Thursday, September 5, 2019

Universal grammar and other pseudosciences

Everyone agrees that astrology is pseudoscience, but this new paper takes on some respected academic subjects:
After considering a set of demarcation criteria, four pseudosciences are examined: psychoanalysis, speculative evolutionary psychology, universal grammar, and string theory. It is concluded that these theoretical frameworks do not meet the requirements to be considered genuinely scientific. ...

To discriminate between two different types of activities some kind of criteria or
standards are necessary. It is argued that the following four demarcation criteria are
suitable to distinguish science from pseudoscience:
1. Testability. ...
2. Evidence. ...
3. Reproducibility. ...
4. The 50-year criterion.
By these criteria, string theory fails to be science. It mentions that a couple of philosophers try to defend string theory, but only by inventing some new category. I guess they don't want to call it "pseudoscience" if respectable professors promote it.

Respectable professors also have a long history of supporting Freudian psychoanalysis.

This claim about universal grammar struck me:
Chomksy (1975, p. 4) argues that children learn language easily since they do it without formal instruction or conscious awareness.
Not only is Chomsky well-respected for these opinions, but Steve Pinker and many others have said similar things.

This puzzles me. I taught my kids to talk, and I would not describe it as easy. I had to give them formal instruction, and they seemed to be consciously aware of it.

The process takes about a year. It is a long series of incremental steps. Steps are: teaching the child to understand simple commands, such as "stop", articulating sounds like "hi", responding to sounds, like saying "hi" in response to "hi", learning simple nouns, like saying "ball" while pointing to a ball, developing a vocabulary of 20 nouns or so, learning simple verbs like "go", putting together subject-verb, putting together subject-verb-object, etc.

All of these steps are difficult for a two-year-old, and require a great deal of individual instruction and practice.

Sure, two-year-olds might learn a lot by observing, but you could say the same about other skills. Some are taught to dress themselves, while others learn by mimicking others. No one would say that children learn to dress themselves without instruction.
Steven Pinker was the first to popularize the hypothesis that language is an instinct. In his influential book The Language Instinct, Pinker asserts that “people know how to talk in more or less the sense that spiders know how to spin webs” (Pinker 1995, p. 18). Pinker’s analogy is striking, since it is obviously incorrect. A spider will spin webs even if it remains isolated since birth. On the other hand, a child who has been isolated since birth will not learn language. In other words, while web-spinning does not require previous experience and it is innate, language does require experience and it is learned.
Chomsky and Pinker are two of our most respected intellectuals today.

Googling indicates that Chomsky had one daughter and no grandkids. Pinker has no kids. I am not sure that is relevant, as many others have similarly claimed that children learn language naturally.


  1. Of course children learn language naturally; the language they learn will depend on their early environment but the function is universal, clearly based upon an innate capacity for both vocabulary and syntax. Children learn to speak at different ages but left to their own devices will start speaking spontaneously having listened to adults particularly their mothers and absorbing the sounds of their native language.The idea that humans evolved vocal cords but not the innate capacity to employ them is extremely unlikely as it does not apply to any other piece of human anatomy.

    As to psychoanalysis, if it had had any validity it would have been invented by a European not an Asiatic and the same would apply to Communism and 12-tone music.

  2. Humans have an innate ability to vocalize sounds. I accept that.

    If speaking is innate, why do some kids learn at different ages? My theory is that some kids are taught better than others.

    If speaking is innate, why did I spend a year teaching it? Why did each advance come after I taught that element of grammar?

    Where is the empirical support for the innate theory? All the evidence seems to point the other way.