•“Chomskyclaims thatthereisa universal,innateDeepStructurethatencodes themeaning of language.”This
reading of Chomsky was responsible for the powerful appeal of generative grammar to philosophers and
psychologists (as well as the general public) in the 1960s: it promised that generative grammar would at last
provide a key to meaning, the holy grail of the study of mind. Unfortunately, Chomsky intended no such
thing. As just observed, Deep Structure itself is not necessarily universal. In addition, it is not meaning: it is
just a levelof syntactic structure. The misinterpretationwas encouraged by the fact that in theAspectstheory,
following a proposal by Katz and Postal (1964), Deep Structure was hypothesized to be the levelof linguistic
structure that, though not meaning itself,determinesmeaning.^40Thereby hangs a tale, if I may be permitted a digression. TheAspectstheory inspired considerable research into
whether Deep Structure couldbebrought stillcloser to meaning. One of theoutcomes was thevery popular theory of
Case Grammar (Fillmore 1968), in which semantic relations such as Agent, Patient, and Goal were coded directly as
“case-markers”in Deep Structure; this formed the basis for much subsequent research in computational linguistics
and cognitive psychology. More radical was the theory of Generative Semantics (McCawley 1968; Lakoff 1971; Postal
1970a), in which Deep Structure was posited to beidenticalwith meaning. This led to a startling expansion of the
overall size of posited underlying syntactic structures, plus a much more elaborate set of derivational rules and
constraints. Outside of Chomsky's immediate circle, Generative Semantics became phenomenally attractive, and with
good reason: if it were correct, the holy grail would have been achieved. However, Chomsky (1972a) and his students
(e.g. Akmajian 1973; Culicover 1972; Jackendoff 1972) attacked it vigorously; the ensuing intellectual melée has with
UNIVERSAL GRAMMAR 73
(^40) This misconstrual is further encouraged by occasional passages like these:“In place of the terms‘deep structure’and‘surface structure,’one might use the corresponding
Humboldtian notions‘inner form’of a sentenceand‘outer form’of a sentence.... The terms‘depth grammar’and‘surface grammar’are familiar in modern philosophyin
something roughly like the sense here intended (cf. Wittgenstein's distinction of‘Tiefengrammatik’and‘Oberflächengrammatik,’1953, p.168)...The distinction between deep
and surface structure, in the sense in which these terms are used here, is drawn quite clearly in the Port-RoyalGrammar.”(Chomsky 1965: 198–9)“The deep structure that
expresses themeaningis common toall languages,so itis claimed [bythePort-Royalgrammarians], beinga simplereflectionoftheforms ofthought.”(Chomsky 1966: 35)
Onecanbefairlycertainthattheauthorscitedreallydid nothavea levelofsyntaxinmindwhentheyspokeof“inner”or“deep”form; theymeant“meaning.”Theproblem
for theseearlier grammarians was that,lackingthetoolsof formal logicand thelike, theonlyway theyhad to talk aboutthemeaning of a complexsentencewas in terms of
simpler related sentences. Chomsky chooses to interpret this practice, for better or for worse, as an implicit version of his own theory of syntax.