Linguistic and Civic Refinement in the N’ko Movement of Manding-Speaking West Africa

Abstract

“Register” has become an essential tool in analyzing languages as sociocultural artifacts. Used in tandem with the concept of language ideology, scholars have elucidated the central role of linguistic work in defining African language and dialect boundaries as we know them today. The role of such ideas in current activist efforts to remake languages and society, however, remains obscure. Here, I focus on the N’ko movement of West Africa, which promotes a non-Latin-, non-Arabic script invented in 1949 for mother-tongue education. Today, through a language register known as kángbɛ ‘clear language’, N’ko activists are altering conceptions of Manding varieties as distinct entities into a single language spoken by tens of millions across West Africa. Such a shift is in part made possible by the compelling sociohistorical linguistic analysis laid out pedagogically in N’ko grammar books and classrooms. Equally important, however, is kángbɛ as a means to discursively cultivate oneself into a new kind of citizen; one that is savvy, hard-working, and just—the opposite of West African elites, who are seen as failing their people. Register is therefore not just an analytic tool but also a resource for cultivating empowering language ideologies to forge new educational opportunities and societal possibilities.

During the summer of 2012, I encountered the writing system N’ko in Burkina Faso for the first time. As a Peace Corps volunteer in-country between 2009 and 2011, I was dedicated to learning Manding, known locally as Jula, or jùlakán ‘trader’s language’ in the language itself.¹ At the same time, I was tied to its promotion; one of my major activities as part of a rural school district’s team was to offer trainings for local associations that had completed a cycle of adult Manding literacy classes. I rarely encountered people reading or writing in the language outside the confines of these subsidized trainings that were often coveted for their per diem stipends. As such, a year after my departure from Burkina, I was pleasantly surprised to discover people doing just that, even if it was in an alphabet that I had only recently been introduced to while taking classes as a Fulbright scholar in France.

N’ko was invented in 1949 by the Guinean “peasant intellectual” Sùlemáana Kántɛ (Feierman 1990).² A non-Latin-, non-Arabic-based writing system designed primarily for Manding, N’ko is written right to left and is a perfect phonological analysis of Kantè’s native Manding variety, with a set of diacritics for marking tone (Vydrin 2001b, 128–29). Trained in the Quranic school of his father, Kantè used his script to write over 100 books spanning linguistics, history, traditional medicine, and Islam—including a translation of the Quran (Kántɛ n.d.)—that continue to be typeset, published, and taught by N’koïsants today.³

In 2012, I met the N’ko teacher Sáliya Tárawele in Burkina Faso’s second city of Bobo-Dioulasso. Over the course of a few intermittent weeks, I participated in a number of his regular lessons that he led after sundown in a dusty classroom at private Catholic high school. One day—surely in response to my interest in his book collection—he provided me with a small piece of notebook paper, which was covered in his own handwritten words. The document came from a radio segment that he had prepared and it was mine to keep, he said. Scrawled across the paper was the following:⁴

Latent within this seemingly banal position about orthographic standards are two intertwined sociological interventions. First, Sáliya uses the term N’ko to refer to not an alphabet but rather a “language” (kán). This usage—N’ko as language—stems from Sulemaana Kantè; he applied the same name to both his alphabet and the language-dialect continuum that Western-trained linguists today refer to as Manding. In embracing this referential pairing, Sáliya recognizes a language that subsumes that which his government and fellow citizens recognize as Jula. He does not, however, imply that “N’ko” or Manding is a single, uniform linguistic code; just like professional linguists, he acknowledges that it has various forms, which, for him, result from a desire to facilitate mutual intelligibility. His second intervention instead is to claim that the language has rules that must be dutifully used and applied when writing. This register of Manding—unnamed, but tacit in his radio remarks, and which he strives to write and teach—is known as kángbɛ ‘clear language’.

N’ko, therefore, is not simply another way of naming the Manding continuum, nor is kángbɛ just another variety under the Manding umbrella. Together, they are ideas behind a transnational grassroots standardization project that aims to establish a single (primarily, but not exclusively, written) linguistic norm and thereby alter common conceptions of Manding varieties as distinct entities into a single language spoken by tens of millions across the subregion (Vydrin 2011). While the majority of speakers have not yet followed suit, thousands of people across West Africa today recognize and embrace these practices of Sáliya and N’ko’s founder Sulemaana Kantè. In doing so, they take part in a social movement that proposes models of personhood and ways of orienting to one’s fellow speakers that together serve as a means of resisting the region’s colonial past and reshaping its neocolonial present.

In what follows, I draw on approximately six months of transnational fieldwork conducted between 2012 and 2017 to take us into the writings, classrooms, and bookshops of N’ko students to offer an account of the spread of this metalinguistic framework, which, through the standard language register of kángbɛ, unites formerly disparate dialects under the banner of N’ko. First, I look at how both the script and register are linguistically compelling in the classroom for Manding speakers of diverse dialectal backgrounds. Second, I turn to the ways in which teachers’ talk about talk and wider discourse tie the learning and use of N’ko and its standard language register to their self-fashioning as “savvy, disciplined, and just citizens,” as enshrined in the common N’ko slogan kólɔn, báara, télen. Connecting such discourse with wider complaints about African postcolonial governments and society, I argue that N’ko’s kángbɛ register is compelling as a discursive means by which its students can shape themselves into the kinds of citizens that they believe their countries and continent currently lack but desperately need.

Conceptual Framework

From a linguistic perspective, Manding⁵ is a language-dialect continuum stretching across West Africa from Senegal to Burkina Faso (see fig. 1; Vydrine 1995–96). The word Manding is a Western adaptation of the word Màndén, the name of both a place and former West African polity, commonly referred to as the Mali Empire, that at its apogee encompassed much of modern-day Guinea and Mali, primarily between the thirteenth and fifteenth centuries (Levtzion 1973; Kántɛ 2008; Simonis 2010, 41–54).⁶

Figure 1. Manding language continuum (author’s illustration, using data from Vydrin, Bergman, and Benjamin 2001).

On the ground, speakers primarily label their speech with a range of distinct proper names (e.g., Maninka in Guinea, Mandinka in the Gambia, Bamanan in Mali, and Jula in Côte d’Ivoire and Burkina Faso), which are variably glossed in Western languages (see fig. 2).⁷ Nonetheless, mutual intelligibility is widely noted, in particular, between Maninka, Bamanan, and Jula (Dumestre and Retord 1981, 3).⁸ Despite both this and linguists’ clear acknowledgment of their connectedness and overlap (e.g., Dumestre 2003; Creissels 2009), national language policies and linguistic work typically treat Manding varieties largely as distinct, albeit related, varieties or even languages (Calvet 1987; see table 1).

Figure 2. Màhamúud Sánkare leads a lesson in N.Fa.Ya’s Wànkáran’ Kàrantá (photograph taken by the author).

Table 1. Major Manding Varieties by Local Name, Etymology, and Foreign-Language Designations

In such a situation, it is hardly surprising that both Sulemaana Kantè and Western academic linguists developed a single hypernym to refer to a range of interconnected and most often mutually intelligible phono-lexical grammatical systems: N’ko for the former, and Manding for the latter. For linguists, Manding is a convenience meant to gesture toward lexical and grammatical congruence of what they understand as freestanding varieties. The ambitions of N’ko’s inventor were far greater.

N’ko

Beyond a script and language, N’ko more broadly denotes a transnational social movement based on Manding-medium literacy and education. Following its invention in 1949, the script has continually spread from its original base in the highlands of Guinea via the historical networks of Manding-speaking Muslim traders and Quranic schools (Oyler 1995, 2005). In recent decades there has been increasing efforts by N’koïsants to move into the formal schooling sector (Wyrod 2003, 2008). Research and commerce related to traditional medicine has also been important vector of the movement’s spread (d’Avignon 2012; Hellweg 2013).

Western scholars have additionally highlighted the movement’s tendency to invoke the historical grandeur of the Màndén empire to promote Manding literacy in N’ko as part of a larger struggle to decolonize the francophone state and Arab-dominated Islam (e.g., Conrad 2001; Amselle 2003). None of the scholarship laid out above, however, ethnographically investigates one of the core features identified by Vydrin (2011), namely, the creation and dissemination of a standard language register that transcends dialectal variation.

Theoretical Framework

My research draws from a critical realist’s approach to language (Cameron et al. 1992; Corson 1997) and linguistic anthropological understandings of “the total linguistic fact” (Silverstein 1985; Wortham 2008) as elucidated through the notions of reflexivity (Lucy 1993) and register (Agha 2007a).

While acknowledging that language, as we know it, is in fact a social phenomenon, “departmentalized linguistics” (Agha 2007b) approaches the study of language as a study of an abstract system (French langue) detached from its use in the real world (French parole). Linguistics then necessarily delineates and studies idealized, pure forms of language that do not in fact conform to the “ways of speaking” of actual people (Hymes 1974). While this is arguably a valid approach for scientists interested in the cognitive side of language structure or creating grammars, it is of little use to those seeking to study language as it is actually used (Cameron et al. 1992). For languages, without a history of top-down standardization, this is especially true; the speech practices and perceptions of Manding speakers, for instance, rarely correspond with the distinct varieties proposed by linguists (Canut 1996, 2001; Donaldson 2016). My approach to language is therefore undergirded by the philosophical paradigm of “critical realism” (Corson 1997)—which combines ontological realism with epistemological constructivism (Maxwell 2012)— in light of the fact that “language is only ever produced or interpreted in a social context” (Cameron et al. 1992, 12).

My study is thus guided by the concept of “the total linguistic fact” (Silverstein 1985; Wortham 2008) stemming from the Boasian linguistic tradition (Boas 1911; Agha 2007b)—which calls for attending to form, use, ideology, and domain when assessing the meaning of any utterance. Form in this sense denotes the lexemes and grammar of language. Use captures the way that forms are often used in unexpected ways to create emergent meaning that befuddles any rule-based account of grammar or pragmatics (e.g., Searle 1975). Ideology and domain account for the fact that no matter how well one dissects the interaction at hand, one cannot ascertain the meaning of an utterance without also appealing to larger circulating models that are known to certain domains or segments of people. These “models of linguistic signs and the people who characteristically use them” (Wortham 2008, 40) are language ideologies (Silverstein 1979; Woolard 1998; Jaffe 1999; Kroskrity 2000).

While ideology conjures up the image of something existing in the head, I instead approach it through language’s fundamentally “reflexive” character (Lucy 1993; Agha 2007a), whereby people continually “remark on language, report utterances, index and describe aspects of the speech event, invoke conventional names, and guide listeners in the proper interpretation of their utterances” (Lucy 1993, 11). Language in this sense always has an inherently metalinguistic character. Whether overt or tacit, every interaction with language over a lifetime provides commentary that determines the stereotypical social values of forms and their uses. Language ideologies are therefore models mediating between the use of language and the social world that are empirically traceable through explicit “habits of evaluation” (Agha 2007a, 17) and implicit patterns of use, which individuals read as metapragmatic commentary.

Agha’s (2007a) notion of “register” inherently links form and use with the reflexive models (i.e., language ideologies) that give speech its social meaning and value. In folk terms and some traditional formulations, a dialect inherently points to the regional provenance of a person, while a register is understood as a situational deviance from a core of denotational forms (Halliday 1964, 1978; Biber and Finegan 1993). Agha’s sense subsumes the two concepts under a single sociologically relevant conceptualization of patterned “fashions of speaking” (Whorf 1956). Registers are not simply different ways of saying the same thing. They are rather “cultural models of action” identifiable by a repertoire (i.e., linguistic features), range (enactable pragmatic values) and domain (a set of users; see Agha 2007a, 55). This article uses the concept of register to advance a social theory of language that aims to account for “how particular systems of speech valorization come into existence in the first place” (Agha 2007a, 15–16). As such, I focus on linguistic forms but primarily in service of investigating how social processes and linguistic grammar are mutually intertwined.

Kángbɛ Grammar

Given Manding’s documented variation, how could Kantè’s N’ko, as an orthography, be all of the language at once? Amselle (1996, 825) suggests that through their so-called cultural fundamentalism, N’ko students aggressively take only the Guinean variety known as Maninka to be correct in spelling and pronunciation. Indeed, the forms metadiscursively prescribed in N’ko documents show evidence of being primarily congruent with Maninka (and more precisely the so-called màninkamóri variety of Kankan; see Vydrin [1996, 2010]; Davydov [2008, 2012]). But Kantè did not clumsily claim that only Maninka was appropriate for writing Manding; he sought to call into being a register that, through his pedagogical language works, would act as a mediating standard between the dialects.

Acquiring N’ko-based literacy typically proceeds linearly.⁹ One of the most important subject matters is N’ko grammar, or what Kantè (2008b, 4–5) terms kángbɛ:

Here, Kantè is clearly developing a technical usage in which kángbɛ is best glossed as ‘grammar (book)’ in the prescriptive sense of the schoolmarm. More broadly, the term refers to the prescribed forms of language—that is, the register—found within Kantè’s grammar books. Etymologically, kángbɛ is a tonally compact compound noun made up of the noun kán ‘language’ and the polysemous adjective gbέ, which can variably be glossed as ‘white’, ‘clean’, ‘clear’ (Bailleul 2007). Through its contributing lexemes, therefore, kángbɛ is naturalized as something that serves to clarify and order language.

This perspective notwithstanding, for N’ko’s founder, the Manding language could never be reduced to a single isolatable phono-lexical grammatical code that a linguist elicits from an informant. For while he relies heavily on the idea that a language has a true or correct form that should be promoted for writing, he also embraced Manding as inevitably composed of distinct registers as made clear in his works in dialectology, language history, and lexicography (Kántɛ 1992, 2007, 2008a, 2008b, 2009). Indeed, the term kángbɛ does not seem to have been chosen randomly; it figures prominently in the preeminent French colonial dictionary and grammar: “These more or less localized dialects aside, a sort of “common Manding” has formed that the indigenous have given the name kangbe (white language, clear language, easy language) and which is understood and spoken by the great majority of the population in addition to the special dialect of each region” (Delafosse 1929, 1:22).¹⁰

Kantè’s selection then of the compound noun kángbɛ serves to tie his prescriptive grammar and its standard language register to an already circulating historically named lingua franca register. What counts then as kángbɛ in N’ko circles today may be largely congruent with a particular Manding dialect, but it is nowhere near a màninkamóri orthography. It is rather the basis for a written standard language register that Kantè sought to bring into being to hold together the named Manding varieties of Maninka, Bamanan, Jula, and Mandinka under a single baptismal hypernym: N’ko.

“We are going for the language, in its pure form”

The above interpretation of Kantè’s oeuvre shines in the N’ko circles of today. In what follows, I explore how this formulation of N’ko as a single language united by kángbɛ circulates among students and activists. Focusing on salient metadiscourse, I investigate two distinct dynamics fueling the N’ko movement and its kángbɛ register. First, it can be attributed, in part, to the compelling sociohistorical linguistic analysis of Manding phonemes and lexemes that is at the heart of the prescriptive grammar register’s teaching. What makes the metalinguistic framework of Sulemaana Kantè compelling, however, are not simply facts of semantics or etymology. Second, I elucidate kángbɛ’s role as the discursive component of an ethos of discipline, logic, and savviness through which students believe they can hone themselves into the kinds of people that so many of them feel their society is desperately lacking.

Learning Letters, Learning Kángbɛ

In practice, it is often difficult to separate the learning of N’ko as a script from that of learning the proper way to write and potentially speak Manding writ large. In the N’ko classroom, adult students right off the bat are given tools of “metalinguistic awareness” (Cummins 1978; Nagy and Anderson 1995). Their education however is not one of being shown how to perform structural linguistic analysis on their own speech. Instead, N’ko lessons introduce them to a metalinguistic framework—diachronically informed—that socializes them into ways of interpreting Manding sounds, sequences, and patterns as dialectal, kángbɛ, or foreign.

In the summer of 2016 I sat in on a class led by Sékù Jàkité, which took place in the morning, twice a week, beneath a tin-roof hangar, seemingly airdropped amidst a flood of outdoor mechanics’ workshops and vehicle carcasses. Poised in front of a long blackboard with a black Robin Hood–style hat, Sékù opened the lesson with the traditional penning of the date using the unique set of day and month names by which Kantè sought to replace the Arabic and French loanwords that one typically hears in Manding today. This, however, gave the students little pause. Instead, Sékù lectured at length about the various phonemes of Manding. According to him, all of the necessary Manding sound categories are captured in the letters of N’ko. This did not mean that speakers of Manding do not make or use other sounds. He picked out /v/ and /z/, two sounds stemming in large part from French loanwords. Vydrin (2016a, 11) analyzes them in Bamanan as follows:

z is a phoneme borrowed from French; French /ʒ/ > Bamanan /z/. zùlùyé ‘July’ [< juillet], zańdármú ‘police officer’ [< gendarme]. In addition, z optionally appears as a variant of ns: zòn ~ nsòn ‘thief’, nsíirin ~ zíirin ~ nzíirin ‘tale’

v is an extremely marginal phoneme that only appears in non-adapted borrowings: vέrì ‘glass’ [< verre], vítri ‘pane’ [< vitre], etc.

Sékù was more blunt regarding the two phonemes, though he spoke in terms of letters: “Our language doesn’t need them” (Án ná kán’ màkó’ tέ ù lá). Nonetheless, given that “we” might occasionally want them for writing other languages, he introduced the N’ko convention of adding superposed dots to consonants and vowels to represent the sounds or letters of other languages (e.g., ߝ߭ߍߙߌ vέri).

While /v/ and /z/ are clearly marginal phonemes emerging from French, Sékù also addressed the case of a nascent Bamanan phoneme, /ʃ/ that likely emerges not from a foreign source but rather from an in-progress phonemic split. Today, one can identify a number of minimal pairs between /s/ and /ʃ/ in Bamanan, but there are also cases of [ʃ] that are contextual realizations of /s/ (Vydrin 2016a, 11). Sékù provided clear instructions regarding the emergent sound: “This isn’t in N’ko” (Nìn tέ Ń’ko lá). While seemingly harsh, such a statement usefully demonstrates how the very learning of N’ko is a first step both in introducing students to etymology and sound change and in opening the door to a disciplining of their written language into kángbɛ. To be clear, Sékù’s statement did not focus on rooting out the pronunciation but instead on introducing the written standard. Neither he nor other N’ko students, for instance, reject the Bamanan forms in example 1. Instead, they recognize them as dialectal deviations (1a, 1b) or loanwords (1c) that one should not attempt to represent directly in writing:

Note that in the Bamanan examples, the etymologies are not blind folk accounts. Instead, they are viable linguistic reconstructions. The Bamanan forms of 1a and 1b, therefore, are not rejected simply because they are Bamanan but rather because they are transparently grasped as instances of linguistic change from forms that still predominate in Maninka and Jula today.

In this instance, it is clear that the very act of learning the (accepted) grapheme-phoneme pairings of N’ko is itself a step toward learning kángbɛ. From the perspective of his own native variety of Maninka, Kantè’s alphabet is regarded as a perfect phonological analysis (Vydrin 2001a). Kantè, however, like any good sociolinguist, recognized his language as replete with various “sub-codes” (Gumperz 1962). As such, even at the level of letters, Kantè engaged with etymology and variation across the sprawling Manding speech community. For instance, in a letter to French linguist and Manding specialist Maurice Houis he wrote, “It must be noted that the letter <g> no longer exists in Manding [mandé], it is only used by races—assimilated at the height of the Manding empire [empire mandé]—that can no longer pronounce the typically Manding [mandén] group <gb> and that they replace by <j> or <g>, for example: jɛman ‘white’, gon ‘gorilla’ which in Manding [mandé] are gbɛman and gbon” (Vydrin 2001a, 138). Not only did Kantè see phonemes (“letters” in his usage here) as historically constituted, but he also delved into accounting for the sociohistorical process that gave rise to such a divergence (i.e., the conquering of later assimilated races [i.e., ethnic groups] during the spread of the Manding/Mali empire).

As such, today, students of N’ko typically embrace and use Kantè’s grapheme ∇ <gb> in writing, even when in their own native variety one finds /g/, /j/, or /w/ in its place.¹¹ A few examples using Bamanan illustrate this dynamic in table 2.

Table 2. Application of kángbɛ at the Phonemic Level to Bamanan Lexemes

Such examples, as well as Sékù Jàkité’s introductory lesson about letters and sounds, demonstrate how learning N’ko is inseparable from learning Manding (i.e., a standard register of it). Of course, no speaker of a Manding variety needs to learn to speak Manding per se. The N’ko classroom’s function in this sense is not to teach people how to speak Manding varieties but rather how to speak a specific register: kángbɛ. Critically, this, in turn, introduces students to a metalinguistic framework that allows them to explicitly understand their own variety of Bamanan, Jula, or Maninka as but component varieties of one single language—N’ko.

In addition to the proto-phonemic ∇ <gb>, Kantè also developed at least one logographic convention that serves as another means for his orthography to transcend the sociological limitations of a purely phonemic orthography.¹² This convention revolves around the notation of intervocalic velars. As Vydrin (2016a, 11) notes regarding Bamanan, “in the intervocalic position, velar phonemes are not contrastive: [-g-], [-k-], [-ɣ-], [-x-] and even a zero consonant, -ø-, are allophones of a single phoneme.” To represent this, Latin-based orthographies vary widely in their preferred grapheme. One may often choose freely between <g>, <k>, or simply dropping the intervocalic velar (e.g., tága, táka versus táa ‘go’). In N’ko, however, Kantè (2011, 15) calls for the use of a single graphemic representation that allows for multiple dialectal realizations to be grouped logographically under one convention. This phenomenon is outlined in table 3.

Table 3. Intervocalic Velar Representation in N’ko Orthography

While this sort of convention may not seem to be very distinct from the de facto and proposed orthographic standard of always marking intervocalic velars with g in Malian Bamanan (see most recently Vydrin and Konta 2014, 24), it circulates as an important feature of N’ko’s pan-Manding iconicity. Take, for instance, this excerpt from an N’ko website that echoes similar comments that I often encountered (N’ko Institute of America 2013):

When Mandens from different sub-groups talk to each other, it is common practice for them to switch, consciously or sub-consciously, from one’s own dialect to a conventional dialect known as N’ko or Kangbe (the clear language). This is even true, sometimes, during conversations between the Bamanans of Mali, the Maninka-Moris of Guinea, and the Maninkos of Gambia or Senegal although pronunciations are practically the same. As an example, the word “Name” in Bamanan is “Toko” and in Maninka it is “Toh.” In written communications each will write it as Tô (ߕߐ߮) in N’Ko, and yet read and pronounce it differently.

Here we see that one of the central appeals of Kantè’s convention is how it allows for the N’ko orthography to ostensibly embrace cross-dialectal variation without neutralizing it or overtly regarding certain dialects as non-standard.¹³ Phonemic orthography is seemingly transcended in a way that allows for pan-Manding unity all while retaining the ability to locate yourself geographically through dialectal realization.

Finally, learning the N’ko script is a lesson in Manding phonology. This stems in part from the fact that it is, in many ways, a more “shallow” (Klima 1972) orthography than any of the official Latin-based ones. In table 4, I outline four different linguistic phenomena of Manding, which are typically marked by <n> in Latin-based orthographies.

Table 4. Phenomena Represented by <n> in Latin-Based Manding Orthography versus N’ko

Note. —In general, the official orthographies promote the use of <ɲ> for the palatal nasal consonant but notable authors, such as Dumestre (2011), eschew this convention in favor of the digraph <ny>.

In the case of N’ko, it is more “shallow” because each one of these phenomena is represented by a distinct grapheme or diacritic, which reduces orthographic depth (Frost and Katz 1992) as measured by how many features a single convention represents. N’ko is thus semiotically iconic as an alphabet because it appears to be a diagram of Manding itself through its more transparent mapping of the language’s sound system. This interpretation is critically solidified by Kantè’s system of diacritics for marking the linguistic phenomena of vowel length, nasalization and most critically tone (Donaldson 2017, chap. 5). Coupling these markings with his unique and (seemingly proto-)phonemic (and logographic) alphabet, Kantè laid the groundwork for a perfect iconic link; N’ko is not just a diagram of Manding, it is Manding.

In sum, studying N’ko as script is itself a first step in learning N’ko as a proper name synonymous with Manding. By learning the letters and diacritics of N’ko, students take their first step toward not only developing synchronic metalinguistic awareness, but also, in the case of ∇ /gb/, a diachronic phonemic lens for understanding the interrelations and history between Manding varieties. This combined with Kantè’s logographic convention for marking non-contrastive intervocalic velars allows for the orthography to be powerfully perceived as capable of inclusively housing—without necessarily standardizing—distinct varieties of Manding. As a cross-dialectal photograph of Manding phonology, the study of N’ko is simply the study of the clear form of the Manding language itself: kángbɛ—a standard language register meant to serve and unite Manding speakers regardless of their own native variety.

Being socialized into the kángbɛ register, of course, also operates at the higher linguistic level of words. N’ko teachers today make compelling appeals to notions of “verbal hygiene” (Cameron 1995), which serve to both harness and solidify a positive metapragmatic stereotype for the kángbɛ register. Today, this most often proceeds through the tacit or overt idiom of “logic.”¹⁴ Let us explore this point by heading back to Bamako.

On a Tuesday in July 2016, I headed to a regularly scheduled one-hour adult language class offered by the N.Fa.Ya¹⁵ association in an outdoor classroom space. Inside, there were four rows of rickety bench-desk combos that students typically occupied according to their progress with the first three primer books. Students slowly filled in as I sat at my desk working on a translation of one of Kantè’s texts. “Áw ní jɔ́’!,”¹⁶ the instructor, Màhamúud Sánkare, greeted us. A prolific N’ko author and the head of N.Fa.Ya, he generally proceeds row by row or student by student, as need be, depending on their progress. Today, the front row was occupied by three men working on parts of speech (kúmaden’ súuya’) of Manding as elaborated in Kantè’s first book of N’ko grammar (2008b). Their lesson focused in particular on “tɔ́ɔnɔ̀dɔbíla,” which Màhamúud readily glossed in French as ‘(personal) pronouns’.

Drawing on their grammar book, the teacher presented pronouns as being sortable by singularity/plurality (kèlenyá ‘singular’ and jàmayá ‘plural’) and by person (kúmala ‘first person’, kúmaɲɔɔn ‘second person’ and gbέdɛ ‘third person’). He did not hesitate to partially explain the terms using French for metalinguistic glosses. Following the book, Màhamúud then introduced the different paradigms of Manding pronouns that exist for Maninka, Jula, and Bamanan (the dominant variety of Bamako and Mali as a whole) as distinct dialects. None of them, however, was selected or upheld as “correct” (ɲúman); rather, they were all explained as “broken” (tíɲɛnen) forms of kángbɛ. The students remained attentive. To make his case, Màhamúud appealed to the plural marker <lu> (ߟߎ߬),¹⁷ a suffix that, he argued, one should simply be able to “attach” (nɔ́rɔ) to singular nouns. “That’s coherent” (Àle tílennen) or “logical” (sáriyama), he posited.

This argument relied not only on the students’ familiarity with the Maninka form (lu pl) but also their implicit recognition of it as a “fuller” (as it were) and thereby older form from which Bamanan had deviated. In the moment, no students spoke up in this regard, but Màhamúud addressed the point directly nonetheless. Specifically, he drew on the example of pluralizing the word c ‘man’. Today the Bamanan plural marker is the clitic /-ù/ (though it is represented orthographically as a suffix-like word final -w):

Logically however, “if you respected the rules” (n’í táara ní sàriyá’ yé) one would use the form lù:

Today, this proposed etymology seems evident in Bamanan’s emphatic form of the third person plural òlú, as well as in Jula’s variability between ánw and ánnù, the emphatic form of the third person plural. Regardless, the “fuller” form circulates as common knowledge thanks to Manding speakers from Maninka areas as well as popular songs and oratorical registers that I regularly encountered in Bamako. One student, a tantie (auntie)–like figure, for instance, spoke up at the end of the lesson and stated that cɛ̌w was simply a faster version of cɛ̌ lù. Màhamúud thus did not need to explain the etymological process that has led to lù being the Maninka equivalent of Bamanan -w today; he simply metapragmatically commented on one form as being in line with logic or the “rules” (sàriyá).

Màhamúud nonetheless conceded that in Bamako people often do not understand things unless they are Bamanan. Putting himself in that category, he acknowledged that “we” deem certain forms as “màninka gírin” (‘heavy Maninka’). Ultimately, however, the language (kán) they all speak is “màninkakán.” Switching to French, he elaborated, “C’est la langue mandingue” (It’s the Manding language) before adding that the language came from “there” (i.e., Màndén) to “here” (i.e., Bamako). In Bamako today, he carried on, people all come with their language. For some it is influenced by “Soninke” (Màrakakán), the language of another major ethnic group in Mali. For others it is influenced by something else. “À bέ tílen cógo dì?,” he asked—how can this be correct or, more literally, straight? Màhamúud supported his implicit argument for written standards with international examples. Other languages are not spoken and written in the same way; take, for instance, the French of Paris and that of Marseille. Moreover, he continued, even the historic Bamanan high form emanating from the precolonial kingdom and modern-day town of Ségou is not one thing.

His takeaway for the students therefore was that they are going “after the language’s true logic” (kán yὲrɛ logique nfὲ).¹⁸ Applying this reasoning to the various dialectal forms of plural pronouns that Kantè listed, as well as to his own knowledge of Bamanan, Màhamúud came to the conclusion that the class’s own third person plural (òlú) and the second person plural emphatic (áw) were not sound. The presumed reasoning behind these points, outlined in table 5, is that neither form was a straightforward derivation from the base singular pronouns (ń, í, and à), as seen above.

Table 5. Màhamúud Sánkare’s Analysis of Bamanan Pronouns

Sánkare’s lesson was far from the only time that myself or others in N’ko circles engaged in a discussion of pronouns. Also in 2016, I interviewed author and bookshop owner Úsman Kùlúbàli (UK in the transcripts that follow), who is known for his fiery rhetoric and books about the history of anti-black racism and slavery (2008, n.d.). One of the striking features of his writings is the use of a particular pronoun form, ߒ߬ߠߋ߬ߟߎ߬ (ǹnelu ‘we’), which I had never seen in print or encountered orally before reading one of his books. When I asked him about the usage, he told me that it is Màndenkó and said that he came to embrace it after having studied Kantè’s first grammar book where he lays out the pronouns systems of the major Manding varieties (2008b, 9). Missing from Kantè’s (2008a) analysis however are the Mandinka or “Mandenko” forms, which he simply does not discuss. For Kúlùbáli, they were key:

He methodically laid out the emphatic and plural pronoun paradigms of Maninka, Bamanan, and Jula before moving on to another variety:

Seizing on the role of lù as the pluralizer (jàmayalán), he concluded as follows about the “Mandenko” system and his decision to adopt the form nelu:

As he makes clear in lines 1429–34, he, along with others, is not interested in writing one single dialect Manding. They are working in pursuit of the language (kán)―in its “pure” (píyɔpiyɔ) form—not a dialect (kánbolon).

Sánkare’s lesson and Úsman Kúlùbáli’s reasoning about pronouns suggests that the concept of kángbɛ ‘grammar’, literally ‘clear language’—predicated on an ideal of logic and cleanliness—is explicitly not meant to be congruent with any one dialect (kánbolon). Rather, regardless of one’s own native variety of Manding, the language’s kángbɛ register must be cultivated and mastered through study, dedication, and perhaps most importantly sound reasoning.

Good Speech, Good Citizenship

Kángbɛ is also actively developed as a denotationally and etymologically logical register for reasons that go beyond compelling linguistic analysis. It is the discursive means by which N’ko students can hone themselves into the kinds of savvy, hard-working and just citizens they aspire to be—and that they believe their countries desperately lack.

N’ko activists actively cultivate an ethos of personhood that is conveniently captured by a widely invoked three-part slogan or hendiatris ¹⁹ that circulates in their circles today: kà kólɔn, kà báara, kà télen (‘to be savvy’, ‘to work’, ‘to be just’, respectively). This tripartite slogan is canonically attributed to the foundation of Màndén. See, for instance, the following typeset excerpt of a speech reproduced in an N’ko instructional textbook:

For the purposes of this article, I will not investigate this slogan’s ties to the historical polity of Màndén or the mythical Manding figure of Sunjata Keïta. Instead, I will focus on the importance of the phrase as a means of understanding the N’ko movement of today and in particular their stance regarding N’ko’s kángbɛ register and the Manding language.

During my fieldwork, I encountered the N’ko hendiatris regularly. In the summer of 2016, the phrase figured prominently on a commissioned truck used to transport a delegation of Bamakois²⁰ to the town of Banamba for a multiday conference and celebration dedicated to N’ko (see fig. 3). In this case, the aim of the slogan was seemingly clearer because it was preceded with an introductory clause:

ߖߡߊ߬ߣߊ ߢߍߕߍ߯ ߞߏ ߛߓߊ߬ ߟߋ߬ ߸ ߞߊ߬ ߞߏߟߐ߲߫ ߸ ߞߊ߬ ߓߊ߯ߙߊ߫ ߸ ߞߊ߬ ߕߋߟߋ߲߫

Jàmaná’ ɲέtaa, kó’ sàba lè, kà kólɔn, kà báara, kà télen

‘The advance of a country is [based on] three things: know-how, work, and justice’

In this article, I will leave aside the intriguing question of which actual or envisioned “country” this use of jàmaná refers to and instead focus on how the line transparently frames N’ko’s hendiatris as one designed to work in service of developing, advancing or moving forward a society or polity.

Figure 3. Truck emblazoned with kà kólɔn, kà báara, kà télen in Banamba (photograph taken by the author).

The subtext behind this slogan is that N’ko activists regularly question the efficacy and work of those that currently staff and lead West African postcolonial states. Such discourse is of course common, but N’ko activists actively view themselves as offering an alternative work ethic. During the summer of 2013, for example, I visited a small Quranic school that operated in N’ko. After the lesson, during which students recited classical Arabic verses of the Quran transliterated into the N’ko script, we were visited by another N’ko activist whom I had been introduced to a few days prior, Yáyà Jàabí. Ethnically Soninke, he had spent eight years working in Angola. His good fortune during this time was manifested by the immaculate and air-conditioned vehicle that we eventually climbed into in order to run a few errands around town. Driving between his brother’s business compound and our next destination, I commented on the poor state of roads as we were jostled about. In response, he insisted that “the government doesn’t work” (tέ báara’ kέ) and that the parliamentary representatives don’t do their jobs. From the back of the car, the Quranic school teacher chimed in that N’ko, “òle yé síra kura’ yé”—that’s the new path.

In other cases, though N’ko activists question the work ethic of not only their government but also their fellow compatriots both nationally and continentally. For instance, in an extended 2015 interview I conducted with Bàbá Màmádi Jàanέ (BMJ), arguably Sulemaana Kantè’s primary intellectual heir, he recounted the following:

In lines 1–9, Bàbá paints the picture of the African continent devoid of its own products or consumer goods. His critique of this in lines 10–16, however, is not one of government or international trade policy; instead, he sees it as a problem of self-imposed African “languor.” Thus the problem with African postcolonial society lies not only in the hands of politicians and bureaucrats, but in those of the general populace as well.

A few minutes later in the interview, Bàbá applied this same logic to language practices.

In line 42 Bàbá uses an imperative, “Don’t rest on your laurels”, which presupposes an interlocutor who deserves such an admonishment. Given the earlier remarks about African society, it is clear that this refers to the kinds of lazy citizens who have become so numerous that Africa does not even produce its own goods for its own markets. In lines 46–54, we get some information on who these kinds of people might be; they are people, presumably Manding speakers, who wield languages willy-nilly. They do not speak French correctly (line 49). They do not speak English correctly (line 50). In fact, they speak Manding with such little care that they essentially have made it “a dialect of French” (line 54). The converse to this kind of practice, of course, would be the use of kángbɛ, even if not made explicit here.

Thus far, Bàbá has painted a picture of two potentially distinct situations and groups of people: discursive misfits that mix French, English and Manding, and Africans that carelessly do not contribute to their society. A bit later, however, he made the link between them more explicit:

In line 82, Bàbá directly links together the two situations that he has presented: “for this reason, we are languishing”. Those that are careless in speech are equally so in life in general. Finally, in line 99, he makes it clear that his critique of his fellow West Africans is similar to that which other N’ko activists made of their government leaders above; they don’t work (kà báara’ kέ). In this interview segment therefore Bàbá implicitly elucidates how the kángbɛ register, beyond compelling etymology, functions as a potential discursive index of a different kind of West African citizen.

Curious how in practice the promotion of a special register instead one’s so-called natural way of speaking functioned, I often asked N’ko teachers why people should write kángbɛ and not their own dialect. One shopkeeper (SK in the transcript that follows) replied with a metaphor while also drawing in my notebook that I handed him (see fig. 4).

Figure 4. Shopkeeper’s drawing of the Manding language tree in the author’s notebook

Making a case similar to those of historical and genetic linguists, he stated that he envisions language as being like a tree in the ways that it starts as a single entity and then develops individual diverging branches as it moves forward through time. His argument for writing in the language—that is, Manding (or N’ko as Sulemaana Kantè would put it)―was one that went beyond etymology. Gesturing toward his sketched tree, he explained,

In these lines, he paints a picture of the true forms of a language being the strongest. Language is comparable to a material good that is available in different grades of quality. While he does not specify the activities for which it is ideal to have the highest one, his publications, books and N’ko activism in general suggest that this form is particularly important when it comes to writing. In other situations, N’ko activists emphasize the kángbɛ register as a means of unifying Manding speakers across state and dialectal boundaries. Here though, SK paints a picture of it as serving in a different capacity. The study of the N’ko orthography and its body of knowledge allows one to pursue a more pure form of the Manding language that if wielded correctly cannot be “contested” or “fail” (lines 2099–2100). In the lines that follow, SK outlines other major world languages like French and Chinese and suggests that, while they may fail, N’ko (Manding) does not. This “narrated event” (Wortham and Reyes 2015) is interesting because it powerfully shows the stakes of reading and writing in kángbɛ—it is the linguistic means by which West Africans can put themselves on equal footing and work to match the accomplishments (and development levels) of other major countries or even civilizations of the world. From this perspective, N’ko and kángbɛ together become a tool to discipline the various earthly forms of Manding that have—like all dialects—deviated from the proper and powerful form that one cultivates in a continual pursuit of kángbɛ.

Conclusion

For both N’ko’s founder and many students today, N’ko often refers first and foremost to the Manding language in its entirety. Today, this conceptualization of Manding as one single language (under the name N’ko)—united by the primarily written register of kángbɛ—continues to spread across areas where people have postcolonially understood themselves as speakers of distinct, albeit related, varieties such as Bamanan, Maninka or Jula. This can be attributed to at least two factors. First, the kángbɛ register—in part, codified into the N’ko orthography itself—is a linguistically compelling analysis of Manding phonology and etymology, as demonstrated by the current words of N’ko teachers and students. Second, the kángbɛ register—independent of linguistic facts—is upheld and embraced as a component of a larger N’ko ethos of know-how, work and discipline (kólɔn, báara, télen). Cultivating themselves to be able to read, write and potentially speak the clear form of Manding is the means by which students and activists can hone themselves discursively into the opposite of people they see as responsible for the disorganized and poorly developed state of the countries and region in which they reside. Unsurprisingly then, even kángbɛ is not a fixed entity or permanent set of linguistic features. It too is subject to scrutiny, improvements and repair. As one N’ko teacher commented following a heated disagreement about some of the conventions of written N’ko or kángbɛ: “fέn bέɛ bέ dílan”—all things can be fixed. Indeed, in the eyes of N’ko activists in postcolonial West Africa, they must be.