Rhonda M. Gonzales. Societies, Religion, and History: Central East Tanzanians and the World They Created, c. 200 BCE to 1800 CE. New York: Columbia University Press, 2009. ix + 257 pp. $60.00 (cloth), ISBN 978-0-231-14242-7.
Reviewed by David L. Schoenbrun (Northwestern University)
Published on H-Africa (September, 2010)
Commissioned by Brett L. Shadle (Virginia Tech)
The Vicissitudes of Language in Writing Precolonial African History
This book (also published as a Gutenberg e-book [http://www.gutenberg-e.org/]) offers the first sustained consideration of the history of a region of East Africa directly adjacent to the Indian Ocean but crossing littoral, hinterland, and upland zones. It will be of great interest to scholars of the Indian Ocean and East Africa during much of the Common Era until the seventeenth century. Rhonda M. Gonzales’s region encompasses a set of ten related languages (“Ruvu languages”) spoken today in east central Tanzania. They are spoken in a distended parallelogram stretching, north to south, from the Ruvu River to the Wami River, and east to west, from the north central Tanzanian coast between Bagamoyo and Dar es Salaam through the Kilombero Valley and into the Nguru Hills, in the interior west. The book opens in the last several centuries BCE, with evidence of “Early Iron Age settlers” (p. 57). But the focus is on later centuries, marked by the formation of a single community of proto-Ruvu speakers existing for a few centuries about 1,500 years ago. In the eighteenth century, its last subgroup dissolved into the languages spoken in the region today. These matters take up the first two substantive chapters (and a fine set of maps available in the e-version helps readers envision the process). The remaining three substantive chapters address specific themes: social organization and worldview (chapter 3); gender, generation, and descent (chapter 4); and healing (chapter 5). A final chapter recaps the others. Gonzales argues that social institutions built around matrifocality and worldviews built on religion governed this long history of settlement, language divergence, and interaction. She further argues that those institutions and cosmology were derived largely from the earlier continental history of Bantu-speaking Africa. Gonzales designs the narrative’s themes and the emphasis on their historical antecedents in the continent’s history to balance the influence of Indian Ocean historiography on the region. For these reasons, Gonzales’s work deserves careful attention.
Gonzales writes using evidence from a variety of historical linguistics enjoying a golden age in the historiographies of regions, like hers, largely without the sources historians commonly use.[1] Conventional historical linguists draw on ancient textual sources for the material they compare. But they reconstruct still earlier linguistic forms of lexis, grammar, and so forth, ancestral to that found in texts. The most famous examples come from Indo-European studies. Like proto-Ruvu, proto-Indo-European is a hypothetical language. It does not exist in either spoken or written form, but it is the hypothetical source from which historically attested languages can be demonstrated to have derived “in both their vocabulary and grammatical form.”[2] Gonzales works in a similar fashion to reconstruct hypothetical Ruvu vocabularies and then to make claims about semantic histories and word transfers (“borrowings”) between different languages. We who use this method in African settings work from vocabularies we elicit in the context of fieldwork together with whatever we can glean from a documentary record. That record is very rarely as old as the seventeenth century and, in Gonzales’s setting, it is not older than the middle of the nineteenth century. Gonzales spent a total of fifteen months in Tanzania between 1997 and 1999 eliciting vocabulary from Ruvu languages and gleaning lexical and semantic evidence from published or archival material, virtually all of which was generated well after the 1880s, the time of imperial conquest in the region. Gonzales draws semantic material most consistently from ethnographic descriptions of various events, things, and attitudes touching on her central themes, generated during the 1940s and 1950s, the latter decades of Tanganyika’s colonial period. She supplements this combination of historical linguistics and comparative ethnography by using archaeological studies of pottery traditions as chronological anchors. Gonzales also draws on the Periplus of the Erythraean Sea, an extremely important document written around AD 50, to set the opening scenes of her story. Working within the sweeping temporalities generated by this approach to language, Gonzales offers a breathtaking story of continuity in many themes of interest to her for much more recent times. Her first foundational statements trace out a settlement history of a region that has been for too long simply a blank spot on historians’ maps of East Africa.
That statement rests on a classification of the ten Ruvu languages into subgroups whose sequential formation and dissolution creates a narrative structure on which she can set out the evidence for continuity and change in vocabulary. The classification is the product of an unusually massive amount of labor, even by the standards of historians who pride themselves on their archival tenacity. Gonzales took a list of one hundred English words (presumably also glossed in Kiswahili) and asked each of her forty-three Tanzanian hosts/informants to translate them into their first language, one of the ten Ruvu languages. She almost certainly did this several times for each of the ten tongues, working with older and younger men and women in town and countryside. In the course of this basic work, Gonzales began to learn about the phonology and morphology of these languages and this helped her when she moved to words that touched on fields of meaning of interest to both her and her Tanzanian teachers. Each time she elicited translations for an English term, she had to record the uttered response using an orthography designed to represent the exact sorts of sounds made in forming consonants and vowels, and to represent the precise rhythms of prosody used to distinguish meaning in tonal languages.
The labor of elicitation is grueling. Most informants see themselves as hosts first. Once they figure out what the researcher desires, they want very much to please her by serving up the discrete translations for individual words that the method favors. Likewise, it can be hard on the researcher to make their hosts expend such effort, even for a decent wage. Often, though, particular fields of meaning--initiation, witchcraft, marriage--prompt an informant to tell stories. These stories help the researcher feel confident that they are asking about matters of interest not only in the academy but also for their host communities. With this taxing, revelatory work concluded, Gonzales took the translations into the ten Ruvu languages of each of the one hundred English words and set them next to one another (appendix). She then compared each gloss in each language for each English meaning, looking for regularly corresponding Ruvu terms that translated the same English meaning. In this case, that labor involved 44 comparisons among Ruvu translations of each of the 100 English words, for a total of 4,400 comparisons. She then counted the number of times two or more languages had regularly corresponding terms translating the same English meaning, for another 4,400 trips over pairs of terms on the lists. She then organized those counts of cognation into a three-sided table with the highest numbers arrayed along the hypotenuse and the lower numbers dropping away toward the box at the elbow of the triangle’s ninety-degree angle (table 1, p. 21). That table contains information supporting a hypothesis about the structure of a sequence of language divergences, a sequence most often represented in a tree diagram (figure 1).[3] This only describes the work of classification, and not the work of lexical reconstruction.[4]
Subgroups in such a classification take shape around the inherited vocabulary measured in cognate counts. But providing calendar dates for their formation and dissolution comes from glottochronology that rests on the assumption that such material is replaced in a language in a random fashion. However vexed that assumption is, the replacements of vocabulary nonetheless tend, over long periods of time, to take the shape of a bell curve (pp. 24-27). The integrity of subgroups as distinct communities of speakers, however, is inferred by discovering shared innovations in vocabulary (or other parts of speech)--either generated internally, through processes of semantic invention, or transferred from different, neighboring languages. In the present study, the field of languages that may be analyzed in this way is called “Bantu,” a name that many readers not steeped in the guild study of African history may recognize either from African history textbooks or from knowledge of the legalized racial categories--“Bantu” was one of them--in place during the Apartheid period in South Africa’s history.
Scholars working in this manner develop an intimate relationship with discrete lexemes, their meaning, and the patterns of sounds that correspond between lexemes in different languages with the same meaning (pp. 18-20). One pursues the spatial distributions of these features to make arguments about their relative antiquity. The historian-linguist grows accustomed to thinking of meanings in an orderly, structured, granular--almost atomistic (Gonzales calls them “shards” [p. 18])--fashion in order to analyze those distributional patterns. A wide gap separates the atomistic sense of meaning produced by lists and the forms of meaning produced in textual, spoken, spatial, or material discourse. This is the main reason that histories built on evidence produced by this method can feel mechanical--a language game played between the scholar, printed sources, and the recorded responses from teachers in the field to our elicitation tests. Ludwig Wittgenstein showed that words are actions. They are not separate from the world they describe but partly constitutive of it. The rule-bound character of that work makes language a game. Wittgenstein invited us to see the sum total of those games through the metaphor of “an ancient city: a maze of little streets and squares, of old and new houses, and of houses with additions from various periods; and this surrounded by a multitude of new boroughs with straight regular streets and uniform houses.” The all-consuming “ancient city” is like life itself.[5] It promises riches to scholars who reverse-engineer the vicissitudes of language that constitute its growth. To understand the past of the “ancient city” of Ruvu languages, Gonzales begins reconstructing its “maze” with the arrival of its first builders in the region.
The story of settlements in the region takes a form familiar to all who work with these sources. Having completed the analysis described above, Gonzales looked for the effects of other languages on their vocabularies--in the form of transfers or loans. These items invite the logical conclusion that a community of people speaking those other languages must have been around at some point in time to provide the transferred material. Identifying transferred vocabulary is a powerful way to bring into the story speech communities no longer around. It is a technique Gonzales uses often in the book, to refer to Southern Cushitic, Eastern Sahelian, and several different groups of (and even individual) Bantu speech communities.
Other speech communities enter the story through inferential side doors. Forms of pottery with distinctive decorative grammars, vessel types, or other technical qualities (such as slips, tempers, and the like) are taken to imply equally distinctive speech communities. This seems an especially irresistible conclusion if the geographical reach of sites bearing them matches inferences about the earlier presence of speech communities ancestral to still-existing ones or to those inferred to have existed from the presence of lexical transfers in still-existing languages. These “correlations” have the intuitive merit of allowing readers to think of potters as speakers, despite the fact that the social production of potting is itself a historical problem worthy of exploration. Study after study has revealed that, at least in the twentieth century, it is not a simple matter to attach a particular kind of potting style to a particular language.[6] The presence of a speech community in a particular area at a particular time may also be inferred directly. One produces lexical reconstructions for specific referents--grain crops, say--present at a particular node in a language classification. If archaeologists have found and dated the remains of that particular grain crop, then the earliest date for that item from the region defined by historical linguists becomes the latest time by which a hypothetical ancestral speech community could have been there.[7]
Whatever conventions of dating we use, the central arc of the story is always provided by the inferences drawn from change and continuity in the speech community that has come to dominate the region in recent times. Nothing can be done about the bias this introduces into the story, except to remind readers of the great “diversity” of speech communities that lived in the region at various removes in the past. These forms of inference limit the narrative tropes Gonzales puts into play, converting her focus on the territories that Ruvu eventually inhabited into two chapters on the largely causeless unfolding of language divergence and settlement chronology. It went like this. Beginning around 200 BCE, four unrelated languages were spoken in the region: Khoisan, Southern Cushitic, Eastern Sahelian, and Azania Bantu. A few centuries later, Upland Bantu languages arrived. A few centuries later still, the Periplus mentions a place called “Rhapta,” describing it as a trade fair of uncertain duration. No actual linguistic evidence is adduced for the Khoisan presence. As the languages of gatherer-hunters, they are implicitly there prior to the arrivals of herders and farmers. Southern Cushitic and Eastern Sahelian languages must have been there, in order to account for the transfers of vocabulary from them into the Bantu languages that arrived later. The Azania Bantu speech community left no descendants anywhere. Its existence is inferred from the presence of a type of pottery with affinities to pottery west and northwest of the region. Upland Bantu speech communities left descendant groups that today speak languages like Chaga, Pare, and Shambaa in the mountain zones of Tanzania, near the border with Kenya. Gonzales infers the presence of Upland Bantu speakers at the coast from the presence there of a type of pottery called “Kwale” found most often (and with the earliest dates) in and around those mountainous areas. These different speech communities were all present at the coast at 50 CE, when Rhapta of the Periplus existed.
Soon after this, in the first few centuries CE, another member of the cluster of Bantu-speaking communities settled south of Lake Victoria (from which the ancestors of Upland and Azania speakers had come) and “settled Tanzania’s hinter-coastal zones” (p. 58). They have been labeled “Northeast Coastal Bantu.”[8] They occupied a rather large area between the southern Kenya coast and the Rufiji River (in Tanzania) and westwards into the Taita Hills (Kenya) and the Pare Mountains (Tanzania), all the way to the Dodoma area, well into the drier lands in the Tanzanian interior. Within a span of perhaps five centuries, Northeast Coastal Bantu had diverged into three subgroups and each of these had diverged into two or more other subgroups. Ruvu speakers lived in the southerly zones of that territory, in the eighth century CE. Swahili emerged in the ninth century CE, at the coast.
Ruvu diverged into three communities, West, East, and Kagulu. The first group, “who migrated westward from their proto-Ruvu homeland” between 600 and 900 CE, transferred vocabulary from Southern Cushitic, Eastern Sahelian, and Njombe (Bantu) speech communities into their own (p. 69). Proto-East Ruvu stayed put. In the tenth and eleventh centuries CE, both East and West Ruvu diverged into new subgroups, a process Gonzales sees as having lasted into the thirteenth century. A final set of divergences may have been influenced by outbreaks of plague (1340s CE), documented at towns like Kilwa on the coast a good distance south of where people who spoke languages descended from proto-East Ruvu lived. Although Gonzales has no direct evidence of such a causal connection between plague and population displacements in the interior, she cites an archaeological survey of the coast, between Bagamoyo and Dar es Salaam, that revealed a break in coastal settlement between roughly 1200 and 1500 as “suggestive” that shifting Indian Ocean mercantile fortunes--and public health disasters--should be taken into consideration in explaining the settlement histories she generates from language evidence (p. 74). By the eighteenth century, Ruvu languages had largely taken the geographical and linguistic shape encountered by outsiders in the second half of the nineteenth century, and by Gonzales very late in the twentieth century.
In seeking chronological anchors for this story outside those generated from the use of glottochronology, Gonzales offers a detailed review of the dating and classifying of pottery traditions from the regions settled by these various Bantu-speaking communities. She studies carefully the many publications--and their sometime contradictory conclusions--of one of the leading Tanzanian archaeologists, Felix Chami. She reads Chami as arguing that the same “Bantu farmers” made the different, successive styles of pottery (p. 62). Chami sees sets of common motifs tying each successive pairs of pottery traditions together. What others have seen as distinct pottery styles, he sees as so many moments of transition tying them together in a single tradition. Chami probably spent as much or more time looking at discrete “shards” of pottery to come up with the sets of motifs supporting his view as Gonzales spent comparing Ruvu glosses for English terms on wordlists. Nevertheless, where Chami sees continuity, Gonzales sees distinction and argues that “ethnolinguistically distinct Bantu people originally manufactured” these various pottery wares (p. 61).
This may seem like the argument of a lumper versus a splitter but their differences lie more tellingly in where they see the geographical origins for “the common Bantu past” (p. 63). For Chami that is east central Tanzania, for Gonzales it is the Southern Lake Victoria Kaskazi cluster of Bantu-speaking communities, several of which hived off and moved into east central Tanzania in the sequence just outlined.[9] Gonzales speculates about the influence of the world of Rhapta, described in the Periplus, where consumption, demand, and circulation might mix and match different styles of pottery in the same locations. But, despairing of knowing the details of these economic relations, without “much more comparative and correlative evidence” of unspecified content, she returns to the conceptual universe in which pottery style and speech community express different aspects of an “ethnolinguistic” distinctiveness, in other words, worldview (pp. 64-65).
According to Gonzales, the book argues “that Ruvu people took into consideration foremost in building their communities that the cosmos was biaxial. In their views, people, animals, and the other features found on the earth’s surface lived in the cosmos’s temporal sphere. The other sphere comprised a world of influential spirits. In Ruvu worldviews, spirits could benefit or hinder peoples’ lives and livelihoods. That precept did not determine the way Ruvu people lived, but it did underpin the rationales that legitimized the sociocultural institutions they established in their communities” (p. 219). African histories of religions other than Islam, Hinduism, or Christianity usually address a category called “traditional religion,” which possesses familiar qualities of religions the world over. Readers may argue with the categories--why are the Abrahamic or Vedic traditions called “religions” when African religions are called “traditional”? But many feel that they know what religion means, traditional or otherwise. And scholars writing in the vein of Indo-European studies, a vein we have already seen bequeath much to African history composed from historical linguistics, do not hesitate to draw on ancient texts and comparative language evidence to describe priests, gods, cosmogony, and religion.[10] So, is it worthwhile to query the equivalencies between religion and traditional religion for what is hidden in them? Is it rude to suggest that the translations of key terms in the one--god, priest, witchcraft, sacrifice, religion itself--are poorly executed because they simultaneously lend too much weight to the terms as they work in so-called World Religions and mask the historical force of that weight on “traditional religion”? Asking these questions goes to the heart of a historian’s task.
Gonzales allows that religion, as a category of historical analysis, does not travel to her region from the academy as easily as the others she studies (p. 90). But the conclusion is announced rather than debated with scholars who have wrestled with its many dimensions.[11] Although Gonzales regularly cites the scholarship of Jean Comaroff and John Comaroff, she declines to take them up on their call for histories of the Christian traditions in Europe that informed the Christianity of missionaries in Africa.[12] The core challenge lies in devising a method for assessing the entanglements of Western prejudice toward and pride in something like religion from those with roots in Africa. Tracking the building of these rooms and houses in Gonzales’s “ancient city” should be possible with her methodology. And the tone of her book strongly suggests that this is a problem she is keen to address (pp. 39-41).
To begin, one must take up problems of translation presented by the sources of language evidence sustaining a historical narrative. For example, for outsiders, religion initially identified what Africans lacked. Soon, Africans’ “analogous” practices came to mark “false religion.” Today, many (including, I think, Gonzales) are comfortable claiming a human universality for religion.[13] Attestations or descriptions of semantic evidence drawn from sources created during each of these different moments in the life of the category must be weighed accordingly. They cannot be treated as having the same qualities of semantic clarity as for tools, houses, or terms of address. Given this premium on records from the opening moves in the evangelical conversation, it is surprising that none of the earliest travelers’ accounts from East Africa is cited in the bibliography.
Bet that as it may, Gonzales says: “because of the way normative definitions of religion have of late included the idea of the ethereal or spiritual in a diversity of forms, I think it is suitable, though perhaps not wholly complete, to consider those moments when people conscientiously seek out or interact with--tangibly or cognitively--the ethereal world to be instances of religious practice” (p. 90). This formulation results to some extent from the legacies of intellectual combat with colonial and missionary conceits and prejudices that are implicated in the translational work that pitted “Religion” against “Tradition” and denied Africans any sort of spiritual (or other) history. To what extent might “the idea of the ethereal or spiritual” result from the actual work of evangelization? That labor process very often unfolded around the production of dictionaries and the translation of bibles into vernaculars. The many different glosses Africans gave to the terms in play in that work reveal just how difficult it is to recognize the entanglements that translate terms as freighted as those at stake in evangelization. Some who wrestle with these issues have gone so far as to state that “the concept (religion) is itself an artifact of Christian encounters with non-Christians.”[14] Such views draw attention to the circularity of using as evidence for a history of religion the very data generated in the process of evangelization. And they warn us to beware of a tendency for “religion” to imply the opposite of “rational.” Gonzales totally refuses the latter danger, but the first problem goes largely unaddressed.
Gonzales turns instead to something she calls “a foremost creative force--something Christians name God and Muslims Allah” in order to establish the reality of “religion” for Ruvu people (p. 91). Gonzales realizes the danger of this string of equivalencies--presumably because it smuggles into the semantic domain of the Ruvu term, *Mulungu, a host of ideas, practices, and forms of knowing with roots in Abrahamic traditions--but still seems to think she can see a pre-Abrahamic horizon from within the language worlds of the late twentieth century.
These labors of translation, especially with respect to the English word “God” and its possible glosses in other languages, engaged Africans in a conversation about evangelism. Did this conversation tap into a preexisting, widespread world of intellectual reflection about the nature of the cosmos and peoples’ relations to it? Christopher Wrigley has written with great erudition about the likelihood that tropical Africa’s longstanding (since at least neolithic times) interactions with West Asians, North Africans, and Europeans in a great agropastoral ecumene could have included “associated institutions and beliefs,” including religion and cosmology. Working to convert a likely scenario into a historical process, Wrigley is at pains to show how the intact whole of institution and belief, developed in Western Asia, gradually lost bits and pieces of itself, in subsequent centuries, which can nevertheless sometimes turn up among the myths and fables of groups and areas involved in the process.[15] His is an arduous and provisional method, but one that is open to formal historical criticism. In any case, then as now, it seems likely that people were given to mulling over the existential conundrums invoked through religion primarily in times of crisis. We cannot, of course, know the answer. But the theological conversations so central to evangelization often took literary shape in the form of dictionaries and biblical translations.[16] That work inaugurated a new semantic geography of distinction familiar to post-Enlightenment thought. Faith and belief were the two most powerful weapons in the missionary endeavor to rid Africans of their “false religions” and lead them to the light of Truth. The proximities of people to God, the necessity or not of priestly intercessors have a history in Christianity and Islam and, it seems, in the past of Bantu languages. The different answers to these questions carried by missionaries into their work of translation matter for those who seek to use those translations as grounds for inferring earlier meanings.
Gonzales knows this story very well. But she conceives the process as one of missionaries “reading Ruvu peoples’ practices” which do not produce “exact renditions of how Ruvu people of long ago ruminated on *Mulungu” (p. 92). That is one reason she (and many of us who write about these issues) routinely warn readers that categories like “politics” and “religion” were not separated from each other in the times and places we write about like they came to be in nineteenth- and twentieth-century European secularisms. Indeed, her book could be read in part as an extended valorization of African beliefs that had been badly distorted by outsiders. But the details of that story of distortion, translation, and evangelization should give Gonzales some pause in historicizing the semantic fields she finds attached to words in Bantu languages that she glosses with such meanings as “first cause or creator” or “the idea of a force responsible for ordering things in a right or good way” (p. 91). The problem with such glosses lies in the fact that they are inextricably bound up with the evangelical conversation that drew speakers of vernacular languages into the translational work of the very missionaries so vilified for denying Africans a religious history. (The fact that *Mulungu is always capitalized masks these difficulties rather than opening them up to critical assessment.) The fact that these glosses are widespread across the field of Bantu languages may reflect the widespread nature of such translational work or it may reflect the durability of these meanings from ancient times or entirely different divisions may be in play, the relations among which might sustain the sort of translations Gonzales proposes without doing too much damage. One such possibility lies in thinking about the “distance” between “God” and people’s lives as a statement allowing doubt as to the “presence” of “God” in general. Gonzales implies that these relational equivalencies existed (and presumably still exist) but does not demonstrate which of them is to be preferred as explanations for her translations. Doing so would bring to life the larger circumstances in which evangelical theological (or political) translations unfolded.[17]
Such a demonstration would not only go far in reconstructing African usages of terms like mulungu. It is a necessary first step in exploring why Southern Kaskazi speakers invented a new term (*Mulungu) in order to apply a notion of “Creator” as moral or “right.” In a footnote, we learn that “divinity” was added to the notion “that a first Creator organized things in a right and good way” as a result of “the introduction of missionary Christian work” (pp. 92, 126n 8). Gonzales does not explain how she knows this. Yet we need to know in order to understand the African perceptions that are at stake, before they became translations.[18] Divergences between Ruvu and missionary understandings of the proximity of *Mulungu to people look very different if the term does not refer to the notion of a single act of “creation.” A rich source of evidence on that ontological mystery tends to be found in so-called genesis stories or myths where ideas and explanations for the “beginning” of the world unfold and often involve rather a larger cast of characters--and moments of creation--than this single term would suggest.[19]
Later in the same chapter, Gonzales introduces the term *mulungu, “spirit force of wilderness, potentially evil spirit.” But the reader may be forgiven for wondering why this term should not be considered together with *Mulungu, “Creator,” as parts of a single semantic field that expanded or narrowed or broke up, under particular historical circumstances (p. 97). Absent are formal phonological or tonological criteria to separate the terms; the difference between their semantic fields is the evidence for their distinction. What if that difference in meaning reflects the work of evangelization, work that often began in circumstances of profound loss of ecological control, from the 1870s? On the one hand, these details might at least have been queried to explore why “recent Ruvu speakers” talk about “*mulungu as sick or hot in nature” and why those same people might have been very busy propitiating *mulungu when missionaries came around, trying to lure Ruvu speakers into debates over theology (pp. 97, 184-185). On the other hand, if it were possible to rule out the chance that the semantic distinctions between the two terms emerged in the course of evangelization, Gonzales does not ask about the historical circumstances that gave rise to the innovations themselves, in far earlier times.
The practical reason that runs through the religion Gonzales writes about saves the day, even if it feels far from the disembodied spiritual core of religion as a field of meaning. The material interests at stake in religious practice may be read in the very sources (what she calls “published ethnographic accounts” [p.10]) that bear problematic traces of the evangelical conversation.[20] In the passages where Gonzales lays this out, she might have drawn her readers into her argument by giving them all the different English glosses that can translate the stem (-lung-) found in the Ruvu terms *Mulungu and *mulungu (pp. 92-93). Why not, for example, include “Nature” or “Wilderness” alongside “God”? Those two English lexemes refer to fields of creation that lay at a remove from people’s ability to influence them to their own ends. Instead, Gonzales observes that such “influence” turns on the use of “medicines” in an arena populated by “ethereal forces” (“other” than the Creator), forces that turn out to be “spirit forces of antecedent generations.” What matters here, as Gonzales says elsewhere, is what “early Ruvu people presumed was true about the world they lived in.” They held that “social wellness depended on maintaining relationships of reciprocity between two spheres of potential power.” “They believed all people, spirit forces, and social institutions were engaged in ongoing dialectical exchanges that crossed the two realms” (p. 89). The practical reason and empirical problems that drove the social work of seeking access to unseen forces are precisely the issues of concern to most of the people who spoke Ruvu languages. Generating historical evidence to fill them in, with attention to specific context and contingency, would shift the focus from system to action. Ethnographic descriptions of action would then offer the chance to see ostensive meaning being produced by Africans. These rich avenues of historical explanation are what lead Gonzales to state: “understanding worldview and religious matters is key to recovering Ruvu history” (p. 90).
The category of “Bantu religion” then must not be allowed to sit apart from the hurly burly of social life in the region. It must be brought down to earth to reveal the ways in which the problems people used it to address, the moral imaginations it embodied as strategies of social reproduction, created circumstances that led to change in “Bantu religion”; or induced some interested parties to nourish continuity in its precepts and practices.[21] Gonzales succeeds in imparting a sense of the antiquity of these entanglements even if their historical specificity eludes the grasp of the method. The elusiveness may in part be an artifact of another category that knits Gonzales’s narrative together, the notion of ethnolinguistic distinctiveness that she discovers from the close-in work of classification, vocabulary reconstruction, and semantic history.
In many places, Gonzales writes about “ethnolinguistic relationships” that invited “renegotiation” despite the absence of any evidence that “ethnolinguistic” named a reality for the people supposed to have carried out the “renegotiation” (p. 228, for example). Sustaining these “identities” was one of the most important achievements of Ruvu peoples in the face of Swahili and Indian Ocean influence to the east and “the absorbing of significant populations of Njombe peoples into their societies” (p. 233). The Ruvu languages spoken today tell us that some sort of “speech community” has existed for the centuries since ancestral Ruvu was spoken. But, if Gonzales intends the term “ethnolinguistic” to name a social reality of groupness, neither its discursive nor its practical significance seems to have warranted lexical signification. Such meaning may well be one of the newer, outlying buildings in the “ancient city” of Ruvu language history. Perhaps, the “steadfastness” Gonzales wishes to celebrate lay beyond the domain of communication or negotiation. It just was.
In chapter 3, we learn that the “largest” social group named by a single lexeme appears to have been the *ikungugo, which Gonzales dated to and locates in the proto-West Ruvu period/speech community, glossed as “super-clan” (p. 227, see also 106-108). Other than that, *kolo, “mother’s matriclan,” appears to have been the largest social unit in play for Ruvu speakers (p. 103). The term was borrowed by Kaskazi Bantu speakers from Central Sudanic speakers, a claim that spurs no reflection on why the distant linguistic ancestors of the Ruvu speakers would have borrowed from linguistic strangers a name (if not the referent) for such an important part of the social imaginary of political life. Later on it turns out that chaudele, “first-born child; a celebration of a grandmother’s granddaughter’s fertility,” another key lexeme for understanding the importance of matrilines, was transferred from a Southern Cushitic-speaking community (pp. 142, 167-168n23). The historical contexts for such a transfer, from a speech community (Southern Cushitic) more closely associated with patrilineal social imaginaries than matrilineal ones, are not explored. At least two other examples of such transfers (*mlao and *lusona) enter the narrative without comment (p. 155). Perhaps these transfers reflect the work and value of building networks of skill and wealth that constituted “clans” or “superclans” or even “lineages.”
In general, the shifting fluidity of “clanship” (even “lineage”) is not explored as an instituted network of interested and unequal members of a community.[22] In her stimulating discussion of the *mwali lifestage and the ceremonial marking the social viability of a young woman, Gonzales has an opportunity to recapture a sense of the larger social networks (and the circumstances shaping their unequal sizes) that convert children into socially viable persons. Those networks are the material dimensions of what Gonzales calls “religio-ritual” practice. If everyone who menstruates may become “*mwali,” not everyone who menstruates bears children (pp. 162-164). But does every woman who bears children become a mother? This reader would have liked to hear Gonzales’s thoughts on the ways in which the social networks making *mwali and dealing with infertility were also key players in sustaining the social reality of motherhood (in which biological reproduction may or may not be the key element). Gonzales makes clear how important grandmothers were to these processes, but they are the only figures she considers. Indeed, Gonzales seems to accept the language of descent as literally constitutive of clanship and lineage. The difference between approaching units of social organization as unstable mixtures of biological and social process has generated an enormous volume of scholarship, none of which is engaged.[23]
Make no mistake, Gonzales mentions much evidence for lexical transfers from non-Bantu or neighboring (but different) Bantu speakers, or from Bantu-speaking communities Ruvu speakers found upon arrival in new areas. She makes powerful use of this bold evidence for the presence of “others” among and near Ruvu speakers. But are cross-linguistic interactions equivalent to “ethnolinguistic relationships”? Even though we are told that “Bantu people never assumed they were the first people on the land,” we are regularly asked to think in terms of a degree of linguistic exclusivity that must have been produced in the face of the multilingual realities Gonzales is keen to keep in play (p. 125). In the world created by the sort of comparative reconstruction on which this book rests, it seems the answer is no or, at least, unknowable from what Gonzales gives us. Why, in effect, are we not confronted with creoles rather than seemingly intact linguistic monads? Gonzales’s understanding of “kinship” or “matrilineage” or “matriclan” might have implications for the dynamics of language acquisition that would have helped with this question.[24] For mainstream historical narratives, which invite readers to simultaneously project themselves into the story and warn them that the past is a foreign country, the difference between “interaction” and “renegotiation” even between “ethnolinguistic” groups, amounts to everything. “Renegotiation” is the territory of aspiration, interest, competition, intention, surprise, and loss. “Interaction” is exceedingly vague, but fitting, given what the evidence bears.
In the end, the compelling case for focusing on Ruvu histories would have been even richer had Gonzales brought them into conversation with the histories of Swahili and Indian Ocean worlds. Doing so would lift the sense, which grows as the book proceeds, that Ruvu societies were largely undamaged by conflict or competitions; unmoved by disaster, violence, or other calamity (plague in the 1340s, being an exception); and undistinguished by rank, inequality, or aspirational imbalances. Gonzales appears not to have devised a grid of research questions (a clutch of English glosses that might have included “slave,” for example) designed to learn about such fractures and fractiousness. She asserts that “sharp socioeconomic differentiation” in Ruvu societies was not likely “before the era of greatly increased commercial enterprise,” which she sees as having begun around 1700 (p. 199). But many scholars of western Indian Ocean economic history point to the tenth and the thirteenth centuries as moments of robust commercial activity.[25] The sixth and seventh centuries CE may turn out to have been another moment of economic vigor.[26] But maybe Gonzales means to direct attention to commodification and its reconfiguring of social organization and individual aspiration?[27] If so, that process is most often studied as part of a long nineteenth century, though it might include a long eighteenth century. [28]
For example, insecurity and enslavement are mentioned: “kidnapping and enslavement is likely to have been a continuous possibility,” a statement meant to explain why Ruvu speakers were vigilant over their daughters (p. 146). Were females, in fact, valued more by slave buyers, and if so, in what period?[29] Answers might lead us toward an actual historical explanation for the persistent value of a matrifocal social imaginary among Ruvu speakers. And it might help explain why matrifocality seems less prominent among societies in the region less exposed to the pressures of enslavement. But this is only speculation and Gonzales neither expands on her sentence nor explores its causes and consequences.
It belongs to other books to build on Gonzales’s pathbreaking work--and the exciting new work of archaeologists--by asking about the ways in which western Indian Ocean histories (including “Swahili” ones) provide points of departure (even, perhaps, arguments as to cause) for explaining the signature continuities and changes she has reconstructed. Gonzales looks in this direction, on several occasions, but not in a sustained manner. The result is that we learn far more about the durability of aspects of Niger-Congo and Bantu speakers’ habits of thought and forms of social life but rather little about the relevant parts of the “ancient cities” of Swahili or Indian Ocean historical context that might be in play in Ruvu histories. These complaints aside, Societies, Religion, and History sets out the basic shape of an “ancient city” of Ruvu language history, joining an exciting new phase in East African historiography when we can expect continental and maritime histories to enjoy a more equal exchange. It is very rare for a first book to achieve such lofty significance.
Notes
[1]. The following monographs use archaeological, ethnographic, historical linguistic, and other documentary material very differently: Christopher Ehret, Southern Nilotic History (Evanston: Northwestern University Press, 1971); Jan Vansina, Paths in the Rainforests: Toward a History of Political Tradition in Equatorial Africa (Madison: University of Wisconsin Press, 1990); Christopher Ehret, An African Classical Age: Eastern and Southern Africa in World History, 1000 B.C. to A.D. 400 (Charlottesville: University Press of Virginia, 1998); David Lee Schoenbrun, A Green Place, A Good Place: Agrarian Change, Gender, and Social Identity in the Great Lakes Region to the 15th Century (Portsmouth: Heinemann Publishers, 1998); Patrick V. Kirch and Robert C. Green, Hawa’iki, Ancestral Polynesia: An Essay in Historical Anthropology (Cambridge: Cambridge University Press, 2001); Kairn Klieman, “The Pygmies Were Our Compass”: Bantu and Batwa in the History of West Central Africa, Early Times to c. 1900 (Portsmouth: Heinemann Publishers, 2003); Jan Vansina, How Societies Are Born: Governance in West Central Africa before 1600 (Charlottesville: University Press of Virginia, 2004); Reinhard Klein-Ahrendt, Die traditionellen Eisenhandwerke der Savannen-Bantu. Eine sprachhistorische Rekonstruktion auf lexikalischer Grundlage (Frankfurt: Peter Lang, 2004); Edda Fields-Black, Deep Roots: Rice Farmers in West Africa and the African Diaspora (Bloomington: Indiana University Press, 2008); and Christine Saidi, Nacimbusa: Mothers and Daughters in East-Central African History, 500 BCE to 1800 CE (Rochester: University of Rochester Press, 2010). See also the repository for reconstructed vocabulary from Bantu languages, edited by Yvonne Bastin and Thilo Schadeberg at http://www.metafro.be/.
[2]. Bruce Lincoln, Death, War, and Sacrifice: Studies in Ideology and Practice (Chicago: University of Chicago Press, 1991), 1. See also James Mallory, “A Short History of the Indo-European Problem,” Journal of Indo-European Studies 1 (1973): 21-65.
[3]. Yvonne Bastin, André Coupez, and Michael Mann, Continuity and Divergence in the Bantu Languages (Tervuren: Africa Museum, 1999), contains the results of a team of scholars who worked for many decades comparing glosses from 546 Bantu languages for 92 French terms for well beyond one hundred millions of such comparisons.
[4]. For a beginning, see Derek Nurse, “The Contribution of Linguistics to the Study of History in Africa,” Journal of African History 38, no. 3 (1997): 359-391; and Axel Fleisch, “The Reconstruction of Lexical Semantics in Bantu,” Sprache und Geschichte in Afrika 19 (2008): 67-106.
[5]. This clutch of Wittgenstein’s ideas was assembled by Geoffrey Galt Harpham, Language Alone: The Critical Fetish of Modernity (London: Routledge, 2002), 2-3. See also Ludwig Wittgenstein, Philosophical Investigations, trans. Gertrude E. M. Anscombe (Oxford: Blackwell, 1997), pt. 1, 18, 8e, on the “ancient city”; pt. 1, 6-7, 4e-5e, on “language games”; and pt. 2, xi, 226e on language and “forms of life.”
[6]. Koen Bostoen, “Pots, Words, and the Bantu Problem: On Lexical Reconstruction and Early African History,” Journal of African History 48, no. 2 (2007): 173-199; Gilbert Wafula, “An Ethnographic Study of Pottery Production from the Rufiji Region,” in People, Contacts and the Environment in the African Past, ed. Felix Chami (Dar es Salaam: University of Dar es Salaam Press, 2001), 30-39; and Mary McMaster, “Language Shift and Its Reflection in African Archaeology: Cord Rouletting in the Uele and Interlacustrine Regions,” Azania 40 (2005): 43-72.
[7]. Vansina, How Societies Are Born, 7-8.
[8]. Derek Nurse and Thomas Hinnebusch, Swahili and Sabaki (Berkeley and Los Angeles: University of California Press, 1993).
[9]. Ehret, African Classical Age, 61-69, 183-193.
[10]. Georges Dumézil, Les dieux des indo-européens (Paris: Presses Universitaires de France, 1952); and Edgar Polomé, “Indo-European Culture, with Special Attention to Religion,” in The Indo-Europeans in the Fourth and Third Millennia, ed. Edgar Polomé (Ann Arbor: Karoma Publishers, 1982), 156-172.
[11]. Rodney Needham, Belief, Language, and Experience (Chicago: University of Chicago Press, 1972); Valentine Y. Mudimbe, The Invention of Africa: Gnosis. Philosophy, and the Order of Knowledge (Bloomington: Indiana University Press, 1988), 186; Paul Landau, “‘Religion’ and Christian Conversion in African History: A New Model,” Journal of Religious History 23, no. 1 (1999): 9-30; and Derek Peterson, “Translating the Word: Dialogism and Debate in Two Gikuyu Dictionaries,” Journal of Religious History 23, no. 1 (1999): 31-50. More generally, see Dipesh Chakrabarty, Provinicializing Europe (Princeton: Princeton University Press, 2000).
[12]. Jean Comaroff, Body of Power, Spirit of Resistance (Chicago: University of Chicago Press, 1985); John Comaroff and Jean Comaroff, Of Revelation and Revolution: Christianity, Colonialism, and Consciousness in Africa (Chicago: University of Chicago Press, 1991); and Jean Comaroff and John Comaroff, Ethnography and the Historical Imagination (Boulder: Westview Press, 1992).
[13]. Landau, “‘Religion’ and Christian Conversion in African History,” 17-18, 21.
[14]. Ibid., 11.
[15]. Christopher Wrigley, Kingship and State: The Buganda Dynasty (Cambridge: Cambridge University Press, 1996), 49-54, quotation on 51.
[16]. The work did not cease with conversion. See E. Dammann, “A Tentative Philological Typology of Some African High Deities,” Journal of Religion in Africa 2, no. 1 (1969): 81-95, which possesses no self-consciousness of translation.
[17]. Steven Feierman, Peasant Intellectuals: Anthropology and History in Tanzania (Madison: University of Wisconsin Press, 1990), chap. 10; and Landau, “‘Religion’ and Christian Conversion,” 23.
[18]. Landau, “‘Religion’ and Christian Conversion,” 14.
[19]. In a massive literature, see Christopher Wrigley, “The Story of Rukidi,” Africa 43, no. 3 (1973): 226-227.
[20]. Pier Larson, “Capacities and Modes of Thinking: Intellectual Engagements and Subaltern Hegemony in the Early History of Malagasy Christianity,” American Historical Review 102, no. 5 (1997): 969-1002; and Derek Peterson, Creative Writing: Translation, Bookkeeping and the Work of Imagination in Colonial Kenya (Portsmouth: Heinemann Publishers, 2004).
[21]. For a wonderful example, see Francis Nyamjoh, “Delusions of Development and the Enrichment of Witchcraft Discourses in Cameroon,” in Magical Interpretations, Material Realities: Modernity, Witchcraft and the Occult in Postcolonial Africa, ed. Henrietta Moore and Todd Sanders (London: Routledge, 2001), 28-49.
[22]. For a recent statement, see Neil Kodesh, “Networks of Knowledge: Clanship and Collective Well-Being in Buganda,” Journal of African History 49, no. 2 (2008): 197-216.
[23]. Wyatt MacGaffey, “Changing Representations in Central African History,” Journal of African History 46, no. 2 (2005): 189-207; David M. Schneider, A Critique of the Study of Kinship (Ann Arbor: University of Michigan Press, 1984); and, though published after Gonzales’s book, drawing on the same kinds of sources as Gonzales, Rhiannon Stephens, “Lineage and Society in Precolonial Uganda,” Journal of African History 50, no. 2 (2009): 203-221.
[24]. Katherine Demuth, “Acquisition of Sesotho,” in The Cross-Linguistic Study of Language Acquisition, ed. D. Slobin (Hillsdale: Lawrence Erlbaum Associates, 1992), 3: 557-638.
[25]. For Unguja Ukuu, see Abdurahman Juma, Unguja Ukuu on Zanzibar: An Archaeological Study of Early Urbanism (Uppsala: Studies in Global Archaeology, 2004); and for evidence of bead grinding to supply a demand issuing from areas including those recently dominated by proto-Ruvu and proto-Seuta speech, see James L. Flexner, Jeffrey B. Fleisher, and Adria Laviolette, “Bead Grinders and Early Swahili Household Economy: Analysis of an Assemblage from Tumbe, Pemba Island, Tanzania, 7th-10th Centuries AD,” Journal of African Archaeology 6, no. 2 (2008): 161-181.
[26]. Adria Laviolette and Jeffrey Fleisher, “The Urban History of a Rural Place: Swahili Archaeology on Pemba Island, 700-1500 AD,” International Journal of African Historical Studies 42, no. 3 (2009): 433-455.
[27]. Jonathon Glassman, Feasts and Riot: Revelry, Rebellion, and Popular Consciousness on the Swahili Coast, 1856-1888 (Portsmouth: Heinemann Publishers, 1995).
[28]. Stephanie Wynne-Jones, “It’s What You Do With It That Counts: Performed Identities in the East African Coastal Landscape,” Journal of Social Archaeology 7, no. 3 (2007): 325-345; and Jeremy Prestholdt, “On the Global Repercussions of East African Consumerism,” American Historical Review 109, no. 3 (2004): 755-781.
[29]. Thomas Vernet, “Le commerce des esclaves sur la cote Swahili, 1500-1750,” Azania 38 (2003): 69-97.
If there is additional discussion of this review, you may access it through the network, at: https://networks.h-net.org/h-africa.
Citation:
David L. Schoenbrun. Review of Gonzales, Rhonda M., Societies, Religion, and History: Central East Tanzanians and the World They Created, c. 200 BCE to 1800 CE.
H-Africa, H-Net Reviews.
September, 2010.
URL: http://www.h-net.org/reviews/showrev.php?id=25644
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License. |