2. A brief history of the Chinese in South Africa

Chinese migration to South Africa dates back to the seventeenth century, gained momentum in the late nineteenth century, and continues to the present day. Darryl Accone (Reference Accone2007), a well-known South-African Chinese journalist and novelist, speaks about South Africa's three distinct Chinas – and senses of ‘Chineseness’ – each of which originated in a distinct, historical migration stream.

The first China: Small numbers of Chinese have settled in South Africa since the seventeenth century, initially as convicts and exiles and later, in growing numbers, as independent artisans and traders. They were attracted especially by the gold-rush of the 1870s. The majority of the nineteenth-century arrivals – a population of just over 2000 in total – came from China's southern provinces, and they were Hakka- or Cantonese-speaking. Initially the two groups kept to themselves: Hakka speakers predominated in the coastal towns; Cantonese speakers settled primarily in and around Johannesburg (Yap & Man, Reference Yap and Man1996: 35).

Between 1904 and 1910, over 60,000 Chinese indentured labourers were brought to South Africa to work in the mines. They were repatriated in 1910 after their second three-year contract had run out. According to Melanie Yap and Dianne Man (Reference Yap and Man1996: 133), it is unclear whether any of these indentured labourers managed to remain in South Africa:

The importance of this migration interlude was that the short-lived presence of large numbers of Chinese led to White public outcry and concerns over high levels of Asian migration. This resulted in strong anti-Asian legislation and racial discrimination against the remaining South African Chinese, i.e. the traders and artisans who had settled in the country in the late nineteenth century.

Under apartheid legislation, these Chinese were classified as non-White. They formed part of the larger ‘Coloured’ group, lived in non-White areas and their cultural practices soon showed considerable hybridity: ‘they live in places like Sophiatown, play the saxophone, speak Chinese and Zulu, dance to Rogers and Hammerstein and maskanda, and eat mielie pap as well as congee’ (Yen, Reference Yen2005: 107).Footnote ¹ Although the Chinese descendants of the nineteenth-century migrants were classified as non-White, they were initially not considered ‘previously disadvantaged’ by the post-apartheid government. However, since 2009, members of this group (about 10,000 in number) are considered Black under equity and affirmative action legislation (Accone, Reference Accone2009).

The second China: In the 1970s and 1980s, the apartheid government encouraged business migration from Taiwan, and over 300 Taiwanese factory owners settled in South AfricaFootnote ² (Hart, Reference Hart2002: 2). As a result of this second migration, the broadly Chinese population grew from just over 4000 in the 1940s to around 20,000 in the 1980s (Wilhelm, Reference Wilhelm, Buhlungu, Daniel, Southall and Lutchman2006; Yap & Man, Reference Yap and Man1996: 418ff.). Looking at the small town of Newcastle-Madadeni in KwaZulu-Natal, Gillian Hart (Reference Hart2002: 6) reports that in 1994 there were around 2000 Taiwanese migrants, who had established more than 60 factories which drew on local African labour. During this time, there was also continued migration of Cantonese speakers via Hong Kong. Unlike those belonging to the ‘first China’ (nineteenth-century migration), Taiwanese and Cantonese migrants were classified as ‘honorary Whites’ under apartheid, and thus granted the same privileges and rights as White South Africans.

The third China: In 1998, the South African government re-established relationships with the People's Republic of China and supported migration via special legislation. As a result, the number of (legal and illegal) mainland Chinese migrants grew dramatically in the last ten years and is approaching the half-million mark. These migrants have been called the ‘New Chinese’ (Accone, Reference Accone2007), and are referred to in this paper as South Africa's ‘new’ Chinese diaspora.

In contrast to the Taiwanese industrialists, migrants from Mainland China are predominantly traders, specializing in clothing, footwear and cheap plastic goods. As in other parts of Africa, these are family-run businesses (Dobler, Reference Dobler, Strauss and Saavedra2009). While the ‘first’ and ‘second’ Chinese communities were close-knit (with their own schools, clubs, etc.), the ‘third’ Chinese community is a highly diverse minority, with numerous internal divisions and, it appears, generally weak inter-community connections. Those belonging to the ‘new’ Chinese diaspora come from across China, although the Fujian province – located in the southern coastal part of China – appears to be some kind of ‘hotspot’. The ‘new’ migrants speak various Chinese languages: those from Fujian speak Min or Hakka; Mandarin dialects are spoken by those who come from the North; Wu is used among migrants originating in the East, and Cantonese among those from Hong Kong.

There exists a noticeable linguistic fault-line between the ‘first’ and ‘third’ China, as those belonging to the former have shifted to English (although cultural practices are often maintained). This is illustrated by a short anecdote related by Darryl Accone (Reference Accone2007: 7):

The size, historical depth and complexity of South Africa's Chinese community notwithstanding, South Africa's multilingual language policy – as enshrined in the constitution – does not (currently) list the Chinese languages as heritage languages, and comprehensive sociolinguistic research is – to the best of our knowledge – non-existent. Can current sociological and sociolinguistic theorizing around migration and globalization in general, and super-diversity in particular, help us to understand the phenomenon and practices of this ‘new’ Chinese diaspora?

3. Super-diversity and sociolinguistics

The term super-diversity, introduced by Steve Vertovec (Reference Vertovec2007, Reference Vertovec2010), has become central to European sociolinguistic debates in recent years. This section provides a brief overview as well as a critique of the term.

According to Vertovec, the past two decades were a time of fundamental social change. Hitherto international migration has supported the formation of fairly well-defined and homogeneous ethnic diasporas: the Turks in Germany, Algerians in France, the Pakistanis in Britain, the Puerto Ricans in the US. This changed in the 1990s when net migration rates increased globally and migrant profiles became ever more diverse:

1. (a) Countries of origin multiplied and in most metropolitan contexts we now find migrants from a large variety of ethnic and geographical backgrounds;

2. (b) Migration channels also multiplied and in addition to long-standing forms of labour and entrepreneurial migration, we see student mobility, family and spousal migration and refugees, as well as undocumented migrants. This has led to a situation where migrants, sometimes from the same country of origin, have different rights and statuses;

3. (c) Whereas past migrations were often socially homogeneous – for example, male Mozambicans with low educational qualifications who came to work in South Africa's mining sector, or Taiwanese industrialists as discussed above – current migrations are diversified with regard to gender, age, education and language;

4. (d) Recent developments in communication technologies (cheap international call rates and improved internet connectivity) have created new opportunities to interact with people from ‘back home’ and to develop transnational communicative spaces (Vertovec, Reference Vertovec2004; Blommaert, Reference Blommaert2011).

Thus, rather than viewing the world as consisting of multiple, distinct and well-bounded ethno-linguistic diasporas, super-diversity invites us to conceptualize such diasporas as internally diverse and multi-dimensional, shaped by individual biographies and characterized by different levels of hybridity and interaction with other groups.

This obviously has linguistic implications: linguistic proficiencies, competencies and knowledges cannot be generalized across a particular group (such as ‘the Chinese’ in the context of this paper), but are the outcome of specific – sometimes individual and sometimes collective – education, work and migration experiences (Blommaert, Reference Blommaert2010). In other words, ‘the other’, ‘the migrant’, becomes unpredictable: he or she cannot be ‘known’ simply by knowing his or her country of origin or ethnicity.

The concept of super-diversity, however, is not unproblematic or uncontested. The ‘newness’ of current migration patterns and the social formations they create remain a matter of empirical investigation. As African scholars we would like to draw attention to the history of colonialism and slavery, as well as African, Latin American and Asian urbanizations. As observed by Anthony King in his introduction to the collection Culture, Globalization and the World System, ‘the culture, society and space of early twentieth century Calcutta or Singapore pre-figured the future in a much more accurate way than did that of London or New York’ (p. 8; our emphasis). Thus, when Jan Blommaert & Ben Rampton (Reference Blommaert and Rampton2011: 13) state that super-diversity ‘speaks of rapid change and mobility’, we need to know if this is indeed ‘new’. Or was it perhaps something which always existed but remained unnoticed (Blommaert, Reference Blommaert2011: 3)?

Given the fact that Vertovec's original work looked at demographic changes in Britain, is it possible that super-diversity as outlined above is a phenomenon of the global North (where it might indeed be ‘new’)? Furthermore, it might be limited to cities and metropolitan contexts only. So far research has focused almost exclusively on urban contexts and ‘the urban’ seems implicit in discussions of super-diversity (e.g. Vertovec, Reference Vertovec2011). At this stage, we know comparatively little about how international migration is affecting everyday life in small towns and rural areas at the global periphery.

But perhaps most importantly, the ontological status of super-diversity is unclear: is it an embryonic theory or a descriptive term? If it is the latter, then we need to be able to determine with some measure of precision when a society moves from being simply ‘diverse’ to being ‘super-diverse’. Is there a quantitative or qualitative threshold which we can identify, a threshold at which the interplay of multiple factors reaches a new level of complexity? If, however, super-diversity is meant to be a theory (or to become a theory), then we need to ask how it articulates with other theoretical approaches in the social sciences, especially notions of late or liquid modernity. Like Vertovec, Anthony Giddens (Reference Giddens1991) and Zygmunt Bauman (Reference Bauman2000) emphasize the unpredictability of contemporary social formations, their complexity and plurality as well as the growing individualization of social life.

The concept of super-diversity has been taken up fairly enthusiastically by European (but not American) sociolinguists. They have welcomed it as an opportunity to re-think multilingualism more generally, and to question traditional ideas of well-defined ethnic minorities which use bounded entities called ‘languages’. It allows linguistics to emphasize the fact that multilingual speakers show considerable linguistic fluidity in their everyday linguistic behaviour, and that fragmentation, rather than unity, defines the experiences and practices of contemporary speakers (see, e.g., Busch, Reference Busch2012).

In the remainder of this article, we will contribute to these on-going discussions and debates by describing the larger sociological/demographic as well as sociolinguistic consequences of the ‘new’ Chinese migration to South Africa within a peripheral locality, the small rural towns in the Eastern Cape Province of South Africa.

4. Frontier linguistics: Chinese traders in rural South African towns

The following discussion reports on preliminary fieldwork which was conducted in three rural towns in South Africa's Eastern Cape Province: Tsolo, Maclear and Cala. The towns are fairly small, with populations ranging from above 10,000 for Cala, to around 6000 for Tsolo and Maclear. They are surrounded by villages and farmers practising mainly subsistence agriculture.

The Eastern Cape is home to over six million South Africans (out of a total population of over 50 million). It remains one of South Africa's poorest provinces and shows high levels of out-migration to the urban centres of, especially, Cape Town and Johannesburg. Nationally it has the lowest percentage of residents who were born outside of the province (Table 1), and seems an unlikely candidate for the study of super-diversity. It is, as indicated in the title to this section, a frontier where international migrants are present in small numbers and live among a mainly isiXhosa-speaking population (with varying levels of proficiency in English).

Table 1: Percentage of population born outside of South Africa's nine provinces (Census 2011, Statistics South Africa 2012)Footnote ³

However, statistics only tell us part of the story (the ‘big picture’) and sometimes things look different in particular localities. International migrants are a highly visible presence in all three towns, where they run various small family businesses. JohnFootnote ⁴ – a Chinese shop-owner in Tsolo – explained that the Chinese and other international migrants have moved to these small towns ‘because people follow the money’.

In other words, business opportunities for independent small-scale traders and entrepreneurs are increasingly becoming limited in the cities, where the market is showing signs of saturation. Over the last ten years, however, rural towns have changed from being depressed and abandoned places, to booming market towns. The reason for this lies in the larger political changes: the South African government has followed a policy of increasing decentralization since 2000. As local government structures were developed, this meant that civil servants brought their considerable spending power to these towns, and employment as well as business was revitalized.

A South African shop-owner in Maclear, who had been in business for decades, expressed concern about the growing, and seemingly never-ending, competition: ‘every day a new shop pops up’. These new shops, run and owned by international migrants, do not cater to a niche ethnic market, as is common in most urban contexts, but to a local African clientele.

Let's take the example of Cala. In addition to the ubiquitous presence of large national chains such as SPAR or Build-It, the main road, which cuts across the town, houses numerous smaller businesses. The hardware stores are typically run by Pakistanis, who also dominate the mobile phone business. Cheap, made-in-China consumer goods are on offer in numerous so-called ‘China-Shops’. These include not only global items such as jeans, leather shoes and stereos, but also goods with a distinctly local flavour (albeit still produced in China): colourful pinafores for the newly-wed makoti (‘bride’); chequered blankets, an important part of traditional attire and used in ceremonial contexts; and simple enamel tableware which has long been common in these parts. Indian migrants tend to be found in grocery and furniture stores, whereas higher-end clothing can be obtained from the Senegalese store. The shoemaker, who runs a mobile business out of a tent, is from Ghana, and the hair-dressers, who operate out of converted ship-containers, are from Nigeria. Local residents run taverns and eateries, find employment in the chain stores and government offices, or work as mobile airtime vendors. Things work more or less the same in the other two towns. Although the discussion in this paper focuses on Chinese migrants, it is important to remember that their presence in these localities is embedded within patterns of broader trans-national mobility.

As we walked along the streets of Cala, Tsolo and Maclear, we wondered: How do these international migrants, especially the ‘new’ Chinese diaspora with very limited competence in English, negotiate the formidable linguistic challenges of trade in these rural African towns? Signage plays an important role in this context and various semiotic devices are used to communicate with potential consumers. The sign in Figure 1 is found above an outside display of clothes, and advertises the shop as specializing in a range of electronic goods. Adding the name ‘Cai Xin’ marks the shop clearly as ‘Chinese’ and thus as ‘affordable’. In interviews with local residents the ‘affordability’ of Chinese consumer goods was emphasized regularly and positively. As one young man explained: ‘the Chinese know, they come from communism, they make sure everyone can afford’.

Figure 1. Shop-sign in Cala (2012).

While all shop signs emphasize the store's identity as a ‘China Shop’, many signs also draw on local semiotics. In Figure 2, the shop sign is modelled on the iconography of the local OK Bazaar brand, a former low-cost store which folded in the 1990s, and had a strong small-town presence. The sign is thus within the realm of the familiar and brings back memories of old times. In conjunction with the Chinese branding (‘Ushun Trading’), it emphasizes, yet again, the promise of affordability to the consumer.

Figure 2. Shop-sign in Maclear (2012).

Signage is obviously an important part of communication in trade settings: it advertises and locates the shop; it serves as an invitation to customers. But what happens inside the shop? How are business transactions carried out?

There are some Chinese traders who have been in the area for close to a decade. John, mentioned above, is an example of such a trader. He originates from the east of China and came to the Eastern Cape in 2004. Upon his arrival in South Africa, he worked in Cape Town for a short while, but then entered into a partnership in Maclear and later opened his own shop in Tsolo. He now owns several shops and is married to a local isiXhosa-speaking woman. He speaks English well and also some isiXhosa.

The situation is very different for those who arrived only recently, often sent down from Johannesburg or Cape Town to man local shops which are owned by relatives. Many of these more recent arrivals are from southern China, with very limited competence in English. It appears that upon arrival in the Eastern Cape three expressions are quickly learned: ayikho ‘it's not available’ (used as a stock response whenever there is a communicative breakdown in the shop), and, even more importantly, sisi and bhuti. The latter two are polite, local terms of address for a woman and a man respectively. They are used by the Chinese shop-keepers to summon local assistants. The routine is as follows: customers would enter the shop and approach the Chinese owner at the till with a request, the latter would shake his or her head, indicating incomprehension, and would then call out sisi (or bhuti). Upon this call a local assistant would come to help the customers – either in isiXhosa or English – and show them the goods they were looking for. The use of such local language mediators is an important communicative strategy, and ensures that economic transactions are carried out smoothly. The linguistic forms sisi and bhuti – even though minimal – show a local embeddedness into cultural norms of address and respect. They differ strongly from the demeaning ways in which White traders would address their local assistants in the past, as boy or girl, irrespective of age and status.

Not surprisingly, given the intense language contact situation, a mixed jargon for basic intergroup communication has emerged. This is illustrated in examples 1 to 4. The speakers are two Chinese women in Maclear, both in their early 30s, whom we asked about their life in South Africa, their accommodation, their visits home and their linguistic proficiency. Their responses show the mixed and structurally reduced, but communicatively adequate, nature of this jargon (English in unmarked font, Afrikaans in italics, and isiXhosa in bold).

1. (1) no batala rent (‘no pay rent’; since they sleep in the shop)

2. (2) you come jonga (‘you come to see’, meaning ‘you are sightseeing/visiting’)

3. (3) we no go china (‘we don't go to China’)

4. (4) speak ncinci (‘speak little’)

Structurally this jargon is similar to what Wolfgang Klein and Clive Perdue (Reference Klein and Perdue1997) described as ‘the basic variety’ in their work on untutored second language acquisition in migration contexts. This basic variety follows a (pre-basic) stage which is largely nominal and makes almost no use of verbs. In the basic variety, verbs are used in their non-finite form (i.e. no use of inflection) and are central to the syntactic structure of utterances. Another noticeable feature in the above examples is the frequent (but not consistent) omission of pronouns and the use of a pre-verbal negation. All of these are well-attested features in early interlanguages.

In situations of social segregation, such jargons can stabilize and develop into pidgin languages. However, this is unlikely to occur in these towns, as interactions between migrants and local residents are plentiful and there are no noticeable barriers to integration. As noted above, in the case of John, inter-marriages are not unusual and inter-group attitudes are generally positive. This supports second language acquisition rather than pidginization.

Let's look at Dongmei, a young woman from southern China, who arrived in Tsolo twelve months ago because she could not find work in Johannesburg. ‘There are too many Chinese [in Johannesburg]’, she says. She now looks after a shop, which is owned by her cousin. Her language is best characterized as a second language variety of English, and not a jargon. Her speech is marked by Chinese phonology as well as by highly localized lexical borrowings. For example, when we asked her about her relationship to the other Chinese person in the shop, she answered without hesitation ‘my sbali’, using a borrowing from isiXhosa, usibali (‘brother-in-law’). Her acquisition of English is thus shaped by local practices. Other lexical features include the frequent use of ja (typical for South African English more generally), and the expression ‘this side’ (for ‘here’). The latter is common in the English of isiXhosa-speakers and is a calque based on kweli'cala (as opposed to kewlaa cala ‘that side’, meaning ‘there’). In terms of phonology, Dongmei – like many Chinese learners of English – commonly replaces /r/ with the lateral /l/. This is a pronunciation which is also common in some local isiXhosa-English varieties, and can thus be accommodated easily. Thus, when Dongmei referred to /lice/ instead of /rice/, one of the authors (who is isiXhosa-speaking) accommodated her without hesitation. In the following transcription, the pronunciation is only indicated for rice, and all other words are spelled according to standard orthography (even though /r/ and /l/ variation remains common throughout). The discussion was about food and the difficulty of getting Chinese rice and flour in Tsolo.

English is clearly an important lingua franca and target of acquisition in these rural towns: it allows not only for communication between the local population and the migrant traders, but also for communication among the traders from China, Pakistan, Ghana and Senegal. At the same time, however, migrants such as Dongmei express a strong desire to learn isiXhosa, and those who have lived in the area for longer – such as John – generally know both English and isiXhosa.

In sum, we can identify different communicative strategies in the absence of a shared lingua franca, ranging from the skilful use of signs and images, to local language mediators, and various processes of language acquisition.