Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives


SRB Archives

This review appeared in Volume 2 (1) of The Semiotic Review of Books.

Tell-Tale Aphasia

by Michel Paradis

The Pragmatic Basis of Aphasia. A Neurolinguistic Study of Morphosyntax Among Bilingual. By Marc L. Schnitzer. Hillsdale, N.J.: Lawrence Eribaum Associates, 1989.x . 213pp.

Contrary to what the title clearly indicates, the author does not claim that aphasia has a pragmatic basis. In fact he demonstrates the opposite: asphasic patients unable to use grammar, have recourse to the pragmatic aspects of language use. Nor does the book investigate bilingual aphasia per se. The fact that the subjects in this study are bilingual is not particularly relevant to the author's argument nor to the interpretation of the findings. Different features of Spanish and English are used as evidence, but the data might just as well have been collected from unilingual speakers of each language. (Some of the subjects are, in fact, unilingual Spanish speakers.) Why the author deliberately chose a misleading title for this book is not explained. A more appropriate title would have been something like The Progression from Syntax to Pragmatics in Aphasia, with, as a subtitle, An Integration of Gizon's Communicative Modes and Brown's Microgenesis Theory.

The volume consists of a detailed description of the linguistic performance of aphasic patients on a variety of tasks provided by a test battery the description of which is the sole subject of Chapter 2. Chapter 3 relates the test administration procedures Chapter 4 describes the subjects, Chapter 5 provides the results, and Chapter 6 interprets those results. Chapter 7, the final chapter, discusses the relevance of the study to related fields such as first and second language acquisition, pidgins and creoles, language death, and "talking apes." A final section discusses implications for theoretical linguistics.

The test battery was comprised of a sentence comprehension test and a grammaticality judgement cum correction test. ln the former, subjects were presented with 8 sentences, each of which was followed by three questions. In the latter, they were presented with 9 sentences and had to say whether they were ungrammatical, and if so, to correct them. Each task was available (though by no means performed by the majority of patients) in each of 4 combinations of input (listening or reading) and output (speaking or writing) modalities in each language.Spanish sentences contained morphosyntactic structures different from the English ones. That is, present past verb morphology, plural formation, and possessive and 3rd person singular present inflections were presented for English, while various verbal paradigms and number and gender agreement forms were presented in Spanish.

Although the sentences used in the Spanish and English sentence comprehension tests were selected according to syntactic criteria, frequencies of correct and incorrect responses failed to reveal a correlation between scores and syntactic properties of the test sentences or the questions concerning them. Pragmatic properties of the sentences did however appear to correlate with item scores.

On the sentence comprehension tasks, there were significantly higher proportions of correct responses (1) to items categorized as transparent (i.e., not requiring lexical recording or syntactic transformation) than to those characterized as opaque (i.e., that could not be answered by following the word order and/or vocabulary of the original sentences); (2) to pragmatically normal questions than to pragmatically odd ones; (3) to questions to which there were no reasonable alternatives than to the ones with alternatives; and, with a somewhat lesser degree of significance; (4) to questions that required a short answer than to those requiring a long answer.

On the syntax correction tasks, overall, subjects responded correctly more often to the two sentences that were morphosyntactically correct than to the seven incorrect ones. On the English verb morphology tasks, subjects performed better on the past tense responses than on the past participle. On the Spanish verb morphology tasks, subjects performed better on the present tense than on the future, conditional, or present perfect. There was also a general tendency to perform better in speaking than in writing.

Results are interpreted to be consistent with aphasia as a unitary disorder involving a regression in communication capability along Givón's communicative continuum. Aphasia is thus considered a regression in linguistic processing and/or representation in the direction of the pragmatic mode. In other words, aphasic patients rely on pragmatics as a compensatory strategy when syntax fails them. The book (and in particular its "Afterword") makes the point that, with pragmatics, one can go a long way towards successful verbal communication. It is really a plea in favour of "the Givón framework" (p. 154) so that in the conclusion (Chapter n evidence is explored that is independent of the study and that derives from fields other than neurolinguistics. The applicability of Givón's (1979) arguments concerning morphosyntactic theory is briefly discussed in view of first language acquisition (pp. 155-162), second/foreign language acquisition/learning (pp. 162-164), pidgins and creoles (pp. 164-165), language death (pp. 165-166), and "talking apes" (pp. 166-182).

Thus, based on his interpretation of the results, Schnitzer makes the basic claim that aphasia tends to reduce linguistic performance to a pragmatic-mode level. Pragmatic-mode strategies are used to compensate for the loss of syntactic -mode processing. He also makes the claim that more automatic, less computational syntactic-mode forms are more resistant to aphasic dissolution than less automatic, more computational forms. Schnitzer's interpretational framework is then based on Jason W. Brown's (1977, 1988) evolutionary and neurological hierarchical model of cerebral language processing and on Givón's communicative modes (monopropositional, pragmatic, and syntactic) which are integrated within Brown's model in that these three modes are considered to represent points on the continuum of phylogenetic, ontogenetic, and even possibly diachronic (with the transition from pidgins to creoles) communicative development.

Aphasia is thus interpreted phenomenologically as a regression along the communicative continuum and neurologically as a regression along the phylogenetic hierarchy. Schnitzer's results could equally be interpreted as supporting the view that linguistic competence (the grammar, Givón's "syntactic mode") is subserved by the left hemisphere and is hence vulnerable to a lesion in the classical language areas of the left hemisphere, while pragmatic aspects of language are subserved by the right hemisphere, and are hence vulnerable to lesions in the right hemisphere, as an increasing body of evidence tends to indicate. According to this account, a left-hemisphere lesion therefore impairs access to linguistic competence but leaves available access to pragmatic aspects of language subserved by the right hemisphere (Deloche and Seron, 1981; Hupet, Seron and Frederix, 1986). Similarly, patients with right-hemisphere lesions have been reported to keep their language (i.e., linguistic competence) intact, with no perceptible deficits in their phonology, morphology, syntax or lexicon in the interpretation of the literal meaning of isolated sentences. On the other hand, these same patients were reported to have considerable difficulty with the pragmatic aspects of language (e.g., Wagner, Hamby and Gardner, 1981; Kaplan, Brownell, Jacobs and Gardner, 1990).

In the context of the recent discredit of the agrammatism/paragrammatism distinction (e.g., Goodglass and Menn, 1985; Heeschen, 1985, Howard, 1985), Schnitzer argues for a unitary explanation of aphasia, according to which both Broca's and Wernicke's aphasia represent a regression along the Givónian continuum. Paragrammatics are considered to simply have a more severe aphasia than agrammatics.

Schnitzer interprets his results as strong support for "a vertical (i.e. hierarchical) approach and complementary to a horizontal (i.e., localizationist) one." (p. 5) While the author remarks in a note that the distinction between "vertical" and "horizontal" does not correspond to Fodor's (1983) distinction between horizontal and vertical faculties, he offers no explanation for the use of these terms in a sense other than their conventional meaning in cognitive neuropsychology. "Hierarchical" and "localizationist" (or "modular") would have been quite adequate, with no need to introduce this confusion in terms.

The investigation reported in this book is considered by Schnitzer to be "an extended pilot study. which offers sufficient food for thought to warrant future studies of Givóns's approach to language evolution as applied to applied to aphasic linguistic dissolution" (p. 5). A pilot study it is, and food for thought it does provide in abundance and variety. Schnitzer does concede that "the conceptual framework of Taby Givón does not account for everything and at best provides a conceptual framework for looking at aphasia, one that needs to be filled in by detailed study of aphasic phenomena in which the framework is used as a guide to research" (p. 154).

One other point that needs to be mentioned is that the book represents a cross-linguistic study rather than a study of aphasia among bilingual. No systematic comparison is made (nor can be made) of each patient's performance in each language because not all sections of the test were administered in both languages to all patients. In fact, most patients took only a subset of the test battery. The assessment ranges from only one test in one language (subject 10, a balanced bilingual) to 4 subtests in both (with 2 patients having responded to 4 subtests in English and 5 in Spanish). Therefore, the statement that "aphasia affects the two languages in much the same way" (p. 7) may be too strong a claim, based as it is on only 7 Spanish dominant and 5 Spanish-English balanced bilingual (12 subjects in all) and on comprehension and grammaticality judgement tasks Of the sentence comprehension tests Schnitzer states, "identical pragmatic factors are shown to correlate in the same way with the results in the two languages" (p. 7). Yet Wulfeck, Juarez, Bates and Kilborn (1986) have found that Spanish and English speakers will rely differently on different grammatical and pragmatic cues (e.g., word order, subject-verb agreement, or animacy), and that bilingual subjects will adopt different strategies when it comes to using these cues (e.g., L1 cues for both, or L1 cues for L1 and L2 cues for L2).It is not clear whether the subjects in this study were relying at all on grammatical cues such as word-order, subject-verb agreement or only on purely pragmatic ones such as animacy. The sample size is probably too small and the nature of the stimuli not appropriate to address this issue fully.

Yet, as Schnitzer appropriately points out, "the differing grammatical structures of the 2 languages do not permit the analysis of identical factors in the majority of the morphological and syntactic tests" (p. 7). Thus the similarity between the effects of aphasia in the two languages is limited to the fact "that pragmatic mode factors of similar kinds relate to dissolution in both languages of the bilingual" (p. n. In other words, as expected, the bilingual did not show evidence of differential aphasia (different symptoms in each language), but the nature of the tests does not speak to the issue of differential recovery (one language recovered better than the other relative to premorbid competency). The author also states, "in both languages we find the kind of performance which can be analyzed as a regression along the communicative continuum." (p. 7) If this had not been the case, it would have been strong counter-evidence for the theory, whether coming from bilingual or from unilinguals.

The nature of the deficit and of the compensatory strategies is the same for both languages, but scores in Spanish (the language used most frequently, even by the balanced bilingual) tended to be higher than in English. Because of the small number of subjects and the large degree of heterogeneity with respect to the various independent variables involved--sex, age, duration of aphasia (from 1 to 160 weeks), level of bilingualism, localization of lesion, etiology, and degree of education--it is difficult to make any strong generalizations about bilingual aphasia. However, in the context of the ongoing controversy regarding the alleged increased participation of the right hemisphere and the consequent expectation of a higher incidence of crossed aphasia in bilingual (Mendelsohn, 1988; Solin, 1989; Paradis, 1990), it is interesting to note that all 13 aphasic patients for-whom localization data was available had left hemisphere damage (3 bilateral--of whom 2 unilinguals).

Schnitzer makes a number of far-reaching, albeit controversial, proposals. He suggests, for instance, that linguistic communication at the syntatic-mode level represents an immense and largely inaccessible hierarchy. According to him langauge cannot be considered as a coherent single system and cannot be viewed as organic wholes. The truly linguistic (i.e., syntactic-mode) part of communicative competence cannot be viewed as a separate system. He concludes that the goal of writing a descriptively adequate grammar is inherently unattainable and he argues that grammatical knowledge (i.e., an individual's "linguistic competence") is intrinsically incomplete because a speaker cannot assign structural descriptions to all and only the sentences of a language.

The book is thus of genuine interest to researchers in the field of linguistics, pragmatics, neurolinguistics, aphasiology, and language therapy. It does not, however, as its subtitle might lead the reader to expect, provide insight into bilingual aphasia per se, or into a theory of bilingualism. The book's major contribution is to invite the reader to cast a fresh look at generally accepted ideas and consider them from new and unconventional perspectives.


Brown, Jason W. (1977) Mind, Brain and Consciousness: The Neuropsychology of Cognition. New York: Academic Press.

Brown, Jason W. (1988) The Life of the Mind: Selected Papers Hillsdale, N.J.: L. Eribaum Associates.

Deloche, Gerard and Seron, Xavier (1981) "Sentences understanding and knowledge of the worid: evidence from a sentence-picture matching task performed by aphasic patients." Brain and Language, 14:57-69.

Fodor, Jerry A. (1983) The modularity of mind. Cambridge, MA.:The MIT Press.

Givón, Talmy (1979) On Understanding Grammar. New York: Academic Press.

Goodglas, Harold and Menn, Lise (1985) "Is agrammatism a unitary phenomenon?" In M.-L. Kean (ed.), Agrammatism. New York: Academic Press. 1-26.

Heeschen, Claus (1985) "Agrammatism versus paragrammatism: afictitious opposition." In M.-L Kean (ed.), Agrammatism. New York: Academic Press. 207-248.

Howard, David (1985) "Agrammatism." In S. Newman and R. Epstein (eds.), Current perspectives in dysphasia. London: Churchill Livingstone. 1-31.

Hupet, Michel, Seron, Xavier and Frederix, Michel (1986) "Aphasic's sensitivity to contextual appropriateness conditions for pragmatic indicators." Brain and Language, 28:126-140.

Kaplan, Joan A., Brownell, Hiram H., Jabobs, Janet R. and Gardner, Howard (1990) "The effects of right hemisphere damage on the pragmatic interpretation of conversational remarks." Brain and Language, 38:315-333.

Mendelsohn, Susan (1988) "Language lateralization in bilinguals: facts and fantasy." Journal of Neurolinguistics, 3:261-292.

Paradis, Michel 1990. "The lateralization of language in bilinguals: enough already!" Brain and Language, 39: (in press).

Solin, Doreen 1989. "The systematic misrepresentation of bilingual crossed aphasia data and its consequences." Brain and Language, 36:92-116.

Wagner, Wendy, Hamby, Suzanne and Gardner, Howard 1981. "The role of the right hemisphere in the apprehension of complex linguistic materials." Brain and Language, 14:15-33.

Michel Paradis is a professor in the Department of Linguistics at McGill University. He has been conducting research on the neurolinguistic aspects of bilingualism for the past 15 years and has developed the Bilingual Aphasia Test which has been transposed into 50 languages so far. Recent publications include The Assessment of Bilingual Aphasia (Lawrence Eribaum Associates, 1987), "Bilingual and polyglot aphasia," in F. Boller and J. Grafman (eds.), Handbook of Neuropsychology, vol. 2, (Elsevier, 1989) and "Cerebral lateralization in bilingual: enough already!"Brain and Language (in press).

Go to Semiotic Review of Books Home Page
Go to SRB Highlights
Go to SRB Archives