[ programme | outline | short biography ]
JEAN-NICOD LECTURES 2011
Outline of the lecturesGergely CSIBRA
György GERGELY
Natural Pedagogy
Lecture 1: Non-Verbal Generics
Non-verbal demonstratives, like pointing to something or showing something up, pick out their referents as particular objects or sets of objects individuated by spatial means. Despite of this, we have repeatedly found in studies with human infants, children, and adults that, when such gestures are used ostensively and the context does not suggest otherwise, they are taken to refer to the object kind that the referent represents rather than to the particular object present in the context. This paradoxical phenomenon of non-verbal demonstrative reference to kinds parallels some core properties of generic linguistic expressions. In particular, generic constructions are unmarked, there is a 'default' bias towards generic interpretation of ambiguous sentences, and the predicates of generic expressions are expected to reflect essential, kind-relevant properties that are tolerant to counterexamples. We argue that non-verbal demonstrative reference shares these properties, which reflect fundamental design features of the cognitive mechanisms subserving human ostensive communication.
Lecture 2: Opaque Knowledge
Instrumental actions performed by non-human primate species pursue a restricted range of goals that directly satisfy basic biological needs of the actor, and employ means that are transparently related to the physical properties of the goal object and the visible situational constraints on action. In contrast, several properties of human goal-directed actions and culturally transmitted instrumental skills render them causally or teleologically ‘opaque’ for a naïve observational learner. Human actions may be teleologically opaque because they are directed to goals that are far detached from their immediate biological needs and because they involve multiple outcomes. Humans also frequently perform hierarchically organized sequences of goal-directed actions where local sub-goals can be identified only in terms of their relation to the final goal. Actions can also be causally opaque because the means they adopt or the manner in which they are performed leaves it unclear to the observer how (or whether) they are causally connected to the ends they achieve. The causal and teleological opacity of human instrumental actions represents a non-trivial learnability problem for purely observational social learning mechanisms. We propose that this learnability problem led to the evolutionary selection of a new type of social learning mechanism provided by communicative action demonstrations, which highlight and actively guide the learner’s attention to the relevant aspects of the cognitively opaque instrumental actions to be acquired. We argue that both producing and interpreting such communicative knowledge demonstrations rely on specific adaptations. Evidence on observational learning of novel instrumental actions by human infants versus non-human primates supports our proposal for a human-specific system of cultural knowledge transfer based on ostensive communicative demonstrations.
In order to learn from others by communication, human infants must interpret certain acts as communicative in nature. We propose that there are at least two ways by which infants are prepared to receive information from others. First, they possess a skeletal format for representing communicative intentions, which are understood as second-order intentions referring some further informative intentions. Setting up such representations is triggered by the reception of ostensive signals, such as eye contact and special intonation, and some of these signals are innately specified. Second, infants assume the referential nature of the signals coming from the source of ostension. Thus, while they have to learn how (verbal and non-verbal) signals refer, they do not have to discover that they do so. We present empirical evidence supporting both proposals.
Ostensive communication evolved as a species-unique form of epistemic cooperation in humans. Communication can induce epistemic gain both by means of ostensive reference to relevant episodic information about a particular referent (when the relevance applies to the ‘here-and-now’ only) or by manifesting relevant generic knowledge about referent kinds. In spite of this symmetry in functional use, we frequently find that non-verbal ostensive communication is assumed to make reference to, and manifest relevant knowledge about, kinds rather than particulars. We hypothesize that this built-in ‘genericity bias’ is a design feature of ostensive communication that is an evolutionary signature reflecting the specialized function that ostensive communication may have been selected for: the fast and direct (non-inductive) transfer of generic and socially shared cultural knowledge about kinds. It is argued that this cognitive adaptation was crucial for making cognitively opaque cultural knowledge easily learneable and efficiently transferable across generations. We propose that by having evolved specific cognitive biases, human infants are prepared to be at the receptive side of communicative knowledge transfer, which, together with adults' inclination to pass on their knowledge to the next generation, constitute a system of 'natural pedagogy' in humans.
Further information:
Sophie Bilardello
Institut Jean-Nicod
Ecole Normale Supérieure
Pavillon Jardin
29, rue d'Ulm
75005 Paris