Letters to Language
Rent from DeepDyve- Letters to Language
Language accepts letters from readers that briefly and succinctly respond to or comment upon either material published previously in the journal or issues deemed of importance to the field. The editor reserves the right to edit letters as needed. Brief replies from relevant parties are included as warranted.
Response to Anderson
February 16, 2009
To the Editor:
In the article derived from his 2008 Presidential address ('The logical structure of linguistic theory', Language 84.795–814, 2008), Stephen Anderson makes two misleading statements about how evidence might bear on language acquisition, one relating to corpus use and the other to typological generalizations. His errors have been made previously in the literature, and should not be repeated in Language without correction.
Anderson discusses a familiar claim of Noam Chomsky's: that children do not learn from experience that the formation principles for interrogatives in English are structure-sensitive rather than string-sensitive. Take the facts in 1.
(1)
a. Everything that wasn't eaten will be thrown away.
b. Will everything that wasn't eaten be thrown away?
c. *Wasn't everything that eaten will be thrown away?
Comparing 1b with its corresponding declarative in 1a reveals that the simple string-sensitive hypothesis in 2 is wrong.
(2) The first auxiliary in the declarative must be positioned initially in the corresponding interrogative.
This hypothesis would predict 1c. I will call sentences like 1b telltale sentences, since their testimony brings out the difference between the first auxiliary in the string and the auxiliary that follows the subject of the clause. If language were learned from the evidence provided by experience, then encountering telltale sentences would help the learner by permitting 2 to be eliminated. But Chomsky claimed that 'A person might go through much or all of his life without ever having been exposed to relevant evidence' (Language and learning: The debate between Jean Piaget and Noam Chomsky, ed. by Massimo Piattelli-Palmarini, Harvard UP, 1980, pp. 40, also 114–15)—that is, evidence that would confirm the correct hypothesis over the tempting but incorrect one in 2. In other words, he claimed that telltale sentences are so rare that you might well never encounter one in your whole life.
Pullum & Barbara C. Scholz 2002 ('Empirical assessment of stimulus poverty arguments', The Linguistic Review 19.9–50; and before that Pullum 1996, 'Learnability, hyperlearning, and the poverty of the stimulus', Berkeley Linguistics Society 22.498–513) probed that claim a little by looking in a readily available body of text, the Wall Street Journal corpus (WSJ). Telltale sentences showed up immediately. However, Anderson remarks: 'one might well question the extent to which the Wall Street Journal is representative of the input to the child' (804).
People have said such things before: Jerry Fodor ('Doing without what's within: Fiona Cowie's critique of nativism', Mind 110.99– 148, 2001; relying on a discussion in Cowie's What's within (Oxford UP, 1999) of the preliminary report in Pullum 1996); Janet D. Fodor and Carrie Crowther Fodor ('Understanding stimulus poverty arguments', The Linguistic Review 19.105–45, 2002, pp. 108–14); and Cedric Boeckx and Norbert Hornstein ('Les différents objectifs de la linguistique théorique [The varying aims of linguistic theory]', Cahier Chomsky, ed. by Julie Franck and Jean Bricmont, 61–77, L'Herne, 2007). People offer their armchair opinion that WSJ could not be relevant, but no one goes back and checks WSJ. The first telltale sentence in WSJ's forty-four million words is the sixteenth interrogative that occurs, and it is not an instance of financial journalistic prose or an editorial about capitalism. It is a nine-word sentence from spontaneous speech, the penultimate sentence of this passage:
Afterward, one of Mr. Tsongas's partners at the Boston law firm of Foley, Hoag & Eliot, told him: 'You've been invited to join your last corporate board.'
Mr. Tsongas says he is puzzled by such observations. 'Is what I'm doing in the shareholders' best interest? Then what's the problem?'
It is just wrong to assert that WSJ cannot provide any evidence that might bear on child language acquisition...
