There are three languages which are of value to the people. The first is vernacular, the second Arabic, the third English. The vernacular is the foundation of the people's life; it guards their customs and their tribal beliefs. Arabic will be the language of commerce and of Local Government; it must be understood by all those who wish to take part in these spheres of activity. The English language is the gateway to scientific knowledge; without this it is not possible to reach higher education.Footnote 1
This statement in 1950 to the Bahr el Ghazal Province Council by its British chairman and province governor reveals a considerable change in the policy of the Sudan government. Before 1947, the British administration of the southern provinces had officially sought to exclude Arabic from local government, and to restrict the entry of Arabic-speaking northern Sudanese. This ‘Southern Policy’, which was suddenly reversed from 1947, has provoked much subsequent criticism from scholars and Sudanese nationalists.Footnote 2 Post-independence rulers would seek to remedy its perceived effects and to forcibly generate national unity through the imposition of Arabic language policies in the south. But the story of Arabic in southern Sudan is more complicated than this narrative of colonial exclusion followed by hastily attempted national integration. What Governor Owen did not reveal was that the Arabic already being used in commerce and local government was not generally the Arabic of northern Sudan, but the creole Arabic of southern Sudan. This language is often known as ‘Juba Arabic’, but this article instead adopts the term ‘South Sudanese Arabic’ to encompass its regional and temporal variations across South Sudan.Footnote 3 It originated as a pidgin language among southern slaves, soldiers, and servants in nineteenth-century trading camps and Turco-Egyptian government stations, drawing its vocabulary primarily from the Egyptian and Sudanese colloquial Arabic of the traders and military officers. But it developed entirely new structures, recognised by linguists as a process of creolisation. This distinct language would survive both the British attempts to suppress its use, and the subsequent efforts to promote Sudanese Arabic in the south.
The military origins of this language and its subsequent use amongst soldiers, police, and other agents of state coercion have given it ambiguous moral and political connotations among South Sudanese. Yet its relationship to state power has never been straightforward, as the British efforts to suppress its use before 1947 indicate. While studies of other pidgins and creoles in Africa have often emphasised their origins in the need for communication between colonisers and colonised, some scholars point out that these languages have always been more important for communication among Africans.Footnote 4 This is particularly evident in the case of South Sudanese Arabic, due to its official disapprobation. Its survival as an unofficial lingua franca of local government, trade, and urban interaction demonstrates that while these ‘spheres of activity’ (as Owen termed them) were produced by and associated with the state, they were never entirely controlled by it.
In exploring the history of South Sudanese Arabic, this article draws on two key arguments from linguistic studies of creolisation, elaborated below. Firstly, scholars have moved away from defining creolisation as the point when a pidgin language – that is, a simplified contact language with restricted vocabulary created from a mixture of languages – becomes nativised as a first language, to defining it as the point when a pidgin language becomes necessary for communication in particular contexts, and hence develops as a main language of interaction. Secondly, linguistic studies of South Sudanese Arabic have argued that its creolisation entailed innovative restructuring, creating a new language rather than representing an attempt to speak a dominant or superstrate Arabic dialect.
The article applies these insights to the historical evidence, to argue that the creolisation of South Sudanese Arabic was driven primarily by the need for communication among local actors, including the state officials who were drawn into using this lingua franca. This also demonstrates the wider relevance of language history to the study of state formation. Scholars analysing African states have increasingly focused on the importance of local-level negotiation of the state among multiple actors who deploy particular discursive ‘repertoires’ or ‘genres’, including ‘popular modes of communication’.Footnote 5 Yet such studies rarely focus on the linguistic media of this negotiation, even when using creolisation as a metaphor for the resulting forms of governance and political culture.Footnote 6 This article begins by exploring the value of a fuller and more precise analysis of linguistic creolisation as both metaphor and means of negotiating the local state. It then traces the history of South Sudanese Arabic from its origins as a military pidgin in the nineteenth century, through its development as a language of local government, trade, and urbanisation, to its current uncertain and ambiguous status as an unofficial national lingua franca of South Sudan. The article suggests that this enduring ambiguity represents the tensions, opportunities and exclusions that have characterised the local state.
CREOLISATION: LANGUAGES AND METAPHORS OF THE LOCAL STATE
Linguists have debated the definition of South Sudanese Arabic: is it a pidgin, a creole, or a ‘post-creole’ Arabic dialect?Footnote 7 This set of categories was once seen to form a standard linguistic trajectory. The first stage of transition from pidgin to creole was traditionally defined by the nativisation and expansion of a pidgin into a mother tongue, a process that has been restricted to a small urban minority in southern Sudan. The situation here has also been complicated by regional variation in South Sudanese Arabic, and by the presence of Sudanese colloquial Arabic, whose influence was assumed by the 1980s to be leading to a ‘decreolisation’ of South Sudanese Arabic. This has not occurred: many South Sudanese can speak Sudanese colloquial Arabic, and use simplified dialects of it as well as, or instead of, the creole South Sudanese Arabic, but this has not eradicated the creole.Footnote 8 New varieties have also emerged since the 1980s, such as the slang spoken by urban youth, combining elements of vernacular language with South Sudanese Arabic.Footnote 9
Increasingly, however, linguists have abandoned the model of a clear trajectory from pidgin to creole to dialect, and have argued instead for a spectrum or continuum of language varieties; some adopt the hyphenated term ‘pidgin-creole’ to encompass these.Footnote 10 Summing up new directions in pidgin-creole linguistics in 1991, Christine Jourdan argued that Solomons Pijin was spoken both as a limited pidgin in rural areas, and as a creole in the city, where it was ‘the main language of urban social interaction’, regardless of whether it was a first or second language.Footnote 11 Jourdan therefore defined the creole not as a mother-tongue but as a ‘main language’ in specific contexts – contexts that, as Peter Pels puts it, made the creole ‘necessary for successful communication’.Footnote 12
This article pursues the idea that creolisation is the process whereby South Sudanese Arabic became necessary in particular contexts. It also draws on another important insight from the linguistic studies: creole languages are distinguished from other ‘mixed languages’ by their distinctively innovative structures, revealing that their speakers were not simply aiming to speak a target superstrate language; instead the priority was immediate ease of communication. South Sudanese Arabic was a new language, distinguished by morphological restructuring such as the adoption of verbal prefixes to denote tense, rendering it largely unintelligible with Arabic dialects.Footnote 13 The creolisation of South Sudanese Arabic did not result simply from a compulsion or desire to speak the language of central state power.
Taking these linguistic lessons together, this article argues that they demonstrate why linguistic creolisation is such an apt metaphor – as well as medium – for wider processes. Much historical and political discourse within and about South Sudan has perpetuated a set of stark dichotomies between state and society, modernity and tradition, and between urban and rural societies and economies. Such binaries have been increasingly questioned in wider scholarship, and yet spatial metaphors of ‘spheres’ and ‘boundaries’ have remained hard to avoid.Footnote 14 Studies of African chiefs, for example, analyse their role as ‘gatekeepers’, ‘brokers’, and cultural ‘interpreters’, mediating between state and society.Footnote 15 Accounts of legal pluralism often employ the term ‘hybrid’, the origins of which imply ‘that original pure and homogeneous elements entered into a combination’.Footnote 16 Creolisation on the other hand is not the straightforward amalgamation of two languages; rather, it is the emergence of something new that, as Justin Willis puts it, is ‘more than simply a blend of its constituent parts’.Footnote 17 Creoles emerge precisely ‘in order that people sharing no language may communicate’.Footnote 18
Similarly the history of local state formation should not be understood as a binary conversation between colonisers and colonised, state and non-state, nor as the hybrid combination of two distinct strands: one urban, cosmopolitan, and governmental; the other rural, mono-ethnic, and traditional. The spread of South Sudanese Arabic far beyond its military origins suggests that we should think instead in terms of broader ‘fields of interaction’, or ‘negotiation arenas’, in which the boundaries between state and non-state actors have been blurred.Footnote 19 Through the interaction and negotiation among these multiple actors, the novel idea and institutions of the local state would emerge.
THE ORIGINS OF SOUTH SUDANESE ARABIC: NEGOTIATING THE VIOLENCE OF THE NINETEENTH CENTURY
Pidgin and creole languages are generally associated with the unequal power relations of colonialism and the dominance of ‘superstrate’ European languages over the ‘substrate’ languages of the colonised. Indeed the wider adoption by non-linguists of the term ‘creolisation’ for imprecise or positive notions of cultural intermingling and globalisation has been criticised for ignoring the political origins and racial connotations of the term.Footnote 20 Colonial domination is very much to the forefront of one historical account of an African creole language, namely Johannes Fabian's study of Shaba Swahili in eastern Congo, where the need for ‘communication with the colonized’ led to colonial attempts to standardise and control this lingua franca.Footnote 21 South Sudanese Arabic was never officially recognised by the colonial state, let alone standardised. But it originated in an earlier context of domination and subordination: in the southward expansion of the ivory and slave trade and of Turco-Egyptian imperialism in the nineteenth century, and in the military slavery on which this expansion relied.Footnote 22
From 1820, the Turco-Egyptian conquest of Sudan pushed the frontier of slave-raiding further south, but vegetation blockages on the Nile prevented the extension of riverain trade until 1840–1, when an expedition sent by the Governor of Egypt, Mohammed ‘Ali, reached the Gondokoro area (near modern-day Juba). Over subsequent decades, multinational traders, with their private armies of northern Sudanese mercenaries and southern slave-soldiers, established fortified stations known as z ara'ib across the Bahr el Ghazal and Equatorial regions, from which they traded and raided the surrounding populations for ivory, supplies, and slaves. From 1869, these territories were annexed to the government and the Egyptian army took over or established its own stations, from which a similar system of raiding largely continued.Footnote 23
In the zara'ib, Sudanese or Egyptian colloquial Arabic speakers formed a small minority in relation to the expanding population of soldiers, slaves, and servants captured or recruited within the south, among whom a military pidgin Arabic would develop into early varieties of South Sudanese Arabic.Footnote 24 Clearly these were contexts of inequality, in which Arabic and Ottoman Turkish were the superstrate languages of imperial power. Yet Jonathan Owens argues that the linguistic structures of South Sudanese Arabic are distinct from either Arabic or the vernaculars, reflecting innovative restructuring by its nineteenth-century speakers. It is difficult to tell from the nineteenth-century sources precisely what kind of Arabic was being used in and around the zara'ib, but the fragmentary references to ‘vehicular’, ‘indigenised’, or ‘slang’ Arabic spoken by soldiers and other southerners indicate that it was not the Arabic of northern Sudan or Egypt. Its distinctive structures would also be preserved in the language of the southern soldiers who later moved into Uganda and became known as ‘Nubi’, and whose creole Arabic began to be recorded in the early twentieth century.Footnote 25
These commercial and imperial forces had recognised the need for negotiation from the outset, and used southern slave-soldiers as interpreters, known by the Turkish term ‘dragoman’. The 1840–1 expedition brought two Dinka interpreters from Khartoum, whose influence was immense: ‘even the expressions of the natives cannot be depended upon, for our black soldiers can translate just what pleases them’.Footnote 26 Local interpreters subsequently emerged along the Nile; by the 1860s, ‘many of the natives in the vicinity of Gondokoro [had] learnt a little Arabic from the traders’, albeit ‘with a peculiar accent’.Footnote 27 In 2008, an elderly Bari politician located the origins of South Sudanese Arabic in the Gondokoro area:
The Baris began to learn the language of the traders, Arabic. Intelligent Baris who learned a bit of Arabic also became translators. There was this mixing up of trade and soldiers coming to the village. The Arabic language in Juba has a lot of Bari words in it. It is more or less half Arabic and half Bari.Footnote 28
Owens acknowledges that ‘Bari substratal influence is observable’ in South Sudanese Arabic though the major source of its vocabulary is Arabic; recent research by Shuichiro Nakao suggests the substratal influence of other vernacular languages too, reflecting the extent of local interaction around the stations.Footnote 29
According to the historian Richard Gray, ‘the Turks became dependent on these riverain Bari to act as middlemen in the search for and purchase of further supplies’.Footnote 30 The relatively peaceful early relations along the Nile deteriorated, however, as the traders established a network of zara'ib across the Bahr el Ghazal region to the west, accompanied by increasing raiding, enslavement, and local warfare. Yet even in this violent context, the need to negotiate local alliances was recognised by the trading forces, who taught young slaves Arabic so that they might become ‘interpreters and commercial agents’.Footnote 31 Local people also saw the value in having their own interpreters: one Zande chief sent a boy with a trading agent to learn Arabic ‘so that he might hereafter be useful to his tribe as interpreter’.Footnote 32 The violent depredations of raiding parties were only ameliorated by entering into relations and alliances with the zara'ib, and a common means of communication was therefore vital to securing protection for local people. Oral histories often trace the origins of chiefship to the bravery of the interlocutors who first negotiated relations with the trading forces so that people settled in the vicinity of the camps.Footnote 33
Linguistic knowledge and skills of interlocution did not simply produce alliances between traders and local groups, but in the process also produced new forms of power and authority. In the 1870s, Zande spearmen trained in the zara'ib were reportedly becoming ‘chiefs’, having ‘learnt to speak Arabic fluently’.Footnote 34 By this time the Turco-Egyptian government was establishing its own network of stations or taking over the trading zara'ib; the ‘Arab-speaking Baris’ retained a privileged position in the stations on the Nile as ‘interpreters, translators and inspectors’.Footnote 35 The role of interpreters was increasingly being formalised as a nascent government chiefship, though strikingly they retained their title of ‘dragoman’. One or two dragomans were said to be placed in every village near the stations:
These men were originally slaves of the old slave-dealers. They all speak Arabic and were trained first as gun-boys, and now are employed as a kind of native police. Each native village is required to support one or more of these men, who are responsible for the peace of the place, and also superintend the payment of the grain tax to the government. Some twenty or thirty of them live near a fort and when porters are wanted, or natives required for work at the station, they are commissioned to collect the requisite number.Footnote 36
According to Robert Felkin, writing in the 1880s, the dragomans were ‘freed slaves, who are employed as police-soldiers, and negotiate with the natives’.Footnote 37
From the outset then, knowledge and use of the emerging South Sudanese Arabic was associated with military force; the local ‘negotiations’ by interpreters were obviously skewed by the threat of violence to which their privileged linguistic knowledge gave them access. But if anything, this gave people the incentive to acquire some of this linguistic knowledge for themselves, and hence South Sudanese Arabic spread beyond the station inhabitants and individual interpreters. General Gordon witnessed a whole village, many of whose inhabitants were Arabic-speaking, moving into the vicinity of a station, and predicted that ‘Arabic must be the language of these countries’.Footnote 38 Concentric circles of settlement and interaction developed around the zara'ib and government stations, creating arenas in which both language and negotiation were shaped by localised needs and tensions.Footnote 39 Relatively peaceful relations might be achieved around some stations, but these could also break down when communication failed.
In the Rumbek area of Bahr el Ghazal, for example, some of the earliest zara'ib had been established from the 1850s with the aid of Dinka interpreters and allies; in 1860, ‘chiefs’ from the Cic and Agar Dinka came to greet the new European owner of the Ronga zariba, and to invoke ‘the trader's alliance’ to fight their enemies, as Gray puts it.Footnote 40 Rumbek later became the province headquarters for the Egyptian government; Gaetano Casati described the ‘friendly relations’ and ‘truce’ that developed here between the Agar Dinka and a particular governor, whom they called ‘our man’.Footnote 41 After this governor left, however, relations around the station deteriorated. In 1881, a ‘deputation of Agar chiefs’ trekked all the way to the government headquarters at Lado on the Nile, to complain to Emin Pasha about the abduction of their children by the station forces, prompting Emin's visit to Rumbek.Footnote 42 In 1883, an Agar war leader, Wol Athiang, led a successful attack on the station and the garrison was subsequently withdrawn.Footnote 43 Yet even Wol Athiang had previously engaged in closer relations with the Rumbek station; his sister was married to, or owned by, an Egyptian officer.Footnote 44 This example demonstrates the importance of local alliances and negotiation to the operation of the trading zara'ib and government garrisons; and it reveals how local interlocutors were prepared to approach the state forces directly to negotiate alliances and to ameliorate the depredations of raiding parties.
Egyptian government rule collapsed in the 1880s in the face of the Mahdist uprising in Sudan, and the last governor of Equatoria, Emin Pasha, was famously forced to withdraw southwards and eventually be ‘rescued’ by Henry Stanley. But the southern Sudanese soldiers remained in Equatoria or what would become northern Uganda; some joined the Mahdist forces in Equatoria, while others were recruited into Frederick Lugard's forces in Uganda in the 1890s.Footnote 45 The latter settled in Uganda or Kenya and became known as Nubi or Nubian, retaining their creole Arabic language, Kinubi. The early origins of this creolisation are evident in the structural and lexical similarity of South Sudanese Arabic and Kinubi, despite their limited contact after the 1880s.Footnote 46
In the wake of the subsequent Anglo-Egyptian reconquest of Sudan in 1898, many of the old Egyptian stations were taken over by the British officers of the new Anglo-Egyptian Condominium of the Sudan, or by their compatriots in the colonial government of Uganda, which administered part of eastern Equatoria from Gondokoro until 1914. Some of the southern soldiers from the old Egyptian garrisons returned to the stations along the Nile; Kinubi was the principal language used in the Ugandan headquarters at Gondokoro, and it was also widely spoken on the west bank in the Belgian Lado Enclave.Footnote 47 The Sudan government relied on southern soldiers and interpreters from the Egyptian or Mahdist armies, such as Rihan Effendi Abdulla, ‘ex-slave native of Gogrial’, who accompanied the Bahr el Ghazal ‘reoccupation’ force in 1901; or the interpreter at Mongalla, El Amin Rabeh, a Bari soldier who had fought in the Mahdist army.Footnote 48 When the Sudan government took over the Lado Enclave in 1910, they again turned to soldiers who were originally from the Enclave to be the most ‘useful and reliable’ interpreters and guides.Footnote 49 Communities of retired soldiers and their descendants formed the nuclei of the small urban administrative centres, particularly along the Nile, living in the malakiya (‘civilian’) ‘native’ quarters and speaking South Sudanese Arabic as their first language.Footnote 50 Other people would be drawn into these Malakiyan communities through military or police service, or marriage, adopting Islam and South Sudanese Arabic.Footnote 51
Though never explicitly sanctioned or acknowledged by the colonial government, South Sudanese Arabic was thus given a privileged place in local administration from the outset, as well as retaining its association with the coercive forces of the state. As in the nineteenth century, language and interpreting were often central to the selection of local chiefs; some of those recognised as chief had been soldiers or employees of previous governments, and many spoke at least some Arabic.Footnote 52 The Uganda government recognised as Bari chiefs men who had been interpreters and allies of the Mahdist forces.Footnote 53 Even in the predominantly Dinka-speaking districts like Rumbek, some early chiefs were Arabic-speaking former police or army conscripts; a later British official in northern Bahr el Ghazal admitted that many chiefs ‘had a knowledge of Arabic as their only qualification’.Footnote 54 The early reports did not specify what kind of Arabic this was, but later reports referred to the early chiefs having gained their positions through their knowledge of ‘bastard Arabic’, as the British termed it.Footnote 55 By the 1920s, British officers would acknowledge the prevalence of a distinctive South Sudanese Arabic in the government stations. Multiethnic settlements were forming around the recognised chiefs and around the government posts, contributing to the need for a lingua franca among these people seeking protection and opportunity in the emerging local state.Footnote 56
The first two decades of colonial rule thus consolidated rather than disrupted the patterns that had developed over the nineteenth century, in which knowledge of Arabic was converted into new positions of power and intermediation. But these intermediaries had not simply sought to acquire the Arabic or English of imperial rulers; nor was there an existing lingua franca like Swahili in eastern Africa for them to adopt. Instead, varieties of a new language had emerged out of the particular circumstances of the military stations and their local relations, produced by the need for communication on the ground more than by the imposition of a superstrate colonial language. This would become even more evident in the mid-colonial period, when the government did seek to replace this creole language with English.
‘BAD ARABIC’? EXPRESSING THE CONTRADICTIONS OF COLONIALISM, c. 1920–50
In the 1920s, British colonial administrators in southern Sudan increasingly reported the ubiquity of what they described as a ‘pidgin’ Arabic in the government stations, army, and police, and noted its wider spread with some alarm. The local state had retained its coercive character as the colonial government sought to enforce taxation and labour demands through an increasingly formalised system of chiefs with their own police and courts. Yet the effects of this state coercion were complex, opening up related arenas of economic activity and social and political interaction which were never entirely controlled by the state. To the dislike of colonial officials, the district headquarters were gradually turning into small towns, as some people discerned opportunity as well as threat in the new arenas of offices, markets, and courts.Footnote 57 The need for communication in these arenas would consolidate South Sudanese Arabic as a lingua franca, so that even British officials were drawn into using it as a language of local government and commerce.
This pragmatic government use of South Sudanese Arabic was increasingly at odds with official policy, however. Unlike the deliberate cultivation or tolerance of Swahili and its standardisation as an intermediary lingua franca by colonial governments in the Congo, Tanganyika or Kenya, the Sudan government sought to suppress South Sudanese Arabic altogether.Footnote 58 The Condominium government feared the spread of subversive, nationalist activity into Sudan after the instability and eventual independence of Egypt in 1922, a fear exacerbated by urban and military unrest in northern Sudan in 1924. Rural Sudanese were therefore to be governed by traditional authorities and insulated from the ‘septic germs’ of such unrest.Footnote 59 The ‘Southern Policy’, articulated in 1930 as the desired creation of ‘self-contained racial or tribal units’, has been seen as a deliberate attempt to separate northern and southern Sudan.Footnote 60 But it was also part of this broader colonial campaign against the perceived effects of education, religious fervour, urbanisation, and migration: effects which were assumed to have the most disruptive potential among hitherto illiterate, non-Muslim southerners.Footnote 61 Northern Sudanese traders still operated under license in the southern provinces, but the administration sought to separate such townspeople from the rural ‘tribes’. Yet the colonial government also required literate junior employees: in turning to the Christian mission societies to produce these, it entered into a long-term struggle over the nature and purpose of schooling; a Christian education was also seen by many administrators as dangerous, if it detached pupils from the supposedly restraining ties of family and traditional community.Footnote 62 Meanwhile, however, colonial labour, taxation, and agricultural policies were having even greater transformative effect on local economy and society, by generating labour migration, market production, and monetisation.Footnote 63
These contradictions and tensions between the conservative visions of traditional rural society and the far-reaching effects of colonial economic and administrative policies were epitomised by discussion of language policy in the southern provinces from the 1920s.Footnote 64 In attempting to insulate southerners from subversive influences, the government sought to exclude Arabic of any kind, and instead to make English and the vernaculars the languages of administration and education in the south.Footnote 65 But there were already warnings of the ubiquity of ‘a debased form of Arabic’.Footnote 66 In 1927, even Governor-General John Maffey, the leading proponent of the conservative ‘Native Administration’ policies, encountered this reality while on a tour of the south:
I have now fully appreciated the difficulties of ever making English the lingua franca of the south. Things have gone much further than I had expected. Wherever I penetrated, whether to the top of the Imatong [mountains] or to the Belgian Congo border, I found Arabic in ready use by the local spokesman of the people … Indeed we shall have to consider whether Arabic, after all, in spite of its risks, must not be our instrument.Footnote 67
A few months later, the education secretary, Matthew, issued a memorandum explicitly identifying a ‘language problem’ in southern Sudan. He argued that ‘Southern Arabic’ was ‘totally unfit to be used for educational purposes’, since it was ‘only a clumsy instrument’. He went on to raise a central concern of the government: that this ‘pidgin’ might start to be written in Arabic script, which risked encouraging Islamicisation and Arabicisation. One solution would be to write it in Roman script, but other British officials argued that teaching ‘bastard Arabic’ using English characters was not ‘education in the true meaning of the term’.Footnote 68
Such debates rumbled on through the 1930s among British officials across all three southern provinces. In 1936, the Resident Inspector of Southern Education declared that the attempt to introduce English ‘as the language of communication’ had been largely a ‘failure’:
Arabic is in fact the lingua franca. It is regarded not so much as the language of Islam but as the language of Government … It is bad Arabic ungrammatical [sic], incapable of exact statement, but it serves the purpose of the natives in their dealings with each other, has a certain prestige and will suffice to get him in touch with Government officials. He learns sufficient for his needs and no more, but there are very few places in which no one knows any Arabic … The market place is a greater factor in the spreading of Arabic than the Government Official and these two are the main factors; but a third is the increasing necessity of some lingua franca between tribes …Footnote 69
This latter point was reiterated by the district commissioner (DC) of Yei, who declared bluntly that the question of a lingua franca had already been solved: ‘Arabic is the language of intercommunication and commerce in the Southern Sudan.’Footnote 70 A number of other officials expressed a similar recognition or resignation at the limited capacity of government to control the use of ‘bastard Arabic’: ‘One cannot dictate to the native what shall become his lingua franca, he will choose what is most useful to him.’ Even in a district like Zeraf, where the entire population spoke Nuer, the DC reported that Arabic was spreading as ‘the language of both trade and wage-earning’. The Upper Nile Province governor added that an ‘almost unrecognisable’ Arabic had ‘already got a strong hold, especially among the Dinka in this province’, and that efforts might better be concentrated on ‘teaching the police to talk understandable Arabic’.Footnote 71
Many officials remained opposed to the teaching or writing of this ‘Southern Arabic’ or ‘Mongallese’, however.Footnote 72 In 1942, one British officer (and later DC), Captain Lyth of the Equatorial Corps, announced that he was translating certain Christian texts into ‘Mongallese pidgin Arabic’, to be printed for use in military services, and for sale in Juba. Lyth had the temerity, and prescience, to question the government's language policy:
Mongallese is the only language – if it may be dignified by such a name – universally known and used throughout the Equatorial Corps, by both black and white alike (as well as in all the market-places in your Province.) Because of this, its users have had to find means to express every shade of meaning which they require in conversation, and this they succeed in doing, if sometimes in a somewhat roundabout way, very satisfactorily … It is interesting in passing, is it not, to notice in books dealing with the period of Emin Pasha's governorship, that an almost identical form of pidgin Arabic was then in use in the Equatorial Province … The policy of the Sudan Government would therefore appear to be almost as optimistic – though no doubt as wholly commendable – as that of the Uganda Government in attempting to eradicate the pidgin form of Swahili in favour of English.Footnote 73
Lyth was forbidden to use or sell the booklet. But even while expressing his disapproval of any written form of Southern Arabic, the education inspector, Hickson, did admit that ‘in fact people do write Mongallese (the local tailor does so when sending me my bill)’.Footnote 74 Some of the early chiefs’ court records in Mongalla Province were also being kept in ‘Arabic in English characters’, though increasingly the government would insist on English or vernacular record-keeping.Footnote 75
British administrators thus remained well aware of the widespread use of South Sudanese Arabic, including in their own communication with traders, chiefs, soldiers, and servants, unless they had learned the vernacular – which was more difficult in linguistically plural districts, and became less common as officials served shorter periods in one district.Footnote 76 Most colonial officers in the Sudan Political Service had studied Arabic. But they too were contributing to the distinctive South Sudanese Arabic by employing their own simplified version, sometimes known as ‘Bimbashi’ Arabic (the lowest rank held by British military officers was that of Bimbashi, or major). As the British linguist Tucker put it:
Any remnant of grammar left is more English than Semitic, the British official finding it much easier to string Arabic words together in an English sentence-order than to attempt a Northern construction which would quite probably be unfamiliar to his hearers.Footnote 77
Unlike the attempts to master and control Swahili in Belgian Congo, the official disapproval of Arabic meant that the creole language was developing without any formal government rules or grammars.Footnote 78
In the courts, South Sudanese Arabic often functioned as an intermediary language, which was (just about) comprehensible to British DCs, and into which the vernaculars were translated by police officers or court interpreters.Footnote 79 As in other colonial contexts, this clearly gave interpreters and chiefs considerable power to control interactions with colonial officials.Footnote 80 As one DC revealed in 1948: in rural areas, ‘chiefs are apt to take advantage of their unsophisticated subjects who speak no Arabic and are all too rarely in contact with a district commissioner’.Footnote 81 It might be assumed then that South Sudanese Arabic acted as a ‘boundary marker’ to elite status, enabling chiefs, police and interpreters to act as powerful gatekeepers to the colonial state.Footnote 82
Yet as we have seen, languages tend to defy rather than to demarcate boundaries.Footnote 83 While South Sudanese Arabic retained its association with state coercion, this only encouraged its wider spread, as Tucker acknowledged:
With the opening of roads and the protection given to travellers, it has spread throughout the Southern Sudan, its pronunciation varying from tribe to tribe. It also had – and still has in some areas – a great prestige as being the language best calculated to win favour with the police … Ultimately, of course, it was a useful language to know, should one's case come before the District Commissioner, since it enabled the plaintiff to evade the court interpreter, who was not always to be trusted to translate fairly unless well bribed.Footnote 84
While literacy in English did remain restricted, spoken South Sudanese Arabic could be more easily acquired, especially through interaction in the small towns and markets. Again this demonstrates the blurring of boundaries, as the colonial government sought – but largely failed – to prevent such interaction or to keep people out of the towns. Even in the predominantly Dinka-speaking districts of Rumbek and Yirrol, administrators reported recurring struggles to evict ‘rural’ people from the town and to combat the influence of ‘detribalised’ and ‘Arabic speaking’ townspeople, including women beer sellers, schoolboys and young men, chiefs, and the families of government employees.Footnote 85 Officials reiterated that South Sudanese Arabic had taken hold as a result of the necessity of and demand for ‘a common means of communication’, not just between government and people, but to enable ‘intercourse at Government centres between people speaking different languages’.Footnote 86 As we have seen, linguists increasingly define creolisation as the development of a main and necessary language in just such contexts.
Even though urbanisation and the spread of South Sudanese Arabic were actively discouraged by the government, these processes – and the wider process of commoditisation that the government did encourage – were contributing to state formation by producing new kinds of local political competition and new disputes, which people turned to institutions associated with the state to resolve. The earliest formal chiefs’ courts established in the 1920s and 1930s were reported to be copying all the militaristic and authoritarian styles of state policing and government offices. But over time they became key arenas in which to negotiate rights and relations, leading colonial officials to comment on the litigious nature of southern Sudanese societies. By 1947, officials in Rumbek were concerned that the appeal court sessions held in the town were attracting two to three thousand people a day, and encouraging ‘all and sundry’ to come to town.Footnote 87
As the courts became ever more crowded arenas, so their use of South Sudanese Arabic expanded beyond any original use merely as an intermediary language. By the 1940s, one DC described ‘Mongallese Arabic’ as the nearest thing to a lingua franca in the courts.Footnote 88 Its use produced an enduring vocabulary that spread beyond the courts and into the vernaculars: Arabic words like ‘garama’ (court fine) and ‘sijin’ (prison) were reportedly replacing vernacular terms in the Bari language by the early 1930s.Footnote 89 The process of linguistic creolisation was ongoing as South Sudanese Arabic became ever more necessary. At the same time a kind of legal and judicial creolisation was occurring in the chiefs’ courts, as they became the site for the negotiation of new rights and regulations. The laws and procedures of these local courts were a product not simply of the hybridisation of colonial with indigenous justice, but of the ongoing, local negotiation and contestation of legal rights and duties among multiple actors with varying degrees of incorporation or relationship to the state. Like the creole language in which some of this negotiation took place, this local justice was thus not a combination of two distinct strands, but a novel, creative production emerging from debates over individual and collective rights in the context of the new colonial economy.Footnote 90
South Sudanese Arabic embodied the contradictions of colonialism: neither the idealised rural vernacular, nor the formal grammar and script of colonial government. Yet it did not simply hang ‘uncertainly between languages and worlds’ like the Arabic words adopted into government discourse in northern Sudan; rather, it took on a life of its own.Footnote 91 South Sudanese Arabic was expanding and undergoing a genuine linguistic creolisation – becoming a main and necessary language in particular contexts – which was beyond the power of the colonial state to prevent or control, even though those very contexts were the product and site of state formation. Revealingly, when the first province councils were established in the late 1940s, ‘Mongallese Arabic’ was the main language used among the chiefs and other members of the Equatoria Province Council.Footnote 92 But from the 1950s, the sudden attempt to integrate north and south through the medium of Arabic would generate new political tensions around the meaning and use of language in southern Sudan.
THE LANGUAGE OF THE SOUTH? DISPARITIES IN THE MEANING AND USE OF SOUTH SUDANESE ARABIC AFTER 1956
Catherine Miller emphasises the frequent disparity between the ‘symbolic importance’ of a language and its actual role in socioeconomic – and, I would add, political – life, a disparity which is often particularly acute in the case of creole languages.Footnote 93 The disparity between the meaning and use of South Sudanese Arabic is paralleled by a set of discursive dichotomies that have been constructed in southern (now South) Sudan since Sudan's independence in 1956 – between state and society, town and countryside, money and non-money economies – which belie long-term processes of local state-making, urbanisation, and commoditisation. Southern political leaders have promoted English as a government language in opposition to the Sudanese Arabic of Khartoum, reflecting both a heritage of Christian missionary education and the increasing links of southern Sudan to neighbouring Anglophone countries. Meanwhile, the vernacular languages have remained associated with rurally rooted, ethnic cultures and traditions and seen as the authentic vehicles of southern identities. South Sudanese Arabic has had an uneasy or unacknowledged position in such political discourse, associated variously with state coercion and the military, with northern Sudanese Arabic, or with urbanisation and the erosion of ethnic culture. Processes of urbanisation, militarisation, and local state formation since the 1950s have continued to generate moral concern and debate among southerners. But the numbers of people participating in these processes have only multiplied, ensuring the increasing necessity of an alternative to either the vernaculars or the oscillating government languages of English and Sudanese Arabic, by which southerners might communicate among themselves in local arenas of the state. Such everyday communication has been more important for negotiating state and nation than the politicised discourses about language reveal.
From the later 1950s, the newly-independent Sudan government promoted Arabic language and Islamic education in the south with increasing forcefulness, a campaign which was revived again from the 1980s, and particularly under the National Islamic Front government in the 1990s.Footnote 94 The southern political opposition therefore associated Sudanese Arabic with northern domination, and instead promoted English; while they could speak South Sudanese Arabic, they did not endorse it as a vehicle of southern identity.Footnote 95 During the peace of 1972–83 and again since 2005, English was recognised as the official language of Southern Sudan and, since 2011, as the national language of the new South Sudanese state.Footnote 96 Unlike Swahili in East Africa, South Sudanese Arabic has not been formally recognised as an official political language of national integration.Footnote 97 But while the political leaders might prefer to promote English for ideological reasons, during both periods of civil war they were forced – like the colonial government had been – to accept South Sudanese Arabic as a military lingua franca for pragmatic reasons, as ‘the only common language which could be relatively easily acquired’ by the ordinary southerners recruited into the rebel forces.Footnote 98
This recent history has generated continuing ambiguity towards South Sudanese Arabic: its military and coercive associations have been retained, and some southerners also associate it with Sudanese Arabic and hence with government policies of Arabicisation and Islamicisation. For those who took refuge in Uganda or Kenya during the civil wars and studied in English, Arabic of any form has carried the stigma of northern Sudanese oppression, whereas English became seen as ‘a medium of communication with other Africans’ as well as among southern Sudanese.Footnote 99 When such refugees returned to South Sudan from 2005 onwards, however, they found themselves criticised for not knowing any South Sudanese Arabic, and referred to as ‘uncivilised’ or baladi (rural/provincial).Footnote 100 The association of South Sudanese Arabic – here subsumed in the category of ‘Simple Arabic’ – with the towns was noted by a newspaper contributor from Juba in 2007:
When I was growing up in this town in the mid 1960s and 1970s I thought Simple Arabic was the language of the people of Southern Sudan. But before long, I discovered it was not. Some of my relatives who visited our family from countryside or village could not speak or understand the simplest version of Simple Arabic. We use to call those relatives Ahaale, meaning villagers or uncivilised.Footnote 101
According to an early 1970s study by Abdon Agaw, a Bor Dinka linguist, ‘[t]he continuous intermingling of people in different towns in the southern region’ had encouraged the spread of South Sudanese Arabic.Footnote 102 He also recognised that it was becoming a first language among children in the bigger towns.Footnote 103 By the early twenty-first century, many urban families – even mono-ethnic ones – were speaking South Sudanese Arabic in the home, and their children were learning it as a first language. This has raised considerable concern about the threat to the vernacular languages. Even – or perhaps especially – in Juba, the town most associated historically with South Sudanese Arabic, many political leaders and other commentators have been bemoaning its spread in recent years: ‘[t]he problem is the encroachment of Arabic deep into the [rural] areas, and people forgetting the Bari language’.Footnote 104 Just as Governor Owen had asserted in 1950, the vernacular languages have continued to be seen by many South Sudanese as vital cultural repositories and mediums of expression.Footnote 105 In contrast, South Sudanese Arabic has retained associations with the moral ills of town life; much scholarship on Sudan also emphasises the moral dichotomies between urban government and the vernacular cultures of the rural areas.Footnote 106
But the debates and uncertainties over the moral and political meaning of South Sudanese Arabic belie the reality of its spreading use and the porosity of ‘the rural-urban imaginary dividing line’.Footnote 107 Studies in the 1970s and 1980s revealed extensive urban-rural linkages, and social and economic interaction, which transcended and incorporated the town boundaries.Footnote 108 Women played an important role in this interaction, being responsible for carrying produce to markets and for alcohol production and sale in towns and marketplaces; interethnic marriages also encouraged the use of South Sudanese Arabic within households. Even in predominantly Dinka-speaking areas of Bahr el Ghazal, this language was becoming a valuable – indeed necessary – investment among local entrepreneurs like cattle traders.Footnote 109
South Sudanese Arabic has therefore also become ever more necessary for negotiating the local state. The Dinka cattle traders’ knowledge of Arabic, for example, enabled them to facilitate communication among local officials, townspeople, and villagers, and often to go on to hold positions as chiefs or court members.Footnote 110 Such individuals have transcended any clear dichotomy between state and non-state actors. Successful traders and businesspeople, teachers, church leaders, chiefs, elders, government employees, and other members of local elites have moved in and out of formal state institutions and appropriated state practices and regulatory orders – such as the formation of ‘committees’ with ‘chairmen’ – into local and familial settings like marriage negotiations or funeral gatherings. The creation of local government councils and committees from the 1970s opened further opportunities for government employment and political participation, even if their autonomy and effectiveness remained limited.Footnote 111
Alongside these councils and committees, chiefs’ courts have remained vital sites for the local negotiation of rights and laws, through which the idea of the state has been produced and reproduced. These courts represent a creolised judicial order, drawing on formal state laws and procedures but producing distinct judicial practices that do not correspond to official state regulations or to customary practices – particularly in their simultaneous awards of compensation and punitive fines or imprisonment.Footnote 112
South Sudanese Arabic was the most common language spoken in the chiefs' courts in Juba in the early 1980s, as it still is nowadays. Even in Bari chiefs' courts, litigants and chiefs have come to speak more in South Sudanese Arabic than in Bari. Miller recorded a distinctively ‘formal’ type of South Sudanese Arabic employed by litigants and chiefs in 1980s Juba, including some old Ottoman Turkish terms as part of an established ‘court vocabulary’. Even a Malakiyan court judge who could speak Sudanese Arabic adapted his speech to that of the local litigants and ‘never tried to assert his Authority through the exclusive use of a high linguistic norm’.Footnote 113 These town court chiefs were not demonstrating their status through knowledge of the Sudanese Arabic of high office; rather they were using and recognising the distinctive linguistic medium of South Sudanese Arabic, which facilitated the most effective communication among litigants and court members. More recently, the chiefs' courts have increasingly been located in the towns and heard cases from multiple language groups. As one MP asserted in 2008, it was vital for chiefs to know South Sudanese Arabic ‘because they hear cases from all the people and simple Arabic is spoken’.Footnote 114 As in the colonial period, the use of South Sudanese Arabic in such courts privileges those with linguistic knowledge and experience of town and government ways. But this category has continued to expand and includes squatters, hawkers, brewers, and prostitutes, as well as government officers, chiefs, and police. By the 2000s, young townswomen in Juba and Yei were using South Sudanese Arabic to argue their cases vociferously in the chiefs' courts, along with a wide range of other litigants. Whether or not this was a conscious decision, it was guaranteed to ensure the widest possible comprehension of their narratives by the chiefs, court officials, police, and (often vocal) multiethnic audiences.
As South Sudanese Arabic has become ever more widely spoken, government officials and political actors have also found it necessary to use this language in local state arenas, as recognised in the 1970s:
Now in Juba … politicians and leaders of popular organizations find it more expedient to speak Juba Arabic than to speak fluent English in addressing a public rally. By so doing, they stand better chances of putting their message across and of winning response from the audience.Footnote 115
Despite the promotion of English as the official government language over the next decade, such use of South Sudanese Arabic had only increased by 1983: ‘As well as being used informally for the normal purposes of social interaction it is now also used over the radio, in big gatherings such as political meetings, and by public officials who wish to communicate with the population.’Footnote 116
Through such interaction, ideas of state and nation were also being produced: in the 1980s, one Bari chief explicitly promoted the use of ‘our Southern Arabic’ in his court and criticised a young man for talking like a northerner.Footnote 117 From this time, some scholars predicted that South Sudanese Arabic would be absorbed as merely a dialect of the dominant Sudanese Arabic, but Miller argues that such a process has been hindered by ‘a tendency to condemn speakers whose speech goes too much in the direction of the Arabic of the north’.Footnote 118 Indeed she argues for the increasing ‘legitimization’ of South Sudanese Arabic as ‘the language of the South’. Both among the displaced around Khartoum and in the South itself, South Sudanese Arabic came to be widely used in cultural and artistic productions. It has also long been used in Christian services, broadcasts, and publications by the churches, thus further eroding any lingering associations with Islam. It was prominent on Radio SPLA, and used for speeches by SPLA leader John Garang.Footnote 119 The idea of South Sudanese Arabic as a unifying lingua franca for the emerging new state of South Sudan was epitomised in its use for the South Sudan Theatre Company's production of Shakespeare's Cymbeline, which was performed in London in 2012. Its translator, Joseph Abuk, located the creole language in the contradictions of colonialism and ironies of Southern Policy: ‘Juba Arabic grew up because the African people in the south were not allowed by the British to speak Arabic.’Footnote 120 Out of the exclusionary practices of the colonial and postcolonial Sudanese state, a creole language thus emerged to negotiate the tensions, threats, and opportunities represented by the state, contributing in the end to the idea of a distinctively South Sudanese state.
CONCLUSION
[In 2013] Juba Arabic is on the radio and in the government ministries. It is the language of the market place and the church hall. Most of all it's the language of the youth, full of humour and ‘street cred’ … In the Konyokonyo marketplace, traders and hawkers from a vast range of linguistic backgrounds use Juba Arabic to do their business. Women from the villages bring their own words and experiment with language to sell their wares and cement crucial relationships. All are using Juba Arabic to varying degrees … It is essential to survival. A whole range of people in the city who barely glimpse English in the distance, need to learn Juba Arabic to make a living here. All the civil servants use Juba Arabic. Enter any government department and the language you hear is Juba Arabic. English hovers around officially and you may well conduct your meetings in it, but when you emerge from the minister's office the staff are speaking Juba Arabic.Footnote 121
The varied uses of South Sudanese Arabic identified in this report are indicative of the multiple actors and arenas that have played a part in constituting the local state and in blurring its boundaries over the past century or more. Some scholars have focused on linguistic knowledge as a means of monopolising and restricting communication with the colonial state, producing powerful intermediary gatekeepers.Footnote 122 But the history of South Sudanese Arabic demonstrates instead that linguistic creolisation opened up, rather than closed down, the opportunities to negotiate the state at the local level, and that language use has defied and blurred any state-society boundary.
The speakers of this language, from government ministers to urban youth or rural women traders, have not been speaking a poor imitation of an elite language; they have been speaking a distinct creole language which emerged out of the imperative to communicate within the new arenas opened up by the establishment of the state. Linguistic studies have defined creolisation as the process by which a pidgin language becomes necessary – essential for survival, indeed in particular contexts. This article has argued that the contexts in which South Sudanese Arabic became necessary have also been the principal arenas of local state formation: arenas in which a wide range of individuals pursued rights and resources, and in the process turned to state institutions and regulatory orders like courts, markets, and local government councils. Attempts to inhibit the use of South Sudanese Arabic or to impose other official languages have failed to prevent its spread: a stubborn sign that a great deal of power and agency in everyday life has come to be located neither in the elite languages of high office nor in the idealised cultural and moral values of the vernaculars, but in the quotidian struggles, disputes and discussions that take place just outside the minister's office – and in the markets, streets, bars, tea-places, courts, police stations, army barracks, and local government offices, in which this lingua franca is so often spoken.
Linguistic studies of South Sudanese Arabic have also emphasised its original innovative structures; creolisation is not simply a mixture of two distinct languages, but the creation of a new language. Again this article has suggested that this is useful device for thinking about local state formation, not as the hybridisation of two kinds of political culture (state and non-state, or modern and traditional) but as the product of multiple negotiations and innovations, and of the need to find a common language of governance. What is perhaps most striking is how both the state and the creole language emerged out of multiple local negotiations to exhibit broadly common features. This no doubt reflects the unbounded nature of local arenas and fields of interaction, and the travel of ideas, discourses, and languages. But some linguists have also speculated that the common features among creole languages more widely suggest universal ‘parallel tendencies of linguistic development in parallel social situations’.Footnote 123 Strikingly of course, these have invariably been situations that arose in ‘the context of the state’.Footnote 124 Studying the history of creolisation perhaps then also offers a way to address the central problematic raised by Gupta: ‘the relationship between the translocality of “the state” and the necessarily localized offices, institutions and practices in which it is instantiated’.Footnote 125 The history of South Sudanese Arabic reveals the processes through which multiple versions of the local state emerged out of the imperative to communicate and negotiate, even in highly militarised and unequal contexts. These processes of local state formation produced locally distinctive articulations of governance. Yet each of these contexts was structured by similar tensions and needs, generating broadly common strategies and institutions – and a developing common language – for dealing with the threats, constraints and opportunities presented by the state.
