Language was proposed as an innate part of humans by both Pinker (32) and Chomsky (Cook and Mark 83). This view was then challenged by Tomasello (“Language Is Not an Instinct” 151) with an endeavor to look at natural language with Cognitive and Functional approaches. In this paper, I propose that language acquisition is not that different from learning how to play chess by displaying creativity and the poverty of the stimulus in our model, the two famous arguments made by Chomsky (Cook and Mark 81) for Language Acquisition Device (LAD). Since there is no way our brain has innate knowledge for playing any human-designed game, this paper agrees with Tomasello’s Cognitive/ Functional paradigm that language is not an innate module in human beings but a cultural artifact (“Language Is Not an Instinct” 152).
In this paper, I model language acquisition as learning chess by watching. To learn chess, human cognition categorizes chess pieces into distinct groups by their features (e.g. color, shapes). By observing others playing chess in progress, a person can pick up rules gradually without any given instruction. Chomsky’s work (Cook and Mark 90) also compares games such as chess and snooker to language and states that humans cannot learn snooker by watching games in progress for lack of negative feedbacks. However, his argument is flawed in two aspects. First, Chomsky’s belief (Cook and Mark 91) that knowing an adequate knowledge of snooker involves knowing what to do and what not to do is not true. The two kinds of rules are always convertible. For instance, “A piece cannot end its move on a square containing a piece with the same color” is equal to “A piece can only end its move on a square containing with a different color” in chess because negating one’s complement derives the other. In other words, rules about knowing what can do is enough if all the rules are acquired. Second, implicit negative feedbacks (Ramscar and Yarlett 927) exist. When watching others play chess, a person generates possible moves in his or her head by applying acquired rules. If some possible moves are never confirmed by others’ moves, rules need to be updated to explain those exceptions. In the following two paragraphs, I will expand the chess model and explain the two phenomenon of language acquisition.
The creativity of a child’s speech is not surprising if we compare language acquisition to game learning. Creativity, according to Chomsky (Cook and Mark 77), is unique to humans, which allows us to generate innovative sentences using generative grammar. However, it is possible to achieve the level of creativity (Cook and Mark 77) without using generative rules. If one observes that one rook can move any number of vacant squares up, down, left, and right in a straight line, then it is natural for this person to generate an “innovative” move of the other rook because the two rook are intuitively categorized into one group by their looks. In a language, words all look different. Humans create common categories for words because of their cognitive and social universals (Tomasello, “Language Is Not an Instinct” 150). Then, a person can then apply rules to words with the same category to create innovative sentences. With the help of categorization, humans can narrow down the valid choices of words from thousands to a few and apply the rules correctly and innovatively. It is also worth noting that one can overgeneralize the rule of the rook that I mentioned above until this person finds out that a rook can have a special move called “castling” with the king under a specific circumstance. This phenomenon is also common in the acquisition of language such as verb forms (Cook and Mark 90).
Also, the poverty of the stimulus (POS) states that children can ignore incorrect grammar and always use correct grammar due to their innate linguistic knowledge (Cook and Mark 81). This is not true neither. Not all moves observed from the players are valid because they might not be proficient with chess rules. Therefore, a person will develop a mechanism to overcome bad examples. First, the person simply imitates what he or she sees others do, which is supported by Tomasello (“First Steps toward a Usage-Based Theory of Language Acquisition” 70). At this stage, the person will start generating abstract rules from observation. Second, a person will receive implicit negative feedback if what he or she expects differs from others’ moves. This process will allow a person to update acquired rules constantly. Third, a person finds that some incorrect examples consistently show up and includes the inconsistent behaviors as exceptions to the rules. Like the game of chess, a person follows the same three steps to acquire language. A person can acquire many versions of a grammar from others, but the person will pick the one with the highest confidence based on frequency and truthfulness. As a strategy, children will also avoid using certain expressions (Tomasello, “Language Is Not an Instinct” 144) if the confidence score for them is low. This kind of statistical learning together with the strategy ensures that children speak grammatically correct languages even though others don’t use perfect grammar.
This paper presents that learning a language is similar to learning chess through observation. The chess model also explains creativity and poverty of the stimulus proposed by Chomsky. Furthermore, this paper endorses Tomasello’s Cognitive/Functional paradigm (“Language Is Not an Instinct” 150) that that language module is not innate in human beings but a cultural product.
Citations
Cook, Vivian J., and Mark Newson. “Chapter 3: General Concepts of Language Acquisition.” Chomsky’s Universal Grammar: An Introduction, 3rd ed., Blackwell, 2007, pp. 74–132.
Pinker, Steven. The Language Instinct. Penguin, 2015.
Ramscar, Michael, and Daniel Yarlett. “Linguistic Self-Correction in the Absence of Feedback: A New Approach to the Logical Problem of Language Acquisition.” Cognitive Science, vol. 31, no. 6, Dec. 2007, pp. 927–960.
Tomasello, Michael. “First Steps toward a Usage-Based Theory of Language Acquisition.” Cognitive Linguistics, vol. 11, no. 1-2, 2001.
Tomasello, Michael. “Language Is Not an Instinct.” Cognitive Development, vol. 10, no. 1, 1995, pp. 131–156.
Word Count: 920