Agrammatism

Mary-Louise Kean

• Publisher: Academic Press
• Year: 2013
• Print ISBN: 9780124028302
• eBook ISBN: 9781483277660
• Topic: Medicine
• Subtopic: Clinical Medicine

Agrammatism provides an overview of the state of knowledge on agrammatism, typically defined as a disorder of sentence production involving the selective omission of function words and some grammatical endings on words. The book opens with discussions of the diversity of the disorder. This is followed by separate chapters that address primarily questions of syntactic structure in agrammatism, from both linguistic and psycholinguistic perspectives. Within these two gross sections there is no consensus among the conclusions reached by the various authors. However, the position is taken that agrammatism is a disorder distinct from other aphasie disorders of sentence structure. This position is reconsidered in the final two chapters. Because of the intrinsically interdisciplinary character of research on agrammatism, it is hoped that the work presented in this volume will be of interest to linguists and psycholinguists working in areas outside the domain of aphasia, as well as to neurolinguists and neuropsychologists who are already involved in the study of language deficits.

1

Is Agrammatism a Unitary Phenomenon?^*

HAROLD GOODGLASS and LISE MENN

Publisher Summary

This chapter provides an overview of the concept of Agrammatism. In the year 1918, Deleuze was the first to provide a description of Agrammatisim, wherein it was described as a remarkable feature in the speech by certain aphasic patients. This phenomenon has recently come under the supervision of both linguists and neuropsychologists owing to the significance of it as a channel linking the linguistic construct to the brain mechanisms of language. However, the description has changed from the prior one during the modern Era, which characterizes solely as in terms of changes in the linguistic structure of speech output.

History of the Term ‘Agrammatism’

Agrammatism, first described as a striking feature in the speech of certain aphasic patients by Deleuze in 1819,¹ has recently come under the combined scrutiny of linguists and neuropsychologists because of its seeming key importance as a means of relating linguistic constructs to the brain mechanisms of language. This essay is written in part because of the swing of the pendulum in the neurolinguistic community to regard agrammatism as a panlinguistic disorder, detectable in all modalities, with the underlying presumption that there is a central representation of grammatical knowledge that governs the processing of syntax and morphology in all modalities of linguistic input and output. Beginning with a review of some key cases that both support and contradict the notion of a central disturbance of grammar, we propose to review the shortcomings of recent approaches and attempt to restate the issues as we see them now.

It is important to note that descriptions of agrammatism from the earliest discussions (e.g., Pitres, 1898) to the contemporary period (e.g., Luria, 1970; Tissot, Mounin, & Lhermitte, 1973) define its features only in terms of changes in the linguistic structure of speech ouput. These changes, as summarized by Tissot et al. (1973), are:

1. The deletion of function words in discourse, that is, the deletion of conjunctions, prepositions, articles, pronouns, and auxiliary verbs, and copulas (Notable exceptions to this are the conjunctions and and because).

2. The predominance of nouns, at the expense of verbs, in some forms of agrammatic speech.

3. The loss of verb inflection, with substitution of the infinitive for finite verb forms.

4. Loss of agreement of person, number, and gender, most notable in inflected languages. Jakobson (1963) points out that in languages with case declensions for nouns, nouns revert to the nominative form.

Descriptions of agrammatism by French, German, and Russian writers show remarkable parallels, with respect to the foregoing features in the three languages, except that in Russian, use of the infinitive for inflected verbs is sometimes reported not to occur (Tsvetkova & Glozman, 1975; but see Tonkonogy, 1968). Since English has few noun and verb inflections, the loss of bound morphemes is less conspicuous. Further, since English verbs do not have an infinitive marking ending, the loss of verb inflection is expressed only by the omission of auxiliaries, of the final s of the third person singular or the ed of the past tense, except in the case of those irregular verbs (e.g., to go) where the infinitive uses a different stem from some of the finite forms. Work by Grodzinsky (1984), on the other hand, suggests that Hebrew verb inflections, which are formed largely by vowel changes (infixation), are not lost in agrammatism.

Individual Variations within Agrammatism

As Tissot et al. (1973) point out, the foregoing features are not constant in all utterances of a patient, nor are they all necessarily found in all agrammatic speakers. Actual counts of grammatical morphemes may show near normal proportions in some agrammatic speakers. Substitution of the infinitive for inflected verb forms is frequent in some cases, sporadic in others. Some samples feature strings of nouns, verbs, and short phrases conveying a narrative by their mere temporal sequence. In these productions, noun–verb order may not be honored because many of the disconnected words have the quality of one-word sentences. In other agrammatic samples the major elements of sentences are present, word order is always maintained within the sentence, but many grammatical morphemes, including copulas, are omitted and verb tense inflections are suppressed or converted to the present. Manifestations of agrammatism in non–Indo-European languages have not yet been clearly described in Western scientific writings.

Older Clinical Accounts and Explanations of Cross-Modal Associations and Dissociations

Clinical case reports of agrammatic patients, prior to the current period of formal psycholinguistic study, by no means ignored the question of the manifestations of concomitant grammatical disturbances in other modalities, nor were early investigators content with superficial impressions. Yet, since their reports are generally based on studies of individual cases or very small series, there are conflicting descriptions. For example, Pitres’ patient, a man of 27 who was stricken while talking on the phone in his office, wrote in a telegraphic style very similar to his speech. Thus a written account of his day’s activity: “Hier; bureau; j’ecrivais; les livres, permis acquits. Grande expedition. Le soir, le diner, puis au cafe, manille aux encheres. Puis je vais me coucher.” (Note the dearth of verbs and of predicate structures in general.) Kleist (1916) also notes that the agrammatism of oral language is generally paralleled by similar changes in the form of written expression and of oral reading of connected text. Isserlin (1922), however, reports that the written syntax of agrammatic patients is often superior to that of their oral output. Ombredane (1951), taking an even stronger position on the autonomy of written from oral expression, holds that the telegraphic style of speech as seen in Broca’s aphasia is a function of conditions specific to the demands of oral communication. He states (1951, 22), “The proof of this is that the patient who speaks in correct telegraphic style [i.e., correct with respect to selection and ordering of the content words] tends to use grammatical forms in written expression.” Ombredane’s explanation for this, however, is based on his psychological intuition that the exigencies of face-to-face communication do not allow time for reflection and for examination of preceding words, which encourage more complete grammatization in written discourse. This explanation is at variance with the observation of Goodglass and Hunter (1970), who found that, while the oral and written expression of an agrammatic patient showed parallel deficiencies, certain agrammatic features were less pronounced in his speech than in his writing, for example, longer runs of grammatically organized words, higher verb to noun ratio, with fewer verb omissions in speech. They attributed this finding to the fact that the slowness of writing and the complete absence of cues from rhythm and intonation deprive the patient of contextual supports for grammatization that are present in speech. Goodglass and Hunter’s interpretation would lead to the position that when agrammatism is more severe in oral than written language, the dissociation is due to a difficulty in the production of grammatical morphemes that is specific to the oral modality-a dissociation whose explanation is not transparent to common sense intuition. A single anecdotal illustration by one of us (Goodglass) can be cited to support the latter view. A young World War II veteran, agrammatic in the course of recovery from Broca’s aphasia, was asked to repeat a series of short phrases, one beginning with If. After several attempts in which he repeated all but the opening word, he expressed his awareness of his difficulty by tracing the letters I-F on the desk.

Although there are cases of dissociation between agrammatism in oral and written expression, the most common configuration is that described by Alajouanine:

Agrammatism is hard to define other than by the essential fact which the patients’ speech makes evident; reduction of the sentence to its skeleton, relative abundance of substantives, almost invariable use of verbs in the infinitive, with suppression of the small words (the function words of language) and loss of grammatical differentiation of tense, gender, number, as well as of subordination; the richer a language is in distinctions of these types, the more glaring agrammatism will appear and it will grow still more apparent as recovery of access to vocabulary takes place (whence the impression that agrammatism seems to increase in severity in the course of reeducation). This agrammatic difficulty is seen also in oral reading and in writing to dictation. (1968, p. 84)

It was Salomon (1914) who first proposed the existence of a disorder of syntactic comprehension. He observed that a patient with a severe mixed aphasia was unable to distinguish well-formed from ill-formed sentences presented to him orally. But subsequent writers (Isserlin, 1922; Kleist, 1916), on the basis of similar examination procedures, rejected the notion that ‘receptive agrammatism’ is associated with the oral agrammatism of Broca’s aphasia. For example, the first of Isserlin’s three patients spoke spontaneously in telegraphic style but understood prepositions, conjunctions, and adverbs and grammatical forms in general. He was quick and error free in distinguishing correct from incorrect grammatical forms. His oral reading, and his repetition after the examiner, tended to maintain the grammatical integrity of the model supplied to him. However, a second agrammatic patient of Isserlin’s failed to distinguish correct from incorrect prepositions and inflections although he demonstrated good comprehension of well-formed sentences, while the third failed both in comprehension and in judgments of grammaticality. Yet Isserlin concludes, as does Kleist 12 years later, that receptive agrammatism is not a concomitant of agrammatic speech but is associated regularly with sensory (Wernicke’s) aphasia. Tissot et al. (1973) make the observation that failure to distinguish between well-formed and ungrammatical speech is a commonplace and undiscriminating feature of most severe cases of aphasia—an observation that is fully in agreement with our own clinical experience.

Thus the period up to 1950 was largely preexperimental and descriptive, with intuitive rather than theoretically motivated interpretations of agrammatism. Yet much of the raw data on which current studies are based were already recorded. Some of it seems to have been overlooked in the experimental work of the 1970s. The primacy of the speech ouput disorder as the defining criterion of agrammatism is consistent. The usual but not constant involvement of written expression and oral reading is noted, while an auditory receptive form of agrammatism is suggested as a feature more regularly associated with Wernicke’s aphasia than with agrammatic speech. The operations defining receptive agrammatism (e.g., judgments of grammaticality), however, have no specific relation to the main features of agrammatic production.

Accounts of Agrammatism

The first attempts to subsume agrammatism in a theoretical linguistic framework are those of Jakobson (1956, 1964). Jakobson saw the polar opposition between the fluent and anomic forms of aphasia on one hand and the nonfluent, agrammatic form on the other as representing the opposition between the paradigmatic and syntagmatic poles of language. The breakdown of the syntagmatic side, which he referred to as ‘contiguity disorder’, is expressed in agrammatism, where the linguistic elements that encode relationships between content words tend to disappear from speech. Jakobson’s interpretation goes beyond the mere renaming of the phenomena with a term derived from a linguistic construct. Where many authors had seen the agrammatic form as the p...

Table of contents

• Cover image
• Title page
• Table of Contents
• Inside Front Cover
• Copyright
• Contributors
• Preface
• Chapter 1: Is Agrammatism a Unitary Phenomenon?
• Chapter 2: A Multicomponent Deficit View of Agrammatic Broca’s Aphasia
• Chapter 3: Agrammatism: Structural Deficits and Antecedent Processing Disruptions
• Chapter 4: The Status of the Syntactic Deficit Theory of Agrammatism
• Chapter 5: Syntactic and Semantic Structures in Agrammatism
• Chapter 6: Two Notes on the Linguistic Interpretation of Broca’s Aphasia
• Chapter 7: On Parallelism between Production and Comprehension in Agrammatism
• Chapter 8: Agrammatism versus Paragrammatism: A Fictitious Opposition
• References
• Author Index
• Subject Index
• PERSPECTIVES IN NEUROLINGUISTICS, NEUROPSYCHOLOGY, AND PSYCHOLINGUISTICS