Speech Development in Three-year-old Children Acquiring isiXhosa and English in South Africa

Michelle Pascoe, Olebeng Mahura, Jane Le Roux, Emily Danvers, Aimée de Jager, Natania Esterhuizen, Chané Naidoo, Juliette Reynders, Savannah Senior and Amy van der Merwe

Introduction

This chapter focuses on bilingual children simultaneously acquiring English and isiXhosa in Cape Town, South Africa. We aim to describe the typical speech acquisition of three-year-old children acquiring both these languages. The chapter starts by providing background regarding the language and demographics of South Africa, together with a rationale for why it is important, but challenging, to identify children with speech sound disorders in this context. This is followed by a description of the two languages. We then move into a review of the literature on monolingual acquisition of isiXhosa and South African English before considering our data from bilingual children.

South Africa is characterized by a culturally and linguistically diverse population. Our progressive constitution recognizes 11 official languages and advocates equal status for them all. These 11 languages include nine indigenous Bantu languages: isiZulu, isiXhosa, Sepedi, Setswana, Sesotho, Xitsonga, Siswati, Tshivenda and isiNdebele, together with the West Germanic languages of English and Afrikaans. Of course there are many more languages and dialects spoken beyond the officially recognized ones, especially by people who have immigrated to South Africa from neighbouring countries. Multilingualism is common in South Africa, with the exact combinations of languages and dialects spoken varying from region to region. In the Western Cape, the main languages spoken are Afrikaans (spoken by 49.7% of the population), isiXhosa (spoken by 24.7%) and English (spoken by 20.3%) (Statistics South Africa, 2011).

The rich linguistic diversity of Southern Africa offers many exciting research opportunities, yet the study of speech acquisition in this context remains relatively unresearched. Some of the Bantu languages, such as Sesotho, isiXhosa and isiZulu, now have small datasets and several papers documenting typical speech acquisition. Much of this work has focused on monolingual children, and where bi- or multilingual children have been investigated, the children's language exposure and abilities across all their languages have not been well documented. This focus on monolingual children is understandable since researchers wish to investigate these under-researched languages in their purest forms, uninfluenced by other languages. There are rural parts of the country where it is common to find children acquiring one language with little exposure to other languages. However, in cities in South Africa, monolingualism is less common and children will typically be exposed to multiple languages (Posel & Zeller, 2016). There are relatively few studies that have focused on the typical acquisition of multiple phonologies at the same time. According to Gxilishe (2004) and Tuomi et al. (2001), the interplay between the languages is often ignored.

Around the world, children with speech sound difficulties (SSDs) comprise a large proportion of speech-language therapists' caseloads. SSDs may affect more children than any other developmental communication disorder and, if left untreated, can result in long-term academic and psychosocial difficulties (Broomfield & Dodd, 2004; Fox & Dodd, 2001). The World Health Organization's (2007) International Classification of Functioning, Disability and Health considers activity and participation to be profoundly affected by speech impairments. Although the prevalence of speech difficulties in South Africa has not yet been documented, in the United States it is estimated to be 7.5% of children between the ages of three and 11 years (Ruscello, 2008) and UK figures suggest that at least 48,000 children are referred for speech difficulties each year (Broomfield & Dodd, 2004).

For speech and language therapists working in South Africa, the lack of knowledge about typical speech development presents a challenge. Clinicians need to identify children with SSDs and assist them and their families. Being able to identify and manage such children requires a baseline of normative data, collected from the same population as that of the child. There are few speech assessments relevant for use with local indigenous languages (van der Merwe & Le Roux, 2014; van Dulm & Southwood, 2015). Many of the English assessments currently used by speech and language therapists in South Africa have been normed on different populations, e.g. monolingual British children, which could possibly result in therapists pathologizing children who are in fact typical (Holm et al., 1999). We are beginning to build knowledge about typical monolingual acquisition of isiXhosa and South African English, but knowledge about the nature of typical bilingual acquisition is limited. As a prelude to a description of current knowledge about phonological development in South Africa, the following sections provide a description of the phonology of isiXhosa and South African English, in...