Chomsky provided a blow to behaviorism with his “poverty of the stimulus” argument. Yet new methods would later help question it.

George Miller explicitly acknowledged Noam Chomsky’s role in the cognitive revolution. Indeed, Chomsky was one of the most damaging critics of behaviorism. Chomsky reviewed Skinner’s book “Verbal Behavior” (1957) in 1959 and suggested that behaviorist models could not explain the lexical explosion observed in young children. To Chomsky, the observed linguistic abilities seemed underdetermined by the evidence of learning opportunities and reinforcements available to children who at a young age already showed the capacity to understand and produce sentences which they have never encountered before. The idea that data and reinforcement is insufficient to explain human language acquisition - and, therefore, that some form of innate competence or knowledge is needed to explain it - has become known as the “argument from poverty of the stimulus” (Pullum, & Scholz, 2002).

Chomsky’s line of reasoning has received two main critiques. First, there has been considerable empirical work since the 1950s to describe the linguistic experience of children and this tends to conclude that the linguistic experience of children is not as impoverished as previously thought (Pullum, & Scholz, 2002).

Second, the new tools of cognitive psychology, in particular advancements in computer-based modeling and simulation, helped create a better quantification of learning and established computational principles that could be behind children’s acquisition of grammatical knowledge. One prominent example of this approach is the connectionist modeling movement that emerged in the 1980s and which aims to explain intellectual abilities using artificial neural networks - simplified models composed of many neuron-like units (cf. Buckner, & Garson, 2019). Simulations with such models have demonstrated that these can acquire remarkable abilities, including the detection of simple grammatical structure. Such models have added plausibility to the idea that at least some parts of grammatical knowledge are acquired rather than innate.

You’ll find a photo of Nim Chimpsky here.

Chomsky argued that humans are able to learn language because of innate abilities and knowledge. In the 1970s, behaviorist-inclined researchers tried to prove that the use of rewards could be used to teach non-human primates - like Nim Chimpsky - to learn a language. The pun proved misguided because Nim never showed the productivity that is characteristic of human language. The issue of which aspects of language are uniquely human and acquired through learning (rather than innate), however, continue to be a central area of research, now aided by new tools of computer modeling and simulation.

---

References

Buckner, C., & Garson, J. (2019). Connectionism. In E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Fall ed.). Metaphysics Research Lab, Stanford University. Retrieved from https://plato.stanford.edu/archives/spr2017/entries/connectionism/

Chomsky, N. (1959). Verbal behavior by B. F. Skinner. Language, 35(1), 26-58. https://www.jstor.org/stable/411334

Pullum, G. K., & Scholz, B. C. (2002). Empirical assessment of stimulus poverty arguments. Linguistic Review, 18(1-2), 9–50. http://doi.org/10.1515/tlir.19.1-2.9

Skinner, B.F. (1957). Verbal behavior. Copley Publishing Group.

Please note that certain links to journal articles are only available within the eduroam network if they concern publications for which the university has a campus license. If you are a member of the University of Basel and want to access university resources from home, you will need to install a VPN client.

Lizenz