Aphasia: A Societal and Clinical Appraisal of Pragmatic and Linguistic Behaviours Glenn M. Goldblum, MA (Sp. & H.Sc.) (UC, Santa Barbara) Department of Speech Pathology & Audiology, University of the Witwatersrand, Johannesburg ABSTRACT Pragmatic abilities of eleven aphasics classified into fluent and nonfluent groups were examined, and compared with global ratings of communicative adequacy. Further, subjects 'pragmatic performance was compared with performance on two standardised linguistic meas- ures. All subjects demonstrated high levels of appropriate pragmatic behaviours and were apparently resourceful in using the context, frequently facilitated by use of compensatory communicative strategies. Similar and different pragmatic deficits were noted for both groups. Regardless of classification and linguistic severity, subjects were less impaired on pragmatic compared to standardised linguistic measures. Implications were highlighted, emphasising the importance of pragmatics relative to other aspects of language function. OPSOMMING Die pragmatiese vermoens van elf afasie-pasiente, wat ingedeel is in 'n vlot — en 'n onvlotgroep, is ondersoek en hul kommunikatiewe vaardighede is onderling vergelyk. Die proefpersone se pragmatiese prestasie is ook vergelyk met hul prestasie op twee gestandaardiseerde linguistiese toetse. Al die proefpersone het bewys gelewer van hoe vlakke van toepaslike pragmatiese gedrag en was oenskynlik vindingryk om die konteks te gebruik vir kompensatoriese kommunikatiewe strategiee. Tekortkominge in die twee groepe was soms dieselfde en soms verskillend van aard. Ongeag die klassifikasie van die groepe en die graad van linguistiese onvermoe, was die proefpersone deurgaans pragmaties minder belemmer as op die linguistiese vlak. Gevolgtrekkings is beklemtoon, wat die relatiewe belang van die pragmatiek tot ander aspekte van taalfunksionering onderstreep. As speech-language pathologists, one of our primary concerns has always been to improve communicative effectiveness of the individu- als we serve. In attempting to meet this goal, the intervention strate- gies we have used have been continually refined and revised-molded by the changing theoretical views of language over the past several decades. Earlier in our history many investigators including John- son (1946) and Van Riper (1939 cited by Prutting, 1982a) stressed the need for our goals to be mutually acceptable to both the client and society. Much of this societalj perspective was lost over the past few decades, perhaps in the name of objectivity and accountability. More recently, researchers in applied behavioural research and in the field of child and adult language disorders, (e.g. Kazdin (1977); Mueller (1983); Prutting (1982a); Prutting and Kirchner (1983); and Prutting, Kirchner, Hassan and Buen (1984)) have pointed out the need to move back clinically to viewing communication dis- orders from a social perspective, taking context into account, rather than attempting to control it. Despite this proposed change in per- spective, clinical aphasiologists still need to concern themselves with identifying behaviours and effecting changes that make socially rele- vant differences in the client's life. As in the discipline of science, the field of acquired adult aphasia has undergone many paradigmatic shifts which have resulted in con- ceptual and methodological reorganisation. At present there is a contemporary and ongoing shift towards a functionalist paradigm which appears to provide promise of a different theoretical frame- work for examining and attempting to understand the aphasic com- munication problem. Holland's observation (1977) that "aphasics probably communicate better than they talk" (p. 173), which views aphasia as a communication (rather than a language) problem, reflects this broadened functional perspective. Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol. 32, 1985 A perusal of the more recent literature reveals that despite the grow- ing interest towards pragmatics in aphasia, a diverse and large list of behaviours are included under the general rubric of pragmatics, in the absence of a comprehensive theoretical framework of natur- al aphasic communication. Such a framework remains to be deve- loped to help clinicians incorporate pragmatics more effectively into the treatment process. _ While most traditional and current assessments of aphasic language focus on the structural aspects of their verbal output, a few re- searchers have begun to be challenged by, and address the assess- ment of diverse communicative functions in aphasia. Using a variety of communication assessment protocols and observational meas- ures a few investigators including Guilford and O'Connor (1982); Holland (1980, 1982); and Penn (1983) have demonstrated gener- ally superior pragmatic abilities with little correlation between tradi- tional syntactic profiles and communicative performance. While their data has furthermore resulted in support for Holland's belief (1982, 1983) in the preservation of communicative competence in aphasia, Penn (1983) has warned that since the social context of communication is so redundant, we may be overestimating the aphasic's ability. Communicative competence compared to linguistic competence may thus best be viewed as less impaired in aphasia rather than intact. The most comprehensively linguistic study evident in this regard appears to be that by Penn (1983), who developed a Profile of Com- municative Appropriateness (PCA) based upon a relatively com- prehensive taxonomy derived from child language literature. Penn used her PCA (Penn, 1983) and the Language Assessment Remedi- ation and Screening Procedure (LARSP) (Crystal, Garman and Fletcher, 1976 cited by Penn, 1983) to obtain and compare com- © SASHA 1985 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) 12 municative and linguistic profiles of hem.sphere aphasic subject, termined by the Boston D J ^ ; ^ 1 9 8 3 ) . T h e r e s u l t s e t - • c i t e d b y p e n n ' 1 9 8 3 ) of the LARSr syntactic j ^ ] e s s e r e x t e n t s e v e n . correlated c.osely ^ . ^ ^ " f ^ o n s t r a t e d essentially idiosyn- skills by indivtdua. subjects, whtah were not n e c e s s a r i l y related to subjects' capacities on a syn- tactic level Severity and type distinctions did not always account for subject clusters on the PCA (Penn, 1983), particularly when individual scales were considered, although severity was postulat- ed to relate more closely to appropriate communicative performance than site of lesion or type of aphasia as predicted by traditional tests of aphasia. Though exploratory in nature, Penn's data (1983) suggesting the relative independence of syntactic and pragmatic aspects of apha- sia are extremely provocative, raising concern with two important issues. Firstly they cast some doubt on the validity of traditional classification schemes and their ability to reflect functional com- municative competence. Alternatively, Penn's data highlights the need to consider structural and functional data together to facili- tate a more comprehensive evaluation of an aphasic's communica- tive competence. In contrast to these data, in the area of child language disorders, Prutting et al, (1984) have recently demonstrated clearly different pragmatic profiles across normal, language and articulation disor- dered groups of children using a societally-based Pragmatic Pro- tocol (Prutting 1982b) inclusive of the behaviours discussed in the literature. Their documentation of a relationship between pragmatic function and linguistic performance, (which contrasts with the limit- ed data reported in the adult aphasic literature) may reflect Prut- ting et al's (1984) use of societal criteria to evaluate the childrens' language use, using a protocol based upon a sound theoretical frame- Glen Goldblum work (Austin, 1962; Searle, 1969 cited by Prutting et al, 1984) to evaluate a range of pragmatic behaviours. Furthermore, Mueller's data (1983) revealing a strong relationship between this protocol and measures of societal judgements, highlighted the apparent value of this tool as a means of approaching the assessment of social com- petence. While Prutting and Kirchner (1983), have emphasised the necessi- ty to gather and analyse data utilising methods which reflect ad- vances in our theoretical knowledge, to date there appears to be no study in the field of aphasia utilising societal criteria to analyse the individual's pragmatic strengths and deficits across a range of pragmatic behaviours. As of yet we have no clear understanding of how pragmatic abilities and deficits stratify across different types of aphasic clients. Furthermore, despite the increasing number of studies examining discrete pragmatic behaviours in aphasic adults, a marked paucity is nevertheless apparent of investigations evaluating pragmatic func- tion in relationship to linguistic performance. The theoretical, clin- ical and research consequences of using a societal Pragmatic Protocol (Prutting, 1982b) in conjunction with traditional clinical measures are apparently far-reaching. Not only would the effects of communicative behaviour in relation to societal values be con- sidered, but attempts would be made to merge clinical and societal goals into our intervention programmes, enhancing the overall com- municative competence of the aphasic individual. In view of these issues, this study examined the pragmatic abilities of eleven aphasic adults classified into fluent and non-fluent groups using the Western Aphasia Battery (WAB) (Kertesz, 1980) taxonomy, and compared them with global ratings of communicative adequacy. Furthermore, subjects' performance on a societal Pragmatic Protocol (Prutting, 1982b) was compared with performance on two clinical linguistic measures - the WAB (Kertesz, 1980) and the Communicative Abil- ities in Daily Living (CADL) (Holland, 1980). Table 1 Descriptive data for subjects used in the Study Subject Sex Age Months post onset Educational Level Premorbid Occupation Premorbid Communicative- ness rating on scale from 1-7* Apraxia rating on scale from 0-7** Dysarthria rating on scale from 0-7** Classification of type of aphasia on Western Aphasia Battery Behavioural Classification Severity rating on Western Alphasia Battery 1 Μ 70 7 High school Retired auto service manager 3 0 0 Wernicke's Fluent Moderate i 2 Μ 58 27 College Retired Navy test pilot 3 2 0 Conduction Fluent Mild- moderate 3 F 51 8 10th Grade Laundromat assistant, bartender 4 0 0 Conduction Fluent Mild I 4 Μ 69 5 8th grade Retired security guard 5 0 2 Anomic Fluent Mild 5 F 51 7 High School Retired legal secretary 3 0 0 Anomic Fluent Mild 6 F 69 3 MA (art) Retired general manager 1 0 0 Anomic Fluent Mild 7 Μ 67 9 Incompleted MA Retired military colonel 4 5 0 Broca's Non-fluent Moderate . · 8 Μ 68 300 8th grade Retired carpet & tile salesman 2 2 0 Broca's Non-fluent Mild ' 9 Μ 65 40 High School Retired plumber 2 3 0 Broca's Non-fluent Mild 10 Μ 54 16 College Retired licensed Land Surveyor 2 3 0 Broca's Non-fluent Mild 11 F 60 120 High School Retired meter maid 2 0 0 Broca's ·' Non-fluent Mild *Key: 1 = superior communicativeness 7 = uncommunicative " K e y : 0 = absent 7 = severe The South African Journal of Communication Disorders, Vol. 32, 1985 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) Aphasia : a Societal and Clinical Appraisal of Pragmatic and Linguistic Behaviours 1 3 sonal clinical experience and the literature (e.g. Holland, 1980; 1983; Penn, 1983) seemed to highlight their importance in aphasia. METHOD SUBJECTS Eleven English speaking aphasic adults were selected for this study a c c o r d i n g to certain criteria. These included confirmed presence of aphasia due to a completed, single, left hemisphere Cerebrovas- cular accident, neurological stability and the absence of gross con- c o m i t a n t problems. All aphasics were required to have absent to moderate symptoms of apraxia and/or dysarthria ranging from 0-5 onaseverity scaleof7(Wertz, 1984). Portions of the Motor Speech Evaluation (Wertz, Weiss, Kurtzke et al., 1978) and the spontane- ous section of the WAB (Kertesz, 1980) were used to determine this. A representation of both fluent and non-fluent aphasics was required as determined by the classification on the fluency subtest of the WAB (Kertesz, 1980). Since fluency appears to be one of the most im- portant factors differentiating the aphasia types, and is a dimen- sion that has been emphasised in more recent behavioural classifications of aphasia, the fluent : non-fluent dichotomy was selected in this study to facilitate comparison of pragmatic abilities in fluent : non-fluent aphasic subjects. Global aphasics were ex- cluded to control for severity of the sample examined (Wertz, 1984). Table 1 illustrates relevant case history information pertaining to the aphasics. Pragmatic Protocol (Prutting, 1982b) is a societal protocol which reflects the theoretical shift to place language within the context of socialisation (Prutting, 1982a). In contrast to traditional meas- ures, the Pragmatic Protocol (Prutting, 1982b) examines the in- dividual's pragmatic strengths and deficits within the conversational discourse, and across a variety of contexts. "The resulting config- uration allows the clinician to evaluate pragmatic function in rela- tion to linguistic performance" (Prutting, et al., 1984, p.24). The Pragmatic Protocol (Prutting, 1982b) has been developed over a four year period and pilot tested at the University of California, Santa Barbara Speech and Hearing Centre (Prutting, et al., 1984). It comprises a pool of 32 behaviours all known to be developed and used appropriately by children entering school, adolescents and adults. This tool was designed to be used while observing individuals engaged in spontaneous conversation during unstructured com- municative interactions (Prutting, et al., 1984). The Pragmatic Pro- tocol^Prutting, 1982b) was developed using the following criteria: theoretical framework, representative range of behaviours found in normal development, and inter et al., 1984). •investigator reliability (Prutting, Theoretical Framework: The Speech Act Theory proposed by Austin (1962) and Searle (1969)|(cited by Prutting et al., 1984), constitutes the framework underlying this Pragmatic Protocol (Prut- ting, 1982b). • " The 32 behaviours were organised within the following speech act categories. The utterance act includes 13 behaviours which form the verbal, non-verbal and paralinguistic aspects of production, characterising the "raw material" of the communicative act (Prut- ting and Kirchner, 1983). The prepositional act is comprised of 4 behaviours which define the linguistic dimensions of meaning. The 15 items comprising the illocutionary and perlocutionary acts constitute the reciprocal behaviours that regulate discourse between speakers. The illocutionary act represents the speaker's intention, while the perlocutionary act represents the speaker or listeners' effects. Modification to the Pragmatic Protocol In addition, the investigator added two behaviours - namely per- sonal appearance and affect (under the utterance act), since per- Die Suid-Afrikaanse Tydskrif vir Kommunikasieafivykings, Vol. 32, 1985 Scoring Since the Pragmatic Protocol (Prutting, 1982b) is a societal appraisal rather than a clinical appraisal, the judgement made is not whether the behaviour is correct or incorrect in a clinical sense, but rather if the behaviour is penalising or not. The investigator must decide whether society will penalise an individual for exhibiting a partic- ular behaviour. Thus, a behaviour may be incorrect but not neces- sarily judged as inappropriate (Prutting, et al, 1984). The following guidelines were used to judge each of the behaviours listed on the Pragmatic Protocol (Prutting, 1982b): - Appropriate: Behaviours are marked appropriate if they facili- tate the communicative interaction, or are neutral. - Inappropriate: Behaviours are judged inappropriate if they detract from the communicative exchange and penalise the in- dividual. - No opportunity to observe: When the evaluator has insufficient information to judge the behaviour as appropriate or inappropri- ate (Prutting and Kirchner, 1983). Scale of Overall Communicative Adequacy A 5 point subjective rating scale (derived from Holland's protocol for adult aphasia (1982)) was used to obtain an estimate of the aphas- ics' and their partners' overall communicative adequacy. Each rating would be converted into a percentage out of a total of 5 for com- parison purposes. Testing Procedure and Schedule Preselection tests and standardised procedures were administered over two sessions. Aphasic clients were seen individually in vari- ous facilities, or in their home environments. A constant task presen- tation order was maintained. Firstly, portions of the Motor Speech Evaluation (Wertz, et al., 1978) and the oral portion of the WAB (Kertesz, 1980) were administered. During the second session the CADL (Holland, 1980) was administered, and each aphasic was videotaped while conversing with a familiar comfortable partner of his/her choice with whom they had some shared history. While c o n v e r s a t i o n a l partner and topic were not specified, the af- filiative nature of the relationship, and topic familiarity were held constant across all aphasics to facilitate and maximise conversa- tional flow. RESULTS PRAGMATIC PROTOCOL DATA While fluent aphasics demonstrated significantly superior pragmatic skills compared with non-fluent aphasics (see Table 2), all individu- als demonstrated a high level of appropriate pragmatic behaviours which manifested as retained social competence (see Figure 1). Table 2 Between group comparison of percent appropriate pragmatic behaviours on Pragmatic Protocol in fluent and non-fluent groups Fluent Group (N = 6) Non-fluent Group (N=5) X Range s.d. 86.27%* 79.41-94.11% 2.07 X Range s.d. 75.88%* 61.76-85.29% 2.95 *t-test significance at ρ <0.05 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) 14 Τ — I — I — I — I — I 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 N u m b e r of inappropriate pragmatic behaviors Figure 1 Number of Inappropriate Pragmatic Behaviours on Pragmatic Protocol for Individual Subjects Within Each Group Table 3 Percentage of inappropriate pragmatic behaviours for fluent and non-fluent aphasics in each speech act category Group Utterance Act Propositional Act Elocutionary Perlocutionary Act Fluent 11.1% 33.3% 11.1% Non-fluent 10.6% 55% 29.3% Glen Goldblum Aphasics were apparently resourceful in using the context, frequent- ly facilitated via the use of compensatory communicative strate- gies. These included word finding strategies, appropriate repair and revision strategies, and a range of largely effective idiosyncratic strategies such as non-talking, to avoid communicative failure. The highest proportion of inappropriate pragmatic behaviour displayed by both groups was in the propositional act category. The non-fluent group demonstrated significantly greater difficulty on the perlocu- tionary/illocutionary act level, reflecting greater problems in manag- ing the dyad (see Table 3 and Figure 2). Examination of profiles of pragmatic deficits across fluent and non- fluent groups revealed interesting similarities and differences. First- ly, the nearly comparable presence of the following four inappropri- ate pragmatic behaviours across fluent and non-fluent groups: specificity/accuracy (100%) in both groups, fluency (50%) and (60%); pause time (50%) and (80%); quantity/conciseness (60%) and (100%) across fluent and non-fluent groups respectively. It is noteworthy that while fluency was rank ordered third as compared to sixth for fluent and non-fluent groups respectively, more non- fluent aphasics (60%) were penalised for this behaviour than fluents (50%). Generally, the prevalence of these four deficits across both fluent and non-fluent aphasics would seem to appropriately reflect the inherent problem of aphasia, irrespective of classification. In contrast to these similarities, two inappropriate pragmatic be- haviours were very prominent in the non-fluent group, and either relatively rare or absent in the fluent group. These were word ord- er (16%) and (100%); variety of speech acts (0%) and (100%) in the fluent and non-fluent groups respectively. These data clearly seem to differentiate the fluent from the non-fluent aphasics sug- gesting that rules of dyadic interaction are most difficult for the non-fluent aphasics in this investigation. In view of the presence of similar and different inappropriate pragmatic behaviours within the fluent and non-fluent groups used in this study, it is apparent that the Pragmatic Protocol (Prutting, 1982b) both does and does not differentiate between these aphasic groups. ro Ο •·-» ο <υ ω in ro . c a . ® < σ> τ - * s Aphasic Type 1 Wernicke's Fluent 2 Conduction Fluent 3 Conduction Fluent 4 Anomic Fluent 5 Anomic Fluent 6 Anomic Fluent 7 Broca's Nonfluent 8 Broca's Nonfluent 9 Broca's Nonfluent 10 Broca's Nonfluent 11 Broca's Nonfluent Group Utterance Act Figure 2 Inappropriate Pragmatic Behaviours Within Each Speech Act Category on Pragmatic Protocol for Individual Subjects in Fluent and Non-fluent Groups The South African Journal of Communication Disorders, Vol. 32, 1985 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) Societal and Clinical Appraisal of Pragmatic and Linguistic Behaviours 15 COMPARISON OF PERFORMANCE ON THE PRAGMATIC PROTOCOL AND THE SCALE OF OVERALL COMMUNICATIVE ADEQUACY Performance on the Scale of Overall Communicative Adequacy (Holland, 1982) revealed the same trend evidenced in the fluent aphasic groups' overall superior performance on the Pragmatic Pro- tocol (Prutting, 1982b) thereby substantiating the Pragmatic Pro- tocol (Prutting, 1982b) as a measure reflecting communicative competences (see Table 4). Despite the aphasic individuals' use of largely effective compensatory strategies to enhance communica- tive success, communicative burden on their respective partners was nevertheless high, resulting in their use of a variety of frequently facilitative strategies (e.g. probing, and encouraging the aphasics' use of strategies). These data highlighted the inherently interactive nature of the communicative process. Table 4 Mean communicative adequacy ratings for fluent and non-fluent groups and their partners on the Scale of Overall Communicative Adequacy The relatively small range of mean percentage scores on the Prag- matic Protocol (Prutting, 1982b) for the fluent and non-fluent groups (79.41-94.11% and 61.76-85.29% respectively) as compared with the greater ranges of scores for these groups on the WAB (Ker- tesz, 1980) and the CADL (Holland, 1980) (see Figure 3) was in- teresting, yielding several implications. This reflected an overall higher group mean of appropriate pragmatic behaviours (social com- petence scores) across fluent and non-fluent groups irrespective of severity. Further, since the Pragmatic Protocol (Prutting, 1982b) is based on societal rather than clinical criteria (used by the WAB (Kertesz, 1980) and the CADL (Holland, 1980), these overall higher scores may also reflect the discrepancies reported in the literature, i.e. related to societal and clinical ratings of aphasic communica- tive competence. Computational correlation coefficients were performed in order to determine and compare the relationship between societal and clini- cal performance of (a) fluent and non-fluent aphasic individuals Fluent Group* Non-fluent Group* X = 3.5 (70%) Range = 3-4 s.d. = 0.55 χ = 2.4 (48%) Range = 1-3 s.d. = 0.89 Partner** Partner** X = 4.6 (92%) Range = 4-5 s.d. = 0.52 X = 4 (80%) Range = 2-5 s.d. = 1.22 Key: 'Rating of 1 = minimal communicative ability Rating of 5 = normal communicative ability ••Rating of 1 = nominal skill in communicating with aphasic Rating of 5 = competence in communicating with aphasic CORRELATION BETWEEN PRAGMATIC PROTOCOL SCORES AND PERFORMANCE ON THE WESTERN APHASIA BATTERY AND THE COMMUNICATIVE ABILITIES IN DAILY LIVING IN FLUENT AND NON-FLUENT APHASIC INDIVIDUALS The relationship between societal'and clinical profiles was examined and'revealed some interesting trends. While significant differences were obtained between social competence scores of fluent and non- fluent aphasics on the Pragmatic Protocol (Prutting, 1982b) (p<0.05), differences between these two groups on the clinical measures (namely the WAB (Kertesz, 1980) and the CADL (Hol- land, 1980)) were found to be nonsignificant (p>0.05). (See Table 5) 100 90 80 70 60 50 40 30 2 0 10 F NF Pragmatic Protocol Aphasic group and measure F NF WAB F NF CADL Key: F = fluent group NF = non-fluent group Figure 3 Percentage group means and ranges of appropriate pragmatic behaviour on the Pragmatic Protocol, and correct responses on the Western Aphasia Battery and Communicative Abilities in Daily Living. Table 5 Percentage group means, ranges and standard deviations «η the Pragmatic Protocol, the Western Aphasia Battery and the Communicative Abilities in Daily Living *t-test significance at ρ <0.05 Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol. 32, 1985 Fluent Group Non-fluent Group Measure X Range s.d. X Range s.d. Pragmatic Protocol 86.26%* 79.41 %-94.11% 2.07 75.88%* 61.76-85.29% 2.95 WAB 74.92% 46.48%-96.26% 18.99 64.56% 42.96%-78.57 % 13.64 CADL 81.25% 52.94%-97.79% 22.95 86.32% 72.79%-94.85% 11.72 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) 16 Glen Goldblum separately and (b) combined (see Table 6). The non- ignificant correlation (for both fluent and non-fluent groups) between the Prag- matic Protocol (Prutting, 1982b) and the WAB (Kertesz, 1980), T d the Pragmatic Protocol (Prutting, 1982b) and the CADL (Hol- land 1980) suggested that these tests are measuring different aspects of an individual's communicative competence, and would not neces- sarily predict his/her performance on one or other test. Table 6 Computational correlation coefficient among three measures of fluent and non-fluent aphasics' communicative abilities Group Pragmatic Protocol + WAB Pragmatic Protocol + CADL WAB + CADL Fluent 0.20 0.34 0.97* Non-fluent 0.83 0.13 0.57 Non-fluent minus Subject 7 0.32 0.28 0.99* Fluent + non-fluent combined 0.53 0.05 0.75 Fluent + non-fluent combined minus Subject 7 0.26 0.15 0.91* *t-test significance at ρ <0.05 In the fluent group, the CADL (Holland, 1980) correlated signifi- cantly with the WAB (Kertesz, 1980) (S = 0.97) (p<0.05) reflect- ing the CADL's (Holland, 1980) almost perfect ability to predict a client's communicative abilities in daily living (Holland, 1980). While the non-fluent group revealed a non-significant correlation between the WAB (Kertesz, 1980) and CADL (Holland, 1980) (S = 0.57) (p>0.05), removal of Subject seven's extremely vari- able data reduced the disproportion and resulted in comparable cor- relational trends between the fluent and non-fluent groups (see Table 6). It is apparent from these data, that the highest correlation ex- ists between the WAB (Kertesz, 1980) and the CADL (Holland, 1980) while performance on the Pragmatic Protocol (Prutting, 1982b) would not necessarily predict performance on the CADL (Holland, 1980) and the WAB (Kertesz, 1980). Similar correla- tional trends occur when the three measures of communicative abil- ities are compared for the fluent and non-fluent groups combined, and when the 'outsider' Subject seven is removed (see Table 6). The overall consistent pattern of generally low non-significant corre- lation of the Pragmatic Protocol (Prutting, 1982b) with the CADL (Holland, 1980), and even more so with the WAB (Kertesz, 1980) (p>0.05) when fluent and non-fluent groups are examined separate- ly, and combined (excluding Subject seven's data) yield some in- teresting implications — most notably the apparent trustworthiness of the data based on the stability of data. Thus it is evident that fluent and non-fluent aphasics perform consistently superiorily on societal as compared to clinical measures. While these data are based upon a small and heterogeneous sample, which may have tended to inflate the obtained correlation coefficients, their implications are provocative. 'DISCUSSION A Pragmatic Perspective of Aphasia In accordance with several investigators (e.g., Davis and Wilcox, 1981, Foldi, Cicone and Gardner, 1983; Guilford and O'Connor, 1982; Holland, 1977, 1982, 1983; Penn, 1983; Wilcox, 1983) the results of this study revealed a high level of appropriate pragmatic behaviours manifesting as retained social competence across all aphasic subjects. Generally despite their linguistic impairments, aphasics are apparently resourceful in using the context effectively in conversational interactions, frequently facilitated via the use of compensatory communicative strategies (including for example, simplification, humour, circumlocutions and interjections to main- tain turns). These strategies reflecting the aphasics' attempts to readapt communicatively to their cerebral insult were found by Penn (1983) to constitute retained communicative competence. While the Pragmatic Protocol (Prutting, 1982b) was able to differen- tiate between the fluent and non-fluent group of aphasics on the basis of the almost exclusive presence of two inappropriate prag- matic behaviours in the non-fluent group, namely, word order and variety of speech acts, in view of the following factors, caution needs to be observed in concluding that this tool clearly differenti- ates between fluent and non-fluent aphasics: the combined presence of four inappropriate pragmatic behaviours across fluent and non- fluent groups (i.e., specificity/accuracy, fluency, pause time, and quantity/conciseness); as well as the scattered presence of prag- matic deficits across individual aphasics, irrespective of classifica- tion. These latter idiosyncratic deficits attest to the perplexing feature of variability that is so characteristic of aphasia (Holland, 1983). Furthermore, the small sample of fluent aphasics manifesting mild linguistic deficits combined with a relative absence of pragmatic deficits may well have inflated the data. Further research using larg- er samples of equivalent types of aphasic subjects is clearly need- ed to clarify these issues. To the present writer's knowledge, Penn's investigation compar- ing syntactic and pragmatic abilities of aphasics appears to be the one most closely related to the research questions addressed in this study. It is therefore interesting to speculate why Penn's Profile of Communicative Appropriateness (PCA) (Penn, 1983), unlike the Pragmatic Protocol (Prutting, 1982b) did not reveal any clearly differentiating pragmatic deficits across aphasic types. Features of the Pragmatic Protocol (Prutting, 1982b) that may account for these differential findings appear to include the following: Firstly, it en- compasses a broad range of pragmatic behaviours which are exa- mined in a more global or molar rather than a molecular manner. Secondly, the Pragmatic Protocol (Prutting, 1982b) employs so- cietal criteria of appropriateness. Furthermore the data base used in this study was spontaneous conversation as compared with Penn's study (1983), where topic content was controlled. Pragmatic Protocol Profiles and Overall Communicative Adequacy Ratings ; The contemporary revival of the awareness of the importance of subjectivity in science (Gould, 1981), resulted in Holland's ;Scale of Overall Communicative Adequacy being used in the present study in an effort to provide a global subjective rating of the aphasic in- dividual's communicative adequacy with their partner. Since all communicative partners used in this study were familiar and af- filiative, the latter being an acknowledged factor contributing to communicative efficiency (Linebaugh, Kryzer, Oden and Myers 1982), it was apparent that overall, the manner and ease with which these partners communicated with the aphasic, and helped share the communicative burden, facilitated greater communicative suc- cess with less frustration in the dyad. More effective partners tended to use strategies such as slowing down, stress, cueing, and encourag- ing the aphasic's use of strategies (e.g. writing, repetition, and a variety of cues such as requesting help from the listener). This use of a range of compensatory strategies by the aphasic was likewise seen to be related to their overall superior performance on the Prag- matic Protocol (Prutting, 1982b). The South African Journal of Communication Disorders, Vol. 32, 1985 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) a Societal and Clinical Appraisal of Pragmatic and Linguistic Behaviours 17 Aphasia The clinical implications of these data are apparently far reaching, yielding support for Linebaugh, et al's suggestion (1982) that it would be valuable to help aphasics (within their limitations) and their partners maximise their share of the communicative burden, and increase the effective use of demonstrated compensatory strate- gies, thereby enhancing the efficiency of the communication in the dyad. Social and Clinical Profiles While the fluent and non-fluent groups demonstrated significantly different pragmatic competencies, these inter group differences were not observed when comparing performance on the clinical profiles (namely the WAB (Kertesz, 1980) and the CADL (Holland, 1980)). Possible reasons accounting for the stable pattern of discrepant per- formance apparent across these pragmatic and linguistic tasks in- clude differential task requirements and criteria used in the evaluation of each aphasic's performance. These data imply firstly that in the sample investigated, overall high pragmatic competence was apparently independent of linguistic severity, and secondly that these tools are therefore measuring different aspects of an individu- al's communicative competence. It is therefore apparent that severity can no longer be measured by linguistic competence alone, but that measures of pragmatic competence need to be considered in rela- tion to these other aspects of language function. This latter interpre- tation was supported by the overall insignificant correlation between the Pragmatic Protocol (Prutting, 1982b) and the WAB (Kertesz, 1980), and the Pragmatic Protocol (Prutting, 1982b) and the CADL (Holland, 1980) in fluent and non-fluent groups separately and combined. While level of linguistic severity was insignificant in the present study, closer examination of the data reveals some apparent con- tribution to communicative competence by severity. For example, examination of the data showing fluent and non-fluent group per- formance on the four subtests of the WAB (Kertesz, 1980) indi- cates overall superior linguistic performance for the fluent group as compared with the non-fluent group. Since these aphasics are generally the least impaired (Holland, 1980), the data again points toward a role played by linguistic severity in overall communica- tive competence. Future research is needed to clarify these issues. / Theoretical Implications The findings of this study clearly demonstrated the aphasic individu- als' retained pragmatic abilitiesInecessary for social competence. Support was thereby rendered for Holland's observation that "aphasics probably communicate better than they talk" (1977, p. 173). All aphasics' social competence scores were above the level predicted by their linguistic impairments. These data therefore in- dicate that despite linguistic limitations, relatively intact pragmat- ic abilities permit effective communication and management with others within the reality of the client's limitations. Despite this, in- terpretation of pragmatic deficits independent of linguistic struc- ture was problematic. As with the language disordered child (Prutting, et al, 1984), it is evident that some 'spill over' exists between these different areas of communicative competence, and that the processes underlying appropriate use of pragmatic skills are not easily explained apart from the aphasic individual's linguistic limitations. However, overall superior pragmatic competence in re- lation to other linguistic skills may well account for the differential progress made by aphasic clients of apparently equal severity. While critical evaluation of traditional theories and methodologies through the telescope of time is crucial to sharpen our focus and reduce distortion, it is equally important to incorporate, rather than exclude the wisdom of our predecessors. It would thus appear valu- Die Suid-Afrikaanse Tydskrif vir Kommunikasieafwykings, Vol. 32, 1985 able to view language within the context of socialisation (Prutting, 1982a) where pragmatics, which is "a pervasive aspect of language which affects the entire communication system" (Prutting and Kirchner, 1983, p.60) is the core. Clinical Implications Since the measures used in the present study served to highlight different communicative competencies and deficits, clearly one needs to consider the range of pragmatic aspects of language in re- lation to other aspects of language (Foldi et al, 1983; Penn, 1983; Prutting and Kirchner, 1983; Prutting et al., 1984; Prutting, 1984; Wilcox, 1983). More all-encompassing, and dynamic intervention goals could thereby emerge, aimed to enhance structural and func- tional aspects of communication in relationship to one another. The value of a societal measure such as the Pragmatic Protocol (Prut- ting, 1982b) as a means of approaching the assessment of social competence and a context for interpreting a variety of communica- tive measures likewise yields several clinical implications. Since competence lies in the relational system, the dyad is neces- sarily the unit of analysis, whose behaviours are judged in terms of societal criteria of appropriateness rather than clinical criteria of correctness. Since these criteria take cognisance of the societal values surrounding the client, facilitating a 'real world prognosis' (Marshall, 1982 cited by Penn, 1983), our criteria for dismissal from remediation will need to shift so that dismissal occurs when the client can manage his/her relationships in a personally appropri- ate and effective manner within the limitations imposed by his/her stroke. Furthermore, intervention goals can emerge directly from the areas assessed by the Pragmatic Protocol (Prutting, 1982b) in which targets are always embedded within this framework (Prut- ting, 1984). Such a framework presupposes interactional, contex- tually based exchanges related to ongoing communicative situations at the level at which the client can meaningfully participate. Since the use of this tool serves to highlight the residual compensatory strengths of each individual, rather than simply faulting them for their inaccuracies and linguistic limitations, emerging intervention goals would be based inherently upon a strength, rather than a deficit model. These data and emerging issues constitute a changing paradigm. While the results of the present study have provided some prelimi - nary answers, they have raised many more questions. The 1980's will undoubtedly continue to add further in-depth understanding of aphasic pragmatic abilities in relation to other aspects of their com- municative competence. In attempting to merge these changing views of theory with practice, aphasiologists need to heed Prutting and Kirchners' apt comment that "new advances always require a respect for time in order to fit them into our existing schemas (1983, p.48). ACKNOWLEDGEMENTS The writer expresses her sincere gratitude to Carol Prutting, Depart- ment of Speech and Hearing Sciences, University of California, Santa Barbara, for her invaluable guidance as the supervisor of the Masters' dissertation on which this paper is based. REFERENCES Davis, G., & Wilcox, M.J. Incorporating parameters of natural conversation in aphasia treatment. In R. Chapey (ed.), Language intervention strategies in adult aphasia. Baltimore, Williams & Wilkins, 1981. R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2) 18 . . „ Γ α r H n p r η Pragmatic aspects of com- Foldi, N.S., Cicone Μ · > & « . f a g s . L s e g a l o w i t z (Ed.), munication in bram-damaged pat ents. ^ Y o r ^ L a n g u a g e junctions and brain organisa G o u T s . Treasure of ma. New York, W.W. Norton and _ £ ° Τ 7 μ 1 9 & O'Connor, J . K . Pragmatic functions in apha- sia J. Comma, Disord., 15, 337-346, 1982. Holland, A.L. Some practical considerations in aphasia rehabilita- tion.In M. Sullivan and M.S. Kommers (Eds.), Rationale for adult aphasia therapy. Nebraska, University of Nebraska Med- ical Center, 1977. Holland, A.L. Communicative abilities in daily living. Baltimore, University Park Press, 1980. Holland, A.L. Observing functional communication of aphasic adults. J. Speech Hear. Disord., 47, 50-56, 1982. Holland, A.L. Spontaneous recovery from stroke: An investigation of its earliest phases. Paper presented at the Academy of Apha- sia, Minneapolis, Minnesota, 1983. Kazdin, A.E. Assessing the clinical or applied importance of be- haviour change through social validation. Behaviour Modifica- tion, 1, 427-452, 1977. Kertesz, A. Western Aphasia Battery. London, Ontario Canada, University of Western Ontario, 1980. Linebaugh, C.W., Kryzer, K.M., Oden, S.E. & Myers, P.S. Re- apportionment of communicative burden in aphasia: A study of narrative interactions. In R.H. Brookshire (Ed.), Clinical Claire Penn aphasiology conference proceedings. Minneapolis, MN, BRK Publishers, 1982. Mueller, S.L. An investigation in social competence using clinical and societal profiles. MA thesis, University of California, Santa Barbara, 1983. Penn, M.C. Syntactic and pragmatic aspects of aphasic language. Doctoral dissertation, University of the Witwatersrand, Johan- nesburg, South Africa, 1983. Prutting, C.A. Pragmatics as social competence. J. Speech Hear. Disord., Al, 123-134, 1982a. Prutting, C.A. Observational protocol for pragmatic behaviours. Developed for the University of California, Santa Barbara Speech and Hearing Clinic, Clinic Manual, 1982b. Prutting, C.A., & Kirchner, D. Applied pragmatics. In T. Gal- lagher and C. Prutting (Eds.), Pragmatic assessment and inter- vention issues in language. San Diego, College-Hill Press, 1983. Prutting, C.A., Kirchner, D., Hassan, P., & Buen, P. A societal appraisal of pragmatic behaviours. Unpublished manuscript, University of California, Santa Barbara, 1984. Wertz, R.T., Weiss, D., Kurtzke, J.F., et al., A comparison of clinic, home and deferred treatment of aphasia. Veterans Ad- ministration Cooperative Study Protocol, V.A. Hospital, Mar- tinez, California, 1978. Wertz, R.T. Personal communication, V.A. Hospital, Martinez, California, 1984. Wilcox, M.J. Aphasia: Pragmatic considerations. Topics in Lan- guage Disorders, 3, 35-48, 1983. The Profile of Communicative Appropriateness: A Clinical Tool for the Assessment of Pragmatics Claire Penn Ph.D (Witwatersrand) Department of Speech Pathology & Audiology, , University of the Witwatersrand, Johannesburg ABSTRACT I The Profile of Communicative Appropriateness — a newly developed profile for the characterisation of pragmatics is described. The theo- retical background to this profile is covered as well as its main components. Its application to a group of eighteen aphasic patients is outlined, results suggesting that patient groupings on the profile could be predicted in terms of severity but not in terms of type of aphasia. Explanations for this finding are discussed and the potential utility of this profile is suggested. 1 OPSOMMING Die Profile of Communicative Appropriateness — 'n nuutontwikkelde profiel vir die karakterisering van pragmatiek word beskryf. Die teo- retiese rasionaal hieragter en die hoofkomponente van die profiel word behandel. Die toepassing hiervan op 'n groep van agtien afatiese pasiente word omskryf Resultate dui daarop dat die pasientgroeperings aanduidend kan wees van die erns van afasie maar nie van die tipe afasie nie. Verduidelikings hiervoor en die potensiele bruikbaarheid van die profiel word bespreek. The clinical profile is a method of characterising language which has become increasingly popular in recent years. Essentially " . . . a linguistic profile is a principled description of . . . those features of a person's . . . use of language which will enable him to be iden- tified for a specific purpose." (Crystal, 1982). The format of such a profile is the presentation of a wide range of variables simultane- ously so that the clinician is able to see at a glance the communica- tive assets and deficits of a patient. The main purpose of such a profile according to Crystal (1982) is to provide not only a comprehensive description of a patient's data but also an adequate basis for remedial intervention. It is not a standardized measure; nor is it an exhaustive linguistic descrip- tion. The profile is, however, a compromise to the clinician faced with the realisation that language disability requires comprehen- sive and individual description. The amount of information con- tained on a profile is determined by the behaviours being measured © SASHA 1985 The South African Journal of Communication Disorders, Vol. 32, 1985 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2)