On considerations of method and theory governing the use of clinical categories in neurolinguistics and cognitive neuropsychology: the case against agrammatism.

The pervasive use of clinical categories of aphasia in neurolinguistic and cognitive neuropsychological research reflects the assumption that these patient groupings represent disruptions of the normal language-processing system along theoretically significant lines. This premise is examined here with particular reference to the status of ‘agrammatism’. It is argued that there are compelling reasons to question the coherence of agrammatism as a psychological entity. To overcome these objections, the clinical intuitions on which this aphasic category is based must be replaced by objective criteria for selecting a theoretically significant patient grouping. To reach this goal, it is especially important that a theoretically motivated distinction be made between withinand across-category variation. It is argued that in the case of categories like agrammatism, there are serious methodological obstacles which make such goals unreachable. It is argued, therefore, that our theories should not take categories like agrammatism as psychological givens, especially if the purpose of our research is to reach an understanding of the mechanisms of language processing or of individual aphasic deficits themselves. Furthermore, it is argued that the single case methodology for the study of aphasia can address these goals by taking patterns of performance on particular linguistic tasks as basic units of analysis, and that this approach avoids the methodological pitfalls faced by studies which take clinical categories as starting points. *This research was supported by NIH grants NS14099 and NS07402 to The Johns Hopkins University. We would like to thank a number of people for favoring us with their helpful criticisms and comments at various stages of our research. These include Howard Egeth, Roberta Goodman, Gary Hatfield, Paul Hofer, Steven Lapointe, Michael McCloskey, John Marshall, and especially Domenico Parisi for discussions which provided much of the impetus for writing this paper. Requests for reprints should be addressed to William Badecker or Alfonso Caramazza at the Department of Psychology, The Johns Hopkins University, Baltimore, MD 21218,