Syntactic processing in agrammatism: A reply to Zurif and Grodzinsky

In Linebarger et al. (1983a), we reported that four agrammatic aphasics were able to judge the grammaticality of English sentences embodying a wide’ range of syntactic constructions. This excellent performance on the grammaticality judgment task was in sharp contrast to their poor performance on standard tests of sentence comprehension. Thus we were led to conclude that the comprehension deficit in these patients is not the result of an across-theboard inability to perform syntactic analysis, nor of a loss of the ability to make syntactic use of the closed class vocabulary. We made certaim quite tentative proposals concerning alternative explanations for the agrammatic sentence comprehension deficit: (a) the ‘trade-off hypothesis’, which claims that agrammatic listeners are capable of both syntactic analysis and semantic interpretation, but achieve their best performance in one only at the expense of the other; and (b) the ‘mapping’ hypothesis, which claims that they are capable of syntactic analysis but are unable to exploit syntactic representations for the purposes of semantic interpretation. In their reply to this paper, Zurif and Grodzinsky (1983) (hereafter called Z&G) argue from two angles against our conclusion that “the comprehension deficit in agrammatism does not reflect loss of the capacity to analyze syntactic structure”. First, they grant our assertion that our agrammatic subjects have in fact demonstrated considerable proficiency in syatactic analysis, but argue that because of various shortcomings of the grammaticality judgment