Agrammatism: A Cross-Linguistic Clinical Perspective

Abstract

You have accessThe ASHA LeaderFeature1 Dec 2005Agrammatism: A Cross-Linguistic Clinical Perspective Barbara O’Connor, Inge Anema, Datta Hia, Teresa Signorelli and Loraine K. Obler Barbara O’Connor Google Scholar More articles by this author , Inge Anema Google Scholar More articles by this author , Datta Hia Google Scholar More articles by this author , Teresa Signorelli Google Scholar More articles by this author and Loraine K. Obler Google Scholar More articles by this author https://doi.org/10.1044/leader.FTR3.10172005.8 SectionsAbout ToolsAdd to favorites ShareFacebookTwitterLinked In Speech-language pathologists who serve people with aphasia must be prepared to evaluate and treat agrammatism. We focus here on fundamental information about this communication disorder, particularly its features in English, dialects of English, and several different languages around the world. It is important to examine agrammatism across dialects and languages, since the disorder is not uncommon, and it is manifested differently, depending on the grammatical structure of the dialect or language in question. Definition Agrammatism is a form of speech production, often associated with Broca’s aphasia, in which grammar appears relatively inaccessible. In severe agrammatism, sentences comprise only strings of nouns; in milder forms, functor words (e.g., articles, auxiliary verbs) and inflectional affixes are omitted or substituted. Some aphasiologists suggest that comprehension of syntax is also impaired in agrammatism. In English The English language has a relatively constrained canonical sentence order: subject, then verb, then object (SVO). Varying that order carries grammatical meaning (e.g., passive). Grammatically speaking, Standard American English (SAE) contains a sizable number of free-standing functor words (i.e., “grammatical words”) and limited inflections. Inflections generally mark tense and plurality in SAE, and, except for irregular forms, are added to the root word without altering the original word structure. Thus, in a sentence like, “She is speaking,” “is” is a free functor, whereas “-ing” is an inflection that marks present continuity. Agrammatism in English manifests itself primarily as the omission of, or substitution for, functors. Agrammatic speakers of English preserve word order, but omit free functors, like “is,” and inflections, like “-ing,” while retaining a telegraphic skeleton (“She speak”). The agrammatic speaker is thus able to produce a degree of connected speech but is missing some required grammatical information. Dialects of English are distinguished from SAE in features that play a role in agrammatism. Although there has been little study to date on how agrammatism is manifested in such language varieties, they are worth considering before turning to different languages. In African American Vernacular English Depending on one’s definition of “dialect,” there are at least 24 dialects of English spoken in the United States. African American Vernacular English (AAVE) is an interesting one to examine, since it differs in functor use from SAE. Some functors that are obligatory in SAE are optional in AAVE, for example, bound morphemes for plural and past tense. Additionally, AAVE contains grammatical forms unavailable in SAE, such as the habitual be (e.g., “He be early to school”) and double/multiple negation (e.g., “I ain’t signing no petition I don’t know nothing about” (Jones, 2002). Thus, determining if an AAVE-speaking client’s output reflects true agrammatic errors or acceptable dialectal variations becomes challenging for the clinician who speaks only SAE. In response to this challenge, Jean Jones (2002) conducted an innovative study examining agrammatism in a speaker of AAVE. Jones’s study included conversation, picture description, and experiments to examine habitual be, double and multiple negation, plural /Z/ marker, and past tense /D/ marker. The results showed that agrammatism could be distinguished from AAVE in spontaneous speech and picture description. There was remarkable preservation of the habitual be observed in its function both as copula and auxiliary, and differences across different types of negative functors (e.g., contractions and negation words such as “never”). These two findings were interesting because cross-language agrammatism studies generally show difficulty with the verb “to be,” and relative preservation of negation functors (Menn & Obler, 1990). Jones’s data suggest that examining the habitual be and negation in AAVE-speaking clients with agrammatism might lead to better dissociation of dialect from disorder. More generally, when appropriate, it is important for clinicians to ensure that clinical interactions include examining the pertinent features of AAVE or other varieties of English in spontaneous communication contexts. In Other Languages In the same way that agrammatism can manifest itself differently in two different varieties of English, it can be manifested differently across languages. To exemplify such differences, we have selected three languages that differ from SAE in interesting ways, two (Dutch and Spanish) relatively related to English, and another (Kannada) from an altogether different language family. Dutch. Dutch is spoken in the Netherlands and is a Germanic language, like English. Despite many similarities, certain differences (e.g., word-order flexibility), distinguish Dutch and English. Interesting agrammatism research is being conducted in Holland, with potential clinical relevance for American clinicians. This particular research is based on Adaptation Theory (Kolk, Van Grunsven, & Keyser, 1985), which holds that the speaker with agrammatism economizes because of an inability to keep planned sentences in short-term storage. This results in communication of only essential sentence elements. The agrammatic speaker compensates and adapts; hence, the name Adaptation Theory. A crucial aspect of this theory is that the resulting telegraphic style is seen as largely intentional. If, indeed, speakers with agrammatism reduce their output intentionally, the elements produced should still follow basic grammatical rules. Indeed, speakers follow these basic word-order rules in spontaneous production, which can be demonstrated using an example involving “ellipsis.” Ellipsis is the linguistic phenomenon whereby certain words can be eliminated from a grammatical construction, for example, “I bought eggs and juice,” instead of “I bought eggs and I also bought juice.” Speakers with agrammatism tend to have reduced output, which includes many elliptical constructions. In Dutch elliptical forms, word order depends on grammatical category. For example, object pronouns, particles, and adverbs follow finite verb forms and precede non-finite verb forms: “Hij eet snel” (finite form of the verb) He eats fast “Snel eten” (non-finite form of the verb) Fast eat Speakers with agrammatism follow these basic word order rules in spontaneous production. In treatment, clinicians then attempt to encourage production of such short, elliptical utterances. These short forms are less sensitive to early decay in short-term storage, and result in more efficient communication. The goal of these elicited short structures is to produce a main verb with its arguments (agent, theme, location). Correct verb conjugation and syntactic morphology are not treatment goals. The underlying Dutch word order, SOV, is used by clinicians in an attempt to generalize to clients’ utterances. An example of such a target utterance, based on a picture cue of a girl blow-drying her hair may be: Meisje haren fohnen (Girl hair blow-dry), instead of: Het meisje fohnt haar haar (The girl blow dries her hair; with correct verb conjugation, surface word order, and syntactic morphology) An advantage of this approach is that the client adapts to the underlying problems of agrammatism. A disadvantage may be the heavy demands it places on executive functioning. Spanish. Unlike English and Dutch, Spanish is an example of a highly inflected Indo-European language. Agrammatism in highly inflected languages has been of particular interest to researchers and clinicians, because the typical agrammatic symptom of functor omission is rarely observable in such languages. Spanish is the primary language of more than 20 countries, yet the aphasiology literature contains very few studies examining agrammatism in Spanish speakers, perhaps because functors are relatively spared. Ardila (2001) summarized the primary findings of the available studies published on Spanish aphasia. He noted that features of agrammatism in Spanish include restricted word order, clitic pronoun omission (e.g., “ella dice” for “ella me dice”-“she tells me”), and overuse of strong pronouns in the subject position (see also Reznik, Dubrovsky, & Maldonado, 1995). In contrast, features that are resistant to agrammatism in Spanish speakers are subject-verb agreement, use of the preposition “a” to distinguish object from subject and definite/indefinite articles (Ostrosky-Solis et al., 1999). Consider the comparison of agrammatism in Spanish speakers to that in English speakers conducted by Benedet, Christiansen, and Goodglass (1998). They found that the hierarchy of difficulty in production and comprehension of grammatical morphemes was similar in the two languages, with two exceptions: 1) Spanish-speaking people with agrammatism had better subject-verb concordance, due to the higher cue validity of subject-verb agreement in Spanish than in English, and 2) comprehension of passive sentences was worse in Spanish than in English, due, the authors maintain, to the greater frequency of passive constructions in English than in Spanish. Additionally, Spanish-speakers with agrammatism demonstrate over-reliance on simpler, earlier-acquired, and more frequently used verb forms (i.e., present tense; preterite), whereas complex verb forms (e.g., conditional) appear rarely during their spontaneous or elicited speech (Centeno & Obler, 2001). Therefore, clinicians treating Spanish speakers with agrammatism should use available grammatical morphemes and verb patterns as a springboard to the recovery of more complex verb forms. Kannada. Kannada is a language used by approximately 44,000,000 speakers in the South of India. It is an example of a non-Indo-European language that has a highly inflected morphological system, like Spanish, yet has very few free-standing functors. Most of its functors exist as inflections on verbs. These verbs are inflected such that if the inflection is dropped, the root morpheme becomes linguistically illegal. For example, unlike in English, where dropping “ing” from the word “speaking” gives rise to a real word, “speak,” in Kannada, dropping “tare” from the word “matadtare” (speaking) gives rise to the morpheme “matad,” which is not a legitimate Kannada word. Datta and colleagues (in progress) collected data, via narrative and picture-description tasks, from a 42-year-old Kannada-speaking individual with agrammatism. As expected, the patient demonstrated no functor omissions. In addition, there were no substitution errors. Low type-token ratios of his narrations, nonetheless (as compared to those of two typical control participants) illustrate that he lacked grammatical variety in his language productions, and tended to repeat the same canonical phrases. We conclude that agrammatism is manifested in a subtle fashion in languages that are inflected, such as Spanish and Kannada. Multilinguals Clinical management of agrammatism is already challenging since this disorder manifests itself differently across various languages and/or dialects. The challenge becomes even greater when patients speak two or more languages and/or dialects. We base our recommendations for assessment and diagnosis for bilingual clients primarily on the writings of Michel Paradis (2001; 2004). Assessing clients in all their languages and/or dialects is a critical part of understanding total communication competence. Ideally, the clinician should be a proficient speaker of their clients’ languages and/or dialects, or, at a minimum, possess enough knowledge about their patients’ languages and/or dialects to infer how agrammatism would manifest itself. Knowledge of pre-morbid function in terms of comprehensive language histories ascertaining, among other things, ages and manners of language and dialect acquisition, mastery level of each language variety, and nature and frequency of their use, are necessary to interpret clients’ current performance. When eliciting and analyzing language output, there are a number of factors clinicians should keep in mind. Formal language batteries, for instance, should be culturally and linguistically equivalent across languages. Direct test translations are generally inadequate, as syntactic structures and other linguistic elements do not necessarily have the same frequency or complexity level from language to language. Clinicians should also keep in mind that clients might perform differently in their languages and/or dialects, depending on the nature of the task. Clinicians should also be aware that there are a wide variety of recovery patterns for bilingual, and, presumably bidialectal, clients with agrammatism. Unidirectional or mutual language interference across grammatical systems may be present. Care must be taken to determine when code-switching is normal for a bilingual and when it is a symptom of aphasia, and whether output is a true agrammatic error or a surface-feature variation of a dialect. Treatment Research Treatment-research studies of agrammatism have historically been conducted primarily on English-speaking participants. Since agrammatism manifests itself differently in different languages and/or dialects, it is important to take an individualistic approach to treatment, by evaluating and treating the specific agrammatic features of the language(s) and/or dialects of our patients. However, there are two general approaches to treatment for agrammatism highlighted in the research literature that can be easily adapted to non-English-speaking patient populations. The first “functional communication approach” focuses on improving the semantic and functional communication content of utterances. The underlying notion here is that the deficit in agrammatism is one of overall expression and communicative effectiveness. The example in our discussion of Dutch agrammatism, where clients are encouraged to use short, elliptical forms, highlights this type of treatment approach. Another example is the training of requesting behaviors (Doyle et al., 1989), which engages clients in structured conversations with volunteer conversational partners. Clients are prompted to ask as many questions as possible about specific topics (e.g., health). The authors report increased use and complexity of unprompted requests over time, as well as generalization to conversational partners unfamiliar to their clients. The second “syntactic approach” emphasizes the retraining of syntax. For example, using the Helm Elicited Language Program for Syntax Stimulation (HELPSS; Helm-Estabrooks & Ramsberger, 1986), agrammatic clients relearn different syntactic structures (e.g., yes/no questions) in a story-completion format, with line-drawn pictures as prompts. The authors report that their English-speaking clients evidenced increased morpheme counts and content units, improved post-test scores, as well as positive gains in receptive language skills using this approach. Lastly, Linguistic Specific Treatment (Jacobs, 2001; Jacobs & Thompson, 2000) teaches clients with agrammatism to identify the components of a sentence (e.g., subject noun-phrase, verb, object noun-phrase) and how to move constituents of one sentence (e.g., active sentence) to derive another (e.g., passive counterpart). For example, a Spanish-speaker with agrammatism would be taught the components of an active voice sentence (“El hombre besa la mujer”-“The man kisses the woman”), and how to generate its passive counterpart (“La mujer es besada por el hombre”-“The woman is kissed by the man”). According to Jacobs (2001), the benefits of this treatment approach have been observed to generalize to narrative discourse in English, and we see no reason that a similar approach should not work in other languages as well. In conclusion, we hope that the information in this article will serve as a useful tool to guide clinicians in their evaluation and treatment of agrammatism. Since research studies on agrammatism have been conducted primarily on Standard English-speaking populations, and corresponding studies on dialectal varieties of English and/or other languages are sparse, clinicians are truly challenged when working with clients with agrammatic aphasia who speak languages and/or dialects other than Standard English. The assessment and therapeutic strategies highlighted in this study could have potential clinical relevance for non-Standard-English speaking clients, if modified based on the grammatical profile of that dialect/language. We encourage clinicians to document clinical cases of agrammatism in speakers of different English varieties, and in non-English-speaking patients, to further our profession’s understanding of this communication disorder. References Ardila A. (2001). The manifestation of aphasic symptoms in Spanish.Journal of Neurolinguistics, 14, 337–347. CrossrefGoogle Scholar Benedet M. J., Christiansen J. A., & Goodglass H. (1998). A cross-linguistic study of grammatical morphology in Spanish and English-speaking agrammatic patients.Cortex, 34, 309–336. CrossrefGoogle Scholar Centeno J., & Obler L. K. (2001). Agrammatic verb errors in Spanish speakers and their normal discourse correlates.Journal of Neurolinguistics, 14, 349–363. CrossrefGoogle Scholar Doyle P. J., Goldstein H., Bourgeois M. S., & Nakles K. O. (1989). Facilitating generalized requesting behavior in Broca’s aphasia: An experimental analysis of a generalization training procedure.Journal of Applied Behavior Analysis, 22, 157–170. CrossrefGoogle Scholar Helm-Estabrooks N., & Ramsberger G. (1986). Treatment of agrammatism in long-term Broca’s aphasia.British Journal of Disorders of Communication, 21, 39–45. CrossrefGoogle Scholar Jacobs B. J. (2001). Social validity of changes in informativeness and efficiency of aphasic discourse following Linguistic Specific Treatment (LST).Brain and Language, 78, 1, 115–127. CrossrefGoogle Scholar Jacobs B. J., & Thompson C. K. (2000). Cross-modal generalization effects of training noncanonical sentence comprehension and production in agrammatic aphasia.Journal of Speech, Language, and Hearing Research, 43, 5–20. LinkGoogle Scholar Jones J. (2002). The manifestation of agrammatism in a bidialectal speaker of African-American Vernacular English and Standard American English. Unpublished manuscript, City University of New York. Google Scholar Kolk H. H. J., Van Grunsven M. F., & Keyser A. (1985). On parallelism between production and comprehension in agrammatism.In Kean M. L. (Ed.), Agrammatism. New York: Academic Press. Google Scholar Menn L., & Obler L. K. (1990). Agrammatic aphasia: A cross-language narrative sourcebook (vols. 1,2,3). Philadelphia, PA: John Benjamins Publishing Company. Google Scholar Ostrosky-Solis F., Marcos-Ortega J., Ardila A., Rosselli M., & Palacios S. (1999). Syntactic comprehension in Broca’s aphasic Spanish-speakers: Null effects of word order.Aphasiology, 13, 7, 553–571. CrossrefGoogle Scholar Paradis M. (2001). The need for awareness of aphasia symptoms in different languages.Journal of Neurolinguistics, 14, 85–91. CrossrefGoogle Scholar Paradis M. (2004). A neurolinguistic theory of bilingualism. Philadelphia, PA: John Benjamins. CrossrefGoogle Scholar Reznik M., Dubrovsky S., & Maldonado S. (1995). Agrammatism in Spanish: A case study.Brain and Language, 51, 355–368. CrossrefGoogle Scholar Author Notes Barbara O’Connor, is a speech-language pathologist, doctoral candidate at the City University of New York (CUNY) Graduate Center, and instructor in Communication Sciences at Hunter College. Contact her at [email protected]. Inge Anema, is atrilingual speech-language pathologist who has worked in the United States, Germany, and the Netherlands. She is also a doctoral candidate at the CUNY Graduate Center. Contact her at [email protected]. Hia Datta, is a doctoral candidate in Speech and Hearing Sciences at the CUNY Graduate Center. She works in both the Neurolingistics Lab and the Developmental Neurolinguistics Lab. Contact her at [email protected]. Teresa Signorelli, is a bilingual speech-language pathologist and a doctoral candidate at the CUNY Graduate Center. Contact her at [email protected]. Loraine K. Obler, is a Distinguished Professor in Speech and Hearing Sciences at the CUNY Graduate Center. Contact her at [email protected]. Advertising Disclaimer | Advertise With Us Advertising Disclaimer | Advertise With Us Additional Resources FiguresSourcesRelatedDetails Volume 10Issue 17December 2005 Get Permissions Add to your Mendeley library History Published in print: Dec 1, 2005 Metrics Downloaded 4,684 times Topicsasha-topicsleader_do_tagasha-article-typesCopyright & Permissions© 2005 American Speech-Language-Hearing AssociationLoading ...