CHAPTER 22
THE SOCIAL MIND Francesco Ferretti
According to a classic model, dear to the anthropology, philosophy and linguistics of a good part of the last century, the human mind is a reflection of the social life of the individual. Two aspects characterize this model: the thesis that the (socio-cultural) factors external to the individual have priority over and autonomy from the individual’s internal (bio-psychic) constituents; the idea that the mind is determined through an internalization process of external factors—following a one-way constitutive path “from external to internal”. The very dualism created by both these aspects calls into question the validity of the classic model. The underlying assumption of this chapter is a unitary, rather than dualistic, conception of the human mind. Against the one-way perspective offered by the classic model we shall claim that the mind is a product of a two-way constitutive process, in which the factors which proceed “from internal to external” have the same relevance as those which proceed in the opposite direction. Our idea is that the human mind is the product of factors which are at the same time both external and internal to the individual. In these terms, such an idea appears to represent a banal truth which holds little appeal. There are good reasons, however, for putting forward the theme of the social nature of the mind in these terms. In the first place because attempts to unify past theories of the mind have had little success. In the second place because the difficulties which have given rise to the failure of these attempts would appear today to give way to more convincing solutions. Contemporary cognitive science and neuroscience may effectively offer satisfactory explanatory models of the bio-cognitive devices involved in the management of the complex inter-individual relationships among the members of a group. These devices, as we shall see, are precious tools for the investigation of internal/external relationships and for reconsidering the theme of the unitary nature of the human mind in a new light. 1. THE PRIMACY OF FACTORS EXTERNAL TO THE INDIVIDUAL Two characteristics emerge from the vision of human nature of the theoretical reflection of a good part of the last century. Firstly, the distinction between sciences of the spirit and sciences of nature founded on the autonomy of social reality and, secondly, the primacy of external factors over internal factors. The two characteristics are closely related. They converge at a dualistic conception of human beings: to consider individuals the product of external (socio-cultural) factors is equivalent to considering biology as a merely marginal aspect of their nature. Only 295 M. Marraffa, M. De Caro and F. Ferretti (eds.), Cartographies of the Mind, 295–308. © 2007 Springer.
296
FRANCESCO FERRETTI
the sciences of the spirit, acting in complete autonomy from the natural sciences, are able the capture the true essence of the human being. 1.1
Autonomy of social reality
According to Émile Durkheim, social reality should be analyzed in its pure form. Social realities have an autonomous and independent status and the work of the sociologist must be aimed at distinguishing two orders of phenomena: individual realities which embody social reality and social reality intended as “a reality sui generis vastly distinct from the individual facts which manifest that reality”.1 Social realities are autonomous and antecedent to individual psychic states: TP
PT
they consist of manners of acting, thinking and feeling external to the individual, which are invested with a coercive power by virtue of which they exercise control over him. Consequently, since they consist of representations and actions, they cannot be confused with organic phenomena, nor with psychical phenomena, which have no existence save in and through the individual consciousness. Thus they constitute a new species and to them must be exclusively assigned the term social. It is appropriate, since it is clear that, not having the individual as their substratum, they can have none other than society […].2 TP
PT
Imposition through education highlights the unidirectionality of the constitutive process: social reality pervades individuals making them social agents. The circle is closed by a precise idea of the mind and, more generally, of human nature. If education is something which moulds the individual, the individual who is socialized through education is a plastic and indeterminate one. Individual natures, from this point of view, only represent “the indeterminate material which social reality determines and transforms”.3 The theory of the plastic and indeterminate nature of human beings was backed up by psychology, anthropology and philosophy for a good part of the twentieth century. The radical version of this theory is known as the Sapir-Whorf Hypothesis. Two assumptions characterize it. Firstly, “linguistic determinism”, according to which the thoughts of individuals are determined by the categories of their language; and secondly, “linguistic relativism”, for which different languages determine different thoughts. Tooby and Cosmides define “Standard Social Science Model” this conception of the human mind.4 We shall not examine in detail the criticisms which it is possible to level at such a conception, suffice it to say, for the purposes of this chapter, the important point is the relationship between the primacy given to external factors and the nature of the mind. When the path uniquely proceeds “from external to internal”, what is indeed required is a markedly plastic and indeterminate mind. But can the mind really be thematized in this way? The only model of functional architecture which the Standard Model appears to refer to TP
PT
TP
PT
THE SOCIAL MIND
297
is something similar to the tabula rasa, an hypothesis on the mind which is literally hopeless.5 For Tooby and Cosmides the “impossible psychology” on which the Standard Model is based, lies in the dualistic distinction between sciences of nature and sciences of the spirit. It is, therefore, towards overcoming this dualism that a conception of the mind which has an effective explanatory value should aim. TP
1.2
PT
Unifying approaches
Linguistic determinism and linguistic relativism are in marked decline, as is the concept of the tabula rasa. And the idea that the time has come for a reconciliation between natural sciences and sciences of the spirit has become the prevailing trend. Some sectors of cognitive science, are now looking with interest at the work of Vygotskij, an author who attempts to put forward a unitary vision of the mind and the human being.6 For Vygotskij, the specificity of the human being is guaranteed by the acquisition of speech. It is this event which determines a completely new organization of the child’s thought. Thus, since speech is an instrument external to the individual, produced by socio-historical exchanges among peoples, the growth which leads to the development of higher cognitive functions in the child is considered as a process of internalization. It is the nature of this process which is of interest here. It is known that “egocentric language” constitutes the decisive mediation device through which social language as an instrument for communication becomes a tool for thought: TP
PT
The great change in children’s capacity to use language as a problemsolving tool takes place somewhat later in their development, when socialized speech (which has previously been used to address an adult) is turned inward. Instead of appealing to the adult, children appeal to themselves; language thus takes on an intrapersonal function in addition to its interpersonal use. When children develop a method of behavior for guiding themselves that had previously been used in relation to another person, when they organize their own activities according to a social form of behavior, they succeed in applying a social attitude to themselves. The history of the process of the internalization of social speech is also the history of the socialization of children’s practical intellect.7 TP
PT
The socialization of thought is, thus, closely linked to the internalization of social language. All higher psychological functions, in fact, should be considered in terms of internalization of social relationships: even psychological factors within the individual are the result of internalized social relationships. The primacy of external factors over internal factors constitutes the basis of the influence of Vygotskij in some sectors of cognitive science. Dennett and Clark, for instance, use the concept of external scaffolding to move the computational
298
FRANCESCO FERRETTI
weight of processing towards the external, in this way removing structures and internal functions from the mind.8 According to these authors, the external scaffolding par excellence is public language, a system of symbols which individuals find ready-made (outside themselves) at birth and which, through an acquisition process, invade their minds. Typically human capacities such as selfreflection and metarepresentation depend on this invasion. Dennet claims that selfreflective thought could never exist without public language; in particular he underlines that capacities such as self-reflection could never exist without a constitutive process which “begins with the overt, public use of symbols and tokens of one sort or another […] and creates practices that later can be internalized and rendered private”.9 The idea that the process is activated by the public use of symbols and is constituted as a phenomenon which is to all intents and purposes an invasion deserves particular attention. To what extent does a theory of the sort stand up to scrutiny? Since the words, rather than the language, are at the center of the theory of Clark and Dennett, it is legitimate to wonder if lexical acquisition models available today conform in one way or another to the theory put forward by the two authors.10 The answer is negative. The prevalent empirical models today go in exactly the opposite direction. Data from cognitive psychology highlight the fact that, among other factors, the acquisition of words is guided by strong internal constraints.11 We have no empirical data in support of the theory in which words invade the mind. The acquisition of lexis requires minds which are internally rich and structured in a complex manner. What Vygotskij and followers attempt to do in bringing about the convergence of mind and society is still within a dualistic conception of the mind, it is still a conception which regards social reality as an autonomous entity outside the individual. The price to pay in maintaining the primacy and autonomy of external factors is, as we have seen, reference to a plastic and indeterminate mind. Yet, such a conception of the mind is not supported by the facts. The great majority of models available today relative to the acquisition, the functioning and the pathology of cognitive skills conceives the mind as a rich and complex whole of processing devices. Thus, since a conception of the mind unsupported by empirical data fails to provide a feasible basis, it is necessary to find alternative paths. Our thesis is that only examining the possibility of a dual constitutive process, proceeding “from the external to the internal” and vice versa, it is possible to truly bring about the integration of mind and society. Now is the time to take into serious consideration the role which factors inside the individual play in the creation of social interaction. TP
TP
PT
PT
TP
TP
PT
PT
2. AN ALTERNATIVE CONSTITUTIVE PROCESS: THE BIO-COGNITIVE FOUNDATIONS OF SOCIETY The human being is not the only social animal. The strategy for understanding the social nature of human beings starting from their peculiar characteristics holds true only on condition that it should be taken together with the study of those
299
THE SOCIAL MIND
characteristics which humans share with other social animals. Society is not an abstract entity. Every society is constituted by the number of individuals which make it up and the network of relationships which they create in relating to each other. From this point of view, human society too, while undoubtedly possessing its own specificity, cannot be understood by abstracting it from the bio-psychological characteristics of the individual: both the number of individuals and the kind of relationships which they can form in exchanges within the group are in fact constrained by the nature of the bio-cognitive system which characterizes each social animal12. With respect to single individuals, the group solves a multiplicity of adaptive problems. Nevertheless, the life of the group also creates tensions: social relations require a nervous system able to manage them. Dunbar underlines the limitation which the volume of the cerebral cortex imposes on the dimensions of social groups.13 The role of the nervous system in the management of relations with other individuals emerges clearly when, following Berthoz, the brain is considered as a machine essentially assigned to predict future, to anticipate the consequences of one’s own or other people’s action, to “buy time”.14 The most advanced evolutionary species are those which have learned to buy time on others, both in hunting prey and in anticipating the attack of predators. The brain is primarily a biological machine with which to stay one step ahead of the others. The ability to anticipate the action of others is the characteristic trait of social intelligence. The difference between the solution of problems imposed by the physical environment and competition with the other individuals who compete for the solution of the same problems is crucial from the point of view of adaptation. Humphrey shows that the genesis of the social intellect is the true evolutionary leap forward in the development of human intellectual capacity.15 With the expression “Machiavellian intelligence”, Byrne and Whiten refer to one of the most pressing evolutionary challenges with which primates had to deal, i.e. the ability to predict and control the behavior of the others, using them as a means to their own ends.16 Possessing a social intelligence means possessing a system of comprehension and prediction for behavior, that is to say a system of interpretation. Now, what is it to interpret behavior? Let us take the case of human beings. Their way of interpreting behavior appears to conform to what Dennett defines the “intentional stance”, i.e. the ability to understand and predict the behavior (one’s own or that of others) by attributing mental states to the agent.17 This is the ability to “mentalize” the behavior or, as is also said, of “mindreading”. In order to highlight the fact that mentalization comes about through a complex and integrated system of knowledge relative to the intentional states of the agent, Premack and Woodruff call this capacity Theory of Mind (ToM).18 Even though the two authors recognized that chimpanzees also possessed such capacities, the prevailing theory today is that ToM: “represents a sort of ‘mental Rubicon’, sanctioning the uniqueness of human cognitive capacities”.19 Effectively, speaking of ToM appeals to a system of propositional structures underlying a sophisticated metarepresentational system which cannot easily be hypothesized for animals which are not human. TP
TP
PT
PT
TP
PT
TP
TP
TP
PT
TP
PT
TP
PT
PT
PT
300
FRANCESCO FERRETTI
Accounting for the capacity to mentalize with exclusive reference to ToM appears to be largely fruitless, since it clearly divides humans from animals and reintroduces unacceptable forms of discontinuity from a naturalistic viewpoint. According to Gallese, the theory that considers the attribution of intentional states “solely determined by metarepresentations created by ascribing propositional attitudes to others, is biologically implausible”.20 So how may we account for the capacity to mentalize typical of human beings from a continuist point of view? To start with, mentalization does not coincide with ToM. It is possible to hypothesize abilities to interpret behavior which do not use propositional attitudes. The simulation theory, which views the interpretation of behavior as putting oneself in the shoes of who carries out the action, offers an important contribution to the study of mentalization from a continuist perspective.21 Some cognitive skills of great social relevance, such as imitation and empathy can, in effect, be explained on the basis of an interpretative model, which is particularly close to the theme of evolutionary forerunners, founded on assumptions different from those used for ToM. One such model has on its side significant empirical backing. Rizzolati et al. have discovered a class of neurons, the “mirror neurons” identified initially in the premotor cortex of macaques, active during the execution of targeted actions.22 These neurons discharge each time the macaque takes a certain object independent, for example, of the fact that it takes the object in its hand or mouth. The common denominator of behaviors of this kind does not consist of the movement that constitutes the action but in its aim. The mirror system, which humans also possess, is activated not only during the execution of an action but also during the observation of someone who is carrying out that action. In these cases, those who observe the action of another do not effectively carry out the observed action. Nevertheless, the motor system is activated as if this were the case: the perception of an action implies the simulation of that action.23 Eliminating the distinction between who interprets the behavior and who effectively acts, the simulation model offers an alternative explanation to the metarepresentational model. The implications of such a move on the level of social intelligence find confirmation in two important cognitive skills: imitation and empathy. Already at 12-21 days infants are able to imitate the extension of the tongue and opening the mouth. The reason for which imitation can be considered a forerunner of the ability to mentalize is that it presupposes the reading of behavior based on the aims, intentions and desires which have determined it.24 The most interesting aspect is that these first forms of imitation give rise, as Gallese claims, to the formation of a “we-centric” space: TP
PT
TP
PT
TP
TP
PT
TP
PT
The importance of early imitation for our understanding of social cognition is that it shows that interpersonal bonds are established at the very onset of our life, when no subjective representation can yet be entertained by the organism because a conscious subject of experience is not yet constituted. The absence of a self-conscious subject does not preclude, however, the constitution of a primitive “self-other space”, a
PT
THE SOCIAL MIND
301
paradoxical form of intersubjectivity without subjects. The infant shares this “we-centric” space with the other individuals inhabiting his world.25 TP
PT
The same mechanisms on which imitation is based also underlie empathy (Gallese 2003c). Observing the actions of others leads to the sharing of emotions and feelings which go beyond the conceptual classification of an action. The understanding of the affective expressions of behavior comes about automatically and without the mediation of higher cognitive faculties. It is once again the reference to embodied simulation which provides an explanatory key to empathy. This reference has more general repercussions on the nature of the mind. Rather than a metarepresentational mind, which interprets the others (taken as external subjects) through the filter of the conceptual system underlying ToM, a conception of the mind as a “resonance mechanism” emerges, a system which simulates within itself what happens externally (Rizzolatti et al. 2002). Gallese is right to say that the exclusive reference to the metarepresentational system fails to exhaust the possibities, human too, of mentalizing behavior. And is right above all in claiming that the existence of ToM can only be in reference to a series of cognitive precursors which humans and animals have in common. Nevertheless, it is not clear what the role of ToM is in the model he proposes. Certain claims seem to admit that ToM can coexist with the capacity to mentalize based on the embodied simulation model: claiming that evolutionary forerunners to ToM exist means in fact admitting the existence of ToM; claiming that metarepresentations are not the only way of mentalizing, signifies that they are, nevertheless, a means for doing so. Other claims seem to go in the opposite direction: in noting that ToM is based on a “disembodied” model of the mind, for example, Gallese maintains that the empirical proof available today leads us to conclude that such a model is biologically implausible. With regard to this last point there are three considerations to be made. The first is that, although the simulation theory offers important indications concerning the cognitive components underlying some relation-building capacities, it is clear that these capacities do not exhaustively explain the way in which human beings relate to one another. To say that human beings need to make use of a common hereditary system common to that of other animals is true, but it still leaves work to be done. It fails to explain what type of specific relationships humans are able to establish in their interpersonal exchanges. When human sociality is the object of analysis, as well as elements of continuity it is also necessary to study the peculiarities which characterize it. The second consideration regards a more general view of what the mind should be. The idea which emerges from a reading of Gallese is that thinking of the mind as a resonance mechanism appears to come into irresolvable conflict with thinking of the mind as a representational-symbolic system. Our idea is that the two perspectives are not mutually exclusive. We shall not examine this question in detail here.26 The third consideration concerns the relationship between metarepresentation and a discontinuist viewpoint. The theory that we shall put forward is that it is possible to tackle the theme of the phylogenesis TP
PT
302
FRANCESCO FERRETTI
of metarepresentations and that, therefore, taking a metarepresentational stance does not mean in itself adhering to a discontinuist conception. ToM represents the peculiar way in which humans interpret behavior. This does not mean that ToM is their only way of mentalizing actions or that ToM can disregard cognitive precursors common with other animals. It means, however, that its specificities await plausible explanatory models. The typical nature of human mentalization is given by the role of language in constituting a specific metarepresentational system. The analysis of the relationship between language and mentalization open the road to the possibility of testing a conception of the mind in which a bidirectional constitutive model is called into question. Dealing with this aspect of the problem we shall demonstrate that the road towards a truly unitary perspective on the relationships between mind and society is that of co-evolution between external and internal factors. Only such an hypothesis can account for both the continuity and the specificity of human cognitive systems. 3. METAREPRESENTATIONAL CONTINUISM Although the metarepresentational system underlying ToM is used as proof of a clear-cut demarcation between human beings and other animals, our idea is that considering ToM as the mentalization system specific to humans does not necessarily mean adhering to a discontinuist viewpoint. Evidence to support this idea comes from studies on the phylogenesis of mentalization. The first data to emerge from these studies is that, effectively, the relationship between metarepresentation and primary representation cannot be thematized in terms of a clear-cut opposition. There are different levels and different forms of metarepresentation. Given that the metarepresentations used in attributing propositional attitudes are typical of humans, it does not, therefore, follow that nonhuman animals relate to their environment only through primary representations. The cue for arguing in this direction is offered by a series of articles which analyze the genesis of metarepresentations starting from the case of non-human animals.27 More specifically, a further cue is offered by an essay by Suddendorf and Whiten in which continuists are invited not to get too depressed about things.28 In the first place a terminological question. Following Perner,29 it is opportune to distinguish two different senses of “metarepresentation”: TP
TP
PT
PT
TP
PT
1. intended as representation of a representation; 2. intended as representation of a representation as representation. Traditionally, metarepresentation in its full sense is that defined in 2. The test of the false belief is the test considered valid in order to identify the presence of representational states of this type. In order to pass this test, in fact, as Perner underlines, one has to master the distinction between what a representation represents and how it represents something as something (how it represents something as being a certain way). Attributing the capacity to mentalize using the false belief test can, however, be a restrictive way, calibrated to human capacities, to
303
THE SOCIAL MIND
pose the question. Since the false belief test imposes constraints which are too strict, as animals cannot pass it, it does not follow that they are unable to employ metarepresentation. It follows only that they are not able to employ metarepresentation in its full sense. An interesting intermediate level of metarepresentation is that of secondary representations. They are characterized by importantant properties, such as the capacity to “suspend”, through the process of decoupling described by Alan Leslie in the pretend play,30 the normal causal constraints which link primary representations to the outside world. This suspension makes secondary representations able to stand for hypothetical or virtual situations, situations for which it is possible to give multiple and alternative pictures, which may be in themselves contradictory, of a fact or an object, such as the possibility in pretend play of representing simultaneously an object for what it is (let us say a banana) and what we pretend it is (a telephone or a pistol). According to Suddendorf and Whiten, such properties have significant repercussions on the nature of the mind: TP
PT
The advantage of simultaneously entertaining a second model is, of course, the ability to collate the two—that is, to bring them into propositional relation. One of us therefore coined the term collating mind to describe the kind of mind that uses secondary representations (Suddendorf 1999). It is the ability to consider a mental model of a situation not currently perceived (be it past, future, or hypothetical). Although this capacity may be a prerequisite to becoming a higher order intentional system, it is a more general skill that can be applied to various areas other than mind reading.31 TP
PT
A collating mind is capable of sophisticated conceptual activity. Not only a primary understanding of “mind” (from which epistemological states are, however, excluded), but also an understanding of the persistence of the object, the means-toan-end reasoning, the empathy, the imitation, the interpretation of external representations and the recognition of self through the mirror. Experimental data show that such capacities, which are present in an analogous way in both the 2-yearold child and the great apes, are justifiable with regard to secondary representations. These representations can be considered a fundamental developmental bridge in the passage from primary representations to metareprentations in their full sense. Data confirming the argument for “evolutionary parsimony” is homology, rather than analogy, as the keystone in understanding the properties of a certain species. From this point of view, secondary representations are characteristics which the great apes share with humans, having inherited them from a common ancestor. A final point remains to be discussed. What allows the passage from a system of mentalization based on secondary representations to a mentalization system able to attribute epistemological states and, above all, false epistemological states? How is it possible to pass from a generic and shared ability to mentalize to that specific capacity, regulated by propositional attitudes, represented by ToM?
304
FRANCESCO FERRETTI
4. CO-EVOLUTION The peculiar trait of the Theory of Mind is the capacity to interpret behavior in a context of false beliefs. In order to understand such behaviors: “the particular content of the propositional attitude ascription is critical to the understanding, because it is content that is not simply reflected in the observer’s own world view”.32 The decisive role played by the ascribed propositional attitudes of this type is due to a certain class of linguistic constructs, i.e. complement structures, sentence structures in which the object of the main mental state verb (think, believe, guess, and so on) comprises a subordinate clause, as in (1): TP
PT
(1) Mary believes that John wants a can of Coca-Cola. Thus, the point which links language to the theory of mind is that very possibility, designated to the syntax, of constructing complement structures: The complement structure invites us to enter a different world, the world of the girl’s mind, and suspend our usual procedures of checking truth as we know it. In this way, language captures the contents of minds, and the relativity of beliefs and knowledge states. These sentence forms also invite us to entertain the possible worlds of other minds, by a means that is unavailable without embedded propositions. Pictures cannot capture negation, nor falsity, nor the embeddedness of beliefs-that-are-false, unless we have a propositional translation alongside. So this special property of natural languages allows the representation of a class of events—the contents of others’ minds—that cannot be captured except via a system as complex as natural language.33 TP
PT
Only language possesses the representational tools necessary to enable the cognitive system to ascribe mental states endowed with propositional contents. The empirical proof of this hypothesis can be found in the case of oral deaf children. Experimental results highlight a close link between their impoverished syntax and the incapacity to ascribe intentional states in the description of events. The case of oral deaf children supports the idea of the constitutive role of language in thought; as de Villiers and de Villiers claim: “an individual with less language or no language would not be able to formulate the appropriate representation of another person holding a false belief and hence would have no basis for reasoning about that person’s actions”.34 Following the same path is Tager-Flusberg.35 Also in her hypothesis, the syntactic structures underlying complementation precede the ability to pass the false beliefs test and must, therefore, have a foundation of deep cognitive devices. Thus, since speaking of language, according to the authors cited above, means referring to public language, i.e. historico-natural languages, it may be possible to maintain that the empirical data provided by them reinforce the theory of language as external scaffolding. Should we, thus, newly admit the primacy of TP
PT
TP
PT
305
THE SOCIAL MIND
external factors over internal factors? And once again, must we return to the old theory of the one-way constitutive process? One way of answering the question is to ask oneself which of the two components, mentalization or language, precedes the other, in other words what factor is constitutive of the other.36 In very simple terms: for language to carry out the role it carries out in its capacity to mentalize, there must be a language. The hypothesis of the priority of language over mentalization comes up against enormous difficulties concerning the genesis of language. This is for a simple reason. When the evolution of language is investigated from more generic forms of communication, shared with other animals, reference to a metarepresentational system seems to be an unavoidable step to take. Sperber and Wilson, following Grice, place the intention of the speaker as the basis of the comprehensionproduction of language.37 Reference to the intention of the speaker highlights the difficulties of the code model (according to which communication is achieved by encoding and decoding messages) in explaining comprehension-production processes in verbal communication. Furthermore, it demonstrates the impossibility of explaining these processes without a device which is able to read the minds of others. Empirical proofs supporting this theory come from the difficulties shown by autistic subjects—affected by mindblindness38—in cases of communication where, as in metaphor and irony,39 the gap between what the speaker says and what the speaker intends is evident. If language is characterized by the presence of a mindreading device, the evolutionary passage from communication to language must, therefore, presuppose such a device. In other words, mentalization must precede language. Indeed, as Origgi and Sperber claim: TP
TP
PT
PT
TP
TP
PT
PT
The function of linguistic utterances […] is—and has always been—to provide this highly precise and informative evidence of communicator’s intention. This implies that language as we know it developed as an adaptation in a species already involved in inferential communication, and therefore already capable of some serious degree of mind-reading. In other terms, from the point of view of relevance theory the existence of mind-reading in our ancestors was a precondition for the emergence and evolution of language.40 TP
PT
Verbal communication rests upon a mechanism originally ascribed to the interpretation of behavior. Only admitting such a device is it possible to account for the advent of language; only admitting this advent is it possible to account for the role of language in the constitution of ToM. Moreover, without the capacity to mentalize, it is possible neither to explain the ontogenesis nor the effective use of language, i.e. the ability to understand and produce linguistic expressions. In other terms it is possible to explain hardly anything about what makes language language. Maintaining the primacy of the mentalization system over language does not exclude the fact that language, once constituted, will have rebound effects on the capacity to mentalize. Complementation, as we have seen, is a significant case in the
306
FRANCESCO FERRETTI
role of language in constituting specific relationship-building skills among human individuals. Sustaining such a theory is not, however, equivalent to a return to the past, revisiting the theory of Vygotskij. Indeed, the dependency of language on the mentalization system excludes the possibility of calling back into play the theory of the primacy of external factors over internal factors. The relationship between language and mentalization must be considered in terms of a co-evolution process. 5. CONCLUSION Investigating the social nature of human beings signifies investigating the specificities which mark them out. Since human beings are not the only social animals, nevertheless, the peculiarities which characterize them cannot be analyzed independently from what they share with other social animals. In order to explain the unitary nature of human beings it is necessary to account at the same time for the elements of continuity and specificity which characterize them. The social mind represents a useful testing ground for this hypothesis. The role of language in constituting the specific mentalization tool which human beings employ is possible, in fact, only by admitting that language itself uses mentalization systems, in its genesis and in its common use, which are phylogenetically antecedent. Language, the principal instrument for the management of interpersonal relations, perfectly embodies the coexistence in humans of peculiar traits and lines of continuity. Attention towards the bio-cognitive foundations of social relations brings out clearly the fact that, beyond declarations of intent, all the hypotheses which, privileging the role of factors external to the mind, continue to appeal to a one-way constitutive process, cannot account for the unitary nature of the human being. Only by accepting the idea of coevolution between external and internal factors is it possible to put forward a conception of human nature in line with a naturalistic viewpoint. As it is easy to imagine, there is still much to be done in this direction. As, concretely and in detail, it is possible to realize the idea of coevolution between external and internal factors is a task which awaits effective explanatory models. At least from a methodological point of view, however, the way is laid. The analysis of the bidirectional constitutive process seems to be the only way to avoid considering the human being in dualistic terms and, therefore, to lay a final stone on the distinction between natural sciences and sciences of the spirit.41 TP
NOTES 1
Durkheim ([1895] 1982, p. 54) Ibid., p. 52. 3 Ibid., p. 132. 4 Tooby and Cosmides (1992). 5 See Pinker (2002). 6 See Frawley (1997). TP
PT
2 TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
PT
THE SOCIAL MIND
7
307
Vygotskij (1978, p. 27). Dennett (2000); Clark (2003). 9 Dennett (2000, p. 21). 10 In restricting the field of investigation solely to the acquisition of lexicon, much credit must go to Dennett and Clark. Putting forward a theory of language acquisition, of all the language, in terms of “invasion” would be a simply unsustainable theory. After Chomsky, every hypothesis of language acquisition must take into account the poverty of the stimulus argument, an argument which appears to exclude the idea of plastic and indeterminate minds (see this volume, p. 7). 11 Bloom (2000, 2004); Landau and Gleitman (1994). 12 Claiming that social relations must be constantly constrained by the analysis of the biocognitive mechanisms which make them possible, we admit, in point of fact, an asymmetric relationship between internal and external factors. In this sense, despite the distinction of the different level of analysis, our investigation is in agreement with the thesis of Kostko and Bickle (this volume, chapter 23), according to which in order to speak in a proper sense of the causal role of social factors in behavior, it is necessary to have a theory of the neurological mechanisms which apply them. 13 Dunbar (1993, 1996). 14 Berthoz ([1987] 2002). 15 Humphrey (1975). 16 Byrne and Whiten (1988); Whiten and Byrne (1997). 17 Dennett (1978). See also this volume, p. 3. 18 Premack and Woodruff (1978). 19 Gallese (2004, p. 165). 20 Ibid., p. 166. 21 Gordon (1996). 22 Rizzolati et al. (1988) 23 Gallese (2004). 24 Meltzoff (2002); Gattis et al. (2002). 25 Gallese (2004, p. 162). 26 For a closer analysis see this volume, chapter 21, in which the compatibility of some simulation-theory models with some theory-theory models is demonstrated. 27 Cosmides and Tooby (2000); Sperber (2000); Suddendorf (1999); Whiten (2000); Whiten and Byrne (1991). 28 Suddendorf and Whiten (2001). 29 Perner (1991). 30 Leslie (1987). 31 Suddendorf and Whiten (2001, p. 630). 32 de Villiers and de Villiers (2003, p. 337). 33 de Villers (2000, p. 90). 34 de Villiers and de Villiers (2003, p. 338). 35 Tager-Flusberg (2000). 36 See Sperber (2000). TP
PT
8 TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
TP
PT
308
37
FRANCESCO FERRETTI
Sperber and Wilson (1986, 1995, 2004). Baron-Cohen (1995). 39 See Happé (1995). 40 Origgi and Sperber (2000, p. 165). 41 Many thanks to Giovanni Iorio Giannoli and Thomas Suddendorf for helpful comments on previous versions of this chapter. TP
PT
38 TP
PT
TP
PT
TP
PT
TP
PT
