Beyond the body: towards a full embodied semiosis
Patrizia Violi
Distributed intersubjective semiosis: ontogenesis, consciousness and semiosis
Patrizia Violi
Department of Communication
University of Bologna, Italy
1. Do we really need consciousness for semiosis? A background sketch
The relationship between consciousness and semiosis has been a constant object of inquiry for semiotics, underlying, often implicitly, the very definition of what a sign is. Indeed, the basic distinction between intentional and non-intentional signs that has characterised the whole history of the discipline since its foundations, presupposes a reference to the idea of consciousness as intentional agency, intentionality becoming, thus, a crucial dimension in the definition of sign typologies.
Even more intriguing is the issue of the relevance of consciousness in relation to semiotic activity in itself. If we are all prepared to agree that we clearly can have signs that are not consciously produced, then can we then speak of genuine processes of semiosis outside of conscious organisms? Does semiosis blend or merge completely with consciousness, or do these two concepts actually refer to at least partially separate domains? Obviously, the answer to this question largely depends on what we mean by the two terms in question and how we decide to define both ‘consciousness’ and ‘semiosis’.
Consciousness is a typical umbrella term that is used to cover a wide range of different concepts, from phenomenal awareness and the distinction between self and non-self, to high levels of reflexive consciousness and the notion of first-person perspective. I shall not begin to go into these complex issues here, but rather concentrate on the other term in our problematic couple, i.e. ‘semiosis’.
One way to do this is to look at the way in which the discipline that studies process of semiosis, i.e. semiotics, defines its object. Clearly, depending on the way in which we decide to define this particular object, we will arrive at very different answers to our principle question regarding consciousness itself.
Semiotics, too, does not have any easy definition of itself: already in 1979 Eco began addressing the issue of whether semiotics ought to be taken as a field or a discipline. Whatever option we decide to settle for, we will have to confront ourselves with the dualistic theoretical foundations of the discipline, rooted as it is on the one hand in the early structuralist approaches of Saussure and Hjelmslev, and on the other in the philosophical pragmatism of Peirce. These two foundational stances differ deeply in respect to our problems in hand, since they construct their reciprocal objects in different ways.
Saussure, and Hjelmslev after him, never refer to the word semiosis, since their attention is not so much in the processes that produce semiotic phenomena, but on the systemic forms that emerge from these processes over time, i.e. structure, and first and foremost, linguistic structure.
In Saussure’s definition of semiotics (sémiologie in his own words) it is quite clear that the realm of semiosis coincides with human beings and their social life, After having described the linguistic system separating langue as structure from langage - the heterogeneous set of the phenomena involved in linguistic activity - he hypotheses a new science to come:
La langue est un système de signes exprimant des idées et par là comparable à l’écriture, à l’alphabet des sourds-muets, aux rites symboliques, aux formes de politesse, aux signaux militaries, etc. Elle est sulement le plus importants de ces systèmes. On peut donc concevoir une science qui étudie la vie des signes au sein de la vie sociale : elle formerait une partie de la psychologie sociale, et par conséquent de la psychologie générale ; nous la nommerons sémiologie. Elle nous apprendra en quoi consistent les signes, quelles lois les régissent. (Saussure 1916: 25)
Although Saussure does not mention the word “consciousness”, his notion of the sign presupposes this, since the sign is a dual entity, comprising an acoustic signifier and a mental signified. As Eco has pointed out:
the signified is something which has to do with the mental activity of anybody receiving a signifier: according to Saussure signs ‘express’ ideas (…) and ideas must be mental events that concern a human mind. Thus the sign is implicitly regarded as a communicative device taking place between two human beings intentionally aiming to communicate or to express something” (Eco 1979: 15) (my emphasis)
It is quite clear that in this approach semiosis coincides with a highly complex intentional system, and it is basically restrained to human actors. The linguistic sign is the model for any other sign; before it emerges, we cannot have any form of semiosis, only an “amorphous mass” that needs to be formed. In this way, albeit implicitly, consciousness becomes an indispensable precondition for semiosis. Hjelmslev (1943) extended this framework and developed the idea of a continuum of semiotically undifferentiated matter that must remain beyond the realm of semiotic inquiry.
As Eco has pointed out, those who share Saussure’s (and Hjelmslev’s, I would add) notion of semiotics “distinguish sharply between intentional, artificial devices (which they call “signs”) and other natural or unintentional manifestations which do not, strictly speaking, deserve such a name ” (ibid).
A quite different picture can be seen within the peircean paradigm, where semiotics is “the doctrine of the essential nature and fundamental varieties of possible semiosis” (C.P. 5.488) and semiosis in its turn is “an action, an influence, which is, or involves, a cooperation of three subjects, such as a sign, its object and its interpretants” (5.484)
According to Eco it is possible to interpret Peirce’s definition in a non-anthropomorphic way: “It does not demand, as part of a sign’s definition, the qualities of being intentionally emitted and artificially produced” (Eco 1979: 15)
If it is not necessary for semiosis to have a human emitter, it is nonetheless necessary, for Eco, to have a human receiver. Precisely on this feature is based Eco’s notion of sign, that recalls the one proposed by Morris (1938): “I propose to define as a sign everything that, on the grounds of a previously established social convention, can be taken as something standing for something else. (…) Semiotics, then, is concerned with ordinary objects insofar (and only insofar) as they participate in semiosis” (Eco 1979: 16, mine emphasis).
For the Eco of A Theory of Semiotics, consciousness - taken as intentional forms of consciousness -appears to be crucial for establishing limits for semiotic activity, which can be clearly seen in his discussion of the notion of unintentional signs. According to Eco, natural and unintentional signs, such as medical symptoms, only become signs when they initiate inferential activity or, in other words, at the moment in which they become interpreted as such by a human agent. Consciousness is thus required, so to speak, only at the tail end of the semiotic chain – i.e. on the part of the interpreter – but it does not need to characterize the production of something that may become a sign only after a “conscious” process of interpretation. Signs in themselves may be unintentionally - or unconsciously - produced, but the processes through which we attribute meaning to these signs are not.
In the same vein, gestures and non-verbal behaviour in interaction become significant only when interpreted by an addressee who decides to recognize them as endowed with meaning. But certainly gestures, as executed by human beings, although often non-intentional, are always part of a complex semiotic behaviour by highly complex organisms that are also conscious. As a result, gestures may be endowed with underlying significant, but nonetheless unconscious, intention. Intentionality does not always have to be conscious in other words, and we always have some aspects of unconscious intentionality in our interactions with others.
As soon as we move towards threshold zones where even a “low” concept of consciousness seems not to apply, this question becomes even more clear cut. Natural phenomena such as the growth of plants, or the turning of sunflowers to follow the sun, to use one of Peirce’s famous examples, are not in themselves semiotic processes according to Eco, who explicitly claims that “It is incorrect to say that every act of inference is a ‘semiosic’ act – even though Peirce did so” (Eco 1979: 17)
Only the discourses that we can enter into regarding natural phenomena and our interpretation of the natural habits these display is a semiotic activity.
2. The question of the origins: the lower threshold
In this vein, Eco sets a number of minimal conditions needed in order to define a semiotic phenomenon, thus fixing the lower threshold in terms of the distinction between physical stimuli and signs.
Since everything can be understood as a sign if and only if there exists a convention which allows it to stand for something else, and since some behavioural responses are not elicited by convention, stimuli cannot be regarded as signs (Eco 1979: 19)
The crucial notion is here one of convention, but it seems hard to imagine a convention without a conscious decision; therefore the lower threshold coincides with the presence of consciousness.
The question of the lower threshold is addressed again by Eco a few years later, on the occasion of a meeting with a group of immunologists (Eco 1988). Immunologists often refer to the immune system as a system endowed with an internal system of communication, where cells communicate with one another and share information. Now the question was: is this only a metaphor, or can we assume that some basic and proto-semiotic functioning is at stake here? In 1988 Eco further elaborated his 1979 position, but did not radically modify it. He maintained the distinction between a stimulus-response process, which is a dyadic one and implies a casual deterministic sequence, and a triadic model where interspersed between the two elements A and B it is possible to envision a third element C, “call it the code, or the process of interpretation implemented through the recourse to the code” (Eco 1988: 9). A semiosic process always implies a C-space, since it is always a triadic process, where C-space is the space of an interval, the space of a possible degree of freedom,
Therefore the conditions for semiosis coincide with the existence of an interval, a temporal gap between stimulus and its response. This implies at least the following: i) stimulus and response do not have to be co-present; ii) there must be a certain degree of freedom in responses; iii) as a consequence of ii) there must be some possibility of choice among different possible responses; iv) responses are not totally predictable; v) context acquires central relevance, since it will contribute to determining the most appropriate, or best, responses. In this way C-space marks the passage from a dyadic to a triadic model, where abductive, hypothetical forms of reasoning can take place.
Although in 1988 Eco seems to allow for the possibility of semiosis without consciousness, for example in the case of sophisticated Artificial Intelligence systems, the relation between consciousness and semiosis still appears crucial. Indeed, since Eco equates semiosis with abductive reasoning, where the latter refers to the capacity to “take difficult decisions when following ambiguous instructions” it is difficult to imagine a total lack of at least some minimal level of consciousness. The turning of sunflowers to follow the sun, to refer once more to Pierce’s famous example, would be equally well ruled out as a phenomenon of semiosis both in Eco (1988) and Eco (1979).
The C-space model sketched out by Eco (1988) has some peculiar features that deserve further analysis. First of all, it is a strongly cognitively oriented space: semiosis is seen as coincident with inferencing processes and abductive reasoning; focus is decisively on cognition, while emotions and affect seem to play no role whatsoever in the initial appearance of semiosis, and the same goes for perception and bodily sensations.
Second, and even more important, the appearance of semiosic processes is envisioned only within the single organism: C-space is the space between stimulus and response that emerges internally, within the single individual taken as a separate whole. We could say that this is a predominantly intra-subjective perspective, rather than an inter-subjective one. I would call it an internalist position, where the origin of semiosis coincides with the existence of an internal space of decision-making and degrees of freedom: on the capacity of the single organism to draw inferences about its environment and to respond in non-deterministic ways to stimuli deriving from external agents. All this seems quite a good approximation of conscious behaviour.
In what follows I will examine the relationship between consciousness and semiosis from a quite specific angle, investigating the beginning of semiotic life in the developmental, ontogenetic dimension.
3. An ontogenetic perspective
There is a very good reason that underlies my choice here: ontogenesis is perhaps the only situation in which we can witness in a direct, verifiable way, and not only speculatively, the simultaneous growth of both semiosis and consciousness. In the development of newborn and infant children, processes of “semioticization” cannot really be separated from processes of “subjectivisation” and the emergence of consciousness. In a way, one could say that to become a subject is to become a semiotic subject. Semiosis and consciousness are, in ontogenetic development, inextricably linked to one another.
But how exactly does the process of becoming a competent “semiotic subject” occur? Through which stages, under which constraints, and by way of which modalities does a newborn infant become a semiotic being?
In the last 30 years a consistent body of research has been developed in this domain using innovative methodologies and achieving some novel and important results. This vast research project is known as Infant Research (I.R. from now on) and has, since the 70’s, seen cooperation between developmental psychologists, psychoanalysts, paediatricians, neurophysiologists to investigate early stages of the life of newborn infants. The methodological and theoretical presuppositions of I.R. are partly different from those of the traditional psychoanalytic approach which is mainly based on speculative assumptions and theoretical hypothesis development. The novelty of I.R. is certainly its empirical methodology which is based both on experimental data and direct observation on the basis of video materials.
The picture that emerges from the impressive data corpus collected within the paradigm of I.R. appears to be very different from the one depicted by classical psychoanalytical approaches, where the infant is seen as autistically closed in a form of primary narcissism (Freud 1914) or fusionally and symbiotically united with the mother (Mahler 1968, 1975).
On the contrary, newborn infants appear to be endowed with a highly specialised capacity for making distinctions and discriminating among a wide range of domains that pertain to different perceptual and sensory sources - the visual, the auditory and even the affective domains. A few days, if not hours, after birth babies are already able to distinguish the smell of their mother’s milk from that of another woman; they can recognize human voices from other sounds that are not voices, and, among different human female voices of the same age speaking the same language, they are able to distinguish the voice of their own mother. Very early the visual motor apparatus of the newborn infant is already developed, and it has already functioning systems of reflexes that control ocular movements that allow it to fixate on objects in space.
Of a particular interest are experiments showing the high capacities of imitation in newborns. As early as a few hours after birth, newborn infants are able to exactly reproduce facial gestures performed by an adult, like protruding of the lips or tongue and opening the mouth, certainly without having seen them before and even without having seen a human face before.
Even more interesting are data concerning the precocious intersubjective competence of neonates, and their capacity to respond to external stimuli and interact in a meaningful way with the environment, as shown in the seminal experiment by Tronick et al. (1978) known as the “still face” experiment. During the course of a normal playful interaction between infant and mother (or father), the adult is asked to stop the interaction and maintain a still face without any particular expression for a few minutes (generally two minutes), altering in this way the expected sequence of actions. At the age of three months, and even earlier, children react in a highly complex and interactive way to this experience. First they generally look disoriented and show a certain level of annoyance, then they try to involve the absent mother by touching her face, smiling, and trying to force an interactive move on her side. When they do not succeed, there is a third phase, which can involve retiring into themselves or crying. The sequence may vary, and there are individual variations among children, but all exhibit a strong interactive competence that is far from the autistic narcissism imagined by Freud. This experiment also cast doubt on the idea of a totally symbiotic fusion with the mother, since newborns appear to be quite aware of the otherness of the mother, and of changes in her behaviour very early on indeed.
So what can we infer from all this evidence? Certainly we are witnessing highly complex forms of semiotic behaviour: babies are much more competent in interacting with their environment and with relevant human figures around them than it was ever supposed before. Their capacity to respond in appropriate ways to external stimuli and to engage in appropriate intersubjective exchanges is striking, and microanalysis using new systems for video-recording and analysis have made it possible to observe this in detail. Stern uses the term “interactive dance” to describe complex interactive behaviour of this kind, which is characterized by the use of highly complex semiotic synchronization devices on both sides of the dance floor.
The crucial question, however, is whether we can also deduce, on the basis of an existent semiotic competence in the behaviour of early children, the existence of high-level consciousness in their mental apparatus.
This question is quite open and has not had, until now, a definite, clear answer, since it is very difficult to infer from specific behaviours the existence of any given mental state. In a way, in the case of neonates we run the same risk that we face in zoo-semiotics while analysing animal behaviour: we risk projecting our own human categories and interpretative schema onto an otherwise different behaviour. If, with animals, the risk is anthropomorphizing them, in the case of neonates we could risk “adultomorphizing” them, but essentially the question is the same. As Nagel (1979) pointed out, the only evidence of subjectivity and consciousness of a given system is being that system yourself, and, I would add, expressing that experience in words. Since this is not possible for either animals or neonates, we are left with no more than mere speculation.
In the field of Infant Research there is not one single definitive answer to this question, although there is common agreement on the early competencies of newborns. Trevarthen (1993) claims for example that newborns’ activities are not reflex actions, and he assumes from birth a certain level of consciousness, while Stern seems to me more cautious on this specific point: the sense of Self for Stern is a stratification of progressively more complex functions extending over the first two years of life.
A crucial point for our discussion here is the neural development of the brain in newborns; at birth the brain is not completely developed, and it undergoes profound transformations during the first year of life. It is far from clear whether the brain of a newborn is already endowed with the necessary functional properties that would make it possible to refer to genuine consciousness. Lacking the possibility of obtaining verbal reports of the state of neonate consciousness, it becomes extremely important to understand whether the brain is in a condition to allow conscious experience. If this is not the case, we can not speak of consciousness in newborns, and alternative explanations must be found to account for their behaviour, starting with their precocious and amazing imitative capacities. Recent research on mirror neurons (Rizzolatti et al. 1996) could for example explain the capacity of newborns to imitate facial expressions, without the need to postulate consciousness. Imitation would simply be a neural semiautomatic response to a given stimulus, due to the existence of some highly specialised neurons – the so-called mirror neurons. These neurons continue to be operative in adult brains too, with only one relevant difference: adults, due to cultural and environmental conditioning, have inhibited their immediate imitative response, while newborns’ behaviour is still characterised by automatic imitation.
However, a description couched purely in terms of mirror neurons does not explain the variety and richness of neonate infant behaviour, and their non-deterministic responses to one and the same stimulus, as Tronick’s experiment has shown. The richness and variety of their responses makes difficult to think in terms of purely stimulus-response patterns, and certainly meets one of the conditions indicated by Eco for semiosis, i.e. a certain degree of freedom in response to a stimulus and a non total predictability of behaviour.
We are facing here an apparent paradox, because on the one hand we can see that neonates are clearly engaged in semiotic behaviours, able to master interactive situations with a high degree of competence, on the other hand there are good reasons to doubt their being endowed with full consciousness, since their brains do not have yet the functional properties necessary for this. Certainly, when at age three months they react to the still face experiment in the complex ways that Tronick has demonstrated, they do not have representations in the classical sense of the term: mental representations that internally reproduce external states of affairs. In semiotic terms we could say that they still do not possess a type on the basis of which they can confront what they see. How can they then manage to recognize anything?
One possible way out of such a paradox is to abandon the classical view of representations, and start investigating different, alternative ideas of representation and mind.
4. The semiotic field
Discussing earlier the notion of C-space as described by Eco, I used the word “internalism” to denote an exclusive focus on processes occurring within each single organism. In such a framework semiosis coincides with the appearance of a space for mediation and distance – which is precisely what C-space is – within the specific organism we are considering, a space that is very similar to consciousness itself.
My suggestion is to now move from an internalist perspective to what could be defined as an externalist perspective. Instead of investigating the inner world of the organism where semiotic processes are supposed to occur, let us turn to the external field where interaction between the individual, the environment and other involved subjects takes place. The methodological advantages of such a move are quite evident: we do not have to worry any longer about the existence of an inner - conscious - state that would be responsible for the semiotic capacity, and whose empirical existence, at that stage, could only be a matter for speculation. Instead, we shall attempt to describe the field in which semiotic events occur and display their sense-making potential; I call this field the semiotic field.
It is the semiotic field in itself that now becomes the object of semiotic inquiry, a portion of the actual world that we are able to isolate and describe with semiotic tools and methodologies. Semiosis starts there, in the complex interactions between body, environment and other organisms; consciousness, if we still want to maintain this notion, is, so to speak, a consciousness that extends to the semiotic field as a whole, rather than merely being an internal state of a single organism.
The crucial question then becomes the following: how can the semiosis and consciousness of the semiotic field develop? To better understand this point let’s consider the imitative phenomena depicted by Meltzoff and his colleagues in their experiments which are so relevant in discussing early semiotic competence of newborns. We can easily accept an initial description in term of a reflex action sequence, based on the existence of specialized forms of neural activity, such as mirror neurons. In this perspective, the imitative behaviour of newborn infants does not require any semiotic consciousness, and can well be the effect of an automatic neural response to some given stimulus. The neonate infant does not have to “understand” the facial gesture or to have any internal representation of it; her behaviour might well be an instinctive answer to an environmental stimulus that probably developed on an evolutionary basis. We can easily imagine that most of the behaviour of newborns depends on the need to ensure themselves necessary attention and care on the part of adults caregivers, and imitative behaviour may well be a device functioning in that direction.
What is at stake in an externalistic perspective is not so much what happens in the mind (or the brain) of the neonate infant, but rather how the semiotic field facilitates the emergence of semiosis and sense on the basis of her behaviour. How, in other words, a – possibly – automatic response is able to become part of a highly complex semiosic activity. In order to understand this, we have to shift attention from the mental apparatus of the infant to the semiotic field as an integrated whole, and take into account not only the single organism (the neonate infant in our case) but also the complex interactive system of the infant and the mother, their reciprocal influences and the system of activities, both behavioural and mental, that emerge out of the interaction of their respective behaviours in the field.
All neonate behaviour, gestures, movements and expressions – independently of their being conscious behaviours or merely reflex actions – are “received” by the mother, interpreted in some way and “responded” to. In this constant “making sense” of what the infant does, which is a basic component of taking care of an infant, there is certainly a projection of intentional schema that might very well belong more to the adult’s world than that of the infant. As Stern noticed, it is impossible to have interactions with a very young children without attributing to them human qualities, intentions, motivations. Adults treat children as if they were already what they will later become. The point is that is precisely such a projection of adult expectations that contributes to their becoming intentional conscious beings, and this projection process is therefore a relevant part of the semiotization process.
Think of all the rich verbal descriptions that mothers perform while dealing with their newborn infants much before these infants can understand them. Much of this talk represent a continuous activity of framing (Goffman 1974) of what is going on, which is at the same time a description, an interpretation and also offers emotional and pathemic support for what the infant is doing. But it is certainly in the corporeal modality of non-verbal communication that the mother interacts most with the infant. Stern has described in very precise ways the complex phenomenon of synchronization that regulates the interactive “dance” of mother and child, that he calls attunement.
In this way some behaviours are reinforced and tend to get repeated over time, others are discouraged and over a certain period of time will tend to become less frequent. The first imitative smile of a new born might very well be a reflex action due to a given mirror neuron, but when it receives a warm answer by the mother who in her turn smiles back at her infant producing a reinforcing effect, a complex semiotic cycle starts to take place. The infant will tend to repeat the same behaviour that turned out to be so “successful” for him or her in terms of environmental response in that it received such a positive and emotionally adequate response. Successful behaviour will tend to be repeated over and over and became a regular form of (inter)action.
Now, the sequence I just described is precisely what Peirce called the creation of a habit. According to Peirce, the habit is a tendency to an action that stabilizes the semiotic process; it is a form of emergent regularity on which the whole semiosic process is centered.
Multiply reiterated behaviour of the same kind, under similar combinations of percepts and fancies, produces a tendency – the habit – actually to behave in a similar way under similar circumstances in the future. (C.P. 1.560-62)
The field of reciprocal mother-infant interaction that Stern and Infant Research have so effectively described is at the same time the intersubjective space where the Self of the child is given a possibility of constituting itself, and it is also where semiosis first takes place: these two processes are indeed two aspects of one and the same process.
In this vein, we can think of the mother’s affective responsive to the neonate infant’s behaviour as an interpretant of that behaviour, i.e. a more developed sign that is created on the basis of the effect on the organism of the mother of the first sign. In the first place we have an emotional interpretant, but emotions in their turn are embodied in gestures, movements, expressions that represent a correspondent variety of pragmatic interpretants. Emotions, feelings and thoughts of the mother, together with her corporeal interactions with her infant child - Stern’s interactive dance - can be seen as a complex system of emotional and pragmatic interpretants that contribute to the creation of habits on the part of the infant, stabilizing his or her behaviour and interactions with the environment in a dynamic cycle where interpretants and habits of both mother and child reciprocally interact with and influence each other.
This is the semiotic field of the interaction, and it is precisely within this field that semiosis first emerges. The emotional, mental, bodily, gestural and linguistic responses of the mother to the child are all ways of giving sense to and making meaningful actions that otherwise might well be seen as no more than simple reflex actions. But if each single action might well be seen as a reflex, the whole field can not. It is important to emphasize that meaning is not in what the child does in itself, nor exclusively in the interpretative activity of the mother; meaning and sense are created in the relational field between the two; indeed, they are this field, which is precisely the semiotic field. In a more general way we could say that semiosis is neither in the object alone, nor in the subject alone, but in the interactive coupling of the two, semiosis is the result of coordinated forms of action – of doing together.
Using Eco’s notion of C-space, we could say that it is within the semiotic field, and not so much in the infant’s mind that we can find the crucial space that is a qualifying condition for the very existence of semiosis. The semiotic field is not a deterministic environment, but rather a domain of degrees of freedom, since quite different habits may develop out of the same initial reflex actions, depending on the way in which the mother-infant interaction process develops over time. The different responses of the mother to what is, initially, a possibly automatic behaviour on the infant’s side, can determine very different habits, reinforcing some courses of actions and discouraging others. This is a fact that is very well known to clinical psychologists, who pay a great deal of attention to early mother child interactions, knowing how relevant and important a mother’s responses to an infant can be for its future development.
The particular perspective on the beginnings of semiosis we are sketching out here will deeply affect our way of framing the interrelated notions of consciousness, mind and representation.
5. The External Mind
Most of the data coming from I.R. pose unsolvable problems for the classical representational theory of Piaget. Consider the still face experiment of Tronick: at two and three months infants already have quite complex expectations regarding their mother’s behaviour, since they immediately recognize a change in her usual patterns of behaviour and react to these in highly differentiated ways. In order for them to be able to do that it is necessary to postulate that they have some memory of the relationship, and of the usual sequence of interactions within it. Now, according to Piaget, full representational capacity develops much later, at approximately 18 months. We are facing here the paradox of infants displaying a complex range of semiotic activity linking memory, expectations, adequate responsive behaviour without their having a developed system of internal representations to support this activity. The way out of this paradox consists, I believe, in a quite different way of looking both at what representations are, and at the relationship between the inner representational world of the infant and the so called “external” world.
In the classical view, representations are internal schema that represent an external reality; they develop gradually through interactions with the environment by way of experience and learning. But Tronick’s experiment suggests an alternative view: the relational memory that very young children already have could not be a representation in the classical sense, but rather a capacity to foresee and anticipate other people’s behaviour and adapt its own functional responses to this behaviour. Rather than a mental schemas we should be thinking in terms of shared intersubjective space, where the mind is not an internal device but, so to say, an active participant in a distributed external mind. If the idea of an external mind can appear somehow puzzling, we can remember that suggestions in this direction can be found also in the work of Peirce, as can be seen in the following passages:
“feeling is nothing but the inward aspect of things, while mind on the contrary is essentially an external phenomenon. The error is very much like that which was so long prevalent that an electrical current moved through the metallic wire; while it is now known that it is just the only place from which it is cut off, being wholly external to the wire” (C.P. 7.364)
“all knowledge comes to us by observation, part of it forced upon us from without from Nature’s mind and part coming from the depths of that inward aspect of mind, which we egotistically call ours; though in truth it is we who float upon its surface and belong to it more than it belongs to us. Nor can we affirm that the inwardly seen mind is altogether independent of that outward mind which is its Creator” (C.P. 7.557)
If the reference to the Creator puts a too metaphysical shadow on the notion of external mind, we could think of external mind more as the cultural environment where our “inward aspect of mind”, what we are used to call – egotistically according to Peirce – “our individual mind”, joins in with other minds in taking its own shape. To go back to our more specific ontogenetic situation, the external mind is what regulates the semiotic field, an intersubjective space that is constructed by both the interacting actors, mother and child, together with their wider physical and cultural environment. The external mind becomes, so to speak, the sedimented joint memory of experiences of interaction and semiosis, rather than merely being an internal propriety of each individual mind: a distributed property of the wider semiotic field, in a way very similar to what we nowadays refer to as “distributed cognition”.
It is in the external world of interaction that semiosis actually begins, and it is only subsequently that the individual mind of the child is able to develop a properly autonomous semiotic system. A perspective of this kinds, which, as we have seen, lies at the very basis of Peirce’s semiotic, is surprisingly close to the position of the psychoanalyst Wilfred Bion (1962, 1965, 1967, 1970). For Bion, intersubjective relation is of fundamental relevance in determining the very possibility of thought and of a thinking apparatus, in other words it is the basis for the possibility of semiotic functioning. Bion (1962) theorizes the existence of a protomental apparatus that does not represent experiences, but can only present them, starting from perceptual data, somatic excitements and sense impressions that Bion called beta elements. Beta elements cannot be used in order to think, they are not mental content yet and, as such, cannot be “digested” by the mental system, but can only be expelled from it. As they are, they are not representations, nor can they become the object of representations, but at the same time these beta elements – i.e. feelings, affects and perceptive data - are the basic elements on the basis of which all thought subsequently will become constituted.
The unstructured experience of beta elements becomes something usable for thinking only when it is transformed by a specific function that Bion calls alpha function, that converts beta elements into alpha elements, which are elements that can become thoughts. The alpha function however is not a function internal to the organism, in that it does not belong to the infant nor to his mental apparatus, but comes from the mother: mind can develop only in relation with another human being. It is the mother who transforms the unthinkable, terrifying experiences of the infant into possible thoughts, into semiotized contents, we could say. In order to do so the mother has to be able to receive and contain the infant’s beta elements, and give them back processed as alpha elements, which is to say as thinkable content. In semiotic terms, such a process could be described as the way to give semiotic form to un-formed matter: a process of semiotization, where the mother acts as the operator of a transformational device. According to Bion she can do so through her own mental apparatus, in the moment she is able to think her infant. Bion uses the French word réverie to refer to this particular form of maternal thinking, a kind of shared dreaming linking the mother’s and infant’s experiential worlds.
Independently of the different theoretical framework and psychoanalytic metalanguage used by Bion, there are clear similarities between his approach and the way in which I have suggested we look at the first formations of semiosis, while also making reference to Peirce’s semiotics. In both cases, thought, semiosis and meaning can emerge only from the intersubjective relation and its capacity to shape and transform infant experience. Both approaches are, in this sense, externalist, since they presuppose an environment, a field of successfully functioning interactional relationships as necessary pre-conditions for the development of an internal world.
From this point of view, Bion’s position is very interesting: according to him, the child first develops thoughts and only later, a mental apparatus that enables it to think. This paradoxical inversion of the usual progression – first the apparatus and then thought – is quite revolutionary, since it enphasizes the intersubjective, interactive nature of mind, which is something that extends far beyond the subject, in a very similar vein to our previously postulated notion of external mind. Our mental apparatus is the result of our thoughts, and not its cause, and thoughts in their turn are generated within the field of intersubjective relationships, and are not the simplistic effect of some ontogenetic need. In a provocative way we might say that relation is the real organon for thinking.
At this point we are in a position to have a closer look at the instant when semiosis actually starts taking place within the single organism, in the child’s consciousness itself.
From habits to consciousness
The picture I have tried to sketch out so far suggests priority of relation over individual mind: semiosis begins in the semiotic field, within an intersubjective relational field, and can only afterwards become interiorised by the infant in a conscious way. The semiotic field precedes the internal mind, but it would be a mistake to see external world and individual mind as two separate domains; there is not a clear cut between the semiosis of the environment and the consciousness of the individual. What at the beginning can be seen as mere reflex actions, mechanical laws of matter in a way, then becomes, by way of the constitution of habits and acquisition of meaning in the semiotic field of interpersonal relationships, a living, psychological reality.
A very similar position can be found in Peirce:
It would be a mistake to conceive of the psychical and the physical aspects of matter as two aspects absolutely distinct. Viewing a thing from the outside, considering its relations of action and reactions with other things, it appears as matter. Viewing it from the inside, looking at its immediate character as feeling, it appears as consciousness. These two views are combined when we remember that mechanical laws are nothing but acquired habits, like all the regularities of life, including the tendency to take habits, itself; and that this action of habit is nothing but generalisation, and generalisation is nothing but the spreading of feelings. (C.P.6.238-71)
The path that goes from reflex actions to habits and conscious behaviours will certainly require a much closer and much more detailed description which far exceeds the purpose of the present work. It is sufficient to emphasize here that in order for semiosis to manifest itself as conscious processes, i.e. for infants to be seen as having developed a fully functional semiotic consciousness, corporeal experience in the semiotic relational field of interpersonal interactions plays a major role.
This was well known to Freud, who claimed that the first Ego is a corporeal Ego (Freud 1923). We could think of such a corporeal Ego as a feeling body, a body made of sensations, feelings and emotions; remembering, too, that for Freud, and also for the pragmatist philosopher William James, student, colleague and life-long friend of Peirce, emotions are essentially, and in the first place, the modality by way of which we best perceive our own bodies.
Even before having developed a fully functional semiotic consciousness, our body is not mere pre-semiotic matter, but a highly complex semiotic system, endowed with feelings, emotions and the capacity to make subtle distinctions and respond in competent and meaningful ways to salient environmental stimuli. The body, and the emotions connected with our own subjective perceptions of it, are the basis for semiosis in that they are already semiosic processes; the origins of both semiosis and consciousness, in infants, are rooted here.
How the body becomes the vehicle for the development of a full semiotic consciousness will be the object of a subsequent chapter of my research. For the time being I would like to conclude by quoting a enlightening passage from Peirce, who towards the end of the 19th century was able to foresee what is today a well recognized fact in Infant Research:
A very young child may always be observed to watch its own body with great attention. There is every reason why this should be, for from the child’s point of view this body is the most important thing in the universe. Only what it touches has any actual and present feeling, only what it faces any actual color; only what is on its tongue has any actual taste. (C.P. 5.213-63)
References
Benzoni, S.,
2004 Il presente discontinuo. Esperienza, coscienza e soggettività nel neonato, Milano: Il Saggiatore
Bick E.,
1964 “Notes on Infant Observation in Psychoanalitic training”, International Jounral of Psychoanalisis, 45, 558-566
Bion W.
1962 Learning form Experience, London: William Heinemann
1965 Transformations: Change from Learning to Growth, London: William Heinemann Medical Books
1967 Second Thoughts: selected papers on psycho-analysis, London: William Heinemann Medical Books
1970 Attention and Interpretation, London: Tavistock Publications
Dennett D.,
1991 Consciousness Explained, London: Little Brown & Co
Eco U.,
1979 A theory of Semiotics, Bloomington, Indiana: Indiana University Press
1988 “On Semiotics and Immunology”, in Servarz e:, et alii (eds.), The Semiotics of Cellular Communication in the Immune System, NATO ASI Series, vol. 23, Belin, New York: Springer-Verlag,
Edelman G.M.,
1989 The rememebered Present. A Biological Theory of Consciousness, New York: Basic Books
2004 Wider then the Sky. The Phenomenological Giflt of Consciousness, New Haven: Yale University Press
Freud S.,
1914 On Narcisism: an introduction, S.E. 14: 67-102
1923 Ego and Id, S.E. 19: 17-25
Goffman E.,
1974 Frame Analysis. An Essay on the Organisation of Experience, Cambridge: Harvard University Press
Hjelmslev L.,
1943 Prolegomena to a Theory of Language ( English tr. Madison: University of Wisconsin 1961)
Mahler M., Furer M.,
1968 On Human symbiosis and the vicissitudes of individuation, New York: International Universities Press
Mahler M., Pine F., Bergman A.,
1975 The Psychological Birth of the Human Infant, New York: Basic Books
Meltzoff A.N.,
1990 “Foundations for Developing a Concept of Self: The Role of Initation in Relating Self to Other and the Value of Social Mirroring, Social Modeling, and Self Practice in Infancy”, in Cicchetti D., Beeghly M (eds.), The Self in Trans-itions: Infancy to Childhood, Chicago: University of Chicago Press
Meltzoff A.N. and Moore M.K.,
1977 “Imitation of facial and manual gestures by human neonates”, Science, 198, 75-78
Morris C.,
1938 Foundations of the Theory of Signs, International Encyclopaedia of Unified Science 1-2: University of Chicago Press
Nagel T,
1979 Mortal Questions, New York:Cambridge University Press
Neisser U. (ed.)
1993, The Perceived Self, New York: Cambridge University Press
Peirce C.S.,
1934-48 Collected Papers, Cambridge : Harvard University Press
Pozzato M.P., Violi P.,
2002, "La messa in discorso delle passioni. Il caso di Segreti e bugie di Mike Leigh", in VS, 93, pp.127-154.
Rizzolatti G. et alii,
1996 “Premotor Cortex and the Recognition of Motor Actions”, Cognitive Brain Research, 3, 131-141
Saussure, F. de,
1916 Cours de linguistique générale, Paris : Payot
Searle J.R.,
1997 The Mistery of Consciousness, New York : New York Review Books
Stern D.,
1985 The Interpersonal world of the Infant: A View from Psychoanalysis and Developmental Psychology, New York : Basic Books
Trevarthen C.,
1991 “Signs before Speech”, in T.Sebeok and Umiker Sebeok J. (eds.) The Semiotic Web, Amsterdam: Mouton De Gruyter
1993 “The Self Born in Intersubjectivity: the Psychology of Infant Communicating”, in Neisser (ed)., The Perceived Self, New York: Cambridge University Press
Tronick E. et alii,
1978 “The Infant’s Response to Entrapment between Contradictory Messages in Face-to-Face interaction”, Journal American Child Psychiatric, vol 17, 95-106
Tronick E. (ed.),
1982 Social Interchange in Infancy: Affect, Cognition and Communication, Baltimore: University Park Press
1. Body is not enough: the semiotic body
The notions of body and embodiment have become more and more prevalent over the last 20 years, in a number of disciplines associated with cognitive science such as philosophy, computer science, psychology, linguistics. Today, the centrality of the body in human cognition, meaning-making and experience is broadly acknowledged and this has provoked a huge quantity of research in this general area throughout a wide range of scientific domains.
This is certainly a more than welcome shift in our traditional Western research paradigm, since it can help free us from the old, seemingly unresolvable dualisms between body and mind, between the internal world of immaterial concepts and thoughts and the external world of objectivist reality. However, the present widespread use of the notions of body and embodiment across different fields and with different meanings makes it particularly important to develop a better understanding and clarification of these two notions, beginning with a rethinking of the first one, "body" which sometimes appears to be, paradoxically, the most misleading.
Body is often taken as a “natural” concept, and one which does not need any further elaboration. Apparently body is something easily accessible, objective, and physically defined. The body seems to be “there”, possessing an immediate self-evidencing character which does not need to be explained.
But this is not the case. The body is not a self evident concept, but the result of the various discourses that construct it. If the phenomenological experience of the body can appear an immediate one, the concept of “body” certainly does not. Rather, it appears to be seen in terms of the construals made of it within any given disciplinary perspective. In other words, the various meanings attributed to the notion of body are the sum of the various effects on sense of the different disciplines as they investigate and define it. The body as described by neurosciences is not the same body as the one described by psychonanalysis, or by experimental psychology, and so on. All these different “bodies” are not reducible to one another; on the contrary they produce a quite “heteroclitic” object, not very different from how language appeared to be when Saussure first started describing it. Many of the differences in the use of the very word “embodiment” that I will discuss in this paper depend on the different discourses that construct “body” in their respective ways as an object of research.
So, the first point to be made here is that there is no such thing as a body "in itself", naively taken as a given, immediate object of inquiry. Body cannot be described outside of the different discoursive practices that define it: to forget this implies the risk of hypostatising the body, as if it were endowed with an inherent essence, independent of the different practices, discourses and cultures that shape it. No "hard" science can escape from this paradox: even the the body as it is described by the most sophisticated technologies - radiography, magnetic resonance imaging and spectroscopy, etc. - is not a more basic level of description that reaches some more essential hypothetical "structure" of the body, but just another way of representing it.
Even the body as studied in medicine is a construal, so much so that different medical practices in different cultures construe as many different bodies as there are cultures: the "Western" body studied in our medical tradition is not the same as the body mapped by chinese acupuncture.
This does not mean a denial of the very exsistence of bodies as material entities, but rather, within a radical constructivist perspective, one which would have appealed to Peirce, to recognize that we can only reach these bodies through different practices and discourses - i.e. through semiosis . "The" body in such a perspective becomes a kind of unreacheable Dynamic Object, to use Peirce's terminology, only approachable through a series of partial descriptions, depending on the particular perspective or disciplinary approach we decide to take. Such descriptions, which we can consider as forming part of an open set of Immediate Objects in Peirce's sense, will not necessarily converge to form a completely homogeneous picture. Rather they may continue to remain highly divergent as, for example, in the case of the phenomenological body we perceive proprioceptively, and the body as it appears to us on the basis of the results of a laboratory experiment.
Body is, then, a semiotic construal, and this remains the case even when we attempt to describe its more basic, material levels of organization, such as neurons or brain synapses, which are certainly "real", but are not the body. If we miss this point we risk a curious paradox, which could be defined as "embodiment without the body". To understand the role the body plays in processes of producing and understanding meaning, i.e. in semiosis, we need much more than this.
In what follows I will discuss the issue of embodiment from a semiotic perspective, starting with a (very brief) look at some of the main contributions to be found in this theoretical field, then going on to review some of the different forms that embodiment has taken in cognitive science, and concluding with a look at what I believe still remains to be investigated.
That the body plays a major role in semiosis is not a total novelty in semiotic quarters. Semiotics, like all the other disciplines already mentioned, has in its recent developments begun to concern itself more and more with issues related to the body, and semiotic investigations have also been started into a related set of problems connected with the role that feelings, emotions, and sensory and perceptual elements play in meaning making processes – in a word: the embodied dimensions of meaning. If such a "corporeal turn" is only quite recent in the post structuralist tradition that gave birth to contemporary generative and narrative semiotics, this is not the case for the other main tradition in semiotics, i.e. interpretative semiotics, as it is commonly referred to today, which may be traced back to the work of the pragmatist philosopher Charles Sanders Peirce. This is not the place to enter into an in depth discussion of the complex philosophical approach advocated by Peirce; it will suffice here to mention just a few points that are relevant for our present purposes. Peirce is often remembered mainly for his cognitive semiotics, and for his important contributions to the logic of abductive reasoning. However I believe that in his phenomenology, which is perhaps less well known than his logic, an important theory of the role of the body in semiosis and a very innovative intuition regarding the nature of the body-mind relation can be found.
Although Peirce does not thematize in an explicit way the role of the body in semiosis, it is quite evident that for him, the body plays an important role: it would be enough to consider that at the very basis of the semiotic processes that enable us to make sense of the world there is, for Peirce, perception with its bodily based inferential processes. Perception, for Peirce, far from being an automatic record of external reality, is a highly constructive process, which requires exactly the same inferential and abductive devices as abstract forms of reasoning do, while being rooted firmly in the basic physiological functioning of our bodies. Therefore, semiosis begins in the body and in its perceptive and proprioceptive processes.
But this is not the only hint of embodiment we can find in Peirce's semiotics. Even more interesting is his theory of interpretants with its implications of a potentially endless process of sign production and interpretation that gives rise to meaning and sense. For Peirce all interpretation implies an interpretant, which is always a sign, produced from a first, preceding sign, as its effect. According to Peirce, there are several kinds of interpretants and more than one classification of these; interestingly enough the first two levels of interpretation, before arriving at the level of logical interpretant, which is the cognitive level of concepts, are the emotional and the energetic interpretants. The first is concerned with the emotions signs evoke in us, the second with the muscular bodily reactions they evoke. Now, all these three levels of interpretants remain active during the ongoing semiotic process, and this means that even in more cognitively oriented tasks, such as abstract reasoning, emotions and bodily reactions are always involved, although with different degrees of relevance with regard to the specific task and situation in hand.
More generally speaking, Peirce does not conceive the mind as something qualitatively different from the body or other forms of matter: there exists a fundamental continuity (referred to in his terminology as "synechism") between these, since both share some natural common characteristics, as we can see from the following citation:
We ought to suppose a continuity between the characters of mind and matter, so that matter should be nothing but mind that had such indurated habits as to cause it to act with a peculiarily high degree of mechanical regularity or routine… This hypothesis might be called materialistic, since it attributed to mind one of the recognized properties of matter, extension, and attributes to all matter a certain excessively low degree of feeling, together with a certain power of taking habits. (CP 6.277)
In this way body, mind and the world are not only connected, but fundamentally interdependent of one another in an endless process of sense making which reminds us of the dynamics of self organizing systems in an ongoing developmental relationship between organism and environment. The classical dualistic relationship between mind and matter is overcome, as well as that between the internal and the external world, which are no longer seen as being dramatically and irreducibly separate from one another. There is mutual interpenetration in all directions.
If the role of the body forms the basis of Peirce's notion of semiosis, then the same cannot be said for classical structural semiotics, rooted in the work of Saussure and Hjelmslev, where a formalistic approach to meaning was dominant. However in Greimas' latest works, as well as in the most recent work by Fontanille the mind-body question is reopened, in particular through a rereading of Merleau-Ponty's phenomenology.
According to Merleau-Ponty, meaning is in the first place articulated in our body, through perception. Also for the French philosopher perception is not merely the simple and passive record of an external world, already structured and pre-given in its configuration; perception is rather the active construction of a world already endowed with meaning and intentionality. Through perception the subject meets the world in the first place and begins to give meaning to it. Phenomenological and perceptive meaning is transformed into linguistic meaning through the corp propre which founds, at one and the same time, the subjectivity of consciousness and the exteriority of the world. Here we can see another possible compatibility with Peirce's philosophy: in Merleau-Ponty's phenomenology, too, external and internal world are not separate and in opposition with one another, but related to each other via the mediation of the corp propre that operates, in a way, as translator of perceptually constructed meaning into linguistc and conceptual meaning.
But the body is also the place where affect and emotion are rooted, as Freud and psycoanalysis have taught us, reminding us that the Ego is first and foremost a corporeal Ego. Recent developments in semiotic theory are insistent on the fundamental role emotions play on the very deep level of sense structuring.
The basic approach to the body that emerges from such a background is not always consistent with the way in which embodiment has been studied in other cognitively oriented research domains. What I shall claim in the present paper is that in order to fully understand the role that embodiment plays in meaning construction and semiosis, we have, so to speak, to go beyond the body itself. To develop a satisfactory theory of embodiment the body is not enough, and we will need to incorporate not only issues related to action and movement, but also those related to affect and emotion, a move that will force us to open up to the crucial issues of subjectivity and intersubjectivity.
At this point, however, it has become vitally important to look more closely at some of the basic tenets of the notion of embodied cognition as developed in various areas of the cognitive sciences, in order to see if we can discover some possible links, overlappings, or differences relative to a more semiotically oriented approach. In particular I would like to claim the following: i) there are today within the field of cognitive studies many very different notions of embodiment, only some of which are of real theoretical interest from a semiotic perspective. It is therefore crucial to distinguish between these in order to specify which type of conception of embodiment might be most productive for semiotics; ii) embodiment is related in an important way to the problem of meaning processes, and it can help in a decisive way to reframe some of the most controversial questions in semantics. A context oriented, encyclopedic approach to meaning, which semiotics intrinsically offers, needs to take into account the role of the body; iii) as I already suggested, the notion of ‘body’ is not a self-evident nor simple one, as is too often assumed in contemporary cognitive science; on the contrary the body is a constructed concept, and as such, cannot be reduced to purely neuro-physiological aspects nor to the brain. The kind of body we need to incorporate into our theory of embodiment is more complex than that; it has to be considered in its full phenomenological complexity, as the place where affect and emotions are articulated, and, maybe more importantly, it must to be tied in with the central issue of subjectivity and intersubjectivity, a topic not often addressed in cognitive approches to embodiment.
But it is now time to have a closer look at what is exactly meant by “embodiment”, and how it might constructively be related to a more specifically oriented semiotic approach
.
2. Different embodiments
In very general terms we could say that the main idea behind embodiment is that mind derives and takes shape from the fact that we have a body that interacts with our environment. Such an assumption is generally seen as drastically opposed to classic representational cognitivism, which is based on functionalism and the computer-mind metaphor. According to functionalism, mind is independent from its material implementation, as the computer-mind metaphor suggests.
Implicitly connected to this position is a theory of concepts and semantic categories which is generally referred to as the “classic” theory, where it is claimed that it is possible to arrive at a precise definition of the semantic categories over and above, and independently from, their uses and contexts of application. In this perspective the body does not play an important role: it is essentially an output device, as often defined, merely executing commands generated in the mind through symbol manipulation.
In the embodied perspective, on the other hand, cognition is seen as depending in a fundamental way on the body and its perception and motor systems, as well as on bodily-based experience and our interactions with the world.
Before going on to discuss these matters, we must immediately point out that there is no such thing as a unique theory of embodiment. On the contrary, the concept of embodiment is a very polysemic one, and different authors use it in quite different ways. Rather than referring to a single theory of embodiment, we ought to refer to different theories of embodiment, often highly divergent from one another, and sometimes having very little in common.
So let us now return to the issue of what might be considered the basic idea underlying the various approaches to embodiment. What exactly does it mean to say that the mind is embodied, and that it emerges and derives from the body? If we look more closely, we can see that there are many different readings of this same thesis, ranging from an extremely weak to an extremely strong, which is theoretically more interesting, but also more controversial. It will certainly prove useful to examine these various positions more closely, since, as has been stated, only some of them will turn out to be of interest from a semiotic point of view.
A first and extremely weak interpretation would simply imply that all cognitive processes have a material basis. This is such a generic option that it would be difficult to disagree with it, but at same time it is so generic that it is not very meaningful. A more interesting assumption would be to say that cognitive processes cannot not have a material basis or, in other words, that cognition is directly connected to the various structures and biological processes that implement it. A somewhat similar version, still rather weak, implies that in order to understand mental processes one cannot ignore the way the nervous system and the brain work. In the last few decades, both neuroscience and neuropsychology have made such a position highly popular, and also widely accepted: today there are probably very few researchers in cognitive science who would disagree with this position, with perhaps the exception of few more orthodox functionalists. From a semiotic point of view, however, this appears to be somehow a more background type of issue, since a semiotic analysis is not directly concerned with these more basic levels of description, but rather with the higher levels of sense organization.
A third interpretation, defined as ‘material’ embodiment (Nùnez 1999: 55), also takes into account –in addition to the idea that the mind depends on underlying neurobiological processes– the constraints imposed on cognition by real-time bodily actions performed by an agent in a real environment. This is a quite popular position today in robotics, where research is focused on low-level cognitive tasks such as visual scanning or motion. Since it has to deal with the construction of robots able to perform real actions in a real environment, robotics must necessarily develop models of vision, perception and movement constrained by genuine perceptual-motor interactions with the environment. Here embodiment means essentially taking into account the spatial-temporal constraints implicit in real bodies, but it does not imply any strong theoretical assumptions. Lakoff and Johnson (1999: 37) distinguish here between embodiment as realization and embodiment as shaping.
Embodiment as shaping, often defined as full embodiment, or radical embodied cognition, is certainly the more popular position in contemporary cognitive semantics, and appears to be the one we should look at more closely from a semiotic point of view. According to this view, all concepts, even the most abstract ones such as those of mathematics are the result “of the way the brain and body are structured and the way they function in interpersonal relations and in the physical world” (Lakoff and Johnson 1999: 37).
Notice that in this quote from Lakoff and Johnson, brain and body are used as substantially interchangeable; this kind of overlapping is found in many fields of research on embodiment. According to Nunez, for example, embodiment explains concepts “in terms of the non-arbitrary bodily experiences sustained by the peculiarities of brains and bodies” (Nùnez 1999: 56).
This is a crucial question, since there is a potential ambiguity in considering body and brain as equivalents –an ambiguity that could produce potentially dangerous levels of confusion. Body and brain are not the same thing, as the phenomenological tradition, both of Husserl and of Merleau-Ponty, has taught us, a tradition to which most researchers today seem to refer. So this would seem to be a vital issue if we want to incorporate an embodied approach in a serious way into semiotics.
The body is something quite different from the brain, and if the latter can be seen as an immediate object for scientific study, the body certainly is not, at least not in any direct and transparent way. Indeed, I have already made the opposite claim, i.e. that the body is not at all a self-evident concept, as it might appear at a first sight.
For the moment I just want to make salient one specific ambiguity of this kind which underlies most work on embodiment. While material embodiment refers to the properties of the brain, and, therefore, in this model the body may be described as a body-brain, when we are speaking of embodied concepts or embodied cognition, a quite different meaning of ‘body’ is at stake, much closer to the notion of ‘corporeal schema’ than to that of the brain. Although embodied cognition might well have a neural plane of implementation, we have here two different levels of description, which do not coincide, and it would be helpful to keep them apart. Semiotics, with its phenomenological tradition, might very well play an important role in clarifying these issues and distinguishing between these two conceptual levels, of which only the second is, as I have already mentioned, of real semiotic concern.
Within the field of cognitive science, the picture is even more complicated, however, since the new paradigm is pursued within different disciplines and by means of different methodological approaches, which do not all necessarily share the assumptions of cognitive linguistics, not to mention those of semiotics.
To simplify, three main research domains relevant for our present discussion might be designated: connectionism (and neo-connectionism), robotics, and cognitive semantics. These domains do not necessarily share the same notion of embodiment.
For example, many of the neo-connectionist models which use a dynamic modelling approach are not at all necessarily embodied, in the sense of having systematic, continuous relations with their actual perception and motor referents. What we have here is rather a conceptual interpretation that has little to do with empirical perceptive states, as Prinz and Barsalou (2000) have shown. Connectionist nets do not guarantee embodiment, neither the radical embodiment of cognitive semantics, nor the weaker notion of material embodiment.
Situated robotics, on the other hand, as I have already pointed out, has necessarily to take into account actual bodily constraints, since, in order to be fully operative the cognitive system underlying a robot must have an efficient interface with perception and action data: a simple abstract computing system would not be sufficient.
Maybe the main lesson we can derive from situated robotics is that to perform perception and action we cannot use only the cognitive system itself, we need also to exploit the resources inherent in the body and the environment. As Clark (1997:36) claims, intelligence is not based exclusively on cognitive abilities –it evolves from the dynamic interaction between brain, body and world.
The concept of embodiment used in situated robotics is also different from the one used in the more theoretical fields of cognitive semantics and contemporary cognitive semiotics, which are crucially concerned with embodied experience. Both cognitive semantics and semiotics see human experience as fundamentally bodily based: concepts and cognition emerge from our experience and are bodily grounded.
To conclude, there are probably more differences than similarities among researchers who explicitly refer to the notion of embodiment. For some, the ‘embodied’ mind is still computational in a literal way, for others it is not computational at all. Some refute completely the concept of representation, generally preferring dynamic systems, others, like Barsalou, refute dynamic systems and still use forms of representation. For some, embodiment exists only in authentically living systems (and not in simulations, not even connectionist ones), for others this is irrelevant; finally for cognitive semantics and semiotics the crucial idea is that of phenomenological bodily experience.
What then do all these different approaches have in common? Well, probably the only real unifying aspect to be found is a critical one.
Embodiment theories are essentially a critical reaction to representational cognitivism, and in particular Fodor’s functionalism. Here, there are two points of criticism: first, the non-consideration of body-based ‘material’ aspects of cognition ; second, the reduction of cognitive processes to purely syntactic symbolic manipulation.
From this point of view, theories of embodiment appear to be a natural development of cognitive semantics and cognitive linguistics of the seventies and eighties. Theoretical antecedents can be traced back to cognitive grammars, especially Space Grammar and Mental Space theory ; research on space and language and Force Dynamics, the system of forces that Talmy (1985) posits as the ground of the linguistic system of modality, which is essentially derived from embodied structuring.
A fundamental antecedent is also to be found in the critical review of the classical category theory that goes under the generic name of prototype theory .
Since these seminal works first arrived, research in this field has continued to advance, reframing in a radical way some of its key concepts, beginning with that of representation.
3. Body and situated meaning
The anti-representational controversy is more properly a controversy against a particular type of representation: symbolic representation, in the Fodorian sense. Such a criticism, as we will see, is not at all contradictory to basic semiotic tenets, rather quite the opposite.
Rosch (1999: 62) for example claims there is a need to distinguish between two types of representation: the first is a device that mediates between mind and world, close to Peirce’s idea of semiosis, connecting the external and internal worlds ; the second is based on a notion used in classical cognitivism, where symbols are seen as syntactic symbols –formal operations within the closed system of a machine (or a mind, which is nothing but a machine).
One of the most important differences between these two models is the different ways they offer for looking at context. Traditional cognitive science sees representations as stable, context-insensitive configurations that cannot be affected by contextual change. The so-called classical theory of categories was based on precisely such an assumption: a category might be a node, a network, a set of features, or a mental world, but it was in any case always a static and immutable entity. In other words the basic idea was that one and the same invariant structure represented one particular concept in all possible contexts.
Now such a conception of the matter seems highly problematic: there is little doubt that natural cognitive systems exhibit a high degree of variety, and that our functioning in the world is much more flexible than any fixed structure could describe. Both our behaviours and our mental states adapt continuously to changing contexts, responding in a highly flexible way to environmental modifications. The traditional concept of representation thus turns out to be radically inadequate.
This is not something new in semiotics: similar criticisms of the classical theory of representation have been developed within a semiotic perspective since the Seventies. Umberto Eco in his A Theory of Semiotics (1976) had already pointed out the fundamental incapacity of any kind of invariant, dictionary-like structure to represent meaning, and successively, in 1984, he elaborated further the general notion of the encyclopedia as the only viable alternative to dictionary based models. From this point of view, semiotic perspectives, at least those developed within a Peircian interpretative framework, and those of cognitive semantics based on prototype theory, are certainly highly compatible, as I have discussed elsewhere (Violi 2001).
At this point, however, my thesis is that developing the issue of embodiment can help us to go even further and to develop a more sophisticated approach to meaning and semiosis, and their relation to context, an approach that is theoretically more radical than that presupposed in Eco’s models.
Concepts are indeed sensitive to contexts because we are embodied organisms and we interact with the environment. Embodiment and interaction are basic features of our semantic system, and more generally, of the ways in which we make sense of all our ongoing experience.
Taking embodiment seriously in describing meaning can help a semiotic approach to overcome some of the limitations that can still be found in the encyclopedic model. Indeed the concept of encyclopedia, as elaborated by Eco, is a cultural construct that can account, in terms of a regulative hypothesis, for all possible cultural and social components of meaning. However it has considerably less to say regarding the phenomenological side of our experience, although it does not in principle exclude it.
I believe that if something such as a cognitive semiotics is to be established as a field of study, it cannot avoid the incorporation of embodiment in its basic definition of cognition, and indeed taking of this very incorporation of embodiment as its starting point.
Among the various embodied approaches we can already find some interesting suggestions in this particular direction. Rosch, for example, emphasizes the role of situation and context in an embodied perspective. According to Rosch (1999: 72), even when concepts appear to be universal and abstract, they always refer to specific and concrete situations. Real situations are events rich in information and should be the real object of study. Generally speaking, psychology tends to see contextual effects as negative elements that invalidate experimental work, but this perspective should be changed, and variations should become the main data for analysis.
Interestingly enough, the adoption of a strong contextualism of this kind parallels some recent positions in semiotics, where focus has been shifted from the system, and therefore from structural regularities, to process and text. The textual turn in semiotics implies making, and considering the text as the real unit of analysis; this is compatible with Rosch’s positions, where the single situation is considered to be the correct object of analysis. In both approaches we can find a common holistic component, which in some semiotic approaches appears to be extremely radicalized.
Today, Rosch’s broader assumptions regarding representations and the nature of concepts are quite different from her previous work on prototypes, and are embedded in a strongly holistic idea of the mind-world whole. Concepts are now seen as intrinsically non-representational: they do not have the function of representing the world in the mind, nor do they mainly have an identifying function, as is generally taken for granted in experimental research on naming tasks. Rather, concepts participate in situations.
“Concepts and categories do not represent the world in the mind, they are a participating part of the mind-world whole” (Rosch: 1999: 72). Their participative nature derives from their being a natural mediation between mind and world, a mediation which is necessarily anchored into specific and locally defined situations.
Concepts are the natural bridge between mind and world to such an extent that they require us to change what we think of as mind and what we think of as world; concepts occur only in actual situations in which they function as participating parts of the situation rather than either as representations or as mechanisms for identifying objects. (Rosch: 1999: 61).
Even those who do not share such a radical position would agree to not conceiving of representations primarily as structures that represent the external world, but rather as control structures for the regulation of interactions with the external world. This shift from mirror or encoding models to action-device models is quite common in current research on embodiment.
In robotics for example, Clark describes representations as control structures: “The idea here is that the brain should not be seen as primarily a locus of inner descriptions of external states of affairs; rather, it should be seen as a locus of inner structures that act as operators upon the world via their role in determining actions” (Clark 1997: 47).
Representations become here oriented toward action, while at the same time describing aspects of the world and prescribing possible actions, in a fine balance between pure control structures and passive representations of the external world.
With respect to the issue of representation it is worth noticing how close an approach of this kind is to the basic tenets of Peirce’s pragmaticism. For the American philosopher, too, concepts (and representations) are always correlated with actions: while concepts, seen as habits of mind, have a regulative function in relation to the internal world, stabilizing the process of unlimited semiosis; on the other hand when operative as beliefs, they also constitute the basis for behavioral and communicative habits, which are nothing but regularities in actions. In this way the very same semiotic structures regulate both the internal world of concepts and beliefs and the external world of actions, acting as a bridging system between the two.
A similar idea can be found in the model for memory proposed by Glenberg (1997: 1-55), where memory does not primarily have a representative function “to store the past”, but is rather an embodied device for facilitating interactions with the environment.
Such a perspective, largely shared among embodiment theorists, focuses on the role of the larger environment and its interactions with the organism, and on the relation between external and internal worlds. This explains a growing interest in Gibson (1979) and his concept of affordances. For Gibson, too, representations and internal states that mediate the relationship with external world, are centred on action, or, to use Gibson’s words, connected to affordances. Affordances are nothing more than possibilities for action and use offered by the local environment to a particular type of embodied agent, equipped with specific bodily features. In this way perception is always contextualized and constructed: the world is essentially perceived by some given organism endowed with its own intentions in some given context, and is seen as affording opportunities for goal directed actions. Perception is therefore always connected to action, and both perception and action are always connected to cognition.
This is a crucial point, because the action-perception-cognition link is perhaps one of the most important acquisitions of embodiment theories. Perception is never seen as a passive recording of information, but is immediately connected to action potentials. Therefore any kind of rigid distinction between perception and cognition disappears, and they become highly integrated and overlapping processes. Not surprisingly, such an approach is very interested in results of neuro-physiological studies that show a connection, even at neuronal level, between perception, action, thought and imagination. Recent research on mirror neurons have shown that in primates, and also in humans, the same neurons fire both when a given action (like grasping a cup of coffee) is effectively executed by some individual, and when it is observed while being executed by an other, and as well as when the subject merely thinks of executing it. Interestingly enough, this does not happen just for any kind of movement, only for intentional actions, finalized to a goal (such as grasping a cup), and thus only for intentional interactions with the environment, or, to use Gibson’s words: interactions connected to precise affordances.
The existence of underlying schemas common to perception, action, language and cognition probably represents one of the most challenging acquisitions of work on embodiment, and it is one that semiotics cannot ignore, since it implies a highest possible level of integration between all these systems. Perception, action, language cannot any more be considered as totally autonomous and independent modules, they must become functional specifications in a common unitary configuration.
This is also the ground of metaphorical concepts, so central in cognitive semantics, in that they represent linguistic and conceptual projections of bodily configurations of various kinds (perceptual, motor, spatial, and so on). Metaphorical projections are always motivated ; this is the second important lesson we can derive from embodiment studies. Together with the motivational aspect, this offers a radical challenge to the dominant view of language as a formal system, totally arbitrary and abstract. An important consequence of this work is a shift from the study of linguistic forms to the study of linguistic substances, a shift fully shared by contemporary cognitive semiotics. As Petitot suggests:
Il s’agit d’abord de rompre avec l’idéalisme sémiotique à l’œuvre dans les approches formalistes du sens qui auront dominé la grande période du structuralisme logico-combinatoire. (Petitot 2000: 84) (What is at stake here is a break with the semiotic idealism of the formalist approaches to meaning that dominated the heyday of logic-combinatory structuralism.)
Idealistic formalism has several important consequences: first of all it implies a totally disembodied approach to meaning :
Le sens perd tout rapport au monde naturel externe et au couplage perception-action qui fonde notre rapport écologique et ethologique à ce monde. (Petitot 2000: 85) (Meaning loses all relationship with the external natural world and the coupling of perception and action that grounds our ecological and ethological relationship with this world.)
Secondly, meaning is deprived of all self-organizing systemic principles and cannot but be purely logical and combinatory. A semiotic approach based on embodiment should pursue a double program that we could define at one and the same time as a de-formalisation and a de-mentalisation of meaning and sense, reintroducing the study of substance as an essential part of its project.
4. Intersubjectivity and the embodied subject
The new field of embodiment has brought to light many interesting concepts and questions of central concern for semiotics: firstly, there is a more realistic idea of the way human beings perceive and interact with their environment, and the way in which meaning emerges from these activities. Next, there is the interconnection between cognition, perception, and action; the crucial relevance of situations and contexts, and a different and more articulated idea of the relationship between external and internal world. Finally, there is the central role of embodied structures in language and cognition, and the embodied nature of metaphorical mappings. All this points to a contextualist and pragmaticist conception of semiosis, in the Peircian tradition, allowing an anti-idealisitic and anti-formalistic shift in semiotics, such as the one advocated by Petitot.
Embodiment allows and indeed requires a superceding of the purely logical and formal approach which had characterized semiotic structuralism in its initial period of development; meaning ceases to be a purely negative value, as it has been conceived in the Saussurian tradition, for it now acquires a living connection with our perceptional, phenomenological and emotional experience of the world. In this way world, experience, body and mind will all come to be seen as much more closely interconnected and strictly related to one another than before, in a way highly consistent with the Peircean tradition, as I have already indicated.
These are all very important acquisitions. However, there are still a few points which will need to be more carefully considered, and where I believe that semiotics will be able to contribute an important series of clarifications to the wider study of embodiment. Indeed, in research on embodiment, there are some possible ‘zones of confusion’ that appear to be particularly crucial in our current situation. The first zone of confusion has already been mentioned and concerns the interchangeable use that is sometimes made of the terms ‘body’ and ‘brain’. It is important to emphasize once again the complete lack of coincidence between these two levels: the body can certainly not be reduced to purely neural forms of activity. A ‘body-brain’ of this kind would exclude the whole phenomenological dimension of experience, that live presence that Husserl called Leib, as opposed to the material Korper.
The second zone of confusion arises in relation to the distinction between body and corporeal schema. The confusion is more implicit than explicit, since corporeal schemas are rarely mentioned, although the notion might represent a crucial concept for the discussion of embodied experience. The concept of corporeal schema was first used by psychiatrists and neurologists towards the end of the nineteenth century, and was then further elaborated by Paul Schilder in the mid-nineteen-thirties (Schilder 1935).
The corporeal schema is not only the general kinaesthetic experience we have of our body, but it is also the spatial dimension that is occupied by the body. According to Schilder, it is neither a sensation nor a mental representation, but rather something intermediate between these two things. Merleau-Ponty (1945) refers to the notion of corporeal schema in order to define the corps propre and its relationship with subjectivity. According to Merleau-Ponty the notion has a gestalt configuration and a dynamic character, implying an intentional dimension. The body is always endowed with a project in the world; it has its own goals deriving from its interactions with the environment.
The notion of corporeal schema seems crucial if we wish to investigate the embodied grounding of concepts, since at that level what is at stake is not the ‘body’ as a material and natural object, but its schematic configuration, as has been well demonstrated in studies on spatialisation in language. On the basis of this type of embodied configuration, the body becomes the first place of meaning articulation, and its embodied schema are the basic structures that organize meaning, even before language, as I will discuss in a moment. However, to fully understand the role of embodied configuration in semiosis, we have first to discuss a very important issue, related to affect and emotion. Bodily states are always, and at the same time, pathemic states, endowed and infused with feelings and emotions. Body is where emotions have their primary space, and if we do not take this aspect of embodiment into account in our analysis, we miss a crucial dimension of meaning making, and risk ending up with a totally inadequate and reduced conception of the body itself.
Affect and emotion are in the body from the very beginning, in all our sensations and perceptions, which are always permeated by an affective-emotional tone. We do not only feel sensations of warmth or coldness: we feel pleasant, unpleasant, or unbearable temperature levels, and the same also holds for perception: what we see, hear, taste or smell is never ‘neutral’, but always endowed with some sort of emotional reaction along the pleasure-displeasure scale. Body is, in other words, never pure ‘soma’, but always soma animated by certain affective and emotional states, in other words: soma and psyche are always simultaneously co-present. Here we can see that it is precisely the notion of psyche that enables the overcoming of body-mind dualism, unravelling the categorial distinction between the two terms.
But this switch from a naturalistic body to a somatic-psychic one also implies that we must enter into the domain of subjectivity and intersubjectivity.
The whole issue of subject and subjectivity is almost completely absent in the North American tradition of work on embodiment. However we can in several cases quite easily find implicit reference to something that we more appropriately would have referred to as subjectivity, but which is not always recognized as such.
Let us take as an example the otherwise excellent article by MacWhinney (1999), where the author analyses some of the different forms in which language emerges from embodiment. According to MacWhinney “language comprehension and production are embodied processes whose goal is the creation and extraction of embodied meanings (…) We can refer to these processes of active embodiment as the perspective-taken system” (MacWhinney 1999: 214).
The embodied perspectival systems operating in language are related to four levels: 1) affordances, where language and cognition are related to individual objects and actions through affordances; 2) spatio-temporal reference frames, which refer to “the set of competing spatio-temporal reference frames” (MacWhinney 1999: 215); 3) causal action chains, most centrally involved in the emergence of grammar and the different perspectives of nominative-accusative language or ergative-absolutive language; 4) social roles, where the perspectival system allows us “to adopt the social and cognitive perspectives of other human beings” (Mac Whinney 1999: 216).
What is of interest here is that all of these systems are not equivalent in their relations to the issues of embodiment and subjectivity. If the first level of affordances is certainly linked to the body and its grounding in the linguistic perspectival system, since all the properties we can think of in relation to an object are affordances grounded in the perspective of our own body, the same does not hold for the other three levels, where it is not so much the body that plays a role, but the point of view of the subject as represented in language. Consider the spatio-temporal reference frames. MacWhinney explicitly mentions three alternative frames, an object-centred, a speaker-centred, and an environment-centred frame. These frames do not depend on the body, but on the way the position or perspective of the subject is framed within discourse. The same is true for the other two systems: both the perspective a given grammatical construction imposes on the action, and the perspective connected to interpersonal and social frames, refer to subjectivity more than to embodiment. What we have in these cases are traces left at sentence level by the process of enunciation. The notion of perspective can be framed in the wider issue of linguistic subjectivity, which, in European post-Saussurian linguistics, has most convincingly been elaborated in the Theory of Enunciation . Such a theory unifies in one and the same framework a family of heavily interconnected issues, ranging from pronominal, temporal and spatial reference systems, to focalization, perspective, point of view, and so on.
So obviously the question is not whether or not we use enunciation theory as formulated in post-Saussurian linguistics, but the possible overlappings that may be found between two different issues, both of which are extremely important. However, they are not necessarily interconnected. Perspectival systems depend on the presence in every sentence of an uncancellable point of view which is the trace of the enunciation process. This is something quite different to embodiment, which is the existence, in semantic structures, of motivated configurations, all of which depend on embodied experience.
Given the extent to which these two issues are not the same, the theory of enunciation removes the issue of embodiment altogether, leaving only reference to a transcendental subject, completely deprived of any form of bodily qualification, gender difference or any other dimension which might be linked to individual subjects . Here we have a deeply paradoxical chiasmus: on the one hand there is a theory of embodiment without the subject, on the other a theory of the subject without a body.
In order to develop a fully embodied theory of semiosis we certainly need a bringing together of body and subject, and to do this we must develop an approach to subjectivity which is quite different from the transcendental Ego that is implicit in the classical structuralist framework. An alternative approach of this kind will need to be more firmly connected to the dynamic dimension of enunicative practices of subjects, and, above all, to the interplay between the embodied subject and the relational dimension of intersubjectivity.
Subjectivity is not the emergence of a transcendental subject revealing himself (and here the masculine pronoun seems more than appropriate), but rather the emergence of a subjective dimension within a complex, relationally grounded interpersonal, social and cultural environment, in other words: the realm of intersubjectivity, in which all embodied organisms necessarily ground their meanings. This implies, in a way, to go beyond the individual subject itself, which cannot manage to exist in any kind of isolated, solipsistic form, and even beyond the body itself, if considered merely as an encorporalisation of mind. An embodied subject is more than a body and more than an individual entity: it is a somatic-psychic organism, constituted by embodied affect and emotions and inextricably enmeshed in a complex world of intersubjective relationships.
To exemplify this last point, I will conclude with some, necessarily very brief, references to my current research on preverbal children. Working on video of interactions of young children (aged less than 12 months) with their mothers it becomes strikingly evident how meaning is inherently embodied, in that it emerges from embodied interactions well before it begins to manifest itself in language. Preverbal babies are already engaged in a complex work of building meaning on the basis of their interactions with their environment and the relationships they are involved in with the adults around them, especially the mother. Their gestures, gazes and movements can all be read as an already articulated kind of ‘language’, where the emotional and mental world of the child manifests itself, not yet through words but through embodied actions.
It is quite intriguing to notice in analyzing these materials the strong interconnections that can be seen to exist between the ongoing intermingling of intersubjective patterns - a kind of relational dance involving both mother and child - and different bodily responses on the part of the child. In order to understand the process of meaning construction at this very early developmental stage it would be quite misleading to look only at the body, without also taking into account the full range of intersubjective practices within which it is created. Meaning seems to emerge as a series of bodily and emotional responses to environmental interactions: a kind of coupling of embodied actions on the part of the individual subject to a wider pattern of intersubjective relations, a process which might be defined as a coupling of subjective and objective components of meaning.
From its very beginnings the embodied subject, far from being either a transcendental ego or a purely neural brain, will emerge as the unique way in which each individual body shapes emotions and feelings in the intersubjectivity of relations with the other.
References
Barsalou, Lawrence
1999 Perceptual symbol systems. Behavioral and Brain Sciences, 22: 577-609
Benveniste, Emile
1966 Problèmes de linguistique générale I. Paris: Gallimard 1974 Problèmes de linguistique générale II. Paris: Gallimard
Clark, Andy
1997 Being There: Putting Brain, Body and World Together Again. Boston, MA: MIT Press
Coppock, Patrick
2002 Semiotics and the body: C. S. Peirce on the mind-body-world relation. Versus. Quaderni di Studi Semiotici 93: 135-167.
Eco, Umberto
1976 A Theory of Semiotics. Bloomington: Indiana University Press
1984 Semiotics and the Philosophy of Language. Bloomington: Indiana University Press
Fauconnier, Gilles
1985 Mental Spaces: Aspects of Meaning Construction in Natural Languages. Cambridge, MA: MIT Press
Fontanille, Jacques
1999 Polisensorialità e autonomia della dimensione figurative. In: PierluigiBasso, Lucia Corrain (eds.), Eloquio del senso. Dialoghi semiotici per Paolo Fabbri, 188-212. Milano: Costa e Nolan
2004 Figure del corpo. Per una semiotica dell'impronta. Roma: Meltemi
Gibson, James
1979 The Ecological Approach to Visual Perception. Boston: Houghton Mifflin
Glenberg, Arthur
1997 What memory is for. Behavioral and Brain Sciences, 20: 1-55
Greimas, Algirdas
1987 De l'imperfection. Paris: Pierre Fanlac
Greimas, Algirdas and Fontanille, Jacques
1991 Sémiotique des passions Des étas des choses aux étas d'ames. Paris: Seuil
Langacker, Ronald
1986 Foundations of Cognitive Grammar. Stanford: Stanford University Press
Lakoff, George and Johnson, Mark
1999 Philosophy in The Flesh. New York: Basic Books
Lakoff, George and Nùnez, Rafael
2000 Where Mathematics Comes from: How the Embodied Mind Creates Mathematics. New York: Basic Books
Mac Whinney, Brian
1999 The Emergence of Language from Embodiment. In Brian MacWhinney (ed.), The Emergence of Language. New Jersey: Lawrence Erlbaum
Merleau-Ponty, Maurice
1945 Phénoménologie de la perception. Paris: Gallimard
Nùnez, Rafael
1999 Could the future taste purple? Reclaiming mind, body and cognition. Journal of Counsciosenss Studies. 6 :(11-12) : 41-60
Peirce, Charles S.
1934-48 Collected Papers. Cambridge: Harvard U.P.
Petitot, Jean
2000 Les nervures du marbre. Remarques sur le "socle dur de l'etre" chez Umberto Eco. In: Jean Petitot e Paolo Fabbri (eds.) Au nom du sens. Autour de l'oeuvre d'Umberto Eco. 63-102 Paris: Grasset.
Prinz, Jesse and Barsalou Lawrence
2000 Steering a course for embodied representation. In:Eric Dietrich, Arthur Markman (eds.), Cognitive Dynamics: Conceptual and Representational Change in Humans and Machines, 51-77 New Jersey: Lawrence Erlbaum.
Rastier François, Cavazza Marc and Abeillé Anne
1994 Sémantique pour l'analyse. De la linguistique à l'informatique. Paris: Masson
Rosch, Eleanor
1999 Reclaiming Concepts. Journal of Counsciosenss Studies 6(11-12): 61-77
Schilder, Paul
1935 The Image and Appearance of the Human Body. New York: International Universities Press
Talmy, Leonard
1983 Language structures space. In: Herbert Pick and Linda Acredolo (eds.), Spatial Orientation: Theory, Research, and Application, 225-282., New York: Plenum Press.
1988 Force dynamics in language and cognition. Cognitive Science 12: 40-100
Varela Francisco, Thompson Evan and Eleanor Rosch
1991 The Embodied Mind. Cambridge Mass: MIT Press
Violi, Patrizia
1986 L'infinito singolare. Considerazioni sulla differenza sessuale nel linguaggio. Verona: Essedue
2001 Meaning and Experience. Bloomington: Indiana University Press
