Professor Kim noted that one of the issues in the debate around human language acquisition is that Chomskyan scholars tend to concede individual points to the cognitive-behaviorist paradigm while holding that if only one aspect of Universal Grammar (UG) is valid, then the theory holds water[1]. These generativists understand theories of UG as logically encapsulating and preceding cognitive-behaviorist arguments. If, they argue, it turns out that human brains come pre-programmed to comprehend one particular syntactic rule, then Universal Grammar and the Language Acquisition device it implies are valid. However, that framing is infelicitous and patently wrong-headed, because the burden of evidence necessary to disprove it (a repository of information on all languages) is too large and one important tenet of the generativist approach to language acquisition is explainable in terms of a cognitive-behaviorist framework. This essay will explore these problems, ultimately arguing that a cognitive-behaviorist understanding reflects a more feasible and appropriate approach to language acquisition.
Bakhtin writes: “verbal discourse is a social phenomenon – social throughout its entire range and in each and every of its factors, from the sound image to the furthest reaches of abstract meaning”. (l982:259) Language is social, not only in the Saussurean sense that it belongs to society, but also in that it only works as an interpersonal and dialogic phenomenon. A UG approach disagrees on this point, noting that Language exists in the brain at birth but merely needs to be triggered to activate the supposed Language Acquisition Device (LAD). It is then primarily a psychological phenomenon which has a neurological incipiency. However, Chomsky described the development of the language faculty as “a procedure that operates on experience acquired in an ideal community”. (Cook & Newson, 1995:79) Because it contradicts the assertion that language is stimulus-free, the notion that experience is a crucial element of initial language development destabilizes the feasibility of a specific LAD and blurs the line between generativist thought and cognitive-behaviorist theories.
The rationale which supported Chomsky’s early description of an internal and specifically linguistic faculty for language development was largely provided by a critique of B.F. Skinner’s behaviorist theory. According to Cook and Newson, Skinner argued that “language is determined by stimuli consisting of specific attributes of the situation, by responses the stimuli call up in the organism, and by reinforcing stimuli that are their consequences”. (76) This, for Chomsky, was reprehensibly deterministic. If that were the case, they queried, how do “people regularly understand and produce sentences that they have never heard before”? (77) Such a line of questioning fundamentally misunderstands the role of sociality in the acquisition of language. I would like to suggest a recast of Skinner’s behaviorist approach.
Language is not stimulus-free. Chomsky cleverly (or perhaps just unwittingly) concedes this point. Recall that they included experience in a description of how the LAD functions. Simply put, experience is stimulus. Take the following scene from the Lion King (1994) for example. Rafiki strikes Simba on the head with his staff and when questioned for his reasoning responds “It doesn’t matter, it’s in the past. [The past hurts] but the way I see it, you can either run from it or learn from it”. When Rafiki moves to strike Simba again, he ducks. Experience (i.e external stimuli) does condition a response. Language is no exception.
In the case of language, we can understand primary linguistic data as stimuli which conditions a response in the uptake of a particular grammar. Chomsky refutes this line of thinking by noting that a child language learner “rarely encounters appropriate external rewards or punishment”. (Cook & Newson, 77) The successfulness of social interactions does in fact constitute a reward system. Misfires (or other types of botched sociolinguistic interaction) thereby are a kind of punishment. These sometimes are explicit (No, Timmy, foots is not a word; you have to say feet), but not nearly always. Human infants are attentive to the social intent of speakers in word-learning situation. (Saffran & Thiessen, 2007:75) When young spoken language learners babble, as they come closer to approximating groups of sounds which agree with the phonotactics of their parent’ native language, they are met with cheers. Indeed, from the moment they are born, children are surrounded by language doers who excitedly await and encourage their development of that faculty. Within that frame of understanding, the conclusion seems inescapable that language is learned, in the first instance, through social interaction.
One might argue that this practice is only applicable to communities which engage in child-oriented accommodation during the critical period of language development. Those persons who engage in parent-ese give targeted attention to the early linguistic production of infants, but this practice is not universal. Scholars have used the existence of communities that require linguistic bootstrapping of their babies as evidence for UG and the innate LAD. However, as Tomasello points out, “Generative Grammar was created to describe English,” and several universalistic claims made by generativists around linguistic features such as X-bar phrase structure, subject/object relations, and long range syntactic movement do not pan out for all languages. (1995:136,139) Therefore, it is not enough to argue that because child-oriented accommodation is not universal then UG theory must hold water. UG has to be a suitable description of all languages first in order to serve as an adequate model of language acquisition. The cognitive-behavioral model, however does create a sufficient platform upon which to understand this process.
There is no such thing as language that is not social. Universal Grammar assumes that there might be, a truly impossible claim to prove. Thus, it is not a useful way of thinking through language acquisition. Because there is sociality in every level of language use, the cognitive-behaviorist framework ultimately provides a stronger foundation to understand that systems of social interaction allow language acquisition.
[1] I use the phrases Chomskyan and generativist synonymously in this essay to mean “associated closely with the Universal Grammar concept and the constellation of ideas it implies”.
References
Bakhtin, Mikhail. 1981. Discourse in the novel. In Michael Holquist, ed., The dialogic imagination: four essays by M. M. Bakhtin. C. Emerson & M. Holquist, trans. Austin, Texas: University of Texas
Cook, Vivian, & Newson, Mark. 1995. Chomsky’s Universal Grammar: An Introduction. New Jersey: Wiley-Blackwell.
Saffran, Jenny, & Thiessen, Erik. 2007. Domain-general learning capacities. In Erick Hoff & Marilyn Shatz, eds., Blackwell Handbook of Language Development. Malden, Massachusetts: Blackwell.
The Lion King. 1994. Walt Disney Pictures. Dir. Rob Minkoff. [online]
Tomasello, Michael. 1995. Language is not an instinct. Cognitive Development, 10, 131-156.
Hi! To be 100% honest, I chose to read your essay because you cited The Lion King. But I also enjoyed it! I do have some thoughts though.
I agree with the basis that all language is social, but I disagree with some aspects of your theories regarding language acquisition. For example, you say:
“The successfulness of social interactions does in fact constitute a reward system. Misfires (or other types of botched sociolinguistic interaction) thereby are a kind of punishment. These sometimes are explicit (No, Timmy, foots is not a word; you have to say feet), but not nearly always.”
However, the “punishments” you describe do not happen often enough to be considered a successful reward system – they definitely do not happen “not nearly always” (?). Parents and teachers occasionally offer correction, whether explicitly as you describe or implicitly (recasts), but sometimes they ignore the incorrect usage or even reinforce it by repeating it because it is “cute.” In addition, explicit correction tends to happen when the meaning of a word used is incorrect, rather than the grammar. (My lack of sources is likely discrediting me – this information comes from my Fall 2016 notes in my Language Acquisition class, also taught by Dr. Kim, and I am having trouble finding the original studies.) Basically: correction can be helpful when present, but it does not happen often enough or consistently enough to have a truly significant effect; i.e. to completely account for the success of language acquisition.
Anyway: I do not believe in Universal Grammar per se, but I do think humans likely have an underlying biological capacity for language. Whether it is rooted in domain-general or domain-specific knowledge, I have yet to firmly decide, though I lean towards a more constructionist view of domain-general abilities that are used for language learning in addition to other knowledge acquisition, while also keeping in mind that experience and input are important. These domain-general capacities include symbolic representation, memory, chunking, probabilistic analysis, etc: all helpful to understanding the world around us, but also likely contributing to successful language acquisition.
Your essay leans towards empiricism, but what do you think about constructionism?