all photos by the author except where otherwise noted
In contrast to the numerous small-scale gerontocracies found in the area we now know as eastern Nigeria, a strong centralized kingship grew up (or was imposed) sometime before the sixteenth century near the Niger-Benue Confluence, nexus of north-south and east-west riverine trade. Its present identity as “the Igala kingdom” glosses over a complex historical entity that has shifting claims of origin among several regional powers over the centuries (Fig. 1). Spatially, it is contiguous with Igbo/Ibaji settlements to the south and western Idoma settlements to the east. Both of these connections are highly visible in Igala masquerade culture. Beyond that, Igala oral histories allege important historical ties to Yoruba, Benin, and Apá/Kwararafa political formations that also figure in Igala political and visual culture (Clifford 1936, Murray 1949, Boston 1969, Miachi 2012). And the oldest, founding tradition is that the first atta appeared in the form of a leopard with transformative powers. With such a complicated past, it is hardly surprising that Igala lacks a unified canonical version of its mask history.
On the early British exploratory trading expeditions, such as Dr. William Baikie's 1854 voyage of the Pleiad up the Niger (“Kwora”) and Benue, only the narrow coastal strip along the east bank of the Niger south and north of the Idda capital is identified as “Igara” (Baikie 1856). All the rest of what we consider Igalaland today is bundled together on John Arrowsmith's map with what is now known as Idoma, both under the general rubric “Akpoto,” a vague term usually interpreted as “indigenous people” (Fig. 2). In the past I have spent many words discussing the “Akpoto question” as it was applied by the British explorers, traders, and administrators to people who think of themselves as Idoma (Kasfir 1979, 2011: 37–39). Here I will cut that argument short and simply state that, despite all the cultural similarities, there is an Igala language (which is more closely cognate with Yoruba than Idoma), there is an Idoma language (also cognate with Igala), and there is no Akpoto language, suggesting that the label applied to the entire region, “Akpoto,” is another name for “unknown territory” on the part of nineteenth century writers. Later, the British administration came to realize that nearly all of the Idoma-speaking lands (aje) demonstrated close historical ties with the Jukun and the federation of early states known as Apá or the Hausa term Kwararafa. The term “Akpoto came to represent the “indigenous people” who predated the arrival of newcomers following the breakup of the Apá federation in the early seventeenth century and the population shifts it caused among the twenty-odd Idoma-speaking lands. In neighboring Igala, these “owners of the land” (i.e., those who were the first to live there) are the Igala Mela, the nine indigenous clans.
The other major component of Igala politics and culture comes from the Igbo, who are among the earliest migrant settlers occupying the Ibaji area directly south of Idah. They share much of the same masquerade culture although they are not a part of the palace retinue.
Economically, the Igbo controlled the lower Niger River trade south of Idah, at which point Igala traders transshipped it northward. The level of trade on the Niger, all carried by canoes, was a source of wonder to the early British explorers (Laird and Oldfield 1837, 1: 165): “I was surprised to learn from Dr. Briggs that there appeared to be twice as much traffic going forward here as in the upper parts of the Rhine ….” The Igala kingdom, until the mid-nineteenth century when Europeans took control of much of the trade, therefore based its wealth and power on transshipment of all manner of goods, from slaves to necessities such as agricultural products and salt and luxuries such as imported cloth and beads. In this sense, it resembles the Cross River and its hinterland despite the very different form of government. And as I will argue, this trade in commodities provided the economic and historical conditions for the movement of masquerades, a very special type of commodity.
The classic version of Igala history in ethnographic writing posits three successive, or possibly overlapping, periods characterized by Yoruba, Benin, and Jukun hegemony. A parallel structure of political authority that balanced the power of the nonroyal clans, the Igala Mela, against that of the king, along with geographical proximity and closely cognate Yoruba and Igala languages, are the basis for an assumption by colonial writers such as Clifford and Boston of early historical and genealogical connections between the two cultures. This is not an argument that Yoruba kingship itself extended into Igala, but that the linguistic and political markers we recognize as Yoruba may have included a wider region than previously thought.
The evidence for a Benin connection is of a different kind and involves neither language nor the structure of palace authority. Rather, it is based in two other pieces of evidence: the Igala war with Benin, which is recorded by the Portuguese as happening in 1517 and won by Benin, and the distinctive brass pectoral mask worn by the atta, which is of undoubted Benin royal provenance and from the same period. (Baikie, writing in 1854, a half century before the so-called Benin bronzes were “discovered” by the British, said that the large “brass plate” closely resembled “the sign of an insurance office”) (Figs. 3–4).
Boston assumes that this sixteenth century interlude, during which Igala was in some way under the suzerainty of Benin, followed a much longer Yoruba connection. The Benin-period hypothesis accords well with what we know of Benin during the reign of Esigie, when territorial wars were frequently pursued. The third, Jukun-dominated phase of the precolonial and pre-Fulani jihad kingship of the mid-nineteenth century is the one espoused most consistently by the Igala themselves but, as we shall see, it is historically all over the place.
As we have seen, the Igala Mela also controlled the very powerful ancestral masquerade, Egwu Afia (cognate in many ways with Yoruba Egungun and even more with Onitsha's Egwugwu and Idoma Ekwuafia, the Igbo and Idoma versions both referred to as “tall ghosts”). To rival this, the atta, as he consolidated his authority over various indigenous or settler clans (the so-called Akpoto and others), purportedly confiscated (or had established—still an unsolved question) a mask from each. These eight masks are known collectively as the Atta Egwu, and represented the palace's ritual dominance and its identity as “the face of the nation” (Sargent 1988).
The public representation of the atta's political dominance was the Ocho Festival, in which the masquerades all appeared in proper ritual order. While an elegant and logical explanation, this is the construct of a professional historian, R.A. Sargent, based on a series of 241 interviews conducted between 1976 and 1978. It is not, in itself, a collected oral account but an attempt to explain an extremely complex masquerade corpus by piecing together many of its parts.
One the other hand, the most recent atta's own version of how these masks were acquired, collected in interviews by Igala anthropologist Tom Miachi in 2012, is completely different and reasserts a well-known oral tradition: The masks were war booty, left behind on the battlefield by a fleeing Jukun army. My purpose in this essay is to compare the validity of these contrasting historical explanations in light of the visual evidence of the masks themselves.
There are numerous collected stories of masquerades accompanying warriors into battle (Yoruba and Idoma as well as Igala), although this supposed defeat of the Jukun army by the Igala is hard to correlate with the widely held popular belief that the present Igala ruling dynasty comes from Jukun. Despite its universality in Igala oral traditions, historian Elizabeth Isichei points out that Jukun imperialism in the seventeenth century (arrived at through king lists supplied by Baikie in 1832) “accords ill” with Jukun interpretations of their own history (Isichei 1983: 140). So we are left with several discrepant positions concerning Jukun hegemony.
The biggest discrepancy bears repeating: Igala, Yoruba, Idoma, and Igbo are all Eastern Kwa languages, while Jukun is usually classed as Benue-Congo. Igala is 60% cognate with Yoruba, there is no indigenous Akpoto language, and there is no Jukun spoken in Igalaland, even by the royal clan. The second discrepancy is the tendency to attribute too much power to Jukun by equating them with Kwararafa. Arnold Rubin, the foremost researcher of Jukun art, as well as Charles Kingsley Meek (1931) government anthropologist under the British Colonial Administration, both pointed out that the historical “Kwararafa (or Kororofa) Empire” along the Benue was not synonymous with the Jukun only, but was a loose federation of small states that included Idoma, Alago (Keana and Doma), Igala, and Afo (Fig. 5). This is borne out by all my field interviews in these places, and this alone can account for masquerade similarities, as well as those of shrine figures.
Despite its historical importance as a crossroads for precolonial trade, war, and migration, Igala has not been investigated through in-depth fieldwork by any art historian. Its primary study has been by colonial officers such as G.M. Clifford (1936) and Kennth Murray (1949), social anthropologists such as John Boston (1969, 2011) (Figs. 6–7), Chike Dike (1976), and Tom Miachi (2012), and historian Robert Sargent (1988), which means that issues of style and visuality—how objects appear and are apprehended visually—have been largely ignored. There is one exception: Cornelius Adepegba (n.d.), who is discussed below but who also did not do fieldwork.
Instead, writers on the historically puzzling Igala masks owned by the royal palace at Idah have argued either for the evidence found in oral accounts or for that seen in their distinctive facial scarification patterns in order to trace their origins. Art dealers, and by extension their clients, who are not typically versed in local histories, almost always use facial markings to identify masks, while anthropologists focus on their social significance, and historians on oral or documentary accounts to explain their meaning.
The masks illustrated in Figures 8 and 9 are actually a nonroyal version, called Agba, whose strong visual resemblance connects them with Jamadeka and other royal masks kept at the palace. They are from the southern reaches of the Igala Kingdom known as Ibaji, a hybrid area settled long ago by Igbo lineages who gradually “became” Igala. But it is also very close geographically to Idah, the seat of the Igala royal palace, which means that its people have long been exposed to the public outings of the Atta Egwu, the royal masks that Agba resembles. Genealogically, this means that we will need to repeat the same arguments for both the royal and nonroyal versions.
One of the more fascinating historical questions is, Did masquerades actually accompany warriors into battle in former times? Where and why? We ask this because, as already mentioned, Igala tradition states that some of the royal masks were Jukun, captured or left behind on the battlefield in an Igala-Jukun war. Let's call this the “palace account” and return to it further on.
By contrast, markings on the face or body surface are a popular art market attribution method when the stylistic identification is ambiguous. It is not as reliable as people usually think, partly because facial and body markings are not only visual symbols for identity but also because they obey the rules of fashion, as well as strategic political alliances. And like fashions, they change over time. In addition, many patterns spread regionally in this way are not exclusive to a single culture. It is for this reason that art dealers and often collectors—even, occasionally, experts—confuse Idoma and Igbo whiteface masks (Figs. 10–11).
The earliest published evidence of facial marking was found in accounts by nineteenth century travelers, such as Siegfried Passarge (1895), followed by early twentieth century colonial anthropologists (Meek 1921), which included illustrations of everything from cicatrization to raised keloid scars to tattoos. In Meek's chart the Igala (Fig. 12) are shown with the same long, curved scars from the corners of the mouth to the temples as are seen on many of the royal masks, suggesting an Igala origin.
However, we know from anthropologist Paul Bohannan (1956) that, for example, the Tiv abaji markings that were all the rage around 1915 (Fig. 13a–b), fell out of favor a few years later with Tiv women, who refused to marry men who had them. This apparently caused great anguish for the men because the abaji were keloid scars, raised welts caused by hooking and lifting, not just incising the skin, and so were impossible to remove. The ekwotame figure seen in Figure 14 was in an Idoma village in the 1970s, before it disappeared into a private collection somewhere, but it had been purchased in Tivland by the grandfather of the then owners in exchange for a slave, a dane gun, and a brass bracelet, dating its purchase to about 1875. It has the characteristic Tiv female navel scarification and the raised facial keloid scars similar to abaji. What this does not tell us is whether or not it was made by a Tiv artist. Iconographically yes, but stylistically, very unlikely. Not only that, but with their shared boundary and frequent cross-cultural borrowing, both navel scarification in a mudfish pattern and abaji scars are also found in occasional Idoma masks and figures, attesting to the designs’ ability to cross boundaries.
The sole work that focuses on Igala mask scarification was written by Cornelius Adepegba, a Yoruba art historian and student of Roy Sieber at Indiana University in the 1970s. After completing a PhD dissertation on scarification practices in Nigeria, based not on fieldwork but on the existing literature, he turned to the matter of using these designs to identify works of art, including the puzzle of the Igala royal helmet masks.
The Igala royal helmet masks look distinctly different from other large Igala masks such as superstructure types, which roughly resemble Idoma and Igbo ones with their combination of birds, people, and animals (Figs. 15–16). The most striking difference is the helmet type itself, which functionally affects its style, since it must be very wide-necked if it is to be carried this way. They also are often fully scarified and wear an abrus-seed covered hairpiece or helmet.
And as Adepegba pointed out, they share certain scarification patterns with several maternity figures found in different places in the Benue Valley between the Niger-Benue Confluence and Wukari. His idea was to identify the Igala royal masks, at least the three or four helmet types, with one of these figures and verify their common origin. I'm afraid the evidence is strewn with further questions, and I'll try to show why.
The best fit of these was a female figure then in the Tishman Collection, now at the Metropolitan Museum of Art (Figs. 10, 17). The supposed genealogical connection lies in the scarification lines extending from the corners of the mouth upward to the hairline, as well as the overall vertical scarification on the face. But here the resemblance really ends, and even if we accept the connection, the origin of the figure itself is not at all clear. I wrote about this figure many years ago in a catalogue of the Tishman Collection (Kasfir 1981) and then again in Central Nigeria Unmasked: Arts of the Benue River Valley (Kasfir 2011). It has been called Afo because of its broad resemblance to several maternity figures found scattered throughout the north bank of the Lower Benue valley (i.e., not in Igalaland) and widely accepted as Afo in William and Bernard Fagg's deductions, even though none were found in an Afo village (Figs. 18–19).1
The problem is that the shared scarification patterns between Igala helmet masks and these figures are found throughout the lower Benue Valley in this period, roughly the 1870s to 1950s. So in the end, all the scarification argument really tells us is that, at the turn of the twentieth century, when many of these figures were collected by German and British scholars or administrators, these similar patterns of facial markings were indeed a widespread regional practice.
Now I want to turn to the other argument, put forward by oral historians and social anthropologists such as Robert Sargent (1988) and, most recently, Tom Miachi (2012), both building upon the earlier writings of John Boston (1969) and Miles Clifford (1936). Robert Sargent, the historian invested in a big-picture approach, says that the Igala royal masks are each associated with a particular local clan that was incorporated, over a long period of time, into the kingdom and thus brought under the atta's authority. They are literally and historically “the face of the nation” (Sargent 1988: 17). The problem is that the reputed clans of origin for the masks are different in Sargent's and Miachi's accounts, both of them based in extensive interviews done in the 1970s in Sargent's case, and intermittently over thirty-plus years in Miachi's. There's another important assumption made in all the oral accounts, namely, that the Igala kingship is, and always has been, of external origin.
Following the periods of Yoruba and Benin influence, the power of Kwararafa, dominated by the Jukun, supposedly expanded and the local population was forced to pay annual tribute to the Jukun. Finally they dug in their heels and refused, at which point the Jukun armies invaded. Here is where things get murky. In the most recent (but also long-running) fieldwork of the Igala anthropologist Tom Miachi (2012), the account of a war with Jukun in which Jukun masks were left on the battlefield as their soldiers fled is recorded from his interview with the atta of Igala himself. But this atta, who died in 2012, insisted that these events happened in the 1400s, while there has been general agreement among scholars that the Kwararafa federation began breaking up two hundred years later—around 1630—following an invasion from Bornu. We know from Portuguese accounts that the Benin period is reliably dated to their victory over Igala in a 1517 war, and the apparent time sequence of Igala dynasties is Yoruba, Benin, Jukun. (I'll explain later what I mean by “apparent”). So the atta's account is off by a couple of centuries and also reverses the order! Not only this, there is a logical inconsistency: If the Jukun lost the war and were driven out, why is the Igala kingship still thought of in the present day as in the hands of a Jukun-derived dynasty?
There is an answer to that question, but I bring this up because it reveals the serious problem with taking oral accounts too literally. Historian Jan Vansina spent his career studying oral tradition in the Kuba kingdom and, by the 1980s, he concluded—contrary to his and others’ views, dating from the early days of African Studies in the 1960s—that oral traditions are not the “raw material” from which the history of nonliterate cultures can be constructed (an idea many historians have let go of reluctantly or not at all). Oral traditions are actually already mediated histories in themselves (Vansina 1985: 22). The extreme statement of this position would be that oral histories are essentially politically useful inventions, or “mythical charters” as Bronislaw Malinowski once put it. John Boston, another anthropologist, was more subtle about this when he stated in 1969 that Igala tradition “takes a synoptic view of the past,” compressing (and, we could add, deleting and sometimes reversing) events that played out over several centuries or, conversely, re-ordering events that were not really sequential at all (like the supposed Yoruba-Benin-Jukun ruler sequence) but were simply different threads of a complex tapestry.
So back to enemy helmet masks left behind on the battlefield: Both the Yoruba and the Idoma have detailed oral accounts, the former collected by John Willis (2011) on Egungun and the latter by me, of masquerades accompanying Idoma-Akweya warriors into battle.
Why bring a masquerade onto a battlefield? The answer to that was brought home to me when I saw and heard for the first time the Akpnmobe masquerade belonging to the Ekpari clan (Fig. 20): it shouts, “Akpnmobe olebe oche!” (loosely, “I eat the meat of human beings!”). It is also about eight feet tall and carries a spear, and to see it moving through the elephant grass some distance away is not something one ever forgets.2
But I would submit, following Boston's and Vansina's views on the matter, that the explanation for the origin of the Igala royal masks probably should not be taken too literally, even if masks did indeed appear on battlefields in precolonial Nigeria. Rather, Boston (1969) argues that the story symbolizes the creation of a fully Igala nationality—or as Sargent says, “the face of the nation.”
Finally, when we look at actual Jukun helmet masks (Figs. 21–22), a few, such as Aku Washenki, bear close resemblance to certain Igala royal masks. But it is not necessary for us to place them on a battlefield in Igalaland to make the historical connection. There is another whole set of Igala traditions—reinforced in neighboring Idoma, Doma, and Keana—that they all, including Igala, were once part of the Kwararafa (Apá) federation further east on the Benue and thus were in close cultural contact with the Jukun. While this is a less exciting explanation than abandonment on the battlefield, it is a good deal simpler and therefore more plausible.
The Afo also claim that connection, and they are both linguistically and geographically very close to Idoma. So, as with the maternity figures that have scarification patterns similar to some of the masks, I prefer to believe that we are looking at a regional, lower- to mid-Benue Valley phenomenon, extending all the way from the Niger-Benue confluence to Wukari, the Jukun capital, and very possibly deriving originally from the Kwararafa federation of states before it broke apart in the early seventeenth century. As Adepegba realized, body markings are seemingly striking evidence, but only as rough boundary markers, and without these complicated oral histories that are not only heavy with symbolism but also at least partly mythical, we would not be able to guess where these distinctive Igala helmet masks originated—or for that matter, the maternity figures with similar scarification.
This paper derives from an earlier version presented at the Baltimore Museum of Art in 2015 during a symposium honoring a new installation of the art of eastern Nigeria.
The one exception was photographed by Bernard Fagg in an Afo village in the 1940s, but its first publications was in the Arts of the Benue (Berns, Fardon, and Kasfir 2011).
There are two versions of this powerful mask. One is now in a private collection in Paris and originated in a neighboring village to Otobi. Both belong to the Ekpari clan of Akweya speakers, who were originally from Ogoja but settled in Akpa district in the nineteenth century or earlier and were Idoma-ized under the British administration.