According to the perspective so far delineated, body, actions, and feelings play a direct role in our knowledge of others. The question remains open as to whether our propositional representations are totally separate from this bodily dimension. Our hypothesis is that they are not. But it remains a fact that linguistic and bodily cognition afford us diversified modalities of epistemic access to the world, even though often such modalities contaminate one another and are inevitably interwoven.
The mind, from the perspective delineated here, is therefore an embodied mind, though it would be more correct to speak of a corporeal mind. The concept of embodiment can induce one to think that a mind pre-existing the body can subsequently live in it, and use it. The truth is that mind and body are two levels of description of the same reality, which manifests different properties according to the chosen level of description and the language employed to describe it. A thought is neither a muscle nor a neuron. But its contents, the contents of our mental representations, are inconceivable without our corporeity. Likewise it is difficult to imagine how the representational format of a propositional type can have developed without our corporeality. Language somehow allows us, as we will see, to transcend our corporeity; nevertheless, we posit that the bond with the body is always present.
A few years ago, Gallese (2000) proposed that we look at the evolution of human language as an exaptation[14] of functional sensorimotor processes, which put them into the service of human linguistic competence. The hypothesis of exaptation was then developed in subsequent papers and later elaborated in terms of “neural exploitation” (Gallese & Lakoff 2005; Gallese 2008), or “neural reuse” (Gallese 2014). “Neural exploitation” consists in the reuse of neural resources, originally evolved to guide our interactions with the world, to serve the more recently evolved linguistic competence. This notion of reuse implies a functional uncoupling of the sensorimotor system from muscular output, to guide the generative-syntactic aspects of language by functionally connecting it to the pre-frontal and, more generally, non-sensorimotor circuits. According to this view, intentionality, the aboutness of our representations, is—in the first place—an exapted property of the action models instantiated by the cortical motor system (Gallese 2000, p. 34). The sensorimotor system, when uncoupled from muscular output, makes available to us a model, or paradigm, of our motor knowledge. As such, not only it houses causative properties but also content properties. And this relation to a content, or aboutness, is a primitive expression of intentionality, then exploited by other forms of representations. This perspective on reuse is acquiring more and more supporters (see Dehaene 2005; Anderson 2010 ).[15]
Compelling evidence shows that humans, when processing language, activate the motor system both at the phono-articulatory and at the semantic level. When listening to spoken words or looking at someone speaking to us, our motor system simulates the phono-articulatory gestures employed to produce those very same words. Furthermore, processing action-related linguistic expressions activates regions of the motor system congruent in somatotopic fashion with the processed semantic content. Reading or listening to a sentence describing a hand action activates the motor representation of the same action (for a review, see Gallese 2008; Glenberg & Gallese 2012). Interestingly, somatotopic motor activation has also been observed during the comprehension of abstract and figurative use of language such as metaphors and idioms (e.g., Guan et al. 2013; Boulenger 2012; see also Gallese & Lakoff 2005 on the bodily foundation of concepts). However, it is important to note that embodied simulation is not always involved in language comprehension, and that there is no contradiction in saying this. There are cases in which language, at least at the content level, is not tied to any form of bodily knowledge. In such cases (e.g., when we talk about the notions such as moral judgement or intelligence) no embodied simulation is likely to be at play.
Nevertheless, the problems that language raises for the embodied perspective on human social cognition are still enormous. As clearly underlined, among others, by the Italian philosopher Paolo Virno (2003, 2011), the common linguistic space shared by a community of speakers proves to be incommensurably different from the pre-linguistic one. The linguistic dimension is based on a distinction between linguistic utterances and facts about the world, be they referable to physical or psychological events. We can say that “today the sun is shining” and be understood, even if outside the window snow is falling. Or we can maintain that “all Italians want to pay taxes”, again being understood and simultaneously contradicted by the factual truth of the enormous tax evasion in our country.
According to Virno, the gap between meaning and denotation (what he calls the neutrality of meaning, namely the fact that the meaning of a word such as, for example, “man”, can be understood apart from any reference to an instance of man) is referable to linguistic reflexiveness, i.e., to the fact that language refers to itself and that with words we can talk about other words. It seems to us that the reflexiveness of language is a product of the symbolic nature of linguistic representations. The symbolic nature of such representations is what allows language to break away from the “here and now”; it is what allows the neutrality of meaning.
In order for a sign to be symbolic it necessarily has to be reflexive. What makes a sign a symbol is its being part of a system in which each term is correlatively defined in relation to the other terms within the system and in relation to the renegotiation of this relationship constantly taking place within the system itself. It is the use of a symbol within a given context that each time redefines relationships inside the language system.
Thus, symbolic relationships are by definition characterized by reflexiveness. Symbols are defined through other symbols. This level of reflexiveness is pre-theoretical; it emerges in the linguistic activity of each speaker and leads to a form of linguistic awareness of a practical character. This practical linguistic awareness has been called the epilinguistic quality and thus it has been distinguished from the theoretical quality that is expressed in the metalanguage of linguistics (Culioli 1968; Lo Piparo 2003). The concept of epilinguistic quality refers to the natural tendency of speakers to reflect on their own language—a tendency made possible by the distinctive quality of language being able to speak of itself.
The uniqueness of human language is also maintained by classical cognitivism and by cognitive linguistics, but for very different reasons. The otherness of human language in comparison with other systems of communication known in the animal world derives from its linguistic recursive quality. In an often-quoted article written some years ago (Hauser et al. 2002) defined the faculty of language in a narrow sense (FLN) as being expressed by recursivity. Nevertheless, this perspective, in addition to suffering from the usual cognitivist solipsism, is exposed to comparative verification in the animal world. If the FLN marks human linguistic uniqueness in terms of syntactic recursivity, the latter must be entirely absent in the extra-human animal kingdom.
Actually, the facts tell us exactly the opposite. Recent studies (Gentner et al. 2006; Abe & Watanabe 2011; see also Margoliash & Nusbaum 2009; Bloomfield et al. 2011) have shown that singing species of birds like starlings or finches demonstrate, both in the production and in the reception of conspecifics’ vocalizations, the ability to produce and to extract recursive syntactic characteristics. The study by Abe and Watanabe also shows that the development of this competence is dependent on social encounters with the vocalizations of other conspecific individuals. Finally, these authors have shown that lesion of the lateral magnocellular nucleus of the anterior nidopallium, a motor structure comparable to the basal ganglia of primates, involved both in the production and the perception of song, prevents finches from discriminating the syntactic-recursive characteristics of the song they hear.
These results show that the best strategy for studying some of the most relevant aspects of human social cognition, even demonstrating the bases of their uniqueness, consists in a preliminary recognition of the mechanisms and faculties that we share with the rest of the animal world. As maintained in the past (Gallese 2003b, 2008), the difference between human and nonhuman nature could originally have been of a quantitative rather than a qualitative nature.[16]