Skinner (remember him?) said that language is behavior, and, just like any other behavior, it is learned. This learning occurs through "reinforcement of successive approximations". Supposedly, a child is randomly making sounds, such as hi-hi, a-a-a-a and says "mi" . The mother, on hearing this, gets very excited, pays attention to the baby, says "Oh, you want milk!" and gives him a bottle of milk. After a while, the novelty of "mi" wears off and mother insists that the baby say "milk" before she provides the reinforcement of praise and milk. After a longer while, milk is not good enough, and the child must say "I want milk" to be rewarded.

    Many years ago, when I was first studying theories of child development in graduate school, I was enrolled in a class with a friend who had majored in linguistics. (My undergraduatre major was business). We had both just read Skinner's explanation of language development for the first time. She asked me what I thought of it and I said it seemed to make sense to me.

She laughed and said, "That's funny, I thought it was the stupidest thing I had ever heard."

I think her sentiment was shared by most linguists, and, as I have learned a little more about child development, I tend to go more with Paula's initial reaction than my own.

    Lovaas, a psychologist well known for his work with autistic children, used Skinner's ideas about reinforcement of successive approximations to teach autistic children to talk. After thousands of trials, for example, reinforcing a child saying "wa" and then "wat" and finally "wa-ter", he succeeded in teaching formerly nonverbal children to talk.

    It was Urie Bronfenbrenner (remember him) who challenged the suggestion that Lovaas had proved anything, saying that, just because a child can learn a certain way in the laboratory doesn't mean a child does learn that way in his or her own home or school.

    There is also the problem, which has plagued a lot of research in psychology, of generalizing results from research on abnormal children to normal children. Jay Haley (a family therapist) gave the analogy of a child with a broken leg. To make him well, we put his leg in a plaster cast and keep him from moving it. That does not mean that we could improve the walking and running ability of all children by putting their legs in plaster casts and immobilizing them.

Okay, enough of that tangent, click here to go back to the page on language development.