Chomsky: main points Flashcards
Skinner wants the relationship between stimulus and response to be lawful: the stimulus
chomsky: There is no way a lawful association would exist, because a stimulus can have an infinite number of properties the organism can respond to.
eg. a red chair could elicit all types of responses based on its various properties (redness, chairness, wood, etc.)
Skinner wants the relationship between stimulus and response to be lawful: the response
chomsky: Response strength cannot be quantified lawfully when it comes to verbal behaviour.
Skinner’s idea would prescribe that frequent occurrence of a response = frequent occurrence of a stimulus, which is not logical for verbal behaviour (pgph 19)
In the lab, working with rats, it can more easily be measured using frequency. However, this does not transfer to verbal behaviour. Sometimes no response or infrequent responses are ‘stronger’ than lots of occurrences of responses (pgph 20).
skinner wants the relationship between stimulus and response to be lawful: describe reinforcement
There are too many ways reinforcement can be attained, making its role in learning (= strengthened S-R associations) unlawful and unclear
Reinforcement is not always identifiable (e.g., self-reinforcement such as children imitating others/sounds; a writer being reinforced by the idea that in the future, his work will be read by many people)
dus samenvatting van punt 1: skinner zijn theorie is vaag
- stimulus kan heel veel properties hebben waarop gereageerd wordt (rode stoel)
- response strength kan niet goed quantified worden (bij verbal behaviour)
- too many ways in which reinforcement can be attained (je kan imiteren, of door het idee dat je succesful wordt etc)
Conclusion: what happens if you take Skinner’s terms (stimulus and response) literally? And what happens if you take them metaphorically?
If you take them literally, you cannot predict verbal behaviour. If you take them metaphorically, the theoretical model has no added value to our understanding of language (acquisition).
Skinner wants to explain behaviour in behavioural terms, not using any ‘black-box’ terms
deprivation argument
chomsky: You cannot quantify deprivation in verbal behaviour, nor can you apply it.
In examples such as “Take me for a ride” and “Let me fix it”, it would be impossible to measure deprivation. Moreover, one has to wonder if these utterances are in fact motivated by any kind of deprivation. Chomsky believes deprivation is just a weird way of saying “X wants Y”, which would make it black box content.
Skinner wants to explain behaviour in behavioural terms, not using any ‘black-box’ terms
why do we need intentions?
You cannot determine the function of verbal behaviour by basing it on the response of the listener
There are too many ways this association could fail. Sometimes, questions are not answered (e.g., rhetorical questions). The question not being answered doesn’t change the fact that it was intended to be a question. Thus, the function of verbal behaviour leads back to the black box of the speaker.
Conclusion: why are behavioural terms not enough to describe / explain language?
Because when applied, they lead to unlawful and illogical associations. By removing the idea of an internal driving force behind verbal behaviour, inferences about functions of behaviour cannot be logically made. We need to combine information from the inner structures of an organism to their behaviour in order to truly understand their (verbal) behaviour.
Why is conditional learning not enough?
Describe the behaviour of children (mimicry).
Children can imitate language without their parents’ feedback
Very young children can imitate language, sometimes very well without any practice, irrespective of their parents’ shaping of this behaviour (pgph 31)
Children do some imitation of language, but not full sentences. They may repeat 1 or 2 words, but not a whole sentence, so there is no build-up of language skills using mimicry
Why is conditional learning not enough?
Describe the behaviour of children (spontaneous learning).
Learning does not require strict Behaviourist circumstances such as feedback. Children can, and do, learn without these contextual factors.
Learning can be done independent of reinforcement and feedback. Humans and other animals have been known to learn for no apparent reason, out of boredom or curiosity.
If children learned from external feedback, the process should be gradual. However, we see that children make sudden leaps around certain ages.
Why is conditional learning not enough?
Describe the role of the parents.
Children’s language development is independent of their parents’ ‘shaping’ or modeling of that language.
Children learn a language so quickly and accurately, despite a lack of strict ‘coaching’ by their parents (e.g. being corrected every time they say something incorrect).
Children can pick up a language on the street, without their parents teaching it to them. They can recognize words and sentences they have not yet learned and make sense of them.
Conclusion: why doesn’t Skinner’s theory describe language acquisition? What does this have to do with the ‘poverty of the stimulus’ argument?
Children learn language much more quickly and accurately than could ever be achieved solely by the feedback they receive from their environment. Parents do not consistently punish and reinforce incorrect vs. correct language use. Moreover, children learn without any external motivator. They can also construct new sentences that are grammatically correct, without ever having been exposed to them. This means that there is some other mechanism that explains children’s rapid language acquisition, which also develops in the same way for most healthy children.
poverty of the stimulus=
the controversial argument from linguistics that children are not exposed to rich enough data within their linguistic environments to acquire every feature of their language.
