INTRODUCTION

Available stuttering literature has repeatedly emphasised the need to adopt a culturally sensitive approach in the management of stuttering ( Robinson and Crowe, 1998, Watson and Kayser, 1994, Cooper and Cooper, 1993). This orientation is critical in the South African context where the profession is currently being challenged to address such issues given that its work is culturally- situated (Pillay, 1997). The overarching need to deliver appropriate services to all sectors of the population, especially to Black South Africans who have - and continue to - receive a poor service, (Pillay, Kathard, Samuel, 1997) has been stressed. Whilst attempts to address this broad service issue has been documented by the profession in South Africa (Jordaan, 1989, Alant and Beukes, 1986, Naidoo, 1994, Parkendorf and Alant, 1997) its application to the management of stuttering is critically minimal. Comparatively, issues of cultural programming in the management of stuttering has received greater attention in the international professional arena (Robinson and Crowe, 1998, Watson and Kayser, 1994). In this regard, there has been an appeal to all practioners/ researchers who work in diverse multicultural contexts to share their research, knowledge and experiences (Watson, 1998), and to add to the existing body of clinical literature in stuttering intervention, which has traditionally been derived from mainstream American populations (Shames,1989). Collaboration in this regard will serve the interests of deriving a better understanding of the complex nature of stuttering, and thereby enhance its management with diverse populations.

Motivation for adopting a cultural focus in clinical practice

Whilst the historical focus on issues of culture and stuttering have been rooted in research (as in prevalence studies) and theoretical perspectives (e.g. diagnosogenic theory), present concerns are related to applications in clinical practice (Robinson and Crowe, 1998, Watson and Kayser, 1994). The conventional focus in the stuttering intervention process has been on the acquisition and development of technical skill /competencies (Dopheide,1987). The present challenge to integrate such skill/ competencies with dimensions of cultural sensitivity in clinical practice. The primary motivation for including cultural considerations is to ensure that assessment and treatment decisions are not inappropriately influenced by cultural variables (Finn and Cordes, 1995).

For the South African context, what might the reasons be for adopting a multicultural practice perspective? An obvious reason is that a multicultural client base exists. However, this client base also has a significant political history. Apartheid's racial classification system has narrowly defined four race groups (76% Black, 13% White, 8.5% Coloured and 2.5% Indian - Nzimande, 1996) elevating colour above its enriching cultural pluralism. South Africa's collage of heterogenous cultures with diverse identities (and languages) has been shaped by the ideological tool of apartheid, serving to establish strong and divisive ethnic and cultural forces. (Coetsee,1996). Such sociopolitical forces have influenced and shaped how culture has been transmitted and maintained. In addition, life experiences of people have been different. e.g. Blacks and Whites have lived in separate worlds, with divergent worldviews impacting on their daily living practices (Kotze,1993) which, in the main, continues even in post-apartheid SA. Why are such contextual factors important to consider when managing stuttering?

The contextual operation of varying worldviews and beliefs has probably contributed to a position of unshared, divergent beliefs about stuttering and its management. In contrast to American and European contexts where such concerns are about minority groups, these concerns in South Africa are about the majority. In this context very little is known about how stuttering is perceived and managed most in communities, especially Black communities (Mohamed and Panday, 1993) because little research has been documented while referral rates for speech therapy in Black communities have also been low (McKenzie, 1992) Furthermore, the professional corps is predominantly White English or Afrikaans speaking with less than 1 percent Black African Language speaking speech-language therapists (Bortz, Jardine & Tshule,1996). This factor coupled with the realities associated with acknowledging SA's 11 official languages and diverse cultural profiles, makes the scenario of providing a cultural-linguistic match for a client and clinician highly unlikely. Hence, cultural sensitivity in clinical practice becomes a realistic goal.

The immediate context

This section reviews the immediate context of clinical practice, emphasing the nature of the student core and client population at UDW. The Stuttering Clinic is located at the University of Durban-Westville (UDW) in the province of Kwa-Zulu Natal. Of a population of 8.5 million people, the majority (76%) are Black African . The clinic draws its clientele from the Durban Functional region where the dominant languages are English and isiZulu (Kaschula and Anthonissen, 1995). While UDW began as an apartheid Indian-ethnic university, student admission policy changes in 1993/4 (i.e. the Social Redress Policy) attempted to redress past inequities by increasing the intake of student-clinicians from Black African backgrounds. This led to changes in the demographic profile of the undergraduate student body engaged in the present clinical stuttering programme. Students are from diverse cultural/ linguistic, ethnic, religious, and racial backgrounds. The transition from traditional ethnically-based 'monocultural' practice to a 'multicultural' one at UDW in the 1990's, has necessitated a critical focus on cultural issues as they influence student training and service delivery.

Enrollment trends of clients who stutter indicate that they also emerge from diverse backgrounds with respect to age, ethnicity, and language and culture. This profile is in stark contrast to the enrollment trends in 1980s where clients, and the University community itself, were predominantly Indian. The most significant enrollment trends in the past three years (1996-1998) reveal that increasing numbers of Black adult clients enroll for therapy. There is also a proportionally higher representation of preschool and school aged children from Indian communities , and to a lesser extent White communities, compared with Coloured and Black communities. The majority of Black clients speak an African language as their first language and English as a second or third language. Clients of other races speak English as their first language.

In this context, all student-clinicians offer services to clients who are of cultural backgrounds different to their own. This practice serves the long term aim of developing competencies that would prepare students to offer relevant and appropriate services in a diverse, multicultural context. Screen and Anderson (1996), point out that all training institutions should engage students in issues that relate to multicultural service delivery At UDW, this has become a norm rather than an exception.

Part Two relates to cultural issues in stuttering intervention in the UDW context. Table 1 serves to orientate the reader to cultural issues in the clinical process. Significant cultural aspects of the process are presented (Taylor and Anderson, 1988), and motivation for considering such issues in the management of communication disorders in general, stuttering in particular, is briefly described in tabular form. The specific issues that are significant in the UDW context are presented and these issues form the main body of the discussion in Part Two.

A. DEFINING CULTURAL BACKGROUND

Robinson and Crowe (1998) have emphasised the value of establishing cultural identity as a prerequisite for appreciating clients cultural circumstances. In the UDW context, two significant issues emerge.

Firstly, clients are from diverse cultural backgrounds, and each client's cultural identity is unique. Culture is a "shared way of perceiving, believing, evaluating and behaving" (Goodenough cited by Culatta and Goldberg , 1995 p.111), which extends beyond ethnicity, and relates to different sources of shared values and beliefs. That clients in the current context are of diverse cultural backgrounds have been confirmed by their explicit ( dress, language) and implicit (values and attitudes associated with spirituality, age, gender, child-rearing practices) behaviour. A client's cultural profile is unique, and is amalgam of values, beliefs and behaviours drawn from the macroculture and microcultures (Gollnick and Chinn, 1993 cited in Culatta and Goldberg, 1995). The macroculture refers to the values and beliefs of the political and social institutions, which, in the South African context, would relate to the values that are enshrined in the country's constitution e.g. equity and democracy. However, within the macroculture there exists many microcultures or subcultures. Individuals may identify with various types of microcultures which include ethnicity, religion, gender, age, exceptionality, urban-suburban-rural, geographic region, and class. Given that a diverse population exists in Kwa-Zulu Natal, it follows that many values and beliefs are prevalent, and hence the context per se is aptly described as multicultural, and it is against this background that stuttering must be managed. Secondly, it is necessary for the student-clinician to establish his own cultural identity before establishing the clients' identity. Within any clinical encounter, underlying cultural influences are reflected in the dynamics of the clinical interaction. When the clinician's values and assumptions become dominant then a deviance model as opposed to a difference model becomes operational, to the disadvantage of the client (Taylor and Clarke, 1994). Therefore, it is critical that the student-clinician is aware of her own values and beliefs, and the impact they are likely to have in the clinical process. Cultural definition at two levels warrant discussion. At the first level, the student-clinicians' identity is derived from his/her social upbringing , whilst at the second level it is influenced by professional training. Student-clinicians are socialised into the culture of their profession (Pillay, 1997), which determines a particular set of beliefs, and norms, and rituals (Spector, 1991). Because clinical practice in South Africa has been influenced by practice models and methodologies from Euro-America contexts, such practice may not be fully applicable to the diverse spectrum of the South African clientele. Therefore, in the UDW context, student-clinicians are required to reflect critically on such issues so that they are aware of how dominant models of practice have developed and shaped their clinical practice. Furthermore, they need to continually evaluate the applicability of these models to their rapidly changing clinical contexts.

Methodological issues

Defining cultural identity can best be appreciated by adopting a process-oriented view (Robinson and Crowe, 1998). This means that at each stage in the clinical process there is opportunity for consideration of cultural factors. The data gathering methodology has traditionally been a structured interview with content probes about the nature and impact of stuttering. Whilst the "traditional" interview does raise issues of cultural identity, the utilisation of the " culturogram " as a supportive tool in the clinical process has been useful. This informal instrument takes the form of guiding questions about culture (Taylor,1994, Congress, E,) and helps to define the clients cultural background. Observation of clients in their natural environment has also been a useful methodology for appreciating cultural phenomenon.

To prevent stereotying in a context where clients are from diverse cultures but share common features like race and language, student-clinicians have been encouraged to adopt a person-centered ethnographic model (Hollan, 1997). This model attempts to draw on the individual's personal and subjective experience or the "near-experience" (Hollan, 1997). The "near-experiences" of the individual are used to gain access to the client's phenomenal field, which are, in turn related to the broader social, cultural, political and economic contexts.

B. FLUENCY AND DISFLUENCY ANALYSES: LANGUAGES AND LANGUAGE USE

The evaluation of stuttering entails an analysis of the overt and covert features (Ham, 1986). The influence of cultural variables must be considered when assessing each of these parameters (Watson and Kayser, 1994) in order to obtain a reliable and valid evaluation that informs the treatment process. An assessment of overt features entails an analysis of observable stuttering behaviour, whilst the covert aspects relate to an understanding of the attitudes that client has towards communication and stuttering. In this process, the clinician is required to consider a variety of cultural variables (interactive patterns, bi/multilingual issues, cognitive learning styles and suitability of assessment tools and procedures) that are likely to influence assessment outcomes (Robinson and Crowe, 1998, Watson and Kayser,1994) In the discussion that follows issues related to the analysis of overt behaviours are raised.

An accurate assessment of stuttering is reliant on understanding of the language/s and communicative norms that operate in a speech community, noting that such norms are culturally situated. The greeting styles, eye contact, and discourse rules and vocabulary used are among the parameters that have a cultural basis (Kashula and Anthonissen, 1996). Tentative observations in the current clinical environment have revealed the following characteristics with respect to languages and social rules of first language English speakers and first language Zulu speakers.

People of Zulu cultural backgrounds differ from first language English speakers with respect to their rules for using eye contact, ways of showing respect, greeting styles, and use of politeness markers. Not maintaining eye contact with certain communication partners is, for the Zulu speaker, regarded as a sign of respect (Kashula and Anhonissen,1996). For the wider English-speaking communities the lack of eye contact may be percieved as indicating avoidance and shiftiness.

Compared with English, Zulu is described as a tonal language, in which there are no monosyllabic words except for a few interjections. Zulu is an agglutinative language with complex and prominent morphology. Hence, words are longer than English words. Unlike English, Zulu has a regular CV syllable structure and no consonant clusters or syllable final consonants in words of Zulu origin. (Bailey, 1998)

Zulu has features of morphophonemic reduplication which is considered a normal part of the language structure. e.g. In some polysyllabic words of more than two syllables, incomplete reduplication of the stem may occur. e.g. "They are singing" -Bayahlabelela and "they are singing a little" Bayahlabehlabelela. Here, the first two syllables of the word are repeated. (Bailey, 1998). This feature is not a regular feature of English verbal morphology. The clinical implications for student-clinicians are : Firstly, it is necessary to be aware of the culturally derived rules of social communication, so that these variations in communicative behaviour are not misinterpreted as part of the stuttering behaviour. e.g. In English language severity judgements are based on the assumption that lack of eye contact is reflective of greater degree of severity. This does not hold true for clients all in the current context. Eye contact is seen as taboo in certain contexts in the Zulu culture. In such instances it is critical that the student- clinician isolate culturally specific behavior patterns from those associated with stuttering. However, it is has also been noted that poor eye contact may be used by Zulu -speaking people who stutter more extensively than the social norm allows for. In such instances, poor eye contact as an avoidance behaviour must also be recognised.

Secondly, they must develop an understanding of the linguistic features of the language/s that their client uses to ensure that their assessment is accurate, and that therapy is planned with knowledge of such features e.g. noting that morphophonemic reduplication is a natural aspect of Zulu and should not be confused with stuttering. Fluency hierachies should be planned noting the nature of word length in Zulu, and with the aim of maintaining the tonal quality when fluency skills are applied.

Thirdly, the methods used for fluency/disfluency analyses must aim for reliability and validity, especially when the clinician is unfamiliar with the clients language. At UDW student-clinicians have attempted to enhance reliability and validity of assessment by observing interaction between other members of the clients speech community, to appreciate social rules that govern common practice. watching video recordings of communication and educating each other (client and clinician) about the social rules for communication, hence including the client in the validation process. selecting methods of analyses that consider the linguistic structure of the language. E.g. Speech rate is a frequently assessed parameter. When calculating speech rate, student-clinicians have reported that syllable per minute calculations, rather than words per minute is more useful especially for Zulu-speaking clients. This is because words are longer than English words, and therefore cannot be compared to existing normative data. Also, direct comparisons of speech rates across languages are not possible. Therefore, syllables per minute is preferred unit for rate calculation. A team-approach to analysing speech fluency/disfluency has been viewed positively. The contributions of the clinician, linguist, client and members of the client's speech community has led to a more accurate understanding of the clients communicative behaviour in general, and of stuttering in particular.

Bi/Multilingual issues

Obtaining a thorough understanding of the clients idiosyncratic communicative behaviour in the current context has highlighted the need to consider issues of bi/multilingualism. . In South Africa, this issue is critical because the majority of the population is bi/multilingual (Kashula and Anthonissen, 1995). The observations in the UDW context have revealed that clients presently enrolled in the clinic speak a variety of languages. All Black African clients currently enrolled speak an indigenous African language as their first language, and English as a second or third language. Clients of other races are primarily first language English speakers. A similar profile exists for student-clinicians.

For clients who are bi/multilingual, the level of language proficiency may vary between languages. The majority of African clients have reported that their proficiency in verbal communication is better in their African languages, compared with English. The frequency of stuttering is imbalanced across language i.e. the frequency of stuttered episodes is greater in one language than another. Preliminary indications are that English appears to be the preferred language in the clinical context.

Language proficiency is influenced by a variety of social, political and cultural and educational factors and it is critical that its impact on stuttering is ascertained. In the UDW context it has generally been found that the frequency of disfluency is imbalanced across languages. Whilst the trend appears to be toward better fluency in languages in which the client was more proficient, individual profiles to the contrary exist. Socio-psychological (Nwokah, 1988), and language competency explanations (Jankelowitz and Bortz, 1996) account for such imbalances.

In the Nigerian context, Nwokah suggested that different languages are associated with different situations which present different demands. e.g. Among friends and family in rural setting, the indigenous African language is used, whilst in formal educational settings English is used. Depending on whether the experience was positive or negative, the disfluency may vary. Similar types of experiences have been reported by African clients in the UDW context. Jankelowitz and Bortz (1996) suggested that linguistic difficulties in language formulation induced more stuttering in a language in which the client is less familiar. In their case study of a bilingual adult who spoke English and Afrikaans, they concluded that when the stutterer does not possess the linguistic competence in a language i.e. the grammatical forms or vocabulary, he is likely to be more disfluent.

In addition to language proficiency, the language preference issue has received attention in the current context. The experience has indicated that proficiency and preference cannot be equated. All clients in the current context have indicated a preference for English within the clinical environment, although it is a language in which they consider themselves to be less proficient. Although this may be influenced to a great extent by the language spoken by the student-clinician, it appears that the choice of English relates to its sociopolitical status. In the South African context, English is perceived as a empowering language (Kashula and Anthonissen, 1996). African clients in particular, are therefore in favour of developing skills in English, and applying skills to other languages, which is contrary what is suggested is recommended in the literature (Nwokah, 1988)

Although it has been routinely recommended that fluency and disfluency be assessed in all languages (Watson and Kayser, 1994), there is no evidence to support the usefulness of such a recommendation . Having engaged in the clinical process in the current context it has become clear that there is value in conducting assessing stuttering in each of the languages the client speaks. Firstly, it has allowed the student-clinician to differentiate between stuttering and disfluencies that are the result of second language learning. Secondly, it is it provides valuable information about the variability of stuttering in relation to the languages used. This information feeds directly into therapy planning as goals can be set in relation to the clients language use and needs across speaking contexts.

C. BELIEFS AND PRACTICES

During the case history/ intake process the clinician aims to consider the impact of stuttering on the clients life. At this point, a significant aspect of cultural programming relates to the specific beliefs about stuttering, and how such beliefs are likely or not, to impact on the intervention process.

The observation in the UDW Clinic has been that there have been varied beliefs about stuttering. Preliminary data gathered indicates that beliefs about stuttering are categorised as etiological and remediational beliefs. Within these categories there appear to traditional/ generational beliefs, those influenced by alternative methodologies, and beliefs which take on professional/academic descriptions. Examples are cited in Appendix A.

A common concern of student-clinicians has been "Why does one need to know about such beliefs?" and "How are they managed in the clinical process?" An understanding of the clients beliefs about stuttering offers the clinician insight into the clients worldviews in general, and specific beliefs about stuttering. The clients belief is compared to the clinicians beliefs, and the extent of congruence is determined. This allows the clinician to appreciate that differing beliefs about the nature of stuttering exist; that such beliefs need to be managed in the clinical process; and that beliefs may impact on treatment acceptance and outcomes.

The therapeutic process is more likely to be successful when the client and clinician share a common phenomenal field i.e. comprehensive experiences and beliefs (Robinson and Crowe, 1998). When the client and clinicians phenomenal field do not match, it may partly be due to the fact that different beliefs systems are operational- as has been observed in the UDW context. . In such an instance, therapy may not be maximally effective, and therefore steps need to be taken to ensure that this issue is appropriately managed. No strict guidelines exist in the literature. However, the following suggestions are offered.

Robinson and Crowe(1998) suggest that beliefs that are obstructive to the clinical process should be eliminated. However, because beliefs /myths may represent the clients honest understanding of stuttering, they should not be disregarded.

In the current context beliefs that are obstructive ( e.g. that stuttering is infectious) are eliminated via the process of informational counselling. However, other beliefs are integrated into a framework of understanding stuttering and its treatment. e.g. consulting a priest and conducting a ritual can be integrated into a broader framework, by acknowledging that the priest may offer assistance at a spiritual level, whilst speech therapy centers around modification of speech behaviour. It is possible for such methods to co-exist and complement each other.

Platsky and Girson (1993) examined the beliefs of African indigenous healers towards stuttering and reported that all healers had names for stuttering, generally identifying it as an inherited disorder. Some causal explanations had their roots in folklore e.g. being left out in the rain or failure to inform the ancestors of a child's imminent birth. The beliefs of community health workers and indigenous healers in a semi-rural community at the Valley of a Thousand Hills, Kwa-Zulu Natal isolated anatomical, physiological and ancestral causes (Pillay, 1992). McKenzie interviewed 113 community members in the rural Timbavati area of Gazankulu, 52 of whom were teachers. She found that 91% knew what stuttering was, while 86% believed that it was infectious. Seventy-seven percent believed that it caused a problem for individuals; and since there was a perception that it could not be cured or treated, there was no reason to refer for help.

It is also the responses from individuals who have chosen not to enroll for speech therapy that highlights the extent to which beliefs serve to influence treatment practices. Anecdotal reports gathered at the UDW clinic reveal that religious explanations about stuttering exist. In such instances, people from varied religious backgrounds have sought help based on their religious values and practices. There is no formal studies which have investigated the complex relationships between stuttering and religion (Culatta and Goldberg, 1995). Other individuals have reportedly considered homeopathy, reflexology, and other such alternative treatment options, based on their beliefs about health and illness in general, and about stuttering in particular.

Clinical implications of these findings for speech-language clinicians are that that when engaging in multicultural practice, it is critical that beliefs about stuttering are examined since they have a direct bearing on how stuttering is perceived, they determine whether or not treatment options gain acceptance, and how beliefs are to be integrated in the treatment process, and who is considered the accepted practitioner. .

Indigenous healers

The beliefs and practices of indigenous healers warrant specific consideration in this discussion because they are described as " ....psychologist, physician, as priest..[and]...tribal historian " (Holdstock, 1979 p 119 cited in Platsky and Girson, 1993), and are consulted by approximately 70% of the Black population in South Africa. Indigenous healers are likely to form an essential core of the primary health care workers in Third World countries. Research on indigenous healers of urban and rural practices from various African ethnic backgrounds, found that treatment methods for stuttering included herbal remedies, slaughtering of animals, breath control, inhalation of smoke from the ashes of medicinal products, prayer, communication with ancestors and parental guidance and prolonged speech (Platsky and Girson,1993). These methods stem from the beliefs held about the nature of stuttering.

Given the need to adopt a community-based model of service delivery in South Africa, speech-language therapists have salient issues to consider given the need to adopt a community-based model of service delivery. Within a community-model of service delivery there is a need to examine the types of methodologies that are likely to be more successful and gain acceptance in a given community, to be cognizant of who the accepted practitioner is likely to be, and to critically evaluate the role of the professional speech-language pathologist within this service delivery model. There are no easy solutions to this complex issue. On the one hand, there is a strong recommendation that health professionals foster a cooperative relationship with indigenous healers and on the other, opposition to such cooperation. Platsky and Girson (1993) have favoured a collaborative model which is intended to gain an understanding of cultural beliefs with an attitude of acceptance, and to develop a systems of referral. While the debate continues and systems are developed, it is critical that the interests of the individual are given priority (Hammond-Tooke, 1989), and that the client is given the opportunity, and the means, to make informed decisions.

D. STUTTERING MODIFICATION ISSUES

Fluency-shaping and stuttering modification treatment approaches are commonly applied in the treatment of stuttering (Guitar, 1998) In the discussion that follows , cultural issues in the application of stuttering modification and fluency shaping therapy approaches are considered.

The observation in the UDW context are : that some clients appear less comfortable than others, with a counseling-based therapy methodology. that the methods, procedures and activities advocated in treatment protocols must be selected and applied in a culturally sensitive manner.

Stuttering modification therapies utilise a counseling-based methodology with the primary aim of developing positive attitudes towards communication (Peters and Guitar, 1991). Whilst the goal has universal application, the process and structuring of the clinical activity requires cultural sensitivity. In the current context, the nature of the counseling process, which has its roots in Psychology, warrants specific mention. Race, language, class-bound barriers, and differing worldviews are common barriers to the counseling process in the South African context (Hickson and Christie, 1994). The issue of African worldviews is discussed here since it has been recognised as a significant factor in the current context of increased enrollment of African clients.

Hickson and Christie (1994) explain that the dominant counseling process has been based on a Eurocentric theory of human reality which has focused on people adapting to the environment. It may be inappropriate to apply this model to all people from diverse cultural backgrounds. In particular, it does not consider the spiritual dimensions of the majority of the Black population (Hammond-Tooke, 1989). Western notions of knowledge have been influenced by ideas of reductionism which focuses on the parts to understand the whole. In this process of counseling, notions of dualism : mind-body, thinking and feeling, experience and behaviour, subjectivity and objectivity (Bodibe, 1992) are prominent. However, in traditional African thinking, knowledge is obtained by going "outward" and establishing relationships with the sky, land, families etc. as opposed to the relationship that people have with their own thoughts and feelings (Bodibe, 1992). The trend in the Western world to be independent, self-sufficient and self-directed may be viewed as unhealty within an African context. Hence, the process of counseling which seeks to encourage the individual to manage his difficulties by focussing on his innerself, may not be applicable to all clients.

The methods and procedures used in stuttering modification therapy seek to eliminate fear and avoidance of speaking, thereby encouraging the development of more positive attitudes to speaking. The procedures used include mirrorwork, confrontation of the disorder via self-disclosure, pseudostuttering, and advertising (Guitar,1998). It has been noted that some clients experience discomfort with these methods which are unfamiliar to them, and as a result are not compliant in the process. Noting that such difficulties exist in the current clinical context, clients are required to indicate their comfort with therapy methods, and to generate activities which are they are comfortable with.. In this way, using the clients experiences and beliefs also guide the selection of therapeutic methods. The therapy process is complex and multidimensional, and it is clear that in a context where multiple worldviews operate, adopting a single operational strategy uncritically, would not serve the best interests of the client.

E. FLUENCY SHAPING ISSUES

The observations have been that fluency shaping therapies methodologies appear to have better acceptance (compared to stuttering modification methodologies) by clients of diverse backgrounds. Commercially available programmes are used, but are modified and adapted to suit individual clients.

Whilst clients from diverse backgrounds have varied expectations about treatment, they commonly report a need to achieve better fluency. In contrast to therapies which advocate attitudinal change, fluency in itself can be managed in a more concrete, direct manner since this methodology has its roots in operant conditioning and programmed instruction. Although clients speak languages other than English, fluency-shaping skills per se have universal application (Cooper and Cooper, 1993).

The feeling of fluency control, which is the end goal in fluency shaping therapies, is a culturally independent construct (Cooper and Cooper, 1998). They maintain that the feeling of control, of which fluency is a by-product, should be perceived as a concrete experience rather than a hypothetical construct, and has therefore become a universally recognised construct that can be communicated with relative ease to people of all cultural backgrounds. The experiences in the present context are congruent with this explanation.

However, most commercially-based programmes require modification to make them applicable in a multicultural context. In the current context, the nature of the adaptations include

• the selection of culturally appropriate stimulus material
• simplification of explanations for use with clients who speak English as a second language
• recording speech samples in a dialect that the client is familiar with, for purposes of modeling.
• structuring of the therapy plan to include opportunity for application of fluency skills in languages and contexts that relate to the clients personal experiences.

CONCLUSION

The experience in the UDW context has revealed that issues of culture should be considered routinely in the assessment and treatment of stuttering. This would serve in the best interests of the client, given that diversity is a distinctive feature of the population we serve. It has also become apparent, that preliminary observations documented need to be investigated thoroughly to isolate the impact of cultural variables on assessment and treatment of stuttering. Whilst we must avoid developing new "cultural" stereotypes in our attempts to consider cultural variables in stuttering, it is equally critical that the definition and management of stuttering are appreciated from the perspective of communities in which they exist. Hence, the individuals in context must set the stage for the understanding the nature of stuttering and developing treatment practices. This implies building on what we have learned from other contexts, and using the present context to define the nature of cultural variables and how these are likely to impact on the stuttering management process.

Alant,E, & Beukes, S.M. (1986). the application of the revised version of the Peabody Picture Vocabulary Test (PPVT-R) on non-mainstream children. South African Journal of Communication Disorders, 33 7-12.

Aron, M.L. (1962). Nature and incidence of stuttering among a Bantu group of school-going children. Journal of Speech and Hearing Disorders, 27 116-128.

Cooper, E.B. & Cooper, C.S. (1993). Fluency disorders. In D.Battle, (Ed.) . Boston: Andover Medical Publishers.

Bailey, R. 1998 Personal communication

Bloodstein, O. (1995). A handbook on stuttering. San Deigo: Singular Publishing Group.

Bodibe, R.C. (1992). Traditional healing: an indigenous approach to mental health problems. In J. Uys (Ed.), Psychological counselling in the South African context. Cape Town : Maskew Miller Longman.

Bortz, M.A., Jardine, C., Tsule, M. (1996). Training to meet the needs of the communicatively impaired population of South Africa : a project of the University of Witwatersrand. European Journal of Communication Disorders, 31 465-476.

Coetsee, T. (1996). Ethnicity - should it be inscribed in the constitution? CSD Bulletin, 2 12.

Congress, E. (1994). The Culturagram : An innovative educational model to increase understanding of cultural diversity, paper presented at the 27th Congress, International Association of Schools of Social Work. Amsterdam

Culatta, R. & Goldberg, S.A. (1995) . Stuttering therapy. An integrated approach to theory and practice. Boston: Allyn and Bacon.

Dophiede, B. (1987). Competencies expected of beginning clinicians working with children who stutter. Journal of Fluency Disorders, 12 15-166.

Goldberg, B. (1997). Tailoring to fit. Altering our approach to multicultural populations. ASHA, Spring , 25-28.

Guitar, B. (1998). Stuttering. An integrated approach to its assessment and treatment. Baltimore : Williams and Wilkins.

Ham, R. (1986). Techniques of stuttering therapy. New Jersey: Prentice-Hall Inc.

Hammond-Tooke,W.D. (1989). Rituals and medicines. Johannesburg: A.D. Donker.

Hickson, J. & Christie, G.M. (1989). Research on cross-cultural counselling and psychotherapy: Implications for the South African context. South African Journal of Psychology, 19 162-169.

Jankelowitz, D.L., & Bortz, M.A. (1996). The interaction of bilingualism and stuttering in an adult. Journal of Communication Disorders, 29 223-234.

Jordaan, H. (1989). Management of language disability in the bilingual/ bicultural child. In SAALED Conference Proceedings: facilitating learning: an ecological perspective, July:217-134.

Kashula, R. & Anthonissen, C. (1995). Communicating across cultures in South Africa: Toward a critical language awareness. Hodder-Stoughton : Wits University Press.

Kotze, J.C. (1993). In their shoes. Understanding Black South Africans through their experiences of lifeKenwyn: Juta and Co. Ltd.

McKenzie, J.A. (1992). The provision of speech and hearing services in a rural district of South Africa. South African Journal of Communication Disorders, 39 5-54.

Naidoo, P. (1994). Test translation an a South African context using the peabody Picture Vocabulary Test-revised .Unpublished M.A. in Speech Pathology. , University of Durban-Westville.

Nzimande, P. M. (1996). The preventative and promotive health digest in the changing Africa. 2nd edition Pinetown: Alberts Publishers

Parkendorf, C. and Alant, E. (1997). Culturally valid assessment tools: North Sotho tracslation of the Peabody Picture Vocabulary Test - revised. South African Jounal of Communication Disorders, 44 3-12.

Pillay,M. (1992) . Communication disorders : beliefs and practices in a Zulu community. Unpublished undergraduate research , University of Durban-Westville, Durban.

Pillay, M. Kathard, H., and Samuel, M (1997) . The curriculum of practice: A conceptual framework for speech-language therapy and audiology practice with a black African first language clientele. The South African Journal of Communication Disorders, 44 109-117.

Platzky, R. & Girson, J. (1993). Indigenous healers and stuttering. South African Journal of Communication Disorders, 30 43- 47.

Robinson, T. L. & Crowe, T.A. (1998). Culture-based considerations in programming for stuttering intervention with Arican-American clients and their families. Language, Speech and Hearing Services in Schools, 29 172-179.

Spector, R.E. (1991). Cultural diversity in health and illness. California: Appleton and Lange.

Taylor, O.L. (1986 ). The nature of communication disorders in culturally and linguistically diverse populations. San Deigo: College-Hill Press.

Taylor, O.L. (1994, June). Cultural diversity in communication disorders. Paper presented at the South African Speech-Language-Hearing Association, Durban.

Taylor,O.L. & Clarke, M.G. (1994). Culture and communication disorders: A theoretical Framework. Seminars in Hearing,.15, 103-114.

Taylor, O.L. & Anderson, N.B. (1988). Communication behaviours that vary from standard norms : assessment and treatment. In Yoder, D. & Kent, R. Decsion-making in Speech-language pathology. Philadelphia: B.C. Decker.

Van Riper, C. (1971). The nature of stuttering. Engelwood Cliffs: Prentice-Hall.

Watson, J.B. & Kayser, H. (1994). Assessment of bilingual /bicultural children and adults who stutter. Seminars in Speech and Language 15, 149-164.

Zimmerman, G., Liljeblad, S., Frank, A., & Cleeland, C. (1983). The Indians have many termsfor it. : Stuttering among the Bannock-Shoshone.Journal of Speech and Hearing Research, 26 315-318.