Every species has a unique way of being in the world, and linguistic communication sets humans apart by providing a vehicle to convey information about physically and temporally displaced objects and events. Human languages utilize general-purpose learning mechanisms that yield a human product unique in the shear quantity and diversity of information. How is this possible?

We start by asking how the human brain gets its structure to the point of being language-ready. The answer to this may be simple. Early evolutionary changes in hominin developmental timing, occurring over three million years ago, produced a larger brain, relatively immature at birth, and open to postnatal restructuring. Changes in developmental timing to permit the birth of a large, skeletally immature infant accompanied a shift in posture and the anterior-posterior narrowing of the birth canal. Progressive prolongation of successive phases of fetal growth delayed skeletal ossification until after birth and resulted in a significant enlargement of the hominin brain relative to the postnatal growth of the body (Deacon Reference Deacon1997). Later divergence of brain from body size, occurring approximately one million years ago, appears to be linked to changes in diet and social organization (Aiello & Wheeler Reference Aiello and Wheeler1995; Milton Reference Milton1999a; Reference Milton1999b; Ragir Reference Ragir and Gyori2001). A larger brain supported more sophisticated manipulations of the environment, fine-motor coordination, expanded short-term memory, and increased speed in information processing (Kien Reference Kien1991). Such systemic evolutionary change provides an explanation for the emergence of capacities that make human brains language-ready.

The extended postnatal development of the human cortex affords information-specific pathways and modules as emergent features of neural organization. Dendritic and synaptic proliferation and pruning restructure the maturing brain in response to the environment and to the community of practices in which development is embedded (Adams & Cox Reference Adams and Cox2002). In humans, passive as well as participatory experiences tune cortical pathways throughout an extended period of brain development and produce increasingly biased perceivers, more efficient learners, as well as behavioral complexity and innovation (Kuhl Reference Kuhl and Gazzaniga2000). Statistical learning involves general learning mechanisms akin to classical conditioning, but some statistics are picked up more readily than others, with learning biases varying across primate species (Newport et al. Reference Newport, Hauser, Spaepen and Aslin2004). Language learning fundamentally relies on memory of sequential and non-adjacent dependencies in the linear ordering of either auditory or visual elements. Auditory processing is especially well suited for tracking information distributed over time (Conway & Christiansen Reference Conway and Christiansen2005), and humans show significant prenatal maturation and adaptations in the auditory system. Enhanced linear processing and Bayesian statistical learning mechanisms, however, are not language-specific, nor are they sufficient for language readiness.

Language is not merely a product of a language-ready brain; it is a cultural product of a community of practitioners. The crucial task is to identify the processes by which shared interpretations of the world are created among members of a community, as the co-construction of shared understanding forms the basis for any exchange of information. Co-constructed activities generate shared understandings of the relationships between vocal, manual, or body gestures that drop out and take meaning from ongoing events. Through social interaction, children learn the repetitive activity patterns of their community of which vocalizations form an integral part (Nelson Reference Nelson2006). As an accompaniment to shared activity, vocalization is public, reflexive, and separates out of ongoing events to stand for all or part of the activity (Carpendale & Lewis Reference Carpendale and Lewis2004). Each learner engages with the language that exists in the community to the extent that he or she is developmentally and socially capable (Ninio Reference Ninio2006; Wray & Grace Reference Wray and Grace2005). One of the challenges for developmental theorists is to identify the entry points to this vast web of cultural knowledge, and how the child's activity within and representations of this cultural system become elaborated over time. Language acquisition requires a constant give-and-take between the child and others about the meaning of communicative exchanges in social interactions (Taylor Reference Taylor1997). Figure 1 adapts C&C's Figure 3b to include the dynamic negotiation of shared meaning occurring in the community.

Figure 1.

Negotiation and Constraint in Language Formation.

Christiansen & Chater (C&C) propose the analogy of language as a “beneficial parasite” (sect. 5.1) symbiotic to the human brain in so far as the systematicity of language is a function of how brains learn. In our opinion, this analogy misleads and ignores the homology between language and all other human inventions. Dynamically negotiated cultural systems such as tool manufacture, music, and play are similarly constrained by body schema and the construction of experience-dependent neural networks, and arise from the needs of users to coordinate their perspectives and goals. Complexity in any cultural system emerges over a historical time frame; hierarchy and systematicity are emergent properties, contingent on critical numbers of communicators and topics (Kirby Reference Kirby1999). As work on the emergence of indigenous sign languages demonstrates, languages are not elaborated without a community of signers (Kegl et al. Reference Kegl, Senghas, Coppola and DeGraff1999; Ragir 2002; Senghas et al. Reference Senghas, Kita and Özyürek2004). As a community increases in size, and members lack intimate knowledge of one another, “rules” emerge from the competing pressures of efficiency, clarity, and expressiveness, and streamline the exchanges of information to reduce cognitive effort (Kirby Reference Kirby1999; Ragir 2002; Wray & Grace Reference Wray and Grace2005). In any cultural system, arbitrary categories are difficult to keep track of, and overlapping, redundant patterns of expression are favored. Languages increase in semantic and grammatical complexity as needed within developmental and historical time-scales. Such systematization is a universal property of culturally negotiated systems, rather than a property of the brain (Bak Reference Bak1996). Similarly, pragmatic principles are not simply constraints, analogous to perceptual-motor or cognitive capacities, but reflect social practices involved in coordinating perspectives for effective communication (Clark Reference Clark, Gumperz and Levinson1996a; Hilton Reference Hilton1995). While C&C suggest that the possession of language might have fundamentally changed the patterns of collective problem solving, we would argue that collective problem solving led to the co-construction of language.

Finally, C&C make an unnecessary capitulation to the speculations of evolutionary psychology in their concession that cognitive pre-adaptations probably evolved through natural selection before language emerged. We suggest that changes in habitat, diet, and social organization co-evolved with the increased nutritional demands of encephalization and the socialization of dependent offspring during an extended childhood. Just as human languages emerged out of the exchanges of information within human societies, cooperation, joint attention, theory of mind, symbolic thought, and pragmatic reasoning are just as likely to have been products of dynamic negotiations within communities.