Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, LINGUISTICS ( (c) Oxford University Press USA, 2016. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy).

date: 29 April 2017

Language and Social Hierarchy in West Africa

Summary and Keywords

In the indigenous sociolinguistic systems of West Africa, an important way of expressing—and creating—social hierarchy in interaction is through intermediaries: third parties, through whom messages are relayed. The forms of mediation vary by region, by the scale of the social hierarchy, and by the ways hierarchy is locally understood. In larger-scale systems where hierarchy is elaborate, the interacting parties include a high-status person, a mediator who ranks lower, and a third person or group—perhaps another dignitary, but potentially anyone. In smaller-scale, more egalitarian societies, the (putative) interactants could include an authoritative spirit represented by a mask, the mask’s bearer, a “translator,” and an audience. In all these systems, mediated interactions may also involve distinctive registers or vocalizations. Meanwhile, the interactional structure and its characteristic ways of speaking offer tropes and resources for expressing politeness in everyday talk.

In the traditions connected with precolonial kingdoms and empires, professional praise orators deliver eulogistic performances for their higher-status patrons. This role is understood as transmission—transmitting a message from the past, or from a group, or from another dignitary—more than as creating a composition from whole cloth. The transmitter amplifies and embellishes the message; he or she does not originate it. In addition to their formal public performances, these orators serve as interpreters and intermediaries between their patrons and their patrons’ visitors. Speech to the patron is relayed through the interpreter, even if the original speaker and the patron are in the same room. Social hierarchy is thus expressed as interactional distance.

In the Sahel, these social hierarchies involve a division of labor, including communicative labor, in a complex system of ranked castes and orders. The praise orators, as professional experts in the arts of language and communication, are a separate, low-ranking category (known by the French term griot). Some features of griot performance style, and the contrasting—sometimes even disfluent—verbal conduct of high-ranking aristocrats, carry over into speech registers used by persons of any social category in situations evoking hierarchy (petitioning, for example). In indigenous state systems further south, professional orators are not a separate caste, and chiefs are also supposed to have verbal skills, although still using intermediaries. Special honorific registers, such as the esoteric Akan “palace speech,” are used in the chief’s court. Some politeness forms in everyday Akan usage today echo these practices.

An example of a small-scale society is the Bedik (Senegal-Guinea border), among whom masked dancers serve as the visible and auditory representation of spirit beings. The mask spirits, whose speech and conduct contrasts with their bearers’ ordinary behavior, require “translators” to relay their messages to addressees. This too is mediated communication, involving a multi-party interactional structure as well as distinctive vocalizations.

Linguistic repertoires in the Sahel have long included Arabic, and Islamic learning is another source of high status, coexisting with other traditional sources and sharing some interactional patterns. The European conquest brought European languages to the top of West African linguistic hierarchies, which have remained largely in place since independence.

Keywords: West Africa, speech intermediaries, registers, honorifics, multilingualism, social hierarchy, Islam, colonialism, sociolinguistic history, oratory

Social hierarchy can be communicated in many ways: through distinctive semiotic forms, including language varieties; through physical distance or barriers; through the use of intermediaries (thus creating communicative distance); or in combinations of these. All these means of expressing hierarchy can be found in West Africa, past and present. This article’s exploration of language and social hierarchy will first concentrate on indigenous communicative traditions; that is, ways of communicating hierarchy that predate the conquest of the African continent by European powers, although they have persisted into later periods. That discussion will occupy the longest portion of the article. Next, we consider the forms of hierarchy, and their connections with language and communication, that followed from the colonial conquest, although earlier traditions persisted in many ways. Finally, we turn briefly to the postcolonial era.

To focus on sub-Saharan Africa, as this article will do, rather than including the Maghreb and Saharan regions, is somewhat artificial, though conventional in the literature. The Sahara was never an absolute barrier sealing off Mediterranean North Africa from the rest of the continent. On the contrary, trade relations and cultural connections across the desert can be traced back for millennia. It is similarly artificial to write of indigenous social or sociolinguistic systems, as if these were isolated from the rest of the world or from each other. Most conspicuously, the arrival of Islam in the Sahel region by the 10th century ce, if not earlier, and its increasing penetration and influence, which continue to the present day, mean that Arabic language and Islamic religious hierarchies have been part of the sub-Saharan sociolinguistic scene for more than a thousand years. Indigenous, then, can only be a relative term, and it applies mainly to the ways of life that antedate European domination of the continent.

It will be useful to organize the discussion roughly by geographical region and forms of polity.

1. Precolonial Sociolinguistic Traditions

1.1. Communicating Hierarchy in the Sahel

We begin with the social hierarchies of the West African Sahel region. This is the area of farmland and pasture just south of the desert. Especially important in the communicative traditions of these systems is the use of intermediaries in communication, elaborated, and mobilized for close, as well as long-distance, relationships. Multilingualism, especially with Arabic, is relevant here too, but close consideration of its connection with social hierarchy is postponed to a later section.

1.1.1. Sahelian Understandings of Hierarchy

Because it is also north of the range of the tsetse fly, the Sahel accommodates social systems in which cattle are an important form of wealth and horses are a military resource. In this region emerged the great medieval empires that controlled the southern end of the Saharan caravan trade, in which salt—scarce south of the desert—and commodities from the north were exchanged for gold, slaves, and ivory (among other things) from the south. Military forays from either side occasionally attempted to control both ends of the trade, with only brief success. Scholars differ as to the political consequences of these campaigns, such as whether invasions from the desert north succeeded in defeating the Empire of Ghana, in about 1076ce, or whether, instead, an already Muslim Ghana cooperated with its northern neighbors in political action against non-Muslims (see, e.g., Burkhalter 1992; Conrad & Fisher, 1982, for opposing interpretations).1 In any case, it was only the growth of the Atlantic slave trade with Europe that ultimately brought about the decline of the Sahel empires.

The social systems that are rooted in these empires and their smaller successor states feature a division of labor accompanied by social segregation, such that the various kinds of economic producers, and an aristocracy, form separate categories whose memberships are ascribed by birth. What is most of interest for our purposes is that the arts of rhetoric—public speaking and artful speaking more generally—are part of the division of labor, hence the prerogative of a special social category. Overall, these social systems are constituted by two intersecting principles of social organization: a principle of caste, which ideologically links a person’s physical essence with professional specialization and conduct (including speaking), and is established genealogically; and a principle of orders, which establishes relationships of menial servitude and a contrast between slave and free (Johnson, 1999; Diop, 1981; Diouf, 2001). The caste principle segregates various categories of artisans—distinguished according to the substances they work with professionally—from aristocrats and common farmers. Interactions and ranking, within and between categories, are subject to caste ideology and regulation. Among these segregated categories are the artisans of the word, bardic categories known across the region as griots.2 Other artisan categories include smiths, leatherworkers, and woodworkers. While the broad categories are endogamous, all these categories can be subdivided and ranked, according to their profession, their conduct, and the rank of their patrons. The details of artisan subdivisions vary from one region to another, as do some particulars of endogamy and rank. Nevertheless, broad similarities across the entire Sahel make for a kind of convertibility among the systems, regardless of dominant language or ethnicity. A traveler can always be located socially within the hierarchy of caste ranks.

Meanwhile, the orders crosscut the caste distinctions, since any social category might have its own slaves.3 For example, the Wolof gewel—praise singers, musicians, speechmakers, messengers, keepers of genealogies—have had slaves called jaamu gewel (‘slaves of griots’) or fat-ndënd (‘drumkeepers’), since one of the duties of these slaves is to take care of the drums used by musicians in their performances. Most slaves, however, were and are attached to free families outside the artisan categories. Most of these non-artisan free families are ordinary farmers. Though they might be considered (and labeled) “noble,” their nobility lies in their supposed moral superiority to artisans and slaves, not in the European sense of an elite minority; in fact, they form a demographic majority. Still, among these “nobles,” there are also aristocratic and princely lineages.

This combination of castes and orders leads to complex calculations of rank and relationship, whose details vary somewhat across sahelian societies. In Wolof traditions, differences in rank have been especially sharp and persistent. All artisans rank lower than the rest of the free Wolof population, known collectively as “noble” (géer). Artisans are excluded from primary land ownership and political office (except for a few posts reserved for them), and are largely segregated in residence; their physical contacts with nobles are highly regulated regarding who may give food to whom and under what circumstances, who may sit on another’s mat, and so on. Among the artisan categories, the griots rank very low—lower than smiths and leatherworkers—and are collectively scorned by nobles, although particular families and individuals have long-term ties that are much valued. In Mande traditions, while the same categories and segregation exist, their ranking seems to be less steep and less suffused with ideas of pollution and scorn. Among Mande, blacksmiths seem to be regarded by the nobles as dangerous to others but not collectively contemptible. A smith could even take over high political office if there is no viable noble candidate (Kassim Kone, personal communication); such an arrangement would be unheard of among the Wolof. Why there should be this difference is not clear. It may have something to do with the greater role of the Wolof, as a coastal population, in the Atlantic slave trade, or possibly with the role of Islam, which may have come with Maghrebi traders to the Senegal River region a little earlier than it influenced the Mande area further east.

1.1.2. Patterns of Communication

The core of the griot’s role and relationship to patrons is transmission (jottali, in Wolof): conveying and embellishing messages composed by others, thus helping to make the message effective in the ears of addressees. The griot is thought of as a kind of conduit from higher-ranking persons, including ancestral generations, to an audience. Both male and female griots serve patrons in this way, working as intermediaries with professional skills in the arts of oratory, beautification, praise poetry and historical narrative, and convivial conversation. This service is deemed necessary even in small spaces if the interactants are major dignitaries, so a conversation between two important chiefs may be conducted by their griots. Some kinds of performance require special training, such as genealogical praise singing, a crucial griot service since it attests to a patron’s high rank and “beauty of birth” (rafet-juddu, in Wolof); this is information transmitted from ancestral generations.4 The griot does not invent the message; his/her role is transmission and beautification, not creation.5

While professional activities like praise singing or participation in a drum orchestra are the griots’ special province, the role of intermediary in less specialized communication modes is generalized to the lower ranks. That is, there is a general principle that interprets social hierarchy as distance, and constructs social distance through communicative distance—longer chains of interaction. If no griot is available to take a message to a neighboring village chief, a slave or possibly some other artisan will do—even, in extremis, a noble from a junior lineage branch, or a teenager.

In Wolof, these ideas about communicative services underlie a differentiation of linguistic registers that rural villagers (in my fieldwork) associated with griot and noble categories (Irvine, 1990). Wolof has relatively little geographical dialect differentiation (apart from urbanism; see Section 2), perhaps because griots, as both messengers and language “experts,” travel frequently and widely. Register differentiation, however, can be considerable. The griots’ characteristic performance style, involving rhetorical elaboration, loud and rapid speech, high pitch, and a particular pitch contour, affects prosody, phonology, lexis, syntactic complexity, and some aspects of morphology. This style tends to carry over into forms of speaking that are not formal public performances but connote griot activities, such as extended narrative, expressions of gratitude and praise, and requests for gifts (in cash or kind). It is important to note, however, that this style of speaking, although locally labeled “griot talk,” is used by anyone who engages in the relevant speech activities, not just by actual griots. It is therefore better understood as “speaking like a griot,” for the purposes of the occasion. Its opposite is a style associated with high nobles: soft, slow, grammatically simple, relatively taciturn, and even disfluent (Irvine, 1990).

Griots and their characteristic performances, widely found across the Sahel, must have existed for a very long time. Described in oral traditions about the origins of kingdoms and empires (e.g., as characters in the story of Sunjata, founder of the Mali empire in 1235), their presence is identifiable in written accounts by Arab travelers and scholars who visited the region’s royal courts. Ibn Battuta, for example, who spent several months at the court of Mali in 1352, describes the king’s griot interpreters, who also served as musicians and praise singers (Ibn Battuta, 1922). Portuguese visitors offer similar descriptions a little later, labeling the category (as gaul, from Wolof gewel) and emphasizing their social segregation (Fernandes, 1940; Fernandes’s account draws on earlier manuscript sources and interviews). Fernandes also describes conversations among dignitaries as carried on via speech intermediaries. Early European visitors sometimes took griots’ social segregation to mean they were Jews, though that origin is unlikely.6

Meanwhile, similar institutions—a social category of bards, historians, musicians, and genealogical praise singers—can be found in precolonial state societies throughout the Sahel (for comprehensive surveys, see Hale, 1998; Tamari, 1997). For example, a classic portrayal of Hausa praise singers by anthropologist M. G. Smith (1957) describes their characteristic public behavior: “the service performed—praise or blame, shouted or sung with all available force—is consumed by all within earshot, as well as by the person to whom it is addressed.” Podstavsky (2004), who elaborates on the complexity of subtypes among Hausa praise singers, notes that they are free to utter bawdy or politically licentious speech without retribution, yet in other contexts they produce elegant performances for high-ranking patrons.7 These descriptions could apply equally well elsewhere in the Sahel. For Haalpulaar’en people (a.k.a. Tukulor or Toucouleur) of the Fouta Toro region, Dilley (2004) describes several categories of praise singers and musicians, as well as other social ranks (see also Wane, 1969); and there are many sources on Mande griots (e.g., Camara, 1976, for Guinea; Conrad & Frank, 1995; Hoffman, 2000, for Mali). Some works offer texts of griot performances. Very few consider linguistic registers outside the formal performance context.8

1.2. Precolonial States South of the Sahel

Although neither griots in particular nor caste-like differentiations in general are part of social traditions south of the Sahel, the pattern of speech mediation as a way of expressing and establishing hierarchy is widespread all over West Africa. Accordingly, many kingdoms and chiefdoms throughout the area rely on intermediaries, including specialists in rhetoric, to create and express social distance between a king (or chief, or high-ranking person) and his/her public. One example, well described in the literature, comes from the Akan group of states. Akan is the name of a set of very closely related languages associated with kingdoms and chiefdoms in Ghana and Côte d’Ivoire, including the Ashanti kingdom. In those states, an official called the okyeame (pl. akyeame) serves as intermediary for a high-status person, especially a king or chief. As Yankah (1995a, 1995b) details, the king or chief does not address his public directly, nor may people speak directly to him. It is the okyeame who relays the chief’s message, either verbatim or in embellished paraphrasing. If the chief’s would-be interlocutors are a group or include a dignitary, they will have an okyeame of their own, who relays their response through the chief’s okyeame. The two intermediaries speak directly to each other, after listening to their principals (chief and visitor[s]). While the role of these intermediaries resembles that of sahelian griots, an important difference is that the Akan okyeame is not drawn from a low-ranking, segregated caste. Although these Akan spokespersons, male or female, are lower in the social hierarchy than the king or chief whose oratory they deliver, they are not conspicuously low-ranking in any other sense. Some are themselves chiefs (of tributary villages), with okyeame of their own. Some have a hereditary claim to their office; others are appointed, on the basis of their rhetorical skill and knowledge.

No matter who the okyeame is, however, when s/he acts as intermediary, the speaking styles of chiefs and okyeame resemble the styles reported for high and low rank in the Sahel. As Yankah (1995b, p. 109) tells us, “royal speech is referred to as adehye kasa; this is … laced with occasional stuttering. In some realizations, adehye kasa is soft (bòkòò), and has a slight nasal resonance, the likely effect of its calculated low volume. A chief may speak in an undertone, whisper, or speak with moderate intensity.”9 This is when the chief speaks in a formal and public setting, speaking to (and therefore through) an interpreter. Otherwise, and unlike the sahelian chief, the Akan chief is expected to have high oratorical skills in his/her own right. Yet, the difference between sahelian and Akan regional patterns looks less great when we learn that the Akan chief too is thought of as an intermediary, between the human and spirit worlds. In formal settings, he speaks with the voice of the ancestors (Yankah, 1995b, p. 95). Sometimes he speaks at length, and in that case the akyeame may utter little more than an affirming response—“it’s true it’s true it’s true” (Yankah, 1995b, p. 109)—although this token relay is still necessary. Meanwhile, those who wish to address the chief must always use third-person constructions, since it is presumed that they may not speak to him directly, only through the interpreter. Even the interpreters tend to avoid second person pronouns when relaying messages; and in general, in Akan it is polite to use third-person constructions when referring to a person who is present, thus alluding to the use of intermediaries when addressing or referring to chiefs (Yankah, 1995b, p. 129).

In addition to the akyeame’s special oratory, speech in the courts of Akan kings and chiefs employs a register reflecting the honor due to the court: ahemfie kasa ‘palace language.’ Apparently it overlaps with such other registers as mpanin kasa ‘elders’ speech’ and opo/obuo kasa ‘polite speech.’ These ways of speaking are full of euphemisms, metaphors, indirection, apologetic formulas, proverbs, and honorifics—conventional titles, appellations, and formalized expressions about the referent’s greatness, achievements, and benevolence (Agyekum, 2003, 2011). For example, among the conventional honorifics for kings is the expression daasebrε‎, ‘the frequent benefactor; magnanimous.’ Agyekum (2003, p. 377) explains that this expression is constructed from “three words: da … ase, a discontinuous verb meaning ‘to thank someone,’ and the verb brε‎ ‘to be tired.’ Daasebrε‎ therefore means ‘I have thanked him and become tired,’ referring to a benefactor (‘the magnanimous one’) who is so liberal and generous that he makes the beneficiary grow weary of returning thanks.” Some honorific constructions are much less transparent, due to their euphemisms, metaphors, and conventional elisions (Nketia, 1971). Moreover, “palace speech” also includes esoteric forms, such as archaic lexicon (Agyekum, 2011, p. 582).

For other states in this part of West Africa, the literature on language and status has focused mainly on the performance of formal genres of poetry, such as the Yoruba oriki, a type of praise-poetry extensively described by Barber (1991). Barber reports that there was only one professional praise singer left in the town where she worked (in the 1980s). The genre persisted, however, in performances by non- or part-time specialists and for narrower publics. Crucial in the politics of reputation, the oriki can be seen as an extended mode of address, naming an individual and expanding on the name (Apter, 1992; Barber, 1991). Through epithets, metaphors, and condensed expressions, it describes the notable qualities and deeds of the named individual and his/her ancestors and connections; in this it resembles sahelian praise performance, but differs in less emphasis on genealogical particulars. While rural Wolof and Mande public praise singing concentrates heavily on laying out the details of a family tree—attesting to the caste purity and rank of the praised individual—with only brief epithets and comments on other deeds and events, Yoruba oriki tend to reverse that relationship.10 The oriki seem to have more in common with the Akan praise genres.

Apart from formal poetry and praise performance, and the “palace language” of Akan, there is relatively little information in the literature about linguistic indexing of social hierarchy in these states south of the Sahel. Wolff (2000) mentions a differentiation in Yoruba between o and e to signal age or social hierarchy, presumably in everyday discourse, but it is not clear how this distinction works.11 For Akan, there is a bit more detail. Agyekum (2003) offers some examples of Akan honorific constructions (address forms and referential epithets) in everyday contexts, including domestic scenes. For instance, a wife asks her husband to buy a piece of cloth for her mother: “Bε‎mpa, Barima Yε‎-Na, Okumpa, anka w’adaworoma wopε‎ ntomasini bi ma me kOma me maame afe yi a, anka m’ani bε‎gye ho pa ara” (‘Great man, incomparable man, good husband, I would be very pleased if by your grace you could buy a piece of cloth for my mother’). The husband replies: “Me DOfo, Obaapa gye w’ahome. Yei nyε‎ den koraa …” (‘My lover, good lady, relax, this is not a problem at all’.) (Agyekum, 2003, p. 381). Meanwhile, the third-person constructions that imply third parties through whom speech is relayed are politeness forms in Akan even among peers. For example, in “Wo gyafo na ε‎te sε‎n?” (‘Your colleague, how is it?’) and “Wo gyafo ε‎nyε‎ o!” (‘Your colleague, it is not good at all’), the expression “Your colleague” implies that the speaker addresses an intermediary, a colleague of the actual addressee, who will relay the message—even though no third party is present. (Agyekum, 2003, p. 380; Yankah, 1995b).

A related situation has been observed in Mooré (the language of Mossi states in Burkina Faso). There, any utterance in the presence of a high-ranking person (king or chief, usually) must be constructed with honorific elements even when it neither addresses that dignitary nor refers to him. These expressions of bystander honorifics—honorification directed at a bystander, rather than addressee or referent—include, in Mooré, pluralizing all pronouns. The more pronouns one pluralizes, the more honorific is one’s speech; so, while pluralizing only second-person pronouns is “polite,” pluralizing all other pronouns in one’s utterance as well is “honorific” (Lehr, Redden, & Balima, 1966). Although the extra pronoun pluralization seems to be particular to Mooré, the broader phenomenon of plural honorifics recalls the structure of address through an intermediary, a three-party interaction. Notice, incidentally, that this information about social deixis comes from a teaching grammar for the Peace Corps. Academic linguistic works on Mooré say nothing about these constructions, even though ethnographic literature on Mossi emphasizes the importance of deferential bodily conduct.

1.3. Language and Social Hierarchy in Small-Scale Societies

So far, the sociolinguistic systems we have considered have been connected with highly centralized states whose political hierarchies culminate in royal courts, and where social statuses are steeply ranked. But there are much smaller-scale and more egalitarian social systems in West Africa too, especially in the forest region and in hilly areas where the armies of the great states could not easily penetrate. One example would be the Bedik, who live in a few small villages in the mountains between Senegal and Guinea. A relatively egalitarian ethos prevails among adult Bedik men (although some women apparently qualify for similar status), but it pertains to living humans. Relations between the living and the dead, and with other spirit beings, are understood as hierarchy. Bedik have a tradition of masked dancers, where the masks represent the visually accessible aspect of spirit beings (Smith, 1984). As in other West African masking traditions, masked dancers move and vocalize in ways that are markedly distinct from their behavior in everyday life. It is not only the costume that masks the ordinary self; the behavior too is different. Since there are various men who might wear a particular mask, even changing off during the course of an afternoon’s ceremony, the individual identity of the wearer matters little. The behavioral pattern pertains to the mask, not the person inside it. Like the communication structures elsewhere, therefore, hierarchy requires mediation: the mask (and its wearer) mediates between an authoritative spirit world and the world of ordinary human beings.

Bedik spirits have auditory representations too, which sometimes take the form of unusual noises or voice quality, and sometimes distinctive linguistic varieties. For example, the kang∧raŋ mask, attended by a flute player whose melody “announces” it, sings the community’s origin story in a deep, rough voice, which is then “translated” in a clear voice by the village’s spokesman—generally a blacksmith who, in other contexts, will relay the words of a village headman (Ferry, 2003, p. 120).12 Similarly, there is another type of mask that “speaks” in a string of squeaks, also translated by the spokesman. Moreover, there are some spirits represented in visible masks that do not speak at all, and still others that are represented only by sound (cries and a set of four small drums), somewhat like the “voice masks” recorded in Côte d’Ivoire.

These communicative patterns are clearly analogous to those found elsewhere in West Africa. The Bedik organization of speech mediation, with special, “incomprehensible” utterances from an authority figure that are relayed through a person from an artisan category, resembles the mediations found in the traditions of other West African societies, including those with royal courts. The difference lies in how hierarchy is understood—among Bedik it pertains to a spirit world and not to a ranking of living men—and in the role of the artisan, since the Bedik smiths are not an endogamous caste, do not have low rank, and do not play the flutes or large drums. The masks and costumes that are part of the mediation among Bedik are scarce in the Muslim societies of the Sahel, but the Bedik and similarly small-scale societies in the forest region are not the only places where masks are found. There are masking traditions in non-Muslim states too, such as among Yoruba.13 What makes an authoritative utterance “incomprehensible” seems to vary from one type of system to another—from esoteric expressions (Akan) to disfluency and whispering (Sahel),14 and to altered voice quality and music (Bedik)—but the public voice of authority is in all these cases supposed to need “translating” and relaying, if it is to reach ordinary people.

1.4. Linguistic Influences of Islam and Precolonial Empires

So far, the linguistic differentiations we have discussed have involved registers and styles of speaking that have been accommodated, in the literature, within named languages such as Wolof, Akan, or Menik (the language of the Bedik). But there are precolonial multilingualisms to be considered as well. While multilingualism is widespread in West Africa, the inclusion of Arabic in some multilingual repertoires is particularly important in regard to social hierarchy. Arabic came south of the Sahara along with Islam, as a language of religion and of learning, including literacy. Competence in Arabic has long been the defining characteristic of a scholarly elite, some of whom had the economic base to abstain from manual labor in favor of learned pursuits. In Timbuktu, the Sankore mosque (built in the 14th century) became an important center of learning, much like the medieval universities of Europe, and attracting Muslim scholars from Egypt and other distant countries. In addition to such scholars, throughout the Islam-influenced area, there were merchants, scribes, and teachers, with varying degrees of Arabic competence.

This specifically Islamic elite had a complex relationship with those aristocrats whose status was based on other sources—on political and military power, on non-Islamic spiritual resources, or on landholding. In much of the Sahel, Islam became the official religion of a ruling aristocracy, but those rulers did not owe their positions to any great claim to orthodoxy, and their regimes were often criticized as inadequately Islamic. Reformist movements and even jihads came in waves, growing along with resistance to the European slave trade. A major reformist movement in the late 18th century established a theocratic state in the Fouta Toro (northeastern Senegal) and then moved eastward, founding caliphates in Mali and northern Nigeria.

Along with the increasing importance, economic and political, of Islam came influences on the languages of sub-Saharan Africa. Languages across the Sahel acquired loanwords of Arabic origin. Wolof, for example, has a large number of words derived from Arabic, mainly in semantic domains concerned with religion, literacy, learning, and the trans-Saharan trade (such as gelem ‘camel’ or jumaa ‘mosque’). Interestingly, many of these words were adopted into a noun class (the JI class) that is no longer productive of loan constructions, thus indicating that these words must have been absorbed into the Wolof lexicon at an early date. At least as large a number of Arabic words entered Pulaar/Fulfulde, the language of the Tukulor theocratic state in Fouta Toro and of the Tukulor/Fulani jihad that swept into Nigeria in 1804. In northern Nigeria, the jihadist regime—a conquest state but supported by many local populations, including some Fulani already residing in the area—established its domination over the Hausa and set about improving the education, religious and otherwise, of the general population. A system of schools, mostly organized and managed by Nana Asmau, the caliph’s sister, offered literacy and religious instruction, not only through instruction in the Qur’an itself, but also via religious poetry in Fulfulde and Hausa (Robinson, 2004). The poetry, which incorporated many Islamic expressions derived ultimately from Arabic, was explicitly aimed at bringing religious concepts to the Hausa-speaking peasantry (Furniss, 1995, 1996; Hiskett, 1969, p. 295), modifying the Hausa language in the process.

It is clear that by this time (the 18th century), a literary tradition already existed in West Africa and that it included writing in local languages, using a version of Arabic script; some recent scholarship places the beginning of such writing much earlier, at least back to the 16th century (Ngom, 2010).15 This literature is called Ajami, a general term for writing languages other than Arabic in Arabic script. Pulaar/Fulfulde, Hausa, Wolof, Tamasheq, the Mande languages, and Kanuri are among the many languages that have been written in a modified Arabic script. In Senegal, some Pulaar-speakers claim that their religious history, as evidenced in the large quantity of ajami literature in their language as well as their early Islamic contacts and later theocracy, gives them superior status compared to Wolof and other Senegalese peoples.

This understanding of social hierarchy, basing social rank on Islamic learning and piety, is distinct from the caste-based understanding discussed in previous pages. There are some tensions between the two sources of status: disagreements as to whether Muslim clerics of low-caste or slave family background should ever have followers of higher caste than their own (i.e., can religious education and piety trump birth rank?); historical periods of conflict, even warfare, between kings and religious leaders; conflicting versions of piety; and questions about whether low-caste persons are eligible for prolonged Qur’anic schooling (see Wane, 1969; Ware, 2014, p. 176).16 As Ware (2014) has argued, however, it would be a serious mistake—although a mistake made often enough in the literature—to see Islam as necessarily and always an incursion, forever external to authentically African ways of life. On the contrary, while purist waves and appeals to Middle Eastern or Moroccan versions of Islam have occurred at various times in West African history, Islam in one form or another is deeply integrated in African life, at least in the Sahel. Accordingly, some of the linguistic practices associated with the caste-based system, described in previous pages, carry over into Islamic contexts. For example, Muslim clerics in Senegal are likely to have griot spokespersons. McLaughlin and Mboup (2010) present a transcript in which an important cleric relays his address through a griot, who repeats it with amplifications, both in content and in loudness. As another example, the understanding of rank as an embodied quality concerns not only people’s blood (genealogy, birth rank) but also their corporeal practices and the substances they ingest, including the special ink with which Qur’anic verses are written on wooden boards in the Qur’an school.

As I have already suggested, the influence of Arabic language on West African languages sometimes came second-hand, through the influence of an intermediary language into which the Arabic expression had already been incorporated. The many expressions from Arabic that made their way into Hausa came through Fulfulde and the authority of Fulani (Fulfulde speakers) after their conquest of the Hausa states. Those Fulfulde speakers were bilingual in Hausa, without reciprocal bilingualism. That is, it appears that the vectors of linguistic influence were one-way. An elite was multilingual, with competence in Arabic and Hausa as well as Fulfulde; ordinary Hausa-speakers were much less likely to control the other two languages. Over time, there has been some language shift, such that many members of that Fulani elite have shifted to Hausa altogether.

Fulfulde (called Pulaar in its western varieties) has not become a “language of wider communication,” despite its wide geographical distribution and elite status in Nigeria, whereas Hausa has. Lüpke and Storch (2013, p. 259) argue “a good explanation seems to be that language contact between Fula and other languages tended to be of an unbalanced nature, and that the exclusivist language attitudes of the Fulbe never made the interethnic usage of their language attractive.” Asserting that Fulfulde is losing ground to Wolof (in Senegal), the Mande languages (in Mali), and Hausa (in Nigeria and Niger), they even argue that it can be considered endangered (p. 260). Actually, what they seem mainly to have in mind is the specialist vocabulary of cattle-herders and the reduction of morphological complexity in urban usage. In those respects, varieties of the languages to which Fulfulde is losing ground are also endangered.

Perhaps there are many monolingual Hausa-speakers, but multilingualism of one sort or another is more common than not, in West Africa. Its patterns reflect local social relations and hierarchies, however, so the question arises whether the situation in northern Nigeria, in which a bilingual elite shifted toward the language of people they dominated politically, is unusual. Looking at forms of political domination before the European conquest, it is difficult to distinguish language shift from population movements, especially since colonial-era (and later) literatures tended to assume language was always attached to biological populations. Still, there are some relevant cases, such as concern the following three “empires”: Mali, Oyo, and Joloff.

Our first case, the medieval empire of Mali, which dominated the Sahel in the 13th and 14th centuries, seems to have left only a few linguistic traces in non-Mande languages: the forms of some political titles (such as those formed with the suffix –tigi) and some praise epithets. It is true that labels for social categories in the caste system look linguistically related in Mande, Wolof, and some other languages. However, given the possibility (likelihood, in my view) that the main caste divisions predate the empire, any similarity of caste labels is not evidence that Mali imposed them.

The Oyo state, our second case, claimed a wide swath of territory before it fell in the early 19th century. Yoruba, a later term for the language of Oyo’s center, spread through the region. But this effect may owe as much or more to the mid-19th century standardization of Yoruba by Protestant missionaries—a standardization that brought many related (“Yoruboid”) languages together—than to any direct effects of Oyo’s rule.

Last, the Joloff empire (Senegambia, 14th–16th centuries) correlates with Wolof language throughout Joloff’s political sphere, at least for administrative and sometimes for domestic use. Subordinate communities—even those where most of the local population would have spoken Serer or a Cangin language, not Wolof, in the home—spoke Wolof in interaction with outsiders. This usage is evidenced even where a Wolof-using aristocracy did not claim ethnic Wolof ancestry. Wolof remained dominant through the colonial period, thanks in part to its geographical position as the African language first encountered by Europeans. Bilingualism with other sub-Saharan African languages was not reciprocal, then or now; not many ethnic Wolof speak other African languages, whereas the majority of Senegalese, regardless of ethnicity, can speak Wolof.

Evidently these three cases—four, when we include the Fulani conquest of Hausa states—are dissimilar, and their histories bring us into the colonial era.

2. Sociolinguistic Effects of European Colonialism

The most obvious sociolinguistic effects of the European conquest involve use of European languages in political and economic domains, as well as in the educational institutions established in colonial regimes. The European powers that held the largest amounts of West African territory by the end of the nineteenth century were Britain and France. Portugal held a small region (“Portuguese Guinea,” now Guinea-Bissau, plus Cape Verde islands), while Germany ruled a sliver of territory, Togo, until 1918.

The usual generalization about European colonial policies, linguistic and otherwise, is that France favored assimilation of its colonial subjects and ruled them directly, through French language, while Britain followed a principle of indirect rule through indigenous African institutions, including language. The facts are more complicated. France’s ability to carry out an assimilationist policy was limited by a lack of adequate resources and by inconsistent support from the metropole. Consequently, some of its colonies, and some regions within its colonies, were actually ruled more indirectly. And although French-run schools did operate entirely in the French language—African languages were not permitted within school grounds on pain of severe punishment—there were far fewer European-style schools in the French colonies than in the British ones.17 Britain, meanwhile, farmed out most of the task of educating Africans to missionary organizations. Since Protestant missionaries in Africa had already, before the military conquest, begun working toward African language-literacy—so that Africans might read the Bible in their own languages—the British colonial government could simply support the missionaries’ efforts. English was only to be taught later in a pupil’s educational trajectory. It proved impractical to develop educational materials in so many African languages, however, so some schools reverted to English.

Despite any policy differences, all these colonial regimes set up a new linguistic hierarchy, with the European language at the top, followed by the African languages of widest distribution or spoken nearest to the main centers of European administration. Languages that were farther away or had fewer speakers came last. The linguistic hierarchy matched the social hierarchy: Europeans at the top; then their closest African administrative assistants; then Africans further down in an administrative hierarchy; finally the general population.

In the French colonies, the dominance of French language initially favored the métis offspring of Frenchmen and African women. Their French fathers were administrators and merchants who were well positioned to enroll their children in French schools; they also brought the language into the children’s homes. As adults, these children became the assistants, interpreters, and clerks in French enterprises and governing institutions. Some métis became prominent and wealthy (and some authored works on local African languages).18 Moreover, in 19th-century Senegal the interaction with French colonists led to new linguistic usages. “Urban Wolof”—a variety larded with French expressions but maintaining Wolof syntax and morphology—began at that time (McLaughlin, 2008).

Lower down in the colonial hierarchy were the servants of French masters and the soldiers recruited into the colonial armies. Known as tirailleurs sénégalais (‘Senegalese riflemen’) whether or not they were actually Senegalese, these African soldiers seldom knew any French at the time of recruitment. Some language training was set up for them, but they were not expected to acquire great fluency. European prejudices assumed neither ordinary African soldiers nor domestic servants, nor indeed most Africans, could be capable of acquiring so rational and civilized a language as metropolitan French. Linguistically, the meagerness of French language training had two results. It fostered the spread of African languages of wider communication, especially Wolof and Bamana (a Mande language); and it gave rise to a simplified variety of French, français-tirailleur, a sort of pidginized variety reported in the colonial literature from the first half of the 20th century. Another pidginized variety, petit-nègre, resembled français-tirailleur, but was attributed more often to servants in civilian life than to soldiers.

It is not clear how widespread among Africans these pidginized varieties actually were, or how much variation they encompassed. As Van den Avenne (2013) has shown, their most stable manifestation is in novels and memoirs by vieux colons, former colonizers who had returned to France. Van den Avenne argues that the existing documents on these varieties more closely represent the colonizers’ ideologized stereotype of African speech than what Africans themselves may have uttered. For European colonial residents (and ex-residents), these linguistic stereotypes served as emblems of the African experience and offered resources for European solidarity and nostalgia. Meanwhile, French educators produced instructional materials for African primary schools, incorporating African-language lexicon (usually from Bamana) for African material culture into a “basic French” (français fondamental). This Africanized version of simple French was all that most schoolchildren were exposed to (Goheinix, 2011; Van den Avenne, 2015). Since very few children went on to secondary schools, only a tiny elite gained exposure to a more standard variety.

In the British colonies, African-language education drew many more Africans to the western-style schools than was the case in the French colonies, but there were problems surrounding which African languages to promote. African linguistic diversity was too great for literacy materials to be developed for every language a schoolchild might speak. While Africans’ multilingualism meant that many students could acquire literacy in a language other than their own, the selection of languages for instruction (and, later, for radio broadcasting) created new hierarchies among African languages. Literacy education and the production of printed texts also required standardizing whichever languages were chosen—a process rife with pitfalls. Unsurprisingly, Africans whose spoken variety was closest to the print standard had an educational advantage (as is the case elsewhere in the world), thus a social advantage in the new economy. Less obvious is how the colonial-era linguists identified these languages in the first place—what ways of speaking, by what African populations, were grouped together as essentially “the same,” constituting one language to be standardized. This problem was particularly acute for Igbo, identified as the language of a large region of southeastern Nigeria, where local ways of speaking were in fact quite different, and not mutually intelligible. Various efforts at standardizing Igbo ran into trouble, each version failing to prove acceptable to enough of the population to be viable (Irvine, 2015). Faced with unusable educational materials in “Ibo,” teachers and pupils alike resorted to English instead. Igbo competence in English soared. When English-literate Igbo people became prime candidates for positions in the colonial administration and business enterprises all over Nigeria, local people who had only African-language literacy felt disadvantaged. Such economic consequences of linguistic hierarchy contributed to the interethnic conflict that has so often plagued Nigeria, focusing, in the Nigerian civil war of 1967, on Igbo relations with the rest of the country.

In general, colonial language policies and language standardization projects, whether the latter were successful or not (or even put into practice), brought with them some European-based conceptions of language, language’s connection to ethnicity, and what it means to speak one language rather than another one. For literacy, only Roman-based scripts could count; Ajami writings in Arabic script were ignored and their practitioners considered illiterate. These ideas ran counter to older traditions that located social hierarchy in honorifics and register differences, and (in Muslim areas) in Arabic education, rather than in competence in a European standard.

3. Postcolonial Trends

If the colonial era in West Africa created status hierarchies emphasizing competence in European languages, the half-century since independence has mostly taken that tendency even further. The people who came to power at independence were mainly well educated in French or British school and university systems. Moreover, their contacts with one another had often taken place within those systems or in Europe itself, and in French or English. That pattern had begun in colonial-era high schools. Not only was secondary education almost entirely conducted in European languages as the medium of instruction, but the small number of high schools at the time meant that the pupils, already a small elite, were gathered together from different regions and ethnicities. They did not share kinship ties or affiliate with the same home language. So the sense that administrative hierarchy and national integration required European language competence was based, not only in an acquired linguistic ideology, but also in the new leaders’ personal experience. After independence, the colonial language remained the language of all official business and of higher education. Where African languages were taught in schools beyond the primary level, they were taught as a subject, not as the medium of instruction.

In the years after independence, the children of elites had the best chance for post-primary education, and therefore for acquiring effective skills in European languages and the lucrative positions that this background favored. The number of high schools increased, but language policies for education changed very little, especially since published materials in African languages and well-qualified teachers remained relatively scarce. Nigerian linguist Ayo Bamgbose, arguing in 1991 for an increased role for African languages in education and media, noted sourly that retaining the colonial language for these functions—supposedly to enhance national integration—meant that “in effect, the interest of the educated elites, who form a minority in each country, is equated with the interest of the nation” (Bamgbose, 1991, p. 19). By the same token the “ethnic” languages easily become associated with a voting bloc—a political leader’s kin and regional connections—hence African languages become symbols of divisiveness.

A few counter-trends are worth noting, however. One is the increased prestige of Arabic and Arabic-influenced varieties among groups who promote greater Islamic piety and who see this move as explicitly anticolonial and, more broadly, anti-Western. For example, in Senegal some proponents of the Muridiyya sect try to avoid French loanwords, and some Islamic purists try to Arabicize the pronunciation of Arabic loans. Meanwhile, more official business, and more national-level economic activity, is carried out in Wolof rather than in French or English. Another trend is the emergence in Côte d’Ivoire of an interethnic urban language, “Nouchi,” originally a language of male street gangs but recently spreading and trending mainstream (Newell, 2009; Roland Kouassi, personal communication). As an emblem of modernity and urban cool, celebrated in Ivorian rap music, Nouchi represents a new kind of prestige and, to some extent, new status hierarchies.

Nouchi incorporates linguistic elements from various indigenous Ivorian languages along with a lot of Ivorian French (and some English). In drawing on the colonial language without actually speaking it, Nouchi bears some resemblance to Urban Wolof, in which the quantity of French is apparently increasing—although Urban Wolof is still considered Wolof, not something new, and has no role for other African language elements. Another twist on these developments is occurring among Yoruba in Nigeria, where the use of English, and of code-switching into English, have become so frequent in urban settings that some people suggest Yoruba language is now actually endangered (Balogun, 2013; Fabunmi & Salawu, 2005).

Meanwhile, there are still praise singers in West Africa celebrating their patrons. But some of their performances have been “folklorized”—put onstage by culture brokers, recorded, framed as “traditional,” and presented for wider audiences than they would previously have commanded. High-ranking people in the Sahel still do not shout, but they may rely on technology (microphones, tape recordings), rather than speech intermediaries, to mediate their communications with a public. The communications industry still recruits personnel from the social categories that used to fill such roles, although they may use new technology in their jobs and be answerable to international corporations. In these examples, and in others there is not space to mention, we can see that many older patterns persist, even if they have sometimes taken modified form.

4. Critical Analysis of Scholarship

Sociolinguistic scholarship on West Africa is relatively sparse, especially in regard to registers and social deixis. Although there are a few studies in linguistic pragmatics, most sociolinguistic research has concerned multilingualism—repertoires and code switching—and is rooted in European schools of sociology of language, including studies of language policy, rather than in American social dialectology or linguistic anthropology (see, e.g., Bamgbose, 1991; Djité, 2008). Of course there are exceptions, which include some important work by African scholars such as Yankah (see also Calvet, 1994). Another scholarly thread is more literary, studying the forms of oral performance as poetry and epic narrative. This body of work is especially rich for Yoruba, Hausa, and Mande traditions.

Researchers of this topic, language and social hierarchy in West Africa, face several challenges. One is the complexity and subtlety of the linguistic forms, potentially including not only stylistic detail but in some cases also more than one African language. Gaining adequate control of these forms is not a simple matter for outsiders. It requires a lengthy commitment. Another challenge comes from the interdisciplinary nature of the topic, within African studies. For various historical reasons, Africanist linguistics and ethnography have been largely separate fields.

Another challenge concerns regional variation and historical change. To what extent can (or should) one generalize over space and time? For the Sahel, these issues—together with differing theoretical positions—arise in debates over caste: how strictly hierarchical are these systems? Is the term caste applicable at all? If applicable, does it attach only to the artisan categories? And how closely do these systems resemble one another? Broad similarities across sahelian social systems may tend to foster additional, perhaps unwarranted, generalizations. Much depends, then, on which region a scholar takes as model, especially whether the scholar is more familiar with Senegal or with Mali; hierarchy seems to be more pervasive and strict in Senegal, especially among Wolof. And what does hierarchy mean? Comparing Wright (1989) and Irvine (1990)—both on Wolof, but one arguing for a much less hierarchical emphasis than the other—one might ask, are their descriptions really very different, apart from what they understand hierarchy to entail? Can one abjure the term caste, substituting “status group,” or a Mande term, as is argued in the introductory essay in Conrad and Frank (1995), or is it artificial and limiting to avoid the term, regardless of how it might apply in other parts of the world, as Tamari argues (Tamari, 1997)?

Similar issues pertain to scholars’ treatments of history. To what extent can social systems originating in precolonial times be described in the present tense, as if isolating them from the effects of colonial regimes and postcolonial forms of urbanization and globalization? To some extent the “ethnographic present” is justified, since earlier social norms and relations persisted long after the colonial conquest. Many aspects of them are known mainly through ethnographic research, though supported by historical documentation. Yet, life in West Africa has not stood still; Mali in 2016 (for example) is not the same as Mali even as recently as 1990; and it can be a challenge to discern which aspects of social and linguistic practice, and which regions, have changed most.

Current research questions include inquiry into the effects of new technologies and globalization on African linguistic hierarchies, as well as the rise of urban ways of speaking. What, for example, is the relationship between particular urban varieties and that country’s social hierarchies, including counter-normative prestige forms (conveying “toughness,” “cool,” or “street-smarts”)? Is English spreading as a sign of status in West Africa, at the expense of French, and/or at the expense of indigenous languages? What is the role of a globalizing and, as some have called it, de-Africanizing, fundamentalist Islam? What effects on linguistic and social hierarchies can there be from language standardization and literacy education—including Ajami literacy—in African languages? Finally, while some linguists have been trying, with limited resources, to produce basic structural descriptions of endangered minority languages in West Africa, a few scholars have noted the endangerment or loss of registers and structural complexities even in languages with large speaker populations (see Essegbey, Henderson, & McLaughlin, 2015). Accounting for endangerment, or even studying where and how it is (or is not) occurring, demands attention to social hierarchy.

Further Reading

Agyekum, K. (2003). Honorifics and status indexing in Akan communication. Journal of Multilingual and Multicultural Development, 24(5), 369–385.Find this resource:

Bamgbose, A. (1991). Language and the nation: The language question in Sub-Saharan Africa. Edinburgh: Edinburgh University Press.Find this resource:

Barber, K. (1991). I could speak until tomorrow: Oriki, women and the past in a Yoruba town. Washington, DC: Smithsonian Institution.Find this resource:

Calvet, L-J. (1994). Les voix de la ville: Introduction à la sociolinguistique urbaine. Paris: Payot.Find this resource:

Camara, S. (1976). Gens de la parole: Essai sur la condition et le rôle des griots dans la société Malinké. Paris: Mouton.Find this resource:

Conrad, D. C., & Frank, B. E. (Eds.). (1995). Status and identity in West Africa: Nyamakalaw of Mande. Bloomington: Indiana University Press.Find this resource:

Diop, A.-B. (1981). La société wolof: Tradition et changement. Les systèmes d’inégalité et domination. Paris: Karthala.Find this resource:

Essegbey, J., Henderson, B., & McLaughlin, F. (Eds.). (2015). Language documentation and endangerment in Africa. Amsterdam: John Benjamins.Find this resource:

Ferry, M.-P. (2003). Masques, initiation et fêtes des femmes chez les Bedik du Sénégal oriental. Journal des Africanistes, 73(1), 111–126.Find this resource:

Hale, T. A. (1998). Griots and Griottes. Bloomington: Indiana University Press.Find this resource:

Hoffman, B. (2000). Griots at war: Conflict, conciliation, and caste in Mande. Bloomington: Indiana University Press.Find this resource:

Irvine, J. T. (1990). Registering affect: Heteroglossia in the linguistic expression of emotion. In C. Lutz & L. Abu-Lughod (Eds.), Language and the politics of emotion (pp. 126–161). Cambridge, U.K.: Cambridge University Press.Find this resource:

Irvine, J. T. (1995). Honorifics. In J. Blommaert, J. Verschueren, & J.-O. Östman (Eds.), Handbook of pragmatics (pp. 1–22). Amsterdam: John Benjamins.Find this resource:

Irvine, J. T. (2008). Subjected words: African linguistics and the colonial encounter. Language and Communication 28, 323–343.Find this resource:

Lüpke, F., & Storch, A. (2013). Repertoires and choices in African languages. Berlin: De Gruyter.Find this resource:

McLaughlin, F. (Ed.). (2009). The languages of Urban Africa. London: Continuum.Find this resource:

McLaughlin, F., & Mboup, B. (2010). Mediation and the performance of religious authority in Senegal. Islamic Africa 1, 39–61.Find this resource:

Newell, S. (2009). Enregistering modernity, bluffing criminality: How Nouchi speech reinvented (and fractured) the nation. Journal of Linguistic Anthropology 19(2), 157–184.Find this resource:

Nketia, J. H. K. (1971). The linguistic aspect of style in African languages. In T. Sebeok (Ed.), Current trends in linguistics (Vol. 7, pp. 733–757). The Hague: Mouton.Find this resource:

Smith, M. G. (1957). The social functions and meaning of Hausa praise-singing. Africa 27, 26–43.Find this resource:

Tamari, T. (1997). Les castes de l’Afrique occidentale: Artisans et musiciens endogames. Nanterre, France: Société d’Ethnologie.Find this resource:

Wane, Y. (1969). Les Toucouleur du Fouta Toro (Sénégal): Stratification sociale et structure familiale. Dakar, Senegal: Institut Fondamental d’Afrique Noire.Find this resource:

Wright, B. (1989). The power of articulation. In W. Arens & I. Karp (Eds.), Creativity of power: Cosmology and action in African societies (pp. 39–58). Washington, DC: Smithsonian Institution.Find this resource:

Yankah, K. (1995b) Speaking for the chief: Okyeame and the politics of Akan royal oratory. Bloomington: Indiana University Press.Find this resource:


Agyekum, K. (2003). Honorifics and status indexing in Akan communication. Journal of Multilingual and Multicultural Development, 24(5), 369–385.Find this resource:

Agyekum, K. (2011). The ethnopragmatics of the Akan palace language of Ghana. Journal of Anthropological Research, 67(4), 573–593.Find this resource:

Apter, A. (1992). Black critics and kings: The hermeneutics of power in Yoruba society. Chicago: University of Chicago Press.Find this resource:

Austen, R. (Ed.). (1999). In search of Sunjata: The Mande oral epic as history, literature, and performance. Bloomington: Indiana University Press.Find this resource:

Balogun, T. A. (2013). An endangered Nigerian indigenous language: The case of Yoruba language. African Nebula, 6, 70–83.Find this resource:

Bamgbose, A. (1991). Language and the nation: The language question in Sub-Saharan Africa. Edinburgh: Edinburgh University Press.Find this resource:

Barber, K. (1991). I could speak until tomorrow: Oriki, women, and the past in a Yoruba Town. Washington, DC: Smithsonian Institution.Find this resource:

Burkhalter, S. (1992). Listening for silences in Almoravid history: Another reading of “The conquest that never was.” History in Africa, 19, 103–131.Find this resource:

Calvet, L-J. (1994). Les voix de la ville: Introduction à la sociolinguistique urbaine. Paris: Payot.Find this resource:

Camara, S. (1976). Gens de la parole: Essai sur la condition et le rôle des griots dans la société Malinké. Paris: Mouton.Find this resource:

Conrad, D. C., & Fisher, H. J. (1982). The conquest that never was: Ghana and the Almoravids, 1076. I: The external Arabic sources. History in Africa, 9, 21–59.Find this resource:

Conrad, D. C., & Fisher, H. J. (1983). The conquest that never was: Ghana and the Almoravids, 1076. II: The local oral sources. History in Africa, 10, 53–78.Find this resource:

Conrad, D. C., & Frank, B. E. (Eds.). (1995). Status and identity in West Africa: Nyamakalaw of Mande. Bloomington: Indiana University Press.Find this resource:

Dilley, R. M. (2004). Islamic and caste knowledge practices among Haalpulaar’en in Senegal. London: Edinburgh University Press.Find this resource:

Diop, A.-B. (1981). La société wolof: Tradition et changement. Les systèmes d’inégalité et domination. Paris: Karthala.Find this resource:

Diouf, M. (2001). Histoire du Sénégal: Le modèle islamo-wolof et ses périphéries. Paris: Maisonnneuve et Larose.Find this resource:

Djité, P. (2008). The sociolinguistics of development in Africa. Clevedon, U.K.: Multilingual Matters.Find this resource:

Essegbey, J., Henderson, B., & McLaughlin, F. (Eds.). (2015). Language documentation and endangerment in Africa. Amsterdam: John Benjamins.Find this resource:

Fabunmi, F. A., & Salawu, A. S. (2005). Is Yoruba an endangered language? Nordic Journal of African Studies 14(3), 391–408.Find this resource:

Fernandes, V. (1940). O Manuscrito “Valentim Fernandes.” Joaquim Bensaúde (Ed.), Lisbon: Academia Portuguesa da História. Originally published in 1509.Find this resource:

Ferry, M.-P. (2003). Masques, initiation et fêtes des femmes chez les Bedik du Sénégal oriental. Journal des Africanistes, 73(1), 111–126.Find this resource:

Furniss, G. (1995). The power of words and the relation between Hausa genres. In G. Furniss & L. Gunner (Eds.), Power, marginality and African oral literature (pp. 130–144). Cambridge, U.K.: Cambridge University Press.Find this resource:

Furniss, G. (1996). Poetry, prose and popular culture in Hausa. London: Edinburgh University Press.Find this resource:

Goheinix-Minisini, A. (2011). Le français colonial: Politiques et pratiques de la langue nationale dans l’Empire (1880–1962) (PhD diss., Institut d’Études Politiques de Paris, École Doctorale de Scicnces Politiques, Paris).Find this resource:

Gutelius, D. (2002). Newly discovered 10th/16th c. Ajami manuscript in Niger and Kel Tamagheg history. Saharan Studies Newsletter, 8, 1–2.Find this resource:

Hale, T. A. (1998). Griots and Griottes. Bloomington: Indiana University Press.Find this resource:

Hiskett, M. (1969). Northern Nigeria. In J. Kritzeck & W. H. Lewis (Eds.), Islam in Africa (pp. 287–300). New York: Van Nostrand-Reinhold.Find this resource:

Hoffman, B. (2000). Griots at war: Conflict, conciliation, and caste in Mande. Bloomington: Indiana University Press.Find this resource:

Ibn Battuta (1922). Voyages de ’Ibn Batoutah. Texte arabe accompagné d’une traduction par C. Defrémy et le Dr B. R. Sanguinetti (Vol. 4). Paris: Imprimerie Nationale. Originally published in 1355.Find this resource:

Innes, G. (Ed. & Trans.). (1974). Sunjata: Three Mandinka versions. London: School of Oriental and African Studies.Find this resource:

Irvine, J. T. (1978). When is genealogy history? Wolof genealogies in comparative perspective. American Ethnologist, 5, 651–674.Find this resource:

Irvine, J. T. (1990). Registering affect: Heteroglossia in the linguistic expression of emotion. In C. Lutz & L. Abu-Lughod (Eds.), Language and the politics of emotion (pp. 126–161). Cambridge, U.K.: Cambridge University Press.Find this resource:

Irvine, J. T. (2015). Language as cultural “heritage”: Visions of ethnicity in nineteenth-century African linguistics. In D. Peterson, K. Gavua, & C. Rassool (Eds.), The politics of heritage in Africa (pp. 191–208). Cambridge, U.K.: Cambridge University Press for the International African Institute.Find this resource:

Johnson, J. W. (1999). The dichotomy of power and authority in Mande society and in the epic of Sunjata. In R. A. Austen (Ed.), In search of Sunjata: The Mande oral epic as history, literature, and performance (pp. 9–23). Bloomington: Indiana University Press.Find this resource:

Kropp Dakubu, M.-E. (2015). The role of language in forging new identities: Countering a heritage of servitude. In D. Peterson, K. Gavua, & C. Rassool (Eds.), The politics of heritage in Africa (pp. 209–221). Cambridge, U.K.: Cambridge University Press.Find this resource:

Lehr, M., Redden, J., & Balima, A. (1966). Moré: Basic Course. Washington, DC: Foreign Service Institute.Find this resource:

Lüpke, F., & Storch, A. (2013). Repertoires and choices in African languages. Berlin: De Gruyter.Find this resource:

McLaughlin, F. (2008). On the origins of urban Wolof: Evidence from Louis Descemet’s 1864 phrase book. Language in Society, 37(5), 713–735.Find this resource:

McLaughlin, F., & Mboup, B. (2010). Mediation and the performance of religious authority in Senegal. Islamic Africa, 1, 39–61.Find this resource:

Newell, S. (2009). Enregistering modernity, bluffing criminality: How Nouchi speech reinvented (and fractured) the nation. Journal of Linguistic Anthropology, 19(2), 157–184.Find this resource:

Ngom, F. (2010). Ajami scripts in the Senegalese speech community. Journal of Arabic and Islamic Studies, 10, 1–23.Find this resource:

Nketia, J. H. K. (1971). The linguistic aspect of style in African languages. In T. Sebeok (Ed.), Current Trends in Linguistics (Vol. 7, pp. 733–757). The Hague: Mouton.Find this resource:

Podstavsky, S. (2004). Hausa entertainers and their social status: A reconsideration of sociohistorical evidence. Ethnomusicology, 48(3), 348–377.Find this resource:

Robinson, D. (2004). Muslim societies in African history. Cambridge, U.K.: Cambridge University Press.Find this resource:

Smith, M. G. (1957). The social functions and meaning of Hausa praise-singing. Africa, 27, 26–43.Find this resource:

Smith, P. (1984). Le “mystère” et ses masques chez les Bedik. L’Homme, 24(3–4), 5–33.Find this resource:

Tamari, T. (1997). Les castes de l’Afrique occidentale: Artisans et musiciens endogames. Nanterre, France: Société d’Ethnologie.Find this resource:

Van den Avenne, C. (2013). Reprise et détournement d’un stéréotype linguistique: Les enjeux coloniaux et postcoloniaux de l’usage du “petit-nègre” dans la littérature africaine. In L. Gauvin, C. Van den Avenne, V. Corinus, & C. Selao (Eds.), Parodies, pastiches, réécritures: La question des modèles dans les littératures francophones. Lyon, France: ENS éditions.Find this resource:

Van den Avenne, C. (2015). “De la bouche même des indigènes”: Les modalités linguistiques de la rencontre colonial en Afrique de l’Ouest. Unpublished manuscript presented for Habilitation certification. Aix-en-Provence, France: Université Aix-Marseille.Find this resource:

Wane, Y. (1969). Les Toucouleur du Fouta Toro (Sénégal): Stratification sociale et structure familiale. Dakar, Senegal: Institut Fondamental d’Afrique Noire.Find this resource:

Ware, R. (2014). The walking Qur’an: Islamic education, embodied knowledge, and history in West Africa. Chapel Hill: University of North Carolina Press.Find this resource:

Wolff, H. E. (2000). Language and society. In B. Heine & D. Nurse (Eds.), African languages: An introduction (pp. 298–347). Cambridge, U.K.: Cambridge University Press.Find this resource:

Wright, B. (1989). The power of articulation. In W. Arens & I. Karp (Eds.), Creativity of power: Cosmology and action in African societies (pp. 39–58). Washington, DC: Smithsonian Institution.Find this resource:

Yankah, K. (1995a). Power and the circuit of formal talk. In G. Furniss & L. Gunner (Eds.), Power, marginality and African oral literature (pp. 211–224). Cambridge, U.K.: Cambridge University Press.Find this resource:

Yankah, K. (1995b). Speaking for the chief: Okyeame and the politics of Akan royal oratory. Bloomington: Indiana University Press.Find this resource:


(1.) The medieval Empire of Ghana was located in present-day Senegambia, Mali, and southern Mauritania. Modern Ghana, named for the empire, is far to the south.

(2.) The origin of this term is not clear. It became current in early colonial-era French, perhaps deriving from Wolof gewel.

(3.) Although “slave” is not a legal status in Senegal or other sahelian countries today, slave ancestry still confers a stigma and barrier to many opportunities. Moreover, some persons with slave ancestry continue the traditional relationship with their “masters,” especially in rural areas. For an interesting Ghanaian case involving slave ancestry and language, see Kropp Dakubu (2015).

(4.) Noble Wolof genealogies tend to be deep. Some genealogies I recorded in a rural setting, in 1971, traced ancestry back to the 16th century. See Irvine (1978).

(5.) These principles also apply to griots’ epic narratives (often punctuated by praise singing), of which the Mande epic of Sunjata is especially famous (see, e.g., Austen, 1999; Innes, 1974). I have not focused on narrative style in this article. It shares some features of the “griot talk” register but not others.

(6.) Unlike the Portuguese sources, the Arabic sources do not emphasize the segregation or low status of griots. Hale (1998) attributes the difference to Portuguese ideas about Jews, to whom they equated the African griots. Possibly, however, the difference is partly due to the lower rank of griots among Wolof—whom the Portuguese encountered first—compared with their rank in Mali.

(7.) On Hausa poetry, oral and written, see Furniss (1996) and other works.

(8.) Hoffman (2000) includes some discussion of jeli-kan, the griot way of speaking.

(9.) In contrast to the high-rank style in the Sahel, which is typically slow, this Akan royal style is “typically hurried” (Yankah, 1995b, p. 109).

(10.) In emphasizing the genealogical focus of Wolof and Mande public praise singing, I exclude narrative genres, which are performed differently (and, for Wolof at least, less publicly).

(11.) Wolff cites a personal communication from Ayo Bamgbose.

(12.) Bedik do not recognize political authorities in the usual sense of a village chief. As Smith (1984) explains, their “headmen” have only a moral authority and specific ritual roles.

(13.) Today, many Yoruba people—perhaps half the population—have converted to Islam.

(14.) The other Sahelian stylistic pole, extreme rapidity and elision, can be incomprehensible too (Irvine, 1990).

(15.) For this earliest date, Ngom cites David Gutelius (2002) on the discovery of a 16th-century Ajami manuscript on Niger and Kel Tamagheg history.

(16.) See Ware (2014, p. 176), and Wane (1969).

(17.) A brief experiment in African-language education, established in Senegal in 1817, lasted only a few years before being denounced in favor of a French-only policy.

(18.) Among the métis authors of linguistic works were David Boilat, a Senegalese priest educated in France, and Louis Descemet, sometime mayor of Saint-Louis.