Published online by Cambridge University Press: 01 February 2004
Theories of children's developing understanding of mind tend to emphasize either individualistic processes of theory formation, maturation, or introspection, or the process of enculturation. However, such theories must be able to account for the accumulating evidence of the role of social interaction in the development of social understanding. We propose an alternative account, according to which the development of children's social understanding occurs within triadic interaction involving the child's experience of the world as well as communicative interaction with others about their experience and beliefs (Chapman 1991; 1999). It is through such triadic interaction that children gradually construct knowledge of the world as well as knowledge of other people. We contend that the extent and nature of the social interaction children experience will influence the development of children's social understanding. Increased opportunity to engage in cooperative social interaction and exposure to talk about mental states should facilitate the development of social understanding. We review evidence suggesting that children's understanding of mind develops gradually in the context of social interaction. Therefore, we need a theory of development in this area that accords a fundamental role to social interaction, yet does not assume that children simply adopt socially available knowledge but rather that children construct an understanding of mind within social interaction.
1. It is not clear that an enculturation approach is directly championed by any theorist within the “theories of mind” literature, but such an approach is discussed by many (e.g., Astington 1996; Astington & Olson 1995) and sometimes attributed to Wittgenstein or Bruner (1990).
2. Rogoff (1997, p. 266) criticized internalization approaches and argued instead for “development as a process of transformation of participation.” Although there is much that we agree with in Rogoff's approach, her view of internalization involves either the transmission or acquisition of information (corresponding to collectivism or individualism, respectively) – both being examples of what Overton (1994; 1998b) termed a “splitting” or “isolation” strategy. Rogoff (see also Matusov 1998) critiqued a particular view of the nature of internalization. We agree with the critique. However, there is a second view of internalization (Lawrence & Valsiner 1993), according to which internalization involves the child's reconstruction of knowledge rather than transmission. Interaction is internalized (or “interiorized,” in Piaget's terminology) to the extent that the child can implicitly perform the act and does not actually have to enact the activity (Carpendale et al. 1996; Chapman 1991; 1999).
3. There are other positions such as Thelen and Smith's (1994; Thelen et al. 2001) dynamic systems approach that are consistent in many ways with the approach we take. Thelen and Smith's account is similar to Piaget's theory in emphasizing the practical and embodied nature of cognition. And Chapman (1992; see also Boom 2004) suggested that Piaget's theory was an early theory of self-organization. However, there may also be important differences. In Thelen and colleagues’ (2001) explanation of the A not B error they appear to assume objectivity, whereas that is what Piaget tried to explain (Müller & Carpendale 2001). This suggests that they started from a different beginning point in development.
4. We cannot survey here the extensive literature on Wittgenstein's private language argument, but we do need to at least respond to Kripke's (1982) interpretation. Kripke began from Wittgenstein's claim that all rules are up for interpretation. This is the paradox that “no course of action could be determined by a rule, because every course of action can be made out to accord with the rule” (Wittgenstein 1953/1968, para. 201). From this, Kripke derived the radically skeptical position that there can be no such thing as rule following, and this also applies to language since word use is rule following. This implies that the apparent meaningfulness of language must be an illusion and language must be meaningless. Kripke ended up in this radically skeptical position because he refused to give up the idea that rules are interpreted. But Wittgenstein was, in fact, arguing against this external explanation that there is a middle step of interpretation in applying a rule. Ironically, Wittgenstein was setting up the paradox that the range of interpretation is infinite to show that rule following cannot consist of interpretation – that “interpretations by themselves do not determine meaning” (para. 198). In the same section that Kripke focused on, Wittgenstein went on to write, “It can be seen that there is a misunderstanding here… . What this shows is that there is a way of grasping a rule that is not an interpretation” ( para. 201). Then, in paragraph 202, “And hence also ‘obeying a rule’ is a practice” (McDowell 1984; Russell 1987). Wittgenstein's position is internalist: that is, it is not possible to separate the rule and the application (ter Hark 1990, Chap. 3).
5. Triadic interaction here refers to interaction involving self, other, and the world, not to interaction among three people.
6. However, Trevarthen took an innatist approach to explaining this development, whereas we do not.
7. We recognize that reference is a controversial issue (Putnam 1988). To be clear about our position, we do not consider reference to involve a dyadic relation between a sign and the thing referred to. This would seem to require a mechanistic view of meaning in which meaning is assumed to be attached to representations (Goldberg 1991). Instead, we consider reference to involve a triadic relation between the self, others, and an aspect of the world (Carpendale 1999b; Chapman 1991; 1999; Sinha 1999). Thus, meaning is not fixed to signs, but signs are used to direct others’ attention and shared meaning is achieved through ongoing social interaction (Turnbull 2003; Turnbull & Carpendale 1999b).
8. Variability in children's performance on false belief tasks is reminiscent of similar debates in other areas of research. The most well-known example is criticism of Piaget's theory concerning “horizontal decalage.” The standard interpretation of Piaget's theory was as a theory of mental logic, according to which reasoning involves the application of a logical rule. A prediction derived from this interpretation is that, once a child has developed such a rule, he or she should be able to solve all problems based on the same underlying logical rule. It is well known, of course, that there is considerable evidence of variability in children's performance on tasks that are all apparently based on the same logical principle. The research literature provides other examples of similar difficulties in explaining evidence of variability in performance on different tasks that apparently should all be assessing the same competence. For example, in research on Kohlberg's theory of moral development, much more variability in the stage of moral reasoning employed was found than had been predicted by Kohlberg (e.g., Carpendale 2000). A particularly good example of the same sort of issue arose in the role-taking literature (e.g., Chandler 2001).
As argued elsewhere (Carpendale et al. 1996; Chapman 1987b), horizontal decalage is only problematic when it is assumed that Piaget's theory is a theory of mental logic. From the perspective of an interpretation of Piagetian theory emphasizing the origin of knowledge in action, horizontal decalage is not a problem (Chapman 1987b; 1988; Lourenço & Machado 1996). Instead, it should be expected. Similarly, variability in performance on different false belief tasks is only a problem because it clashes with a common and implicit assumption about the nature of reasoning – that is, the view of reasoning as being based on the development of rules or principles which are then applied to problems to generate solutions (in domains such as moral reasoning or reasoning about the physical world, and now the social world).
9. Pretend play has been considered as a possible facilitating context for the development of social understanding, that is, as a “zone of proximal development” (Lillard 1993a; Youngblade & Dunn 1995). The overall amount of pretend play has not been found to be associated with false belief understanding, but false belief understanding is associated with specific types of pretend play: when children make joint proposals in their pretend play and when they explicitly make role assignments to themselves and their partners in play (Astington & Jenkins 1995). Also, young children's tendency to role enact is associated with belief understanding seven months later (Youngblade & Dunn 1995). From our perspective, we would expect that increased social understanding would facilitate children's ability to engage in cooperative pretend play, and that this social interaction could also serve as one context, among others, for further social development. We would not expect that any facilitative effect would depend only on the fact that pretense is involved, but rather also on the fact that such situations would require cooperative interaction in which children must coordinate their activity toward shared goals. This would require talking about human activity and the need to reach mutual understanding because the children are enacting some event together.
10. In considering the role of the child's abilities we recognize that our approach needs to be integrated with domain general approaches to children's reasoning about the mind (e.g., Frye et al. 1995a; Gordon & Olson 1998; Mitchell & Riggs 2000). The child must have an ability to pay attention to important aspects of social interaction and must be capable of achieving some distance between himself and the situation so he can reflect on it rather than act impulsively (Moses 2001). Furthermore, reasoning about situations that are not immediately present would require the ability hold in mind and to imagine aspects of situations (Harris 2000). Clearly, there is some distance to go in developing this aspect of our approach, but we suggest that a likely candidate for further study would be attentional capacity, which in Chapman's (1987b) approach has a role in understanding and reasoning.
11. Concerning autism, our position is consistent with Hobson's (1993; 2002) view of autism as due to a disruption in the child's ability to engage in affective interaction. In this article we have spelled out in further detail the nature of this interaction beyond infancy and the role of language in social cognitive development. However, we have not applied the issue of autism to our analytic framework because it is associated with so many other social and cognitive problems and differences from the typically developing population that no conclusive statement can be made here.
12. This focus on the role of relationships in development, as well as what the child brings to the relationship, allows us to think about both atypical development and research with nonhuman primates. Here our approach is consistent with Tomasello's (1999b) position that what is required in typical social cognitive development is both normal neurological development as well as the right social and cultural conditions. In children with autism we see biological abnormalities hindering the children's ability to engage with others in ways that are essential to normal social cognitive development, even though they have access to normal human social interaction (Hobson 1993). “Enculturated” apes – that is, chimpanzees and bonobos raised in a human linguistic environment – do develop more social cognitive and language skills than wild chimpanzees, but they can only go so far – not much beyond the level of a 2-year-old human child (Savage-Rumbaugh et al. 1993).