere dominance for processing

<DEST"bej-r40">

vocalizations in adult, but not infant rhesus monkeys: Field experiments”. Proceedings

of the National Academy of Sciences 91: 3946–3948.

Heyes, C. 1993. “Imitation, culture and cognition”. Animal Behaviour 46: 999–1010.

<DEST"bej-r41">

Hodos, W. and Butler, A.B. 2001. “Sensory system evolution in vertebrates”. In G. Roth, and

<DEST"bej-r42">

M.F. Wullimann (eds), Brain, Evolution and Cognition. New York: Wiley, 113–134.

Jeannerod, M. 1994. “The representing brain: Neural correlates of motor intention and

<DEST"bej-r43">

imagery”. Behavioral and Brain Sciences 17: 187–246.

Jerison, H. 1988. “Evolutionary neurology and the origin of language as a cognitive adapta-

<DEST"bej-r44">

tion”. In M.E. Landsberg (ed), The Genesis of Language. Berlin: Mouton de Gruyter,

3–9.

Jones, S.S. 1996. “Imitation or exploration?: Young infants’ matching of adults’ oral

<DEST"bej-r45">

gestures”. Child Development 67: 1970–1987.

Karmiloﬀ-Smith, A. and Inhelder, B. 1974. “If you want to get ahead, get a theory”. Cogni-

<DEST"bej-r46">

tion 23: 95–147.

Kintsch, W. 1998. Comprehension. A Paradigm for Cognition. Boulder, CO: Cambridge

<DEST"bej-r47">

University Press.

138 Teresa Bejarano-Fernández

Kuhn, D. 1990. “Education for thinking”. In M. Schwebel, C.A. Maher, N.S. Fagley (eds),

<DEST"bej-r48">

Promoting Cognitive Growth Over the Life-Span. Hillsdale, NJ: Lawrence Erlbaum,

25–45.

Leslie, A. 1988. “Some implications of pretense for mechanisms underlying the child’s theory

<DEST"bej-r49">

of mind”. In J. Astington, P. Harris and D. Olson (eds), Developing Theories of Mind.

Cambridge: Cambridge University Press, 19–46.

Lorenz, K.Z. 1966. Evolution and Modiﬁcation of Behaviour. London: Methuen.

<DEST"bej-r50">

Luria, A.R. 1981. Language and Cognition.NewYork:J.Wiley.

<DEST"bej-r51">

Mackenzie, J. 1985. “No Logic before Friday”. Synthese 63: 329–341.

<DEST"bej-r52">

Marler P. 1991. “The instinct to learn”. In S. Carey and R. Gelman (eds), The Epigenesis of

<DEST"bej-r53">

Mind: Essays on Biology and Cognition. Hillsdale, NJ: Lawrence Erlbaum, 37–66.

Mascolo, M.F. and Fischer, K. W. 1998. “The development of self through the coordination

<DEST"bej-r54">

of component systems”. In Ferrari and Sternberg (eds), 332–384.

McCabe, A. and Peterson, C. 1997. “Meaningful ‘mistakes’: The systematicity of children’s

<DEST"bej-r55">

connectives in narrative discourse and the social origins of this usage about the past”. In

J. Costermans and M. Fayol (eds), Processing Interclausal Relationships. Hillsdale, NJ:

Lawrence Erlbaum.

McGilvray, J. 1998. “Meanings are syntactically individuated and found in the head”. Mind

<DEST"bej-r56">

and Language 13: 225–280.

Meltzoﬀ, A.N. and Moore, M. K. 1983. “Newborn infants imitate adult facial gestures”. Child

<DEST"bej-r57">

Development 54: 702–9.

Millar, M.G. and Tesser, A. 1986. “Eﬀects of aﬀective and cognitive focus on the attitude-

<DEST"bej-r58">

behavior relation”. Journal of Personality and Social Psychology, 51: 270–276.

Millikan, R.G. 1998. “A common structure for concepts of individuals, stuﬀs, and real kinds:

<DEST"bej-r59">

More Mama, more milk, and more mouse”. Behavioral and Brain Sciences 21: 55–100.

Newton, P., Reddy, V. and Bull, R. 2000. “Children’s everyday deception and performance

<DEST"bej-r60">

on false-belief tasks”. British Journal of Developmental Psychology 18: 297–317.

Nichols, S. 2001. “Mindreading and the cognitive architecture underlying altruistic motiva-

<DEST"bej-r61">

tion”. Mind and Language 16, 425–455.

Nielsen, M. and Dissanayake, Ch. 2000. “An investigation of pretend play, mental state terms

<DEST"bej-r62">

and false belief understanding: In search of a metarepresentationallink”. British Journal

of Developmental Psychology 18: 609–624.

Papineau, D. 2001. “The Evolution of Means-End Reasoning”. In D. N. Walsh (ed), Natural-

<DEST"bej-r63">

ism, Evolution and Mind. Supplement to Philosophy, 99:145–178.

Parker, S. T. 1998. “A social selection model for the evolution and adaptive signiﬁcance of

<DEST"bej-r64">

self-conscious emotions”. In Ferrari and Sternberg (eds), 108–134.

Perner, J. 1991. Understanding the Representational Mind. Cambridge, MA: The MIT Press.

<DEST"bej-r65">

Perner, J. and Wimmer, H. 1985. “‘John thinks that Mary thinks that…’. Attribution of

second-order beliefs by 5- to 10-year old children”. Journal of Experimental Child

Psychology 39:437–471.

Piaget, J. 1959. La formation du symbole chez l´ enfant. Neuchâtel: Delachaux & Niestlé.

<DEST"bej-r66">

Povinelli, D. J., Bering, J. M. and Giambrone, S. 2000. “Toward a science of other minds:

<DEST"bej-r67">

Escaping the argument by analogy”. Cognitive Science 24: 509–541.

Povinelli, D. J. and Prince, C.G. 1998. “When self met other”. In Ferrari and Sternberg (eds),

37–107.

Metarepresentation and human capacities 139

Pulvermuller, F. 1999. “Words in brain’s language”. Behavioral and Brain Science 22:

<DEST"bej-r68">

253–336.

Récanati, F. 2000a. Oratio obliqua, oratio recta. An Essay on Metarepresentation. Cambridge,

<DEST"bej-r69">

MA: The MIT Press.

Récanati, F. 2000b. “The iconicity of metarepresentations”. In Sperber (ed), 311–360.

Reddy, V. 1991. “Playing with others’ expectations: Teasing and mucking about in the ﬁrst

<DEST"bej-r70">

year”. In A. Whiten, (ed), Natural Theories of Mind. Oxford: Blackwell, 143–158.

Rieser, J., Garing, A.E. and Young, M. F. 1994. “Imagery, action and young children’s spatial

<DEST"bej-r71">

orientation”. Child Development 65: 1262–1278.

Risjord, M. 1996. “Meaning, belief and language acquisition”. Philosophical Psychology 9:

<DEST"bej-r72">

465–475.

Rizzolatti, G. Fadiga, L. Gallese, V. and Fogassi, L. 1996. “Premotor cortex and the recogni-

<DEST"bej-r73">

tion of motor actions”. Cognitive Brain Research 3: 131–141.

Robinson, E.J. and Mitchell, P. 1995. “Masking of children’s early understanding of the

<DEST"bej-r74">

representational mind: Backwards explanation versus prediction”. Child Development

66: 1022–1039.

Rochat, P. 2001. “Dialogical nature of cognition”. Monographs of the Society for Research in

<DEST"bej-r75">

Child Development 66: 133–143.

Sabbagh, M. and Callanan, M. 1998. “Metarepresentation in action: 3-, 4-, and 5-year-olds’

<DEST"bej-r76">

developing theories of mind in parent-child conversations”. Developmental Psychology

34: 491–502.

Searle, J.R. 2001. Rationality in Action. Cambridge, MA:The MIT Press.

<DEST"bej-r77">

Sperber, D. 2000. “Metarepresentations in an evolutionary perspective”. In Sperber (ed),

<DEST"bej-r78">

117–137.

Sperber, D. (ed). 2000. Metarepresentations. Oxford: Oxford University Press.

Sterelny, K. 1998. “Intentional agency and the metarepresentation hypothesis”. Mind and

<DEST"bej-r79">

Language 13: 11–28.

Stern, D.N. 2001. “Face-to-face play”. Monographs of the Society for Research in Child

<DEST"bej-r80">

Development 66: 144–149.

Stich, S. and Nichols, S. 1995. “Folk psychology: Simulation or tacit theory?”. In M. Davies

<DEST"bej-r81">

and T. Stone (eds), Folk Psychology. Oxford: Blackwell, 123–158.

Suddendorf, T. 1999. “The rise of the metamind”. In M. C. Corballis and S. E.G. Lea (eds),

<DEST"bej-r82">

The Descent of Mind. Oxford: Oxford University Press, 218–260.

Taylor, C. 1999. “Sympathy”. The Journal of Ethics 3: 73–87.

<DEST"bej-r83">

Tomasello, M. 2000. “Primate cognition: Introduction to the issue”. Cognitive Science 24:

<DEST"bej-r84">

351–361.

Tomasello, M., Kruger, A.C. and Ratner, H. H. 1993. “Cultural learning”. Behavioral and

Brain Sciences 16: 495–552.

Turnbull, W. and Carpendale, J.I. M. 1999. “A social pragmatic model of talk: Implications

<DEST"bej-r85">

for research on the development of children’s social understanding”. Human Develop-

ment 42: 328–355.

Visalberghi, E. 1993. “Capuchin monkeys: A window into tool use”. In K.R. Gibson and T.

<DEST"bej-r86">

Ingold (eds), Tools, Language and Cognition in Human Evolution. Cambridge: Cam-

bridge University Press, 138–150.

140 Teresa Bejarano-Fernández

de Waal, F.B. M. 1998. “No imitation without identiﬁcation”. Behavioral and Brain Sciences

<DEST"bej-r87">

21: 689.

Wegerif, R. and Mercer, N. 2000. “Language for thinking: A study of children solving

<DEST"bej-r88">

reasoning test problems together”. In H. Couvie and G. van der Aalsvoort (eds), Social

Interaction in Learning and Instruction. Oxford: Pergamon, 179–192.

Whiten, A. 2000a. “Chimpanzee cognition and the question of mental re-representation”. In

<DEST"bej-r89">

Sperber (ed), 139–167.

Whiten, A. 2000b. “Primate culture and social learning”. Cognitive Science 24: 477–508.

Whiten, A. and Ham, R. 1992. “On the nature and evolution of imitation in the animal

kingdom: Reappraisal of a century of research”. In P. J. B. Slater, J. S. Rosenblatt, C. Beer

and M. Milinski (eds), Advances in the Study of Behavior. San Diego: Academic Press,

239–283.

Wilson, D. 2000. “Metarepresentation in linguistic communication”. In Sperber (ed),

<DEST"bej-r90">

411–448.

Wimmer, H. and Perner, J. 1983. “Beliefs about beliefs: Representation and constraining

<DEST"bej-r91">

function of wrong beliefs in young children´s understanding of deception”. Cognition

13: 103–28.

Zentall, T. R., Sutton, J.E. and Sherburne, L. M. 1996. “True imitative learning in pigeons”.

<DEST"bej-r92">

Psychological Science 7: 343–346.

Author’s address

Teresa Bejarano Fernández

Salto de Alvarado nº 1 (antiguo 66)

41007 Sevilla

Spain

Email: tebefer@us.es

About the author

Teresa Bejarano teaches Philosophy of Language at the University of Sevilla (Spain). Her

research has mainly dealt with the relationship between pragmatics and cognition and with

the origin of the ‘theory of mind’. Among other topics, she has published articles on Frege’s

puzzle of identity, laughter and interpersonality, and ﬁctionally triggered emotions.

</TARGET"bej"></TARGET"bej">

¿Por qué es placentera y perceptible la risa?

Article

Full-text available

May 2011

Teresa Bejarano

Why is laughter pleasant? Since all laughter attends to an already frustrated expectation, when a child laughs, he or she would be putting into practice a pre-training of a crucial ability-that of thinking contents which the child knows to be merely mental-. In fact, laughter is ideal for that exercise, as (I intend to prove) it involves just those merely mental contents which are the easiest to grasp-others' expectations-. So laughter, like any other pleasure, guides the child towards behaviours which are somewhat useful. Why is laughter perceptible? Of course, as it is well-known, shared laughter reinforces the links within the group, but also and more specifically, would exercise the same type of ability which leads to the grasping of such mental contents which constitute shared social rules. In the Third Section, I put these proposals in a general, anthropological frame. Introducción ¿Por qué es placentera la risa? Si el concepto de placer lo colocamos dentro del marco de la evolución y de las adaptaciones ventajosas, entonces se nos aparece como un 'mecanismo enseñante' que guía al organismo hacia los estímulos y las conductas apropiadas para su éxito biológico. El placer puede señalar tanto conductas que por sí mismas ya son intrínseca e inmediatamente ventajosas como igualmente otras conductas, en sí mismas inútiles, pero que servirían para ejercitar y potenciar destrezas extremadamente ventajosas. Este último tipo, que es el ligado al juego, podría ser también el que está envuelto en la risa. Así que reformulando la pregunta obtenemos: ¿Qué habilidades se estarían ejercitando en la situación placentera que supone la risa? Tras distinguir (en 1. 1) tres tipos de comicidad-situación cómica, broma y chiste-, nos preguntamos cómo ha de ser una expectativa frustrada para que

Mentiras, relevancia y teoría de la mente

Article

Full-text available

Nov 2010

Victoria Camacho-Taboada

Mentir intencionadamente es una de las características más salientes de nuestra especie. En este artículo discutimos cuáles son las herramientas cognitivas necesarias para el arte de mentir y manipular al otro. Concretamente argumentaremos que, además del módulo lingüístico, es necesaria la capacidad de atribuir estados mentales al otro. Dicha capacidad sería el resultado de un módulo cognitivo complejo en el que están incluidos, entre otras habilidades, el principio de relevancia (Sperber y Wilson 1986/1995), la habilidad de asignar estados mentales a otros recursivamente (Perner y Wimmer, 1985) y el mecanismo de detección de tramposos (Cosmides, 1989; Tobby y Cosmides, 1989, 1992, 2000).

Types of technology

Article

Jun 2010
TECHNOL FORECAST SOC

Robert Aunger

Technology is a concept rife with confusion. Here, I argue that technologies can be distinguished as a combination of type of producer and an idealized artefact life history. Using this definition, a number of technologies are identified. The first technology historically, in the Protostomes, was the production of individual or family dwellings. Next came objects such as spider webs for trapping prey. Stigmergy followed, with the social insects, as a collective endeavour to construct a mega-structure using simple rules of accretion. Some birds and primates began to make tools, or simple technological objects whose function is closely related to their form. Humans are distinguished by their ability to make machines. Traditional technology took place once people voluntarily organised into groups with specialised knowledge to produce more complex objects and structures. Monumental objects like ceremonial pyramids came with the command economies of the early agrarian societies, which also resulted in a new category of artefact, the network. Finally, with modern civilizations came ad hoc accretion, or population-level adding-on, to make truly complex technological systems. Developing a theoretical framework within which artefacts, production processes and ways of interacting with them are identified should help the study of technology to become more scientific.

Language for thinking: A study of children solving reasoning test problems together

Article

Full-text available

Jan 2000

Meaningful "mistakes": The systematicity of children's connectives in narrative discourse and the social origins of this usage about the past

Article

Full-text available

Jan 1997

A chapter in J. Costermans & M. Fayol (Eds.), Processing interclausal relationships: Studies in the production and comprehension of text. (pp. 139-154). Erlbaum, 1997

Pragmatics and the Philosophy of Mind: Vol. I: Thought in Language

Book

Jan 1983

Marcelo Dascal

This volume deals with the relation between pragmatics and the philosophy of mind. Unlike most of the books written on the subject, it does not defend the view that a specific form of dependence holds between language and thought, to the exclusion of all other possible relations. Taking pragmatics in its original sense of “that part of semiotics that is concerned with the users of a semiotic system”, the book analyses the nature of the mental processes and states mirrored in language use. Drawing on results from cognitive psychology, the philosophy of mind, the philosophy of language, linguistics, etc., a unified view of the mental dimension in the use of language, both as an instrument of communication and as an instrument of thought, is offered. After offering a tour d’horizon of the relationship between language and mind, this volume deals with the way thought is manifested in language.

Origins of the Modern Mind: Three Stages in the Evolution of Culture and Cognition

Article