The Debate: On Style and the Premises of Artifact Production

Several observers have remarked that the Luso-African ivory saltcellars are not all carved in the same style. And on this basis they have suggested that the ivories must not all come from the same place. I have argued that almost all of the ivories should be attributed to artists whom sixteenth- and seventeenth-century Portuguese writers refer to as “Sapes.” Here, I shall demonstrate that the argument for Nigerian origins for the saltcellars has several flaws. First, it is based on the misunderstanding of who the “Sapes” were and where they lived. Second, it entails an oversimplification of stylistic analysis as method and a misunderstanding of its potential efficacy.

The provenance debate stems in part from terminological confusion. The “Serra Leoa” of sixteenth-century Portuguese sources is not the same as modern Sierra Leone. Serra Leoa extended from present-day Sierra Leone northwest along the coast, to the Rio Grande (in Guinea-Bissau). And the “Sape” were not an ethnic group. The word was an umbrella term that encompassed several ethno-linguistic groups who in turn inhabited the Upper Guinea Coast from northern Sierra Leone, through Guinea-Conakry, possibly into southeastern Guinea-Bissau (to use the post-independence appellations). The “Sape” comprised several groups – Bulom, Temne, Nalu, Landuman, some Baga, and, by the late sixteenth century, a cultural admixture of “Manes” (Mande-speaking people).

The Upper Guinea Coast with “Serra Leoa” broadly indicated

The argument for Nigerian provenance of the post-1530 ivories (Bassani) or the post-1550 pieces (Curnow) is also chronologically flawed. It disregards the fact that “Sape” ivory carving, as documented by contemporary Portuguese and Cape Verdean sources, lasted for a period of over one hundred and fifty years. Written documentation exists, in the form of five reliable eyewitness observers (four merchants and one missionary), plus a sixth writer who himself relied upon informants who traveled on the Upper Guinea Coast. These men, Portuguese and Luso-Africans,^{Footnote 6} wrote, respectively, in 1505 and 1506,^{Footnote 7} 1593–1594,^{Footnote 8} 1615,^{Footnote 9} 1625,^{Footnote 10} and 1669/1684.^{Footnote 11} All of the writers through 1625 used the present-tense to describe the society of the “Sapes” and their commerce in ivory. Four of them specifically described ivory carving.

Arguments derived solely from apparent stylistic similarities cannot serve to refute dated, definitively located evidence provided by multiple primary written documents. The argument for Benin origins lacks all but the most cursory contemporary written documentation (see below), while documentation for “Sape” origins is detailed and precise. “Sape” artists carved ivory for export by the Portuguese, who highly prized the work, from the late fifteenth century through at least the early seventeenth century.

For stylistic comparisons to tell us anything meaningful about provenance, it is necessary first to establish the specific historical and cultural context that gave birth – and meaning – to the objects in question. Where contemporary written documents directly address the historical and cultural context and even discuss the artworks themselves, those sources must be given priority in establishing provenance and chronology of the artworks in question. When, furthermore, multiple independent written sources exist, enabling us to corroborate the individual descriptions, it would be both ahistorical and irresponsible to ignore those sources.

As art historical method, stylistic comparison dates to the late nineteenth century; its classic expression is in Bernard Berenson’s lists, which attribute Italian Renaissance paintings to individual artists on the basis of style. Berenson, however, began with a firm body of work: the paintings were Italian, “trecento” and “quattrocento,” their provenance and dating already broadly known, and many of the works were firmly documented. So there already existed a scaffolding, to which Berenson attached additional works using stylistic analysis. Berenson does not use style to compare unknown to unknown; he begins with broadly identified works.^{Footnote 12} Thus, only given sufficiently precise contextual documentation, stylistic comparison can be a highly effective art historical tool.^{Footnote 13}

No such historical information exists to frame a stylistic analysis of the saltcellars. More important, existing historical data does not confirm the theory of Benin origins, except for some spoons. In a recent publication, Jean-Michel Massing, author of catalogue essays for the Smithsonian and for other major museums, attempts to document the export of Benin ivory carving to Portugal, citing the early sixteenth-century Portuguese traveler Duarte Pacheco Pereira. Or rather, Massing cites Ezio Bassani, who, in 1988, referred to Pacheco Pereira – Massing does not cite the original source. As Luís Afonso and José da Silva Horta have recently observed, what Pacheco Pereira actually writes in Esmeraldo de Situ Orbis, is: “There are many elephants in this land, and we often buy their teeth, which are called ivory (‘marfim’).”^{Footnote 14} Afonso and Horta observe that Pacheco Pereira also gives this same passage, verbatim, for Fernando Pó, but he observes that in Serra Leoa “[t]he most refined [spoons] are made here of ivory, the best work that is to be found anywhere.”^{Footnote 15}

As the Governor and Captain of the Portuguese trading fort at El Mina, Pacheco Pereira was the official responsible for commerce throughout the region. He had also visited Benin. If any European observer was familiar with the production of carved ivory, it was he. As Afonso and Horta state: “He clearly prizes the quality of Serra Leoa carving, while for Benin he only mentions the availability of raw ivory.”^{Footnote 16}

Massing points out that two contemporary sources (1588 and 1621) mention ivory spoons from Benin.^{Footnote 17} It is likely, as he asserts, that some surviving spoons did originate among Bini carvers. However, the inherent weakness of arguing solely on the basis of “style” becomes clear when Massing rejects a late-sixteenth or seventeenth century “Sape” origin for any surviving spoons, this despite the eye witness account of Father Alvares (1615), the evident high regard the missionary held for these objects, and the fact that Alvares offers iconographic interpretations that fit some of the spoons.^{Footnote 18} Massing writes: “That the later surviving spoons, of a homogenous type, were produced in Sierra Leone cannot be accepted, as one of them (…) is directly related to a figure on a Bini-Portuguese saltcellar.”^{Footnote 19} Massing refers here to a group of seventeen saltcellars for which he argues, on stylistic grounds, for Nigerian origins. However, no contemporary source establishes the date or provenance of any of these works.^{Footnote 20}

Quite aside from the fact that the saltcellar (in the Ethnologisches Museum in Berlin, see Figure 1) to which he compares the spoons has no documented origin,^{Footnote 21} there is a fundamental fault in Massing’s logic. He attributes a large group of spoons to Benin. I find the argument (including the shape of the bowl of the spoons) weak.^{Footnote 22} But even were he correct in this attribution, he has no basis for arguing that this group of “Benin spoons” comprises all surviving late sixteenth/early seventeenth century spoons. Here is an example of the logical difficulty the art historian faces, who would have the argument from style trump existing, contemporary written documents.

Figure 1. Saltcellar, ivory, Serra Leoa, sixteenth or early seventeenth century. © D. Graf. Ethnologisches Museum der Staatlichen Museen zu Berlin –Preußischer Kulturbesitz. Identi. Nr. III C 4890 ab.

I recently spent time at Berlin’s ethnographic museum, comparing the Benin leaded-bronze images (one free-standing figure and one plaque) of horsemen, and a saltcellar depicting cavalrymen (see Figure 1). I could see no close stylistic similarities between the two Benin pieces and the saltcellar. The Benin animals and riders are hieratic, frontal and immobile, images of royal authority. The saltcellar cavalry are, in marked contrast, dynamic. The horses’ heads thrust diagonally across the surface, energizing the composition. As for the figures – who, with their beards and long hair, may represent Europeans – it is difficult to compare a subtractive medium (ivory) to an additive medium (the plaques were made using the lost-wax process, so that details are built up). For this reason, a close comparison of such details as eyelids or nostrils is almost impossible. The eyes in the ivory are outlined and defined by incised curving lines, while the eyelids on the bronzes are thin, tubular protrusions. I see no close similarity between the treatment of the bronze faces, and the hatchet-faced Europeans on the ivory. Fagg is correct!

The Argument for Serra Leoa (“Sapes”) Provenance

We know that some of the spoons and saltcellars are from Serra Leoa because they are documented in the Lisbon customs records (only the records for 1505 survived the earthquake and fire of 1755), and because they are described by Pacheco Pereira (who spent time as the Governor of the Fort of El Mina), and by Valentim Fernandes (his informants were Portuguese ship captains who had been to Guinea). Both sources also date to 1505–1506. The stylistic similarity of figures on some of the saltcellars to the “nommoli,” stone figures, of unknown date, which are occasionally unearthed in coastal Sierra Leone, also supports a Sierra Leone provenance.^{Footnote 23} The stylistic comparison, once made, allows us to consider both saltcellars and olifants, because of close similarities in the treatment of human form, especially the exaggerated details of faces – eyes and nostrils – that are common to all of these works.

I have reattributed the post-1550 works to Serra Leoa and specifically to the “Sapes.” Bassani and Curnow attribute them to Benin. My re-attribution relies on and is made by reference to: a) the context of Portuguese-West African commercial contact and of Luso-African settlement in Serra Leoa; b) an apparent absence of direct commerce between Portugal and Benin from 1520 to about 1580; and c) contemporary Portuguese eyewitness accounts that clearly document the continued production and trade of Serra Leoan ivories into the seventeenth century.

Contemporary documents written by observers who were in Serra Leoa (especially Father Manuel Alvares) are paramount. Stylistic comparison is reliable only after one can delimit the chronological and spatial connections. Without any documentation of a connection, perceived stylistic similarities could be due to any of a number of factors: 1) coincidence; 2) our perception of “style” as opposed to the artists’ different aesthetic criteria; 3) historical factors such as common influence by other, perhaps lost, intermediary objects. To give two examples of reasonable stylistic comparison, I can safely and with good results compare Mantegna to early Giovanni Bellini, or an ivory hip mask from Benin to a bronze plaque from Benin. In both cases, the chronological and spatial ties are independently established.

If one looks at the process whereby the initial Benin attribution was made, it becomes clear that Curnow failed to consult Portuguese-language historical sources, and failed to establish an a priori case for why the ivories should have stopped coming from Serra Leoa. In fact, on that point she is dead wrong. The historical reinterpretation of the “Mane invasion” – once thought to have destroyed “Sape” society ca. 1550, now understood to have been a migration followed by cultural assimilation of the “Manes” into “Sape” society – cuts the legs out from under her argument, which Alvares’ testimony then completely disproves.

Let me return to the matter of apparent similarities in the representation of “Portuguese” in Benin plaques and in some of the ivories from Serra Leoa. The sudden appearance in two cultures, that have little or no direct connection with one another, of strangers with a strange physiognomy may engender similar – or superficially similar – artistic representations. The same aspects of the foreigners’ visage appear most extraordinary, and hence worthy of depiction. The resulting images may emphasize or exaggerate the same elements. In late medieval representations of Black Africans in the Rhineland and in Italy, for example, there are representations of Africans with exaggerated thick lips and with similar treatment of the hair. And the reason is not surprising: European artists in diverse locales depicted Black Africans in similar manner, because they were most struck by the same elements in an unfamiliar physiognomy. One could say the same of West African depictions of Europeans during the first hundred or hundred-and-fifty years of contact.

All of us who study African art are familiar with the stereotype of pre-colonial or “traditional” art as unchanging or static. And none of us would deny the existence of regional, if not local, or even individual stylistic variations. Yet, in the case of the Luso-African ivories, some scholars have argued that there is an anomaly. The ivories were carved by “Sapes” peoples, that is, Bulom, Temne, Nalu, Landuman, Baga, but to explain stylistic variation among the ivories, some scholars have felt it necessary to attribute them to a distant and completely different culture. And this is despite abundant contemporary eye-witness documentation that the works come from Serra Leoa, that is, from Sierra Leone, Guinea-Conakry, and southern Guinea-Bissau. These scholars argue that all of these ethnic groups should/would have developed one common carving style. Their implication is that this one homogenous style should/would have remained static over a period of hundred-and-fifty years. To the contrary, I argue in this aricle, it would be inconceivable that all of these groups shared only one style across the period from 1500 to 1650.

The Ivories as Images of Global Commerce

A key to the stylistic differences that may be observed in some of the Sierra Leone ivories, and which have led observers to attribute some works to Benin, may be found in one of the olifants discussed by Bassani.^{Footnote 24} This piece is number 105 in his 1988 catalogue raisonné. It was in the Kugel Collection, in Paris. The bell of the hunting horn is adorned with a relief of an elephant with a howdah^{Footnote 25} on its back.

The Indian theme is probable evidence of a date after 1580. Bassani, however, who attributes all post-1530 pieces to Benin, engages in intellectual contortions to ignore that evidence. He writes: “The elephant with the howdah that appears near the wide end is typically Indian. No Indian miniatures predating 1600 with this subject have been found.”^{Footnote 26} To avoid the later dating, he proposes earlier European engravings as models, even though these engravings do not show a convincing similarity of motif to the ivory. This association allows him to identify the piece as “Sapi-Portuguese, ca. 1490–1530.” In his desire to fit “Sapi” pieces into his chronology of “pre-1530 = ‘Sapi;’ post-1530 = Benin,” Bassani misses the historical evidence that the howdah provides.

The Portuguese sea route to Asia, the “Carreira da India,” included a stop for water on the return voyage in Sierra Leone. Commerce from India also flourished to Senegambia and Cacheu, where Indian swords and tapestries were highly prized and are documented on at least three occasions between 1610 and 1620.^{Footnote 27} Father Manuel Alvares, in 1615, writes that the King of Bussis, who was the largest slave trader on that part of the coast, owned some of these Indian silk tapestries, known as “colchas.”

The ivory elephant with a howdah on its back does reflect Indian influence. Most probably that influence was conveyed through “colchas.” These textiles were imported from India to Upper Guinea beginning around 1580. The ivory horn, with its distinctive style, is indeed from Sierra Leone.

This olifant may also help us to correlate stylistic variation with dating. Adjacent to the elephant is a hunter blowing an olifant and walking beside a quadruped. These “Sape” artists clearly enjoyed a visual joke; they frequently use ivory, as here, to represent itself. The face of the hunter is in a style that is frequently assumed to reflect extra-Sierra Leonean origins. This style is characterized by incised eyes, integrated features, and a “Greek” nose profile. This olifant is probably nearly a century later than the “Sape” style of flared nostrils, bulbous eyes, and protuberant facial details. Here then, by correlating the subject matter and the style, we may have a key to the later chronology of the “Sape” ivories.

Beyond the Debate over Provenance: Creole Art and Intercultural Mixing

Ironically, the debate over provenance – Nigeria vs. Serra Leoa – may itself be an oversimplification. The Luso-African ivories are indisputably the product of inter-cultural mixing. All scholars of the ivories acknowledge the presence not only of local African imagery, but also of European themes, including Portuguese royal insigniae and Christian subject matter. Further, the presence of themes from Indian “colchas” reflects the fact that this is an art of globalization. Yet, the possibility of inter-African cross-fertilization is excluded by the effort to establish a single, specific ethnic origin for all the ivories, or one for the early ivories and another for the later works.

These objects were highly portable, and they travelled. Throughout the sixteenth and early seventeenth century, Portuguese and Luso-African traders or lançado traders engaged in coastal traffic that linked different African communities. We know that “Sape” ivories traveled north to The Gambia; it is conceivable that some pieces were brought to Benin. The Portuguese, who were responsible for the growth of an overseas ivory market from Serra Leoa by 1506, may actively have encouraged the development of ivory carving elsewhere in West Africa to meet European demand. Clearly, the Portuguese brought illustrated prayer books to Serra Leoa to serve as models for “Sape” artists. If we accept the predominant – but to date undocumented – view that indigenous Benin ivory carving already existed by the sixteenth century, then might not Portuguese merchants have brought examples of “Sape” ivories to Benin, to serve as models for local carvers? We might even go so far as to hypothesize the movement of individual artists, under Luso-African patronage, between Nigeria and Upper Guinea, or between both places and the Cape Verde Islands. The field for future research – or at least for hypothesis – is wide open.