Recreating a Language: a socio-historical approach to the study of Shaba Swahili


Swahili is a Bantu language spoken in a wide area of Africa. In East Africa, it is spoken in Tanzania, Kenya, Uganda, and the southern part of Somalia; in central Africa, one hears it in Burundi, Rwanda, and the Democratic Republic of the Congo (DRC), also known as the former Zaire; in Southern Africa, it is spoken in Zambia and Mozambique. Used by people in ethnically, culturally, and linguistically diverse areas, it provides a wide continuum of dialects (for dialect details, see Nurse & Hinnebusch, 1993).

The focus in this paper is on Shaba Swahili (hereafter ShS), a dialect of Swahili used in Shaba -- known today as Katanga -- located in the southern province of the Democratic Republic of the Congo. Today, Katanga has an ethnic distribution reflecting social realities that existed throughout the development of ShS. The bulk of past research focusing on the analysis of Katanga Swahili assumed that, because of its deviations from the norms of East African Swahili (EAS), it was a pidginized variety that later developed into a creole variant. Unfortunately, scholarship was not based on reliable or empirical data. (Harris, 1956; Polomé, 1968, 1969, 1971; Fabian, 1982, 1986, 1990; Heine, 1970; Bokamba, 1977) For the most part, the theory was speculative in nature, or it relied on non-native speakers of ShS. (Kapanga, 1991) To complicate matters, there are no extensive written accounts of the development of this variety, or of the original variety that gave rise to the one spoken today in Katanga. These realities have meant that the determination of whether ShS was a pidgin that later developed into a creole has been one of the most difficult tasks faced by linguists. Recent inroads in the field of linguistics, however, may bring some closure to these hypotheses.

In a given historical period, social variables are indexed such that some are more prominent than others; the most potent variables are assigned the highest indices. A change in historical period, which generally derives from (a) social event(s), generally leads to a reordering of the variables in terms of vitality, determining the direction of linguistic change. (Kapanga, 1998) By isolating linguistic variables pertaining to each historical period, we can reconstruct the variety of ShS that served as the backbone for the present-day variety. In doing this, we find that ShS is not a pidgin that developed into a creole language, but rather a language that was acquired as a second language and underwent the normal processes of language change and variation.

Katanga as a Political Entity

Given that most linguistic changes do not take place in a social vacuum, historical facts may be used to see how certain linguistic attributes interacted with the socio-cultural and socio-psychological variables of the following periods:

-- the pre-colonial, characterized by the spread of Swahili into the predominantly Sanga areas of Bunkeya and Likasi;

-- the colonial, typified by the creation of the mining company and the arrival into Lubumbashi of migrant workers from the Kasai provinces in the DRC and neighboring countries (i.e., Rwanda, Burundi, and Zambia); and

-- the post-colonial, known for its adoption of French as the language of education, of government, and of socio-economic advancement.(1)

The early history of Katanga -- characterized by trade -- involved local ethnic groups and foreigners from East Africa. Among the latter were the Waswahili and the Wanyamwezi, porters for the Arab traders who headed the caravans (Polomé, 1968), and who, along with their East African workers, spoke a language known as Swahili. This they introduced into Katanga; it was adopted later as the language of trade and inter-ethnic communication.

At this time, the major trading center was Bunkeya, located in the Likasi area and inhabited by people belonging to the Yeke and Sanga ethnic groups, of whom the latter were in the demographic majority. Having come to Bunkeya with traders' caravans, the Yeke or Wanyamwezi were less significant numerically. Very highly valued, however, were the caravan heads (i.e. the Arab merchants) with the most linguistic prestige; they spoke Kiswahili (Polomé, 1968), which was adopted as the language of trade in Bunkeya and learned by locals.

The advent of colonization in the Congo, in general and in Katanga in particular, led to the possession of the Congo by the Belgian King, Leopold II. The Belgians' arrival corresponded to the discovery of minerals in the southernmost part of Katanga, which in turn led to struggles for dominance and to conflict between Arab traders and Belgian colonizers. In subsequent confrontations between the two groups, the Arabs were defeated. This defeat, coupled with the development of mining, led to the creation of a new economic center: Lubumbashi, known at the time as Elisabethville. With the rapid expansion of mining, large numbers of laborers were needed. Earlier, however, manpower had been depleted by Arab slave traders. Lack of personnel led the colonizers to hire the Roberts Williams Company, whose duty was to recruit workers from other areas of the Katanga province, i.e., the Kasai provinces and the neighboring countries of Rwanda, Burundi, Zambia, and Zimbabwe. Recruits included the Luba-Katanga, who spoke a Luba variety of Kiluba, and the Luba-Kasai and Lulua, who spoke Tshiluba. Workers from Zambia and Zimbabwe spoke Bemba, other languages, or English.(2)

In the aftermath of Arab defeat, and with the abolition of slave trading in the Congo Free State, the Belgians decided to keep Swahili for wider communication. (Polomé, 1968) They expanded it into various areas of the province, and they prohibited non-Arab Swahili people in the Congo from returning to East Africa. They also encouraged the use of the language in the school system, and signed a deal with Catholic missions to spread Swahili. Thus, at the creation of the Congo Free State in 1908, three agents -- mining, administration, and missions -- were responsible for the expansion of Swahili.

In their campaign to spread Swahili, the Belgians were affected by a drastic curtailment in contact between East Africa and Katanga. Agents relied upon the small number of EAS speakers and on a large number of non-native speakers from local ethnic groups. Of importance here was the fact that the Luba ethnic groups and the Bemba represented large numbers. Their demographic advantage, coupled with their higher levels of education (due to missionary education), allowed the Luba from the Kasais to occupy influential positions (as teachers, office workers, and artisans, for example). (Poloreé, 1969; Heine, 1970) As a result of their socio-economic situation, and because they proved to be zealous learners, the Luba were employed as agents of linguistic expansion throughout the major industrial centers, and played a crucial role in it. (Polomé, 1968; Heine, 1970)

With the advent of independence, Swahili was widely spoken and used throughout Katanga as the sole lingua franca. Its spread did not lead to significant language shift, however, because Swahili co-existed and shared certain communicative functions with other ethnic languages (which continued as conduits for intra-ethnic communication). Daily linguistic contacts, which were cultural as well, led to mutual linguistic coloring, and certainly contributed to the evolution of the original variety of ShS.

In the long run, however, although Swahili was considered important, its prestige dwindled. French, instituted as the official language of the country, was adopted as the language of government, court, education, and economic development. For all purposes, and especially after independence, when education was accessible to a large portion of the population, French's importance increased.

Linguistic Vitality in the Shabian Context

Recent developments in sociolinguistics and psycholinguistics have given rise to a new approach -- known as the socio-psychological approach -- to the study of linguistic variation/change. One of its basic concepts is the notion of ethno-linguistic vitality, identified as one of the socio-cultural determinants of variation in language (Ryan and Giles, 1982), and a derivative of linguistic vitality (Giles and Powesland, 1977). Linguistic vitality is measured by the range and importance of metaphorical functions in terms of fulfillment for users. In other words, the social functions that some linguistic variables perform for the users of a language and the scope in the usage of such linguistic variables is what determines their vitality in a given social context. In general, the language or variety that has the most important functions in the community, as well as the greatest number of them, will usually have the highest degree of vitality. Ethnolinguistic vitality, referring to the forces that bring about change in language use and symbolic value, assumes that the vitality of a given ethnic group depends on three factors: status, demographic strength, and institutional support.

During each of the aforementioned historical periods, one or another ethnic group exhibited a high degree of vitality and was likely to determine the direction of linguistic variation/change. (Kapanga, 1998) Each period corresponded to one dominant ethnic group that had the highest vitality in terms of demography or status. As a consequence, features associated with that group had high indexation markings that could influence variations or changes attributed to that period. Each subsequent historical period led to the re-ordering of indexation markings.(3) Given the historical background of Katanga, one can speculate that the social groups with the highest degree of vitality were the Sanga (the original inhabitants of the area) during the slave trade; the Luba during the colonial period; and the Bemba in post-colonial times.

During the pre-colonial era, the Likasi-Bunkeya region constituted the first Katanga area where Swahili was used as a lingua franca. This was a predominantly Sanga region. Moreover, as a second or third language, Swahili was acquired by the dominant group (i.e., the one with the largest population). It stands to reason that this linguistic group had a high degree of vitality, which translated into the transfer of many Sanga features into ShS. This explains, for example, ShS's adoption of the phonological features attributed to Sanga speakers.

During the colonial period, the prestige enjoyed by the Luba from the Kasais placed them in enviable role-model positions, and other groups with no ethnolinguistic variables looked to them for leadership. The speech of these groups consequently underwent the process whereby social or linguistic groups who have little power or prestige tend to imitate the speech of community members who do. (Bloomfield, 1933) In essence, this explains the high frequency of Luba features in the local dialect.

Later, many changes influenced linguistic patterns. The Likasi area lost its influence to Lubumbashi, where Bemba speakers constituted the majority ethnic group. Ultimately, however, because of the prestige that French was accorded in Congolese society, its speakers -- Congolese intellectuals -- constituted the social group with the highest degree of vitality.

Applying the theory of ethnolinguistic vitality, we see that variables that were dominant in the Likasi-Bunkeya area became marginal as a result of major immigration movements, the use of Luba speakers from the Kasai provinces as agents for the spread of Swahili in the province, and the creation of the capital of Lubumbashi in a predominantly Bemba area. The ethnolinguistic vitality variable that played a major role in influencing Swahili was demographic. Some of the features that found their way into ShS include the realization of certain pronunciation features that correspond to the ones produced by the learners of EAS as a second language.


Today, although some social forces that govern life in Katanga have changed, one can still equate existing forces likely to lead to linguistic variation/change with those extant during the province's developmental stages. The Luba-Kasai speakers (Luba and Lulua) continue to maintain the highest number of intellectuals, to constitute the single most influential ethnic group in all spheres, and to control the economic life of the region. In the Likasi area, the Sanga, together with the Luba speakers, predominate demographically; in Lubumbashi, on the other hand, the largest numbers exist among the Bemba and the Luba. Swahili's prestige, however, is second to that of French among the inhabitants of the province, and French has drastically increased in importance as the language of socioeconomic advancement.

A multi-faceted approach suggests that some of the changes found in East African Swahili have resulted from the different languages found in Katanga through the interference process; items from ethnic languages found in Katanga have simply been transferred to ShS. Historical, social, psychological, and cultural parameters that characterized the province of Katanga have allowed us to predict the changes -- and the stages of change -- for Shaba Swahili.

(1). Editor's note: In the next part of this article, Dr. Kapanga provides a historical reconstruction of Swahili, suggesting that the variety introduced into Shaba bears more features of East African Swahili (EAS) than of any other variety spoken in East Africa. Kapanga argues that historical reconstruction, however, is not able to provide the social, cultural, political and economic factors that may have led to the creation of ShS. Kapanga asks whether an examination of the history governing the development of Likasi, Lubumbashi, and Kolwezi, three important regional centers that served as the bases for the development of Swahili in Shaba, can be useful in attempting to predict the original variety on the one hand, and the successive stages of change that took place thereafter, on the other. To answer these questions, the author examines Shaba's political development.

(2). This explains the presence of some English words in ShS; these include simoko `smoke', kona `corner', and mbekeci `bucket.'

(3). It should be noted that changes that take place at a given period of time do not necessarily disappear once there is reindexation of social variables likely to cause linguistic variation. (Kapanga, 1998) Features well entrenched can resist change despite the fact that the social, cultural, or socio-psychological factors closely associated with them have drastically weakened.

References & further reading

Bloomfield. L. (1933). Language. New York: Henry Holt.

Bokamba, E.G. (1977). The impact of multilingualism on language structure: the case of Central Africa. Anthropological Linguistics 24, pp 14-50.

Fabian, J. (1982). Scratching the surface. Observations in the poetics of lexical borrowings in Shaba Swahili. Anthropological Linguistics 24, pp 14-50.

Fabian, J. (1986). Language and Colonial Power. The appropriation of Swahili in the former Belgian Congo (African Studies Series, 48). Cambridge: Cambridge University Press.

Fabian, J. (1990). History from Below. Amsterdam: John Benjamins Publishing Company.

Giles, H. & Powesland, P. (1977). Social psychology and applied linguistics: towards an integrative approach. Review of Applied Linguistics 33, pp 27-42.

Harris, L. (1956). Congo Swahili. Tanganyika Notes and Records 44:3, pp 8-45.

Heine, B. (1970). Status and Use of African Lingua Francas (Afrika-Studien, 49). Munich & London: Weltforum Verlag.

Kapanga, A.M. (1991). Language Variation and Change: A Case Study of Shaba Swahili. Ph.D. Dissertation. Urbana-Champaign: University of Illinois.

Kapanga, A.M. (1998). Impact of Language Variation and Accommodation Theory on Language Maintenance: an analysis of Shaba Swabili. In Endangered Languages: Current issues and future prospects. Lenore, L.A. & Whaley, L.J., Eds. Cambridge: Cambridge University Press. Pp 261-288.

Nurse, D. & Hinnebusch, Th.J. (1993). Swahili and Sabaki: A linguistic History. Linguistics 121. Berkeley, Los Angeles & London: University of California Press.

Polomé, E. (1968). Lubumbashi Swahili. Journal of African Languages 7, pp 14-25.

Polomé, E. (1969). The position of Swahili and other Bantu languages in Katanga. Texas Studies in Literature and Language 11, pp 905-913.

Polomé, E. (1971). Multilingualism in an urban center: the Lubumbashi case. In Language Use and Social Change. Whiteley, W., Ed. London: Oxford University Press. Pp 364-375.

Ryan, E.B. & Giles, H. (1982). Attitudes Towards Language Variation. Social and applied context. London: Edward Arnold.

Article copyright Cultural Survival, Inc.

CSQ Issue: