In taking their cues from the extant primary-source materials, scholars have written about African “ethnic” communities in the colonial Americas since almost the moment that they began writing about the transatlantic slave trade and its origins. Researchers today are occasionally surprised to discover that even scholars of the Jim Crow-era, such as Ulrich B. Phillips, wrote about these various “ethnic” groups in the Americas. As early as 1918, Phillips gestured to a theory of ethnogenesis—the idea that distinct African identities underwent a collective transformation on American plantations. “Ceasing to be Foulah, Coromantee, Ebo or Angola, ” Phillips wrote, African people in the diaspora became “instead the American negro.” This statement was one of the earliest expressions of what the historian Michael Gomez has more-recently called the “process whereby Africans [in the Americas] moved along a continuum from ethnicity to race.”[1]

The discussion around African “ethnicities” has a long history in the literature of American slavery. Nonetheless, as a scholarly conversation, it has received an unprecedented amount of attention over just the past quarter century. Case studies by authors like David Littlefield and David Wheat (Rice and Slaves, Atlantic Africa and the Spanish Caribbean), surveys by authors like Michael Gomez and Gwendolyn Midlo Hall (Exchanging Our Country Marks, Slavery and African Ethnicities in the Americas), and compilations by editors like Paul Lovejoy and David Trotman (Trans-Atlantic Dimension of Ethnicity in the African Diaspora) have all contributed to a renewed interest in studying African diasporic identities through the framework of “ethnicity.” For many of these historians, “ethnicity” serves the simple function of moving our dialogue beyond homogenous portrayals of African peoples in the diaspora. “Ethnicity” helps scholars avoid speaking in the analytically flat categories of “African,” “Black,” or “Negro.” In this sense, the conversation is both well-intentioned and necessary. However, in another sense, the language of “ethnicity” brings with it a series of assumptions that threaten to limit our ability to understand African identities. I address a couple of those limitations in this essay. In doing so, I argue that that framework of “ethnicity” is useful, provided scholars localize their studies, interrogate their sources, and emphasize the inherently creole, dynamic, fluid nature of all diasporic groups.[2]

Vocabulary is a reasonable place to begin. Discussions about African “ethnicities” almost invariably turn to the complications of judging terminology in the historical record. The reality is that labels used by contemporaries to create and/or mark out specific African identities in the Atlantic World did not have either precise, consistent, or stable meanings. This statement holds true for regions of the West African continent as well as for the groups of people who originated from in and around those regions. This challenge continues to prompt debates and disclaimers among scholars about the meanings of the very vocabulary they must engage with in order to make their claims about “ethnicity.” In Exchanging Our Country Marks, for example, Gomez expresses uncertainty about the origins of the term “Gullah.” This word refers to an African-descended culture in coastal South Carolina and Georgia; however, scholars remain uncertain about whether it derives its name from Africans shipped out of the Portuguese colony of Angola in West-Central Africa and/or from the Gola communities in the interior regions of modern-day Liberia. Similarly, the Americanist historian Gwendolyn Midlo Hall and the Africanist historian Robin Law have carried on a debate for some time about the meanings of the label “Mina” as it appears in the historical record. Does this word refer to African peoples exported from the Portuguese (São Jorge da Mina) and later Dutch (Elmina) port on the Gold Coast? Or does it refer to a more specific “ethnicity,” originally exported from mining centers and then concentrated in and around the “Costa da Mina” of the neighboring Slave Coast? On this debate, historians agree the word had different meanings in places like Portuguese Brazil and French and Spanish Louisiana.[3]

In general, the vocabulary that contemporary writers used to distinguish between African peoples in the historical record of the early-modern period is not easy to decipher. Moreover, the words change their “meaning depending on the perspective of the author and the era of their usage. To cite an example of how this can happen, let us take Gomez’ brief mention about African people from the “Windward Coast.” This term was sometimes an explicit reference to the littoral of present-day Sierra Leone and/or Liberia. Other times, writers employed the term in reference to what was elsewhere called the “Upper Guinea Coast.” In the former sense, “Windward Coast” was a geographical or a physical location, a spot that was unchanging. In the latter sense, “Windward Coast” was a place relative to other regions of Africa the writer was familiar with. It was a large catchment area that existed to the windward direction of the Lower Guinea Coast. Put more simply, traders to the southeast needed to sail “into the wind” in order to get there. This point of reference makes sense if we recall that the majority of English traders were conscious of the fact that they did much of their work to the leeward of Senegambia and Sierra Leone, in ports that are now located within present-day Côte d’Ivoire, Ghana, Togo, Benin and Nigeria.[4]

African labels and/or identities are often ambiguous in the historical record. For this reason, it is not unusual to find scholars making arguments about the “real” or “true” meanings of a particular identifier, as when John Thornton argued that the term “Angola” in South Carolina actually referenced Africans who came specifically from the Kingdom of Kongo. Some historians, like Gomez and Walter Rucker have implied that African “ethnicities” may be identified based on physical features, especially teeth filing and ritual scarification. Yet the evidence they cite in their own works suggests that this is an oversimplification. Africans described as hailing from very distant regions of the coast—the Kissi of Senegambia, the Eboe of Biafra, and some labeled simply as from “Guiney”—are referenced by Gomez as having their teeth cut and/or filed. Similarly, Africans described as being from the Gambia of Senegambia, the Tarkar of the Gold Coast, and Angola of West-Central Africa are all described as having several “country marks” on their cheeks. On the question of teeth filing and scarification, it seems at least two facts might be true: not all Africans identified with a particular “ethnicity” engaged in bodily manipulation, and multiple unrelated “ethnicities” engaged in the same or similar patterns of manipulation.[5]

African “ethnicities” are a notoriously hard thing to locate and define in the historical record. The limitations involved were captured by Deborah Jenson in a recent article about the Haitian Revolution. Jenson writes that, while “identity categories…were sometimes bona fide labels for the largest demographic rubrics,” they were “more often…rhetorical figures [used] for a general idea of Africanness or blackness.” Jenson clarifies this statement through a discussion of the word “Congo.” First, the word “Congo” might be used to label and/or identify where the majority of African-born people had come from (i.e. the Kingdom of Kongo or a larger catchment area of Congo-Angola in West-Central Africa). Second, the term “Congo” might be employed as “a symbolic association based on the contiguity of one category to another.” In this form, the word may function for the user as a coded way of saying a person was “not Creole,” “not Lower Guinean,” or “not Upper Guinean.” Finally, the word “Congo” might function as “synecdoche” or “a symbolic association based on the relationship of the part to the whole.” Colonial authors might use “Congo” to reference an expression of “Africanness” that seems distinct to them, even if they have no intention of linking that identity specifically to the Kingdom of the Congo. [6]

Jenson has captured something crucial in her discussion of “Congo.” It is the protean nature of African labels and/or identities in the Americas. In many ways, this protean nature mirrors the inconsistency of much knowledge that was produced in the early-modern era, as well as the fluid nature of all peoples’ identities. Just as colonial writers were trying to understand—and simultaneously rearrange—African ideas about identity, modern historians are now attempting to parse out the original meanings of terminology intended to differentiate between Africans in the historical record. Consider, as a final example, the case of the term “Guinea” and its many derivatives. As Jenson writes, a derivative of “Guinée” (Guinen) now functions as a synonym for “Africa” in the language of Kreyòl. This usage parallels a particular lineage of writers in the eighteenth century (at least, though not exclusively, in the English Atlantic). White colonists, such as Thomas Thistlewood and Edward Long (circa 1750s-1780s), used “Guinea” or “Guiney” in reference to the entire African continent. Yet this usage reflected a change from earlier centuries of the transatlantic slave trade. When the Luso-Hispanic Empires dominated the Slave Trade out of the Upper Guinean Coast, they employed the term “Guinea” as a particular label for that greater region. As Gwendolyn Midlo Hall explains, the word “Guinea” referred to the more-specific region of Greater Senegambia during at least the first two-hundred years of the slave trade. James Sweet basically supports Hall’s analysis, observing that “almost all African slaves in the Portuguese world were described as ‘Guinea’ slaves” before the seventeenth century.[7]

The point I am belaboring is this: African labels and/or identities are relativistic. Most people interpreted these labels within the context of their own colonial experience. As authors like Paul Lovejoy have reminded us, African identities must therefore be historicized in time and place. Otherwise, there is simply too much potential to make either false or misleading assumptions. The same word—for instance, carabalí—may reference two completely different ports in the Bight of Biafra (Old Calabar or New Calabar) depending on the period in which the term was used (before or after the mid-seventeenth century). Similarly, one might assume that authors like Thistlewood and Long used the term “Guinea” in the same way as the Jesuit missionary Alonso de Sandoval had done over a century prior. In fact, Sandoval employed the word in reference to a much more specific area of Western Africa. In short, working with African labels and identities demands that historians localize their studies. As Alexander X. Byrd summarizes in a lengthy but well-articulated sentence, “Neglecting the history and substance of slaves’ consciousness of the nation, the details of when, how, where, and whether Africans and their descendants nurtured and articulated a certain ethnic or national consciousness, cannot help but produce an analytically flat and relatively ahistorical understanding of what it meant to be Igbo, or Nago, or Coromantee, or Congo, in particular times and in certain places across the Atlantic world.”[8]

Bryd makes a point worth elaborating here. For many, the purpose of studying “ethnicity” to begin with was not only to recapture the way that white and black historical actors perceived of African people in the early-modern period, but also to move beyond (and before) the consuming racial categories of “African,” “Black,” and/or “Negroe.” Indeed, as Phillips’ quote implied at the opening of this essay, scholars wanted to historicize, rather than assume, the emergence of these unifying racial categories. That being said, it would be defeatist for researchers to exchange these words for other labels of African identity that are equally fixed, “analytically flat and relatively ahistorical.” When we imply, as Gomez does throughout the early chapters of Exchanging Our Country Marks, that the Eboe people are despondent and more suicidal, the Malagasy and Gambians are good with rice, the Bambara are rebellious, the Akan are good field laborers who favor centralized authority, the Fulbe and Mandingos are educated and literate, and the Angolans are especially prone to absconding, we employ the language of  “ethnicity” to the disservice of enslaved peoples’ more-complex identities. We reproduce a few of the stereotypes of colonial writers who labeled enslaved people and, in doing so, preserved their identities in the records as they existed relative to their own concerns. Even more importantly, we risk exchanging one “analytically flat” category (“African,” “Black,” or “Negroe”) for a set of them.[9]

This line of reasoning leads me to a more-detailed critique of how Gomez uses “ethnicity” in his book. In Gomez’ own words, his book focuses “on the development of an African-based society in America from its inception to 1830.” More particularly, Gomez discusses all of the major regions of the West African coast in subsequent chapters. He adds to these chapters a comparable discussion of the selected areas of North America where the African peoples in question likely came in the greatest numbers. These are all regions that eventually became part of the US South—specifically, the Chesapeake, Lowcountry, and Louisiana. Gomez relies heavily upon runaway slave advertisements and WPA slave narratives to make connections between “ethnicities” in Africa and their cultural descendants in the US South. For his thesis, Gomez argues that classism in the modern African-American community was built on a solid foundation of “ethnic” difference and privilege. This argument in implied in the title of his book, as Africans ostensibly exchanged their “ethnic” “country marks” for new ideas of racial and class consciousness.[10]

Gomez’ thesis is about ethnogenesis. It is about how African people moved on the “continuum from ethnicity to race.” Gomez cites the preferences of colonial planters for certain “ethnicities” in order to build this argument. He contends, “The unavoidable conclusion is that labor differentiation within the African-based community was directly informed by the legacy of ethnicity.” For example, “The Akan were universally acclaimed as industrious and [so they] were certainly well represented in the skilled positions” of Virginia, Maryland, South Carolina, and Georgia. By contrast, “The Igbo” were a minority in the Lowcountry, and so they were “probably relegated to the fields” with the rest of the lesser-desired “ethnicities.” To summarize this argument, certain in-demand “ethnicities” were privileged with easier work assignments as domestics and tradesmen in colonial and antebellum societies. These were generally “ethnicities” that “most closely approximating the European phenotype.” Almost all other “ethnicities” were sent to work in the field. As time progressed, other factors like miscegenation and manumission exacerbated this supposed division between ethnic “House” and multi-ethnic “Field” slaves. Eventually, the descendants of the former group emerged as a privileged class of black elite, who were less interested in embracing their “African” past. Meanwhile, “those relegated to the category of [the American] underclass” today carry on the multi-ethnic legacy of the latter group. Supposedly, these are the black people who “continue to represent Africa to America.” [11]

I am going to critique this argument. Before I do, however, I want to offer a disclaimer. I am not criticizing all of Gomez’ work. His discussion of “ethnicity” is only one part of the book. Gomez emphasizes many other historical factors that shaped African-American identity, such as distinctions between African- and “country-born” peoples, distinctions based upon previous rank or status in Africa, and commonalities based upon a shared experience of “social death” à la Orlando Patterson. Gomez devotes entire chapters, for example, to the contributing effects of Islam, Christianity, and folkloric memory (although readers may find it troubling that he associates the folk with both field labor and African purity). Notwithstanding all of these nuances, the primary role that “ethnicity” plays is clear and disturbing. Gomez traces out a rough constellation from certain “ethnic” groups during the era of the slave trade to the present-day black elite. Although he never explicitly puts all of the pieces together, readers are encouraged to assume the following lineage: Fulbe-Muslim Africans from the Greater Senegambia, whose “features were believed to be phenotypically closer to Europeans than other Africans” and who were seen as “more intelligent,” were privileged as drivers on American plantations. Here they presided over the majority of generally darker-skinned and less-desired “African” “ethnicities,” thereby embracing privilege and starting a centuries-long journey toward the ranks of the modern-day black elite.[12]

This part of Gomez’ argument is an example of the kind of “analytically flat and relatively ahistorical” interpretations of “African ethnicity” that Byrd warned us about in his article. Whether Gomez would stand by this interpretation or not, one cannot blame an uninitiated reader for concluding that lighter-skinned and/or mixed-race peoples were the descendants of the most-privileged “ethnicities,” like the Fulbe and certain Mande speaking peoples, while darker-skinned Africans were the descendants of a more multi-ethnic tradition of “field slaves” who had to work harder and who never forgot (or never sacrificed?) their African cultures. Here we see the terms “African” and “non-African” being employed as symbolic referents. Gomez seems to ascribe the labels to people in ways that fit with his ideological agenda. The implication appears to be that “field slaves” were often darker-skinned in the early-modern period and, therefore, they were somehow more “African.” By association, black people in poverty today are also often darker-skinned. They are also, according to this simplistic logic, more “African.”[13]

Crucially, Gomez’ tenuous and troubling link between “ethnicity” and class division rests upon an abiding confidence in a series of terms that, as we have already discussed, were unstable and ambiguous. In order to tie “ethnicity” to class in this way, we must first believe that writers correctly and consistently identified “ethnicity” in the documentary record. We must then believe that colonial and antebellum societies obediently adhered to the preferences expressed by a handful of planters in their writings. Finally, we must avoid grounding our studies in local processes and contexts, lest we discover that “ethnic” words like “Fulbe” had very different meanings, and evolved those meanings in different ways, throughout the Atlantic World.[14]

As Gomez’ work demonstrates, a common misstep in working with “ethnicity” in America is assuming the fixed nature of someone’s identity based on the presence of a word we think we can definitively trace back to Africa. In writing about historic connections between Cuba and the Bight of Biafra, Stephan Palmié has observed that words such as “Enllenisón  (generally taken to mean Africa) not so much index…topographic data as impose sacred value on specific situations and utterances.” In other words, communities of African and/or African-descended peoples in the diaspora might use terms, like Enllenisón, that denote specific linguistic traditions or geographical regions while masking entirely new identities—cultures in which the participants actually blended, borrowed, created, and even smuggled in diverse cultural influences. Palmié  places this interpretation in contrast to that of Ivor Miller and Fernando Ortiz, who argued that the Cuban word Enllenisón has a direct counterpart in an Efik-speaking region of southwestern Cameroon and southeastern Nigeria called the Cross River region, and that this specific “ethnic” concept was transferred over to colonial societies, like the abakuá, in Cuba.[15]

Palmié is far from the only scholar who has argued that historians should understand diasporic communities as fundamentally unique or different from their African counterparts. Richard Price has observed the same phenomenon in his work on the Saramaka Maroons in the former colony and modern-day nation of Suriname. Building off of his foundational work with Sidney Mintz in their co-authored book, The Birth of African-American Culture, Price concludes the following in a recent chapter about creolization in The Cambridge World History of Slavery: although “the dance/drum/song performances called “Loangu,” “Nago,” “Papa,” or Kromantí” on the eighteenth-century plantations could have included a number of individuals who actually came from the parts of Africa these names refer to…the cultural complexes associated with each had already been separated/isolated from the actual origins of the practitioners, and they developed and changed through time.” In essence, a new or “creole” identity might cloak its sui generis cultural components behind a name that suggests a more definitive geographic origin. This is one reason why historians working with African “ethnicities” should be careful to emphasize the inherently creole nature of all diasporic groups, even while they search for African influences.[16]

For Price, the distinction between plantation slaves and maroons was significant. The former group had escaped the surveillance of whites on the plantation and, therefore, had the freedom to produce a new cultural identity out of the heterogeneity of its fugitive members. Nonetheless, I believe that historians can make similar conclusions about performances on plantations and pens. During holidays and Sundays, Thistlewood sometimes wrote about activities that he labeled in African terms. Memorably, he wrote about a “rope trick” called the Tabrabrah, led by the “Coromantee.” Since Thistlewood identified this performance as a “Coromantee” act, historians might be tempted to assume that only African people who came from the region associated with that port and its hinterlands (Koromantyn on the Gold Coast of West Africa) had participated in the performance or that, like Miller’s interpretation of Enllenisón, the performance had remained relatively unchanged in the diaspora. I believe this would be a mistake.[17]

Knowing that an act was labeled as “Coromantee” does not tell us much about the performance. What reasons have we to believe that this “Coromantee” performance was not similar to the Kromantí drum celebration referenced by Price or the invocation of Enllenisón mentioned by Palmié—an act which masked ethnic diversity under a specific African referent? Barring further evidence, with what confidence can we really claim that the meaning of “Coromantee” remained stable in the diaspora, that in this case the category did not include other participants who associated with it? With what confidence can we suppose that a white colonial author like Thistlewood was capable of recognizing the “ethnic” origins of this performance? Perhaps some of the Africans who participated thought of themselves as “Coromantee” in Jamaica, but they had not associated with its alleged ethnic or geographic counterparts—like Fanti or Asante—when they had been in Africa. Maybe a person who identified themselves as “Eboe” brought their own cultural expression to the rope trick. Overall, if we conceive of “Coromantee” as a form of religious congregation and/or mutual-aid society with flexible boundaries, rather than as a fixed “ethnicity” transferred to the Americas, then diasporic communities becomes more adaptive.

Most scholars of precolonial Africa and Latin America are well acquainted with this concept of adaptive creole societies. As Gomez reminds us, historians in the former group have written quite extensively about African “male and female societies” (often called “secret societies” by non-Africans) that transcended and overlapped kinship, village, political, and “ethnic” identities. Some of the most common gendered societies include the female Sande and Bundu and the male Poro in the present-day region of Sierra Leone. Meanwhile, scholars in the former group often write about how Africans in urban areas of Latin America were relatively more free to create black spaces that functioned as cultural centers, places of worship, and/or mutual-aid societies. These were often called cofradías (confraternities or brotherhoods) or cabildos de nación (councils of the nation). Like gatherings in plantation and maroon settings, these urban clubs or lodges often named themselves after specific African referents. As one example, Palmié writes about an “officially credited association of first-generation Africans” in eighteenth century Cuba called the cabildo de nacion carabali bricamo apapa efi. As suggested earlier, the term carabalí in the title suggests an association with the Bight of Biafra, particularly the areas surrounding one of the two ports of New or Old Calabar. But Palmié does not believe this to be the case. By that time period, he writes, the term carabalí had no clear “spatio-temporally localizable African referent.” Instead, it is more likely that the African people who participated in this particular society masked an entirely new culture beneath a more specific geographic label.[18]

I strongly believe that performances and communities in the Americas labeled by writers as Carabalí, Coromantee, or Congo were at least partially diverse, sui generis, and unique. However, even if I was wrong, and everyone who participated in these performances and societies actually came from the regions suggested by their African referent—that is, everyone who participated in the “rope trick” was from the Gold Coast—then the communities would still be creole in nature. David Northrup, David Wheat, and Rebecca Shumway are among the many scholars who have tackled this question. In his article “The Gulf of Guinea and the Atlantic World,” Northrup draws on the work of Sandoval, who discussed the term “Caravalies.” He says that the people labeled as “Caravalies” were not really from a homogenous group transferred to the Americas.[19]

According to Sandoval, the “Caravalies” label was divided in two separate groups. There were the “‘native or pure’ Caravalies from the eponymous port town (at that time known as Elem Kalabari but later called New Calabar) and its vicinity. But there was also the ‘caravalies particulares.’” This latter group, Sandoval said, were not technically “Caravalies.” He only gave them that name because they had a tradition of traveling to trade with the peoples around Elem Kalabari. They were comprised of Africans from over forty different communities around the region. Some of these Africans spoke dialects of the same language and others spoke entirely different languages. Similarly, Wheat has observed much the same phenomenon in regards to other labels used as markers for African identities, particularly “Angola” (a catchall toponym used for those exported from the Portuguese colony in West Central Africa), and “Zape,” a group whose diversity he suggests remains “open to interpretation.” As a final example, Shumway lamented that certain historians of the diaspora continue to conflate the various social, linguistic, and political identities of the Gold Coast region under what she calls the “myth of Akan unity.”[20]

Northrup has also taken up this issue with greater clarity in a much-cited article entitled “Igbo and Myth Igbo.” He argues that “people from the region who were forcibly transported to the Americas…did not possess the ethnolinguistic ‘tribal’ identities of today.” As a result, historians need to have “greater sensitivity” when dealing with the many “ethnic” identities of West Africa, recognizing that the “labels commonly used by Europeans enlarged and simplified African identity groups in the era of the Atlantic slave trade.” Northrup levels criticism at Gomez’ work among others. In Exchanging Our Country Marks, Gomez had briefly acknowledged the “ethnic” diversity of West African regions like the Bight of Biafra, but he then used the labels that flattened these identities uncritically, talking about “Akan,” “Yoruba,” and “Igbo” groups when these terms did not appear in the runaway advertisements he analyzed. In reality, planters identified as “Eboe” an amalgamation of African “ethnicities” from the Biafra. This amalgamation included diverse Efik-, Igbo-, and Ijo-speaking peoples. Many of them, like the caravalies particulares mentioned by Sandoval, spoke vastly different dialects of these languages. In perhaps his most scathing criticism of Gomez, Northrup writes, “he assumes that those individuals called ‘Eboe’ by European colonists in the Chesapeake already possessed the pan-Igbo identity of modern times, as though neither time nor circumstance made any difference.”[21]

As previously discussed, the documentary record of the Americas makes it challenging to move beyond these broader categories. Nonetheless, every once in a while, a rare and fascinating discovery is made in the archives that allows us to do so. David Wheat published just such a find in his work Atlantic Africa and the Spanish Caribbean. This case presents an ideal opportunity to see how European labels flattened the complex identities of African peoples. In 1607, an African slave named Luis Congo became the subject of litigation in the colonial courts of Cartagena de Indies. Two owners were engaged in a dispute about who the slave legally belonged to, and the burden of proof ended up revolving around the court’s ability to determine Luis’s ethnicity. Different parties in the case insisted that Luis belonged to one of the three “ethnic” labels most commonly attributed to African peoples who came out of West-Central Africa: “Congo,” “Angola,” or “Anchico.” People at the court read Luis’s body as well as the alleged bill of his sale. They brought in witnesses to testify and native speakers to translate. First, they used Luis’s language and lack of scarification marks as apparent evidence that he was not “Anchico.” But this alone did not resolve the issue. The parties involved still could not agree whether Luis was “Congo” or “Angola.” Initially, Luis declared himself to be a “Congo.” Upon further questioning, however, he clarified this statement. He said he was actually from a community in a southeastern province of the Kingdom of Kongo that was called Wandu and so he was more accurately of the “Vando nation,” also known as “Mosi Ovando” to many West-Central African people.[22]

The story of Luis Congo takes up less than ten pages of Atlantic Africa and the Spanish Caribbean. And yet, it speaks volumes about our ability to study African identities in the era of the slave trade. One point of the story is that multiple labels made sense as descriptors for Luis’s identity. Being from a community that paid tribute to the Kingdom of Kongo, and that spoke the same language as many people from that kingdom, it made sense for Luis and others that he was labeled that way. And yet, being shipped out of the Portuguese port of Angola, it also made sense that he was labeled “Angolan,” a term which did not have a definite African counterpart anyway. Crucially, if settling Luis’ ownership did not depend on determining his “true ethnicity,” then the court probably never would have inquired beyond either of these broad labels. They never would have discovered that neither of these terms fully captured how Luis understood his African roots. As writers like Northrup and Shumway suggest, Luis’s identity was more particular than the African kingdom his community paid tribute to or the European port out of where he was enslaved. Is this true for most Africans who entered the Atlantic World? Does Luis’ story offer context to the historian Gregory O’Malley’s statement, in his book on the transshipment slave trade, that “even when we can be fairly sure that a group of captives was Aja or Igbo with regard to language or regional background, all we have really recovered…is a general ‘type,’ not individuals.” In other words, we have not necessarily recovered Luis’s identity.[23]

There is another crucial lesson that we can learn from the court case of Luis Congo. That is this: our ability to study African identities through the framework of “ethnicity” is greatly determined by the circumstances of our sources. Europeans recorded “ethnicity” when it related to their interests. Historians today need to remain conscious of how their sources—and the legacy of the colonial contexts in which they were created—structure the conversations that they end up having. It is no coincidence that studies which take “ethnicity” as a main category of analysis—works like Walter Rucker’s The River Flows On, Wheat’s Atlantic Africa and the Spanish Caribbean, Gomez’ Exchanging Our Country Marks, and Hall’s Slavery and African Ethnicities—often cover some rather similar topics. These authors generally conclude that certain African “ethnicities” were more rebellious than others, that certain African “ethnicities” were better at certain types of labor than others, and that certain African “ethnicities” were more receptive to European acculturation. I will not dive into the individual arguments of these authors here. I will only state that, in Africa in America, Michael Mullin once urged scholars to think about how Africans provoked colonial writers to speak about “ethnicity” at certain moments and not at others. The answer is, at least partly, that distinguishing among Africans was relevant to authors who wanted to improve acculturation, explain slave rebellion, rationalize commercial production, and/or resolve financial disputes. We tend to write about African “ethnicities” in the context of assimilation,  labor assignments, and slave revolt not because peoples’ identities were more relevant in those situations, but because those are the situations in which colonial authors saw relevance.[24]

To conclude, groups like the Carabalí, Coromantee, and Congo were not simply African “ethnicities” transferred to the Americas. Rather, they were labels applied to diverse and similar groups of people rapidly undergoing creolization. When modern-day historians project anachronistic labels back onto the past, they further obscure this initial process of African creolization—what Price called “the miracle” of creolization” following Michel-Rolph Trouillot. In searching for the cultural descendants of African identities that developed their meanings in the nineteenth century and later, historians make the mistake of projecting those identities back in time. They write of the “Akan,” “Yoruba,” and “Igbo” in reference to peoples who were actually called by other names—Coromantee, Papah, Eboe—in American records. They write of a “Congo” person who thought of themselves as being from Mosi Ovando. In writing about “ethnicity,” historians run the risk perpetuating a long tradition of obscuring an initial process of creolization. Consider the opening quote of this essay by Phillips. When Phillips wrote that Africans transported to the Americas were “Ceasing to be” Coromantee and, therefore, starting to be “the American negro,” he had skipped over an earlier process in which individual Africans ceased to be Fante, Akyem, and Asante and became “Coromantee.” This process happened at different places and at different moments in history. Unfortunately, it was rarely captured in the records.[25]

African peoples’ identities in the early-modern era were much more complex than historians are able to parse out by using the framework of “ethnicity” alone As Gomez explained in his book (though he seemed to forget at moments), enslaved African people united along many lines of common experience, of which “ethnicity” was only one. From shared experiences of imprisonment in the coastal barracoons, to bonding with “shipmates” in the Middle Passage, to undergoing “seasoning” together on the plantation, slaves forged connections both because of and despite of ethnicity. They united through pidgin languages, through religion and worship, through mixed-ethnic marriages that were both arranged and chosen, through a shared sense of suffering, and through the development of fictive kinship, attempts to run away or create maroon societies, and common labor assignments. For historians to argue that performances labeled as “Coromantee” (like the Jamaican rope trick), “Kromantí” (like the Suriname drum music), or “Efik” (like the Cuban Enllenisón), or that groups labeled as “Carabalí” (like the Cuban cabildo de nacion carabali bricamo apapa) were restricted to the single “ethnicities” implied by their names is to deny or to de-emphasize these many intersecting bonds. It is to combat the perpetuation of fixed stereotypes about race by putting in their place fixed stereotypes about ethnicity.[26]


Provenance: This essay was originally written for a graduate course in the UC Davis History Department. The course was called “HIS 202H: Creole Identities in the Atlantic World,” and it was taught by Professor John Smolenski.

[1] For an example of Phillips discussing these categories as early as 1914, see Ulrich B. Phillips, “A Jamaica Slave Plantation,” The American Historical Review Vol. 19, No. 3 (April, 1914): 545. For “Ceasing to be,” see Ulrich B. Phillips, American Negro Slavery: A Survey of the Supply, Employment, and Control of Negro Labor as Determined by the Plantation Regime (New York: D. Appleton and Company, 1918), 291; Michael A. Gomez, Exchanging Our Country Marks: The Transformation of African Identities in the Colonial and Antebellum South (Chapel Hill: University of North Carolina Press, 1998), 185.

[2] David C. Littlefield, Rice and Slaves: Ethnicity and the Slave Trade in Colonial South Carolina (Champaign-Urbana: University of Illinois Press, 1991); David Wheat, Atlantic Africa and the Spanish Caribbean, 1570-1640 (Chapel Hill: The University of North Carolina Press, 2016); Paul Lovejoy and David Vincent Trotman (eds.), Trans-Atlantic Dimensions of Ethnicity in the African Diaspora (London: Continuum, 2003); and Gwendolyn Midlo Hall, Slavery and African Ethnicities in the Americas: Restoring the Links (Chapel Hill: University of North Carolina Press, 2005).

[3] Gomez, Exchanging Our Country Marks,” 102, 153. Hall, Slavery and African Ethnicities, 47, 112-125. Hall and Law have engaged in this debate in numerous forums. For an example of Law’s perspective, see Robin Law, “Ethnicities of Enslaved Africans in the Diaspora: On the Meanings of “Mina” (Again), History in Africa, Vol. 32 (2005): 247-267. For a summary of the unstable nature of the term “Mina” in the Atlantic World, see Walter Rucker, Gold Coast Diasporas: Identity, Culture, and Power (Bloomington: Indiana University Press, 2015), 235-236 n. 7. Sweet defines “Mina” in Recreating Africa: Culture, Kinship, and Religion in the African-Portuguese World, 1441-1770 (Chapel Hill: University of North Carolina Press, 2003), 16.

[4] Michael A. Gomez, Exchanging Our Country Marks,” 40.

[5] John K. Thornton, “African Dimensions of the Stono Rebellion,” The American Historical Review Vol. 96, No. 4 (October, 1991): 1103. Gomez, Exchanging Our Country Marks,” 39, 103, 104, 124, 140, 175. See Walter Rucker, Gold Coast Diasporas, 2-3. The implication here seems to be that having three country marks on your cheeks was a signifier of being “Coromantee.” Interestingly, there is some evidence that other Africans labeled as “Coromantee” had three “country marks” on their cheeks. However, as is clear from the runaway slave advertisements featured in Gomez’ work, Africans labeled or identified by other “ethnicities” also had these same types of “country marks.”

[6] Deborah Jenson, “Jean-Jacques Dessalines and the African Character of the Haitian Revolution,” The William and Mary Quarterly 69, No. 3 (July 2012): 619-620.

[7] Jenson, “Jean-Jacques Dessalines and the African Character of the Haitian Revolution,” 619-620; Hall, Slavery and African Ethnicities in the Americas, 80-82. James H. Sweet, Recreating Africa: Culture, Kinship, and Religion in the African-Portuguese World, 1441-1770 (Chapel Hill: University of North Carolina Press, 2003), 19

[8] The distinctions between the ports of “Old Calabar” and “New Calabar,” also known as Elem Kalabari, are discussed in David Northup, “Igbo and Myth Igbo: Culture and Ethnicity in the Atlantic World, 1600-1850,” Slavery & Abolition 21, 3 (2000): 2, 8-9. See also Stephan Palmié, “Ecué‘s  Atlantic: An Essay in Methodology,” Journal of Religion in Africa 37 (April, 2007): 284; Paul E. Lovejoy, “The African Diaspora: Revisionist Interpretations of Ethnicity, Culture and Religion under Slavery,” Studies in the World History of Slavery, Abolition and Emancipation, II, 1 (1997): 7; Alexander X. Byrd, “Eboe, Country, Nation, and Gustavus Vassa’s Interesting Narrative,” William and Mary Quarterly 63, no. 1 (January, 2006): 147.

[9] Gomez, Exchanging Our Country Marks,” passim. Gomez discusses each of these broad stereotypes of trends in subsequent chapters dedicated to various regions of West Africa. Chapters 3-6 cover the “Senegambia and the Bight of Benin,” “Islam in Early America,” “Sierra Leone and the Akan,” and the “Igbo and West Central Africa,” (38-153).

[10] Gomez, Exchanging Our Country Marks,” 185.

[11] Gomez, Exchanging Our Country Marks,” 185, 220-221, 223, 291-292.

[12] Gomez, Exchanging Our Country Marks,” 82, 223, 230. When I suggest that Gomez associates the folk with both the field as well as African purity, consider the following lengthy quote: “Among the consequences of such developments was that privileged Africans and African Americans tended to lack proportional representation of certain ethnic groups, whereas the ranks of the field laborers were filled with representatives from all ethnicities. It would therefore have been at the level of the field worker, at the level of the folk, that race as a unifying concept would have been more persuasive, for it was on this plane that the question of identity could be resolved with as little interference from white society (the coercive sphere) as possible. To put it another way, the largest portion of the African-American composite identity was fashioned by rural folk deeply influenced by the African antecedent,” (223).

[13] Even while Gomez makes this argument about “ethnicity” and its alleged connections to privilege in the house rather than the field, he does not present it as dogmatic. In other words, he acknowledges moments of collaboration between “darker-skinned” “field slaves” and “lighter-skinned” or “mixed-race” “house slaves.” See (237-239).

[14] Gomez’ analysis is most troubling in the final line of his book, where he implies a connection between American poverty and the true character of “Africa.” He writes, “The black elite manage to stay afloat with great difficulty. In any event, their condition is not so critical as those relegated to the category of underclass, the least of these, who continue to represent Africa to America,” (292). Byrd also addresses this distinction between defining African “ethnicities” as tending towards certain “traits,” “actions,” and “behaviors,” and as defining them as ongoing “social processes” in the context of Gomez’ work. See Bryd, “Eboe, Country, Nation, and Gustavus Vassa’s Interesting Narrative,” 134-135 n. 37.

[15] Stephan Palmié, “Ecué‘s  Atlantic,” 275-277.

[16] Richard Price, “The Concept of Creolization,” in The Cambridge World History of Slavery, ed. David Eltis and Stanley L. Engerman (New York: Cambridge University Press, 2011), 524.

[17] For the Tabrabah, see Thomas Thistlewood Papers, 1, 532, 26 December 1750; Richard Price, “The Concept of Creolization,” 524.

[18] Michael A. Gomez, Exchanging Our Country Marks, 94-96.  Stephan Palmié, “Ecué‘s  Atlantic,” 283-284.

[19] David Northup, “The Gulf of Guinea and the Atlantic World,” in The Atlantic World and Virginia, 1550-1624, ed. Peter C. Mancall (Chapel Hill: University of North Carolina Press, 189-190.

[20] David Northup, “The Gulf of Guinea and the Atlantic World,” 189-190; David Wheat, Atlantic Africa and the Spanish Caribbean, 50, 155; Rebecca Shumway, The Fantee and the Transatlantic Slave Trade (Rochester: University of Rochester Press, 2011), 21.

[21] David Northup, “Igbo and Myth Igbo,” 1-3, 5. Gomez’ treatment of the Igbo in the US South is about one half of his sixth chapter, “I Seen Folks Disappeah: The Igbo and West Central Africa,” 114-153.

[22] David Wheat, Atlantic Africa and the Spanish Caribbean, 132-139.

[23] David Wheat, Atlantic Africa and the Spanish Caribbean, 132-139; Gregory E. O’Malley, Final Passages: The Intercolonial Slav Trade to British America, 1619-1807 (Chapel Hill: University of North Carolina Press, 2014), 21.

[24] Walter Rucker, The River Flows On: Black Resistance, Culture, and Identity Formation in Early America (Baton Rouge: Louisiana State University Press, 2006); Michael Mullin, Africa in America: Slave Acculturation and Resistance in the American South and the British Caribbean, 1736-1831 (Urbana: University of Illinois Press, 1992), 5.

[25] Richard Price, “The Concept of Creolization,” 514, 534. Price is citing the following work, Michel-Rolph Trouillot, “Culture on the Edges: Caribbean Creolization in Historical Context,” in Brian Keith Axel (ed.), From the Margins: Historical Anthropology and Its Futures (Durham, NC, 2002), 189–210.

[26] Gomez discusses these lines of common experience in the later chapters of his book, particularly “Talking Half African: Middle Passage, Seasoning, and Language,” 154-186.