The Logical Problem of Language Acquisition

Language... Language Acquisition The Logical Problem of Language Acquisition • Adults have a very sophisticated knowledge of their native language. • This is knowledge that we don’t have conscious access to, but that we have nevertheless. • Children become adults. Children wind up with this same sophisticated knowledge. How does this happen? Judgments Judgments • Some things are English. Some things are not. Adult native speakers know which is which, and agree. 1) Dogs three cat a chased. 2) Cheddar, I like, but brie, I can’t stand. 3) Eight hungry capybaras danced enthusiastically. Grammar 1) You cannot stop a philosopher with a thesis. 2) Mary saw her in the mirror. 3) Mary saw her duck in the mirror. Isn’t this obvious? • Adult native speakers uniformly and overwhelmingly agree. • To know English is to have knowledge of (at least) which sentences are possible in English and which are not. • There is a (hidden) system, a grammar. • How do people acquire the grammar? • What’s the big deal? Kids hear what their parents do and imitate that. Why do we need 15 weeks to learn about that? And besides, English is easy. Not like Chinese. Or Hungarian. No, not really obvious. • Imitation by itself could never work. • The knowledge we’re talking about differentiates English sentences from nonEnglish sentences. (Or whatever language) • Which ones are good? The ones you hear? 1) The first platypus to eat twenty-five yellow gummi bears will win a prize. Sure, ok. You generalize. • So, maybe kids hear what parents say and recognize the patterns, and come up with general rules. • That’s not so hard, is it? • Let’s try it out. That’s easy • It looks like you make yes-no questions by taking the second word and putting it in the front. Though if there is no word like should, then you just put did in the front, and use a bare (untensed) verb. • Wh-questions are formed by removing something and putting who or what in front of the yes-no question form. Things adults don’t say • Also, kids say things an adult has never said: 1) Me playing. 2) What do you think what the puppet has eaten? • And they are learning some kind of rules, some kind of system. They say wugs. Making questions 1) Sue should borrow my guitar. 2) Sue borrowed my guitar. 3) Should Sue borrow my guitar? 4) Did Sue borrow my guitar? 5) What should Sue borrow? 6) What did Sue borrow? See? Simple. 1) My roommate should borrow my guitar. 2) Roommate my should borrow my guitar? 3) Roommate my did borrow my guitar? 4) Who did borrow my guitar? 5) What roommate my should borrow? 6) Why person any would think this is hard? Poverty of the stimulus “Structure dependence” • The grammatical properties that the • The actual procedure for forming a yes-no • Yet, children always acquire the same rules. • 1, 2, 3, —, —, —? • 4, 5, 6? — 5, 7, 11? — 5, 8, 13? — 3, 2, 1? • All syntactic operations have this property “Structure dependence” I hear (that) these could also be tricky children are acquiring are underdetermined by the evidence available to them. • So, even though the yes-no questions that children hear are compatible at least with two kinds of question-formation operation, children never even consider the nonstructure-dependent one. A)Move the auxiliary after the subject to the front. B) Move the second word to the front. Wanna contract? question in English does not refer to the second word, but rather the auxiliary (possibly do, if there is no other auxiliary) that follows the entire subject constituent. —they refer to the structure and not the linear sequencing. 1) Sue said that Mary borrowed my guitar. 2) Sue said Mary borrowed my guitar. 3) What did Sue say that Mary borrowed? 4) What did Sue say Mary borrowed? 5) Who did Sue say borrowed my guitar? 6) Who did Sue say that borrowed my guitar? Learning Principle B? 1) Who do you want to invite? 1) Mary saw her in the mirror. 2) Who do you wanna invite? 2) Mary saw her duck in the mirror. 3) Who do you want to dance? 4) Who do you wanna dance? 5) Whoj do you want PRO to invite tj ? 6) Whoj do you want tj to dance? • Why can’t her be Mary? (Except if it’s Mary’s duck—but it can’t be Mary who is ducking). How is a kid supposed to deduce this? Negative evidence • The thing that is most problematic about these examples is that, based only on hearing grammatical sentences, we might expect a child to generalize to an incorrect rule. • If the child also had evidence that the ungrammatical sentences were ungrammatical (like linguists do), these rules should at least be learnable. Constraints and negative evidence • Binding Theory imposes constraints against certain meanings. • The only way such a constraint could be learned is through negative evidence— evidence that the missing meaning is in fact missing (even though the sentence itself is perfectly fine, with a different meaning). • Yet: it is at least not obvious that children have access to negative evidence. Maybe they’re taught? • The clearest form of negative evidence would be the sort where an adult corrects a child who has gotten it wrong—or, alternatively, the adult simply instructs the child that He saw John can’t mean John saw himself. • Who here remembers having that conversation with your caretakers? Explicit teaching Braine (1971) reports: Want other one spoon, daddy. You mean, you want the other spoon. Yes, I want other one spoon, please Daddy. Can you say “the other spoon”? Other...one...spoon Say “other” Other “Spoon” Spoon “Other spoon” Other...spoon. Now give me other one spoon. Explicit teaching McNeill (1966) reports: Nobody don’t like me. No, say “nobody likes me.” Nobody don’t like me. (repeats 8 times) No, now listen carefully: say “nobody likes me.” Oh! Nobody don’t likes me. Explicit teaching Cazden (1972) reports: My teacher holded the baby rabbits and we patted them. Did you say your teacher held the baby rabbits? Yes. What did you say she did? She holded the baby rabbits and we patted them. Did you say she held them tightly? No, she holded them loosely. No negative evidence. • Thing is, kids don’t really get negative evidence. Not about the grammar. • In general, when adults correct a child, it is because what they said was false. When a child says something true (even if ungrammatical), adults will act pleased. • And that couldn’t have been the whole story anyway, because different parents behave differently. But the kids all still achieve the same grammar. Maybe there is implicit negative evidence...? 1) Jane said Mary lost her keys. 2) Jane said Mary voted for her. • Listen, listen, listen—but never does her turn out to be Mary in the second kind of sentence. • So, conclude that the sentence/meaning is ungrammatical? But after how long? It could be the very next sentence we hear. How long would a child have to keep track of this (and the rest)? Negative evidence by other channels? Parents differ, and stop • Approval? • Brown & Hanlon (1970) found that adults • Eve and Sarah’s parents ask clarification • Comprehension? • Adults understand 42% of grammatical • After about age 4, Adam’s and Sarah’s express approval to 45% of grammatical sentences, and 45% of ungrammatical sentences. sentences and 47% of the ungrammatical ones. Statistics (Marcus 1993) • Maybe the child can work with the odds, where there is a difference, even if noisy. • Suppose a response happens 20% of the time to an ungrammatical sentence, and 12% of the time to a grammatical sentence. • So, kid says something. Response, 63% chance it was ungrammatical; no response, 52% chance it was grammatical. questions after ill-formed wh-questions. • Adam’s parents ask clarification questions after well-formed wh-questions, and after past tense errors. parents stopped responding differently to ungrammatical utterances, but the children still made errors. Then what? Statistics • Having a 52% or 63% chance of being correct about the status of an utterance isn’t really good enough to build a grammar (with the kind of reliability that kids do). • To improve the chances, simply try the utterance again and see whether there is a response. To reach 99% confidence, it only needs to be tried a mere 446 times. Hmm. Errors • There’s another big problem with negative evidence (at least in the form of negative feedback) as a determining factor in acquisition: There are a lot of errors that kids just don’t make—so nothing to provide negative feedback on. Only positive evidence • So, we’re left with the conclusion that however children acquire language, it is just on the basis of the positive evidence—things that they hear that they take to be part of the language they are trying to acquire. 1) Roommate my should borrow my guitar? What are we left with? Universal Grammar • Of course, you know where this is going: • The generally accepted view is that there is Having language = being human. • Rocks, ferns, cats, apes don’t soak up language when surrounded by it. some kind of template for language, that’s just built-in. Knowledge of language can only take a certain form. • Birds have wings, people have arms. • What determines whether you’re a bird is • Languages vary, of course, but what children UG explains how this could be possible A “critical period” for language acquisition • Children acquire language • very quickly • following the same course across languages • under varying circumstances • without explicit teaching • with a uniform result • It also seems that there is a unique window whether your parents are birds. are actually doing is just determining which among the permissible variations the language they are acquiring has. of opportunity for acquiring language early in life—which closes at a certain point. • If you are going to acquire language, it must happen while you are young. • There are a few (tragic) cases documented that seem to show this. Genie Critical period • The most often cited case is “Genie” (Curtiss • Other cases where children had been • Though—caution is required here. Genie also • Isabelle (starting at 6), Jim (starting a little Other critical periods Critical periods in the visual system • Beyond language, there are plenty of • The visual system is a favorite example. In experiments done on macaque monkeys, it was determined that there is a critical period for development of binocular vision cells in the visual cortex (tested by monocular deprivation) • Recovery after CNS damage is disappointingly limited in the adult brain, but can be nearly 100% in the immature nervous system. 1977), who was essentially locked in her room until she was 13, without language input. She was rescued, and underwent a great deal of language therapy, but never really managed to acquire the syntactic structures of English. discovered in such conditions and brought into language contact do seem to show that if language input comes soon enough, the normal language acquisition process can still occur. had a great deal of other psychological trauma. after 3 was caught up by 6) examples of critical periods in the biological world, so it’s not really that strange. Particularly on the view that the fundamental structure of language is biologically determined. From “Principles of Neural Science” Other critical periods • Vision studies replicated in cats (Hubel & Wiesel 1962, 1970). • Similar effects can be observed in human vision as well— there seems to be a visual critical period at around age 6, after which providing previously delayed visual stimuli is of no use. (Congenital opacities of the cornea; surgery performed on juveniles or adults does not restore sight) • Imprinting in birds; just after birth, they “become attached” to a prominent moving object in their environment (typically, the mother). This attachment persists. But it can only be done sometimes in the first few hours, for some species. • …The development of form perception and the binocular vision necessary for depth perception proceed in stages after birth. Each stage culminates in one or more developmental decisions, many of which are irreversible. In each stage, appropriate sensory experiences are necessary to validate, shape, and update normal developmental processes. Consequently, the effects of sensory deprivation are most severe during a restricted and well-defined period early in postnatal life when these developmental decisions are still being made. (Kandel, Schwartz, Jessell 3d ed. 1991, p. 956) From “Principles of Neural Science” • Broca’s area ...Critical periods of development generally do not have sharp time boundaries. Different layers within one region of the brain may have different critical periods of development, so that even after the critical period for one layer has passed, rearrangement of the layer may still be possible because the entire region has not yet fully developed. For example, 8 weeks after birth layer 4c in the visual cortex of the monkey is no longer affected by monocular deprivation, whereas the upper and lower layers continue to be susceptible for almost the entire first year. (Kandel, Schwartz, Jessell 3d ed. 1991, p. 957) Brain development? • Lateralization? (Lenneberg’s idea—but seems to be complete by 5, too early) • Mylenization? (Perhaps more promising, but probably too late—still going on after 15) • Still not really clear, but something in the process of neural development seems likely. Myelinization & neurons