Navigation – Plan du site

AccueilNuméros18Brass Working and Mforowa Manufac...

Brass Working and Mforowa Manufacture among the Akan of Coastal Ghana during the 17th–20th centuries

Travail du laiton et production de mforowa parmi les Akan de la côte du Ghana du xviie au xxe siècle
Christopher R. DeCorse
p. 11-38
Traduction(s) :
Travail du laiton et production de mforowa parmi les Akan de la côte du Ghana du xviie au xxe siècle [fr]

Résumés

Cet article traite de la fabrication et de la datation des ornements, outils et boites en laiton produits par les Akan dans les régions côtières du Ghana, en particulier les mforowa (sing. forowa), récipients en feuilles de laiton spécifiques utilisés pour y conserver le beurre de karité et qui sont associés à des rituels mortuaires. Les fouilles archéologiques ont permis de retrouver une variété d’artefacts fabriqués localement à partir de laiton européen importé, souvent en étroite association avec des objets de traite européens. Comme les gammes de production de nombreuses manufactures européennes sont connues et peuvent être bien circonscrites dans le temps – souvent à quelques décennies près –, les marchandises de traite offrent un moyen de dater précisément les artefacts qui leur sont associés et de manière plus assurée que ce qui est souvent possible pour les sites archéologiques africains des 500 dernières années. Cet affinage chronologique apporte un élément nouveau sur les traditions de travail des alliages de cuivre parmi les Akan côtiers, en particulier l’utilisation et la réutilisation du laiton importé. Cet article passe en revue les connaissances actuelles sur les origines de la métallurgie et la transformation du métal dans la région côtière du Ghana, et sur le développement du travail du laiton avec l’avènement du commerce européen. Les données archéologiques provenant de la ville africaine d’Elmina suggèrent que la production de mforowa a débuté au xviie siècle et que ses origines stylistiques trouvent leurs racines dans la tradition antérieure des nkuduo (sing. kuduo) en laiton coulé. Le contexte des découvertes archéologiques fournit également un aperçu des cadres culturels dans lesquels les mforowa ont été utilisés.

Haut de page

Texte intégral

Coastal Akan Metal Working

  • 1 Akan refers to an ethnolinguistic group that includes the Akwapim, Akyem, Anyi, Aowin, Asante, Bono (...)
  • 2 Syracuse University’s Central Region Project has surveyed 100s of sites in central and western coas (...)

1The focus of this article is the metal working traditions of the coastal Akan, particularly the Fante who inhabit the Central and Western regions of coastal Ghana (Fig. 1).1 The Akan are known for distinctive artifacts made from both cast and sheet brass, including vessels, implements, and gold weights (e.g., Cole 1979; Garrard 1979, 1980a, 1989; Cole & Ross 1977; Herbert 1984; Quarcoopome 1997; Ross 2002, 2009, 2014; Silverman 2014). Metal working, including iron smelting, brass and other copper alloy casting, and gold working in Ghana long pre-date the arrival of Europeans on the coast in the late 15th century. The earliest dates for iron smelting in coastal Ghana are from the early to mid-1st millennium AD, ages that are comparable to those currently available for central and northern Ghana (e.g., Chouin 2009: 719–727; DeCorse 2005: 46; Reid 2022a: 60, 2022b; Shinnie & Kense 1989: 207–210; Spiers 2007: 206–207; Stahl 1994: 63–64).2

Figure 1 – Map of coastal Ghana showing locations mentioned in the text

Figure 1 – Map of coastal Ghana showing locations mentioned in the text

© C.R. DeCorse

  • 3 Aside from the isotopic characterizations of the Elmina and Eguafo brass vessels discussed, the ele (...)
  • 4 This observation was made by Timothy Garrard (1980a: 104), who also noted that early castings may h (...)

2Evidence for the working of copper alloys and copper metallurgy dates substantially later than iron production.3 The current supposition, based on archaeological data, art historical evidence, documentary sources, and oral traditions suggests that the casting of copper alloys, as well as gold, was introduced to the Akan area during the 14th century, probably via the trans-Saharan gold trade (Garrard 1979, 1980a: 3 & 102–104, 1988: 4; Silverman 1983a: 175–181, 2019). The production of brass and copper alloy objects during the pre-Atlantic period was limited by the comparatively small amount of metal that could be transported across the Sahara. Imported brass would also have been more readily available closer to the trans-Saharan trade networks, than in the southern Akan forest and coastal margins.4 The scarcity of these copper alloy objects likely limited their use to specialized ritual and mortuary contexts (Chouin 2009: 720).

Figure 2 – Two crotula (paired bells) excavated at Elmina from 18th and 19th century mixed fill contexts

Figure 2 – Two crotula (paired bells) excavated at Elmina from 18th and 19th century mixed fill contexts

Artifacts such as these were likely produced in northern Ghana and hint at the wider trade networks with which the coastal emporia were connected

© C.R. DeCorse

  • 5 Stylistically, the crotula are similar to ethnographic pieces associated with the Frafra in norther (...)

3The advent of European trade and the subsequent increasing availability of brass undoubtedly led to an increase in Akan casting, but evidence for casting from the coastal hinterland is limited. Examples of vessels, gold weights and other cast objects have been recovered, but these do not necessarily represent local production in the coastal hinterlands and manufacturing sites further in the interior may be more likely (e.g., DeCorse 2021a: 129–130; Nunoo 1957: 15; Spiers 2007: 202–204). For instance, two crotula from Elmina likely represent northern Ghanaian production (Fig. 2; DeCorse 2021a: 130).5 The best evidence for the casting of brass and copper alloys on the central Ghanaian coast comes from crucibles and mold fragments from 17th century contexts at Nsadwer Bosomtwi, two kilometers north of Dutch Komenda. Chouin (2009: 718–719) suggests that the concentration of slag and casting debris represented may be indicative of a specialized iron smelting and smithing complex associated with this settlement during the 17th century. He further observes that such a production center would have been advantageous for a small-scale casting industry utilizing scraps of European imported brass, copper, and lead (Chouin 2009: 718).

  • 6 The more complete of the Elmina crucibles has an exterior diameter of 5.5 cm and was at least 6 cm (...)
  • 7 De Marees referred to kacraws or kakaras, a type of faux gold dust used as currency at Elmina and s (...)

4Additional evidence for the reuse and casting of copper alloys is provided by fragments of two crucibles from the site of Elmina, both from pre–1873 mixed fill contexts (Fig. 3).6 Another late 19th century context at Elmina produced a gather of partially melted brass, including a brass musket side plate, a copper bracelet fragment, and brass filings. These remains may relate to the production of faux gold dust.7 Notably, the melted brass material was found east of Elmina Castle in a context more likely associated with the European garrison than the African settlement.

  • 8 Similar crucibles, mold fragments and casting debris were recovered from the sites of Asebu in the (...)

5The above evidence indicates the presence of brass and other copper alloy metallurgy on the Central Region coast during the 17th century. The crucible fragments from Bosomtwi and Elmina are small and similar in form to examples recovered from other Ghanaian sites, as well as from other parts of West Africa. However, here it should be noted that the amount of casting material recovered from the central coast thus far is limited compared to some interior sites, particularly Begho, where over 500 crucibles used for casting copper alloys were recovered (Garrard 1980a: 41–42; Herbert 1984: 88, 94–95).8

Figure 3 – Crucible fragment from a pre–1873 mixed fill context at Elmina

Figure 3 – Crucible fragment from a pre–1873 mixed fill context at Elmina

The crucible’s paste is highly vitrified and distinct in appearance from other local ceramics

© C.R. DeCorse

  • 9 This observation is based on the examination of museum pieces attributed to the Akan, as well as ob (...)
  • 10 There is occasional confusion in the classification of nkuduo and mforowa in museum collections, so (...)

6Akan casting technology was primarily based on cire-perdue (lost wax), the cast pieces finished by sanding and polishing.9 The Akan produced a variety of cast brass objects including spoons, basins, gold weights, and nkuduo (Silverman 2014). Nkuduo are ritual vessels that may have subsequently inspired the manufacture of mforowa, locally constructed sheet brass vessels that partially overlapped with nkuduo in the cultural contexts in which they functioned (Cole & Ross 1977: 65–68).10

  • 11 Forming alluvial gold such as that found in coastal Ghana does not require smelting or melting. Lar (...)

7Akan artisans also worked with gold. Beginning in the 15th century, the first European accounts describe people of the Ghanaian coast bedecked in gold ornaments. However, early evidence for the fabrication and, especially, the casting of gold artifacts from coastal sites is limited. The earliest gold artifact recovered in Ghana is from the Coconut Grove site west of Elmina, which produced a single gold bead and iron artifacts securely dated by a series of thermoluminescent and C14 dates to the 1st millennium AD (DeCorse 2005: 47, 2021a: 126–127). Notably, the gold bead was produced using formed gold foil crimped together, as opposed to casting.11 Another gold bead from a probable pre-15th century context at Bantama west of Elmina was also made from cold-worked gold foil (Calvocoressi 1977: 130; DeCorse 2021a: 127). In contrast, cast gold and copper alloy beads have been recovered archaeologically from a number of 18th through 19th century contexts in central coastal Ghana and beyond (e.g., Agorsah 1975; Bellis 1972: 61–62; DeCorse 2021a: 126–128; Ehrlich 1989: 52; Garrard 1980b: 85–155; Spiers 2007: 183–184, 231–232; Kankpeyeng & DeCorse 2004: 117).

  • 12 Garrard (1980a: 105–107) discusses this and other 17th century documentary references to gold objec (...)
  • 13 Barbot (1992: 527) states that there were many goldsmiths at “Boutrou [Butre], Comendo [Komenda], M (...)

8Limited European documentary references to metal working also suggest local casting may have only become more common on the Ghanaian coast during the 17th century. Prior to the late 17th century, the method of manufacture used to make gold artifacts is unclear. Among the earliest descriptions, a mid-16th century account, specifically refers to “beaten gold” (Blake 1967: 343). Writing early in the 17th century, de Marees (1987: 65) suggested that the knowledge of metallurgy on the coast was limited at that time, with some metal working techniques having been introduced by the Portuguese. He refers to “remarkably beautiful gold Chains and other ornaments, such as Rings” produced by the people of Senya Beraku, a settlement in the eastern Central Region associated with Dutch Fort Goede Hoop. Although the method of manufacture is not described, these items could have been readily produced by hammering and cold forming.12 In contrast, the gold beads and other ornaments described and illustrated by Jean Barbot (1992: 494, 527, Pl. 43, 542–543 note 41) in the late 17th century are clearly pieces produced by lost wax casting.13 These references and the relative paucity of early brass casting evidence from coastal sites compared to those in the interior may indicate that metal forming by cold working methods predominated in central and western coastal Ghana until the 17th century, the major casting centers located further from the coast in the Ashanti and Brong-Ahafo regions, and beyond.

  • 14 Alpern (1995: 11–18) provides a compilation of the varied finished metal goods and unprocessed meta (...)
  • 15 The cargo of the Elmina shipwreck, provisionally identified as the Groeningen, included rolls of le (...)
  • 16 Two iron molds for lead musket balls and lead spew were recovered from late 19th century contexts a (...)

9In addition to iron, copper alloys and gold, other metals were brought to the coast and utilized by Akan metal workers.14 A pendant in silver or electrum was excavated from a late 17th–early 18th century mortuary context at Nsadwer Bosumtwi (Chouin 2009: 486, 730). Cast silver objects attributed to the Akan are also known from ethnographic and museum collections (e.g., Cole & Ross 1977: 35–36). In addition, lead was an early import to the coast and its low melting temperature, malleability, and potential use in alloys would have made it attractive to African metal workers.15 Akan metal workers used lead to fill small defects in brass castings and to adjust the weights of castings (Garrard 1980a: 185, 233). However, on the coast, the only direct evidence of lead casting consists of musket ball molds from 19th century sites associated with European trade.16 Lead was also used for fishing net weights and, occasionally, to fashion buttons. However, these items required no melting or casting, but rather were produced by cold hammering, folding, and cutting.

  • 17 Iron smelting, casting and the cold working of brass sheet metals seem to have been traditionally r (...)
  • 18 Even during the following centuries, brass casting on the coast may have remained limited compared (...)

10The preceding review of the evidence for metal working and metallurgy in coastal Ghana provides important context for brass working among the coastal Akan during the Atlantic period.17 On one hand, the evidence for iron smelting and the cold-forming of gold, demonstrate that the peoples of coastal Ghana were well familiar with varied metal technologies by the 1st millennium AD. The methods employed in the Coconut Grove bead’s manufacture are comparable to some of the methods later used in reshaping European sheet brass. The techniques represented—forming using hammering, bending and crimping, with stamped decorations— likely represents the common, if not the predominant, methods used in the manufacture of metal objects on the Akan coast until the 17th century and these techniques remained important in the succeeding centuries. On the other hand, the limited evidence for early casting suggests that this technology may not have been well established on the western and central Ghanaian coasts prior to the 17th century.18 Casting, brass working, and the development of craft specialization, may have expanded on the coast as part of the general trend of increasing trade, commodity production, and the emergence of a merchant class during the 17th century, a major factor being the growth of the Atlantic slave trade (DeCorse 2021a: 22–28; Kea 1982: 171–205).

Brass Working and the European trade

  • 19 See examples of trade lists and reviews of the brass trade in Alpern 1995: 12–18; Garrard 1979, 198 (...)
  • 20 Chouin (1995, 2009: 272–282) has documented large copper or copper alloy basins of likely 16th or 1 (...)

11The advent of the European coastal trade in the 15th century dramatically increased the availability of copper and copper alloys, which subsequently facilitated the elaboration of Akan metal working and casting (Figs. 4, 5; DeCorse 2021a: 103–104; Garrard 1979: 39; 1980a: 104; Silverman 2014: 6). Brass goods were among the most important imports to the coast from the opening of the trade, including raw materials in the form of unworked rods and cast brass items such as manillas, as well as an extensive array of pans, basins, plates, kettles, pots and jugs.19 Describing the trade on the Ghanaian coast in the early 17th century Pieter de Marees (1987: 52) noted a wide variety of brass basins and that “Such Brass Basins, which the Ships bring there in large quantities, have become so common in the Country that people often sell brass-ware as cheaply (to the Negroes or their Landsmen) as it is bought in Amsterdam.” Early pieces of European metalware also occur in archaeological contexts.20

Figure 4 – Illustration from Pieter de Marees (1602) depicting how men dress on the Gold Coast

Figure 4 – Illustration from Pieter de Marees (1602) depicting how men dress on the Gold Coast

- A (at left) is described as “a Nobelman or great Monsieur, whom they call Brenipono, as he goes through the streets every day, with a cap on his head, like a Bonnet; his mantle or wrapper is of Linen.”
- B. “shows a Merchant, whom they call Batafou, coming from distant Lands to trade on the seashore, wearing a hat made of Dog-skin on his head and a roll of Cotton or Linen around his body; in one hand he has assegai and in his right hand a copper Basin.”
- C. depicts “an Interpreter, who comes with the Merchants or Peasants to buy things on the Ships, having on his head a little Hat made of Sugar-cane.”
- D. (in the background) “shows how a Merchant, having done his trade with the Dutch, returns home with his Catious or Slaves, loaded with his Merchandize.” De Marees 1987: 33, from de Marees 1602: Plate No. 2, reproduced courtesy of Leiden University Libraries, Digital Collections https://digitalcollections.universiteitleiden.nl/​search/​Pieter%20de%20Marees?type=edismax

Figure 5 – Brassware from the Elmina Wreck, provisionally identified as the Groeningen, a Dutch West India Company ship that burned and sank in front of Elmina Castle in 1647

Figure 5 – Brassware from the Elmina Wreck, provisionally identified as the Groeningen, a Dutch West India Company ship that burned and sank in front of Elmina Castle in 1647

The ship was loaded with cargo arriving on the coast, including stacks of nested brass basins, pewter bowls, and manillas

© Nicole Hamann

  • 21 See Cole & Ross 1977: 65, 85 note 23; Garrard 1979: 43, 1980a: 186, 1983: 54; Ross 1974: 45, 1983:  (...)
  • 22 See discussions and references in Aitken 1866: 231–232, 310; Day 1973; Cook 2012: 182–191; Hamilton (...)
  • 23 Jones (1983: 4) observes that the majority of metal goods the Portuguese and Dutch traded in West A (...)

12European technology for the production of sheet brass and for drawing rods, wire and tubing steadily advanced between the 15th and 19th centuries, reducing costs and increasing availability. In particular, it has been suggested that the expansion of the British brass industry during the late 18th and early 19th centuries, and its subsequent availability in West Africa, were contributing factors in the appearance of mforowa.21 The emergence of Britain as a major producer and exporter of brass undoubtedly facilitated the availability of brasswares, especially in areas of British trade. Yet, here it is important to note that the technology for production of brassware was introduced to Britain from the Continent where it was well established by the 15th century.22 Indeed, during the 17th century the metal working industries of the Netherlands and Germany emerged as the world’s major consumers of copper, a development that coincided with the United Provinces ascendancy in maritime trade (Wallerstein 1980: 50–52, 205–206). Quantities of sheet brass from Continental sources were reaching West Africa well before the late 17th century—particularly at sites such as Elmina which was the Dutch headquarters on the Guinea coast between 1637 and 1872.23

13Although metal goods were a major import to the West African coast, documentary sources often lack sufficient detail to identify the specific materials, forms, and manufacturing methods represented. W.C. Aitken’s discussion of the mid-19th century British brass industry provides one of the most detailed descriptions of the raw materials Europeans brought to the West African coast. He writes:

  • 24 Aitken is confusing in his references to West Africa. While he refers to the Gold Coast (modern day (...)

14“A considerable quantity of the brass wire made in Birmingham finds its way to the Gold Coast, to Old Calabar, in the form of what are called ‘guinea rods,’ one hundred of which, each three feet in length, of Nos. 4 and 5 gauge in thickness, packed up in deal cases, and being at their destination, sold in exchange for palm oil, &c., are used as the ‘circulating medium’ by the natives, and at the death of the possessor are interred with the body. An influential Birmingham merchant states the orders from that country frequently amount to from five to twenty tons each. Large numbers of rings made of solid brass wire, about seven-sixteenths thick and three-and-a-quarter inches diameter, made of wire, are also sent to the Gold Coast. A smaller size of brass wire (a little thicker than ordinary pin wire) is converted by being wound around spits into spirals like an ordinary check bell spring, and is also exported to the locality named for purposes of ornament and personal decoration” (Aitken 1866: 319).24

  • 25 The rods recovered archaeologically at Elmina were approximately a foot in length (30.5 cm). Althou (...)

15Unworked copper and copper alloy rods were readily reshaped into a variety of jewelry, and implements. A bundle of brass rods was recovered from an 1873 floor context at Elmina (DeCorse 2021a: 134). Although these may have served as a ‘circulating medium’ on parts of the coast, as Aiken suggests, it is more likely that many were purchased as raw material and locally used in the fabrication of bracelets, rings, and other manufactures.25 A wide range of artifacts found archaeologically, and still found in markets today, were made by bending and simple modification of imported brass and copper.

  • 26 For example, while tons of manillas were sent to Elmina, only two fragments were found in archaeolo (...)
  • 27 Notably, de Marees (1987: 52) made a similar observation in the early 17th century, “Although these (...)

16While raw material in the form of rods and wire were imported to the coast, no documentary references to the importation of unworked sheet brass to the Gold Coast have been found prior to the 20th century. References to such imports may yet be found, but Akan metalworkers working with sheet brass likely often—perhaps principally—relied on finished brassware vessels for their raw material. European basins, pots, pans and jugs were reformed anew (see discussion of the 18th century forowa below; also see Chouin 2009: 729; Garrard 1980a: 105). Brass vessels and sheet brass scraps could also be melted down and recast. This use and reuse may explain the relative paucity of finished European metal objects recovered in archaeological contexts on the coast.26 Yet, while a great deal of brass was undoubtedly reprocessed within the greater Akan region, the lack of brass found archaeologically may also be indicative of the wider interior trade networks represented (Garrard 1979: 39).27

  • 28 Local manufacture in or near Elmina is perhaps suggested by the prevalence of these items at Elmina (...)

17Excavations at the African settlement of Elmina, recovered a variety of artifacts made from imported European metal, particularly wire, rods, and sheets of copper and copper alloy (Fig. 6; also see DeCorse 2021a: 134–135). While securely dated pieces of 17th century sheet brass artifacts have been found, the vast majority of material recovered archaeologically dates to the 18th and, especially, the 19th centuries. This is, however, perhaps less an indication of expanding European trade in brass than of the archaeological contexts represented, contexts of later dates being more fully represented. The Elmina excavations recovered numerous portions of copper and copper alloy necklaces chains, bracelets, and rings. Many may have been made or modified in coastal Ghana, if not in Elmina itself.28 They were simply formed by cutting and bending wire and rods to the appropriate size. Three rings have soldered ends, a technique that may suggest European manufacture. While Akan metal workers used lead to patch nkuduo and other pieces of cast brass, the soldering of brass requires substantially more heat than copper. Many pieces are undecorated and ornamentation present on the rings and bracelets consists of simple, stamped or engraved geometric designs. Similar ornaments in both copper alloy and iron have been recovered from 17th-19th century contexts at a number of other coastal Ghanaian sites (e.g., Bellis 1972: 71–73; Chouin 2009: 729; Garrard 1980b: 85–155; Nunoo 1957: 15; Shaw 1961: 57–58, Pl. iv; Spiers 2007: 192, 200–205). In their simplicity in manufacture, many of the pieces are indistinguishable from items still sometimes sold in West African markets. They can be readily contrasted with ornaments produced in Europe, which by the 19th century often had machine-engraved decoration (Fig. 7).

Figure 6 – Artifacts possibly locally made or modified in coastal Ghana from 19th century contexts at Elmina

Figure 6 – Artifacts possibly locally made or modified in coastal Ghana from 19th century contexts at Elmina

Rim of imported brass basin with engraved decoration (left); a copper wire bracelet (upper right); brass ring and bracelets; a forowa bottom made from a European vessel (lower right)

© C.R. DeCorse

Figure 7 – European bracelet made from brass tubing with machine engraved decoration

Figure 7 – European bracelet made from brass tubing with machine engraved decoration

© C.R. DeCorse

  • 29 For other examples of sheet brass objects related to the gold trade see Bellis 1972: 71–73; DeCorse (...)
  • 30 Riveting, a procedure already common in antiquity, was widely used in the manufacture of European b (...)

18Ghanaian craftsmen also made a wide range of items from European sheet brass, including a wide range of implements specifically associated with the gold trade, such as spoons (saawa, atere, atire), blow pans or shovels (mfamfa), gold dust sieves (huhuamoa), scales (nsania or nsenia), and gold dust boxes (adaka, adakawa) (Garrard 1980a: 178–188, 358–359).29 The sheet brass was cut, bent and beaten into shape as needed, with pieces of metal joined by folding, crimping, and riveting (Fig. 8).30 Decoration consists of simple engraved, punched, repoussé, chased, and cutout designs. The type of decoration varies in the different artifact categories represented. For example, shovels were painted black and engraved with concentric compass circles, while repoussé and chased decorations were used on gold dust boxes and mforowa.

Figure 8 – Artifacts related to the gold trade made from imported sheet brass: a gold dust shovel and a spoon of heavy sheet brass, both from 19th century contexts at Elmina; sheet brass gold dust box and spoon with engraved and repoussé decoration

Figure 8 – Artifacts related to the gold trade made from imported sheet brass: a gold dust shovel and a spoon of heavy sheet brass, both from 19th century contexts at Elmina; sheet brass gold dust box and spoon with engraved and repoussé decoration

© C.R. DeCorse

Mforowa and the Coastal Akan

  • 31 Surveying mforowa from documented provenances in the collections of the Ghana National Museum, Crow (...)

19The most distinctive objects produced by the coastal and near-coastal Akan are mforowa, sheet brass containers primarily associated with the Fante.31 While made of imported European material, they are distinct in form. Mforowa are lidded containers with basal stands, generally cylindrical, ranging in diameter from less than 4.5 centimeters to over 35 centimeters, though other shapes occur (Fig. 9). More elaborate, possibly later examples, include hidden compartments in the base or secondary containers on top of the lid. They are characteristically unsoldered or unwelded, but rather fastened together by folding, crimping, riveting and, less commonly, stapling. Notably, the decorative inventory of mforowa is more elaborate than other sheet brass objects, including chased and punch-work designs, and repoussé figures of animals, symbols, and geometric forms. The bases may also have geometric cutouts. In some examples, the brass sheets used for the vessel lids have been hammered to form a dome. However, as will be discussed, domed lids are likely a late 19th–20th century innovation.

Figure 9 – A forowa with a domed lid

Figure 9 – A forowa with a domed lid

This is the most common form seen in historically and ethnographically documented examples and likely dates to the late 19th and early 20th centuries. The grid in the background is 1 cm squares.

C.R. DeCorse collection

  • 32 It is possible that mforowa were inspired by European forms. However, as there are no obvious paral (...)
  • 33 The only kuduo from a coastal archaeological site is the example from Eguafo (Fig. 10).

20Stylistically and functionally, mforowa are closely related to the Akan nkuduo, which date earlier and are made from cast brass (Silverman 1983a: 180–199, 1983b, 2014: 5). The stylistic inspiration of the nkuduo were 14th–15th century Middle Eastern vessels brought via the trans-Saharan trade.32 The casting of nkuduo may, therefore, have been closely connected with the introduction of brass casting into the Akan forest. Well documented pieces are principally associated with the interior Akan, particularly the Ashanti Region, and they may not have been produced on the Akan coast.33 Stylistically, nkuduo are more diverse than mforowa (Cole & Ross 1977: 67). The similarity in the two forms is most evident in the bases, some nkuduo having open vertical ribs, a feature that appears duplicated in the repoussé grill work decoration often found on the bases of mforowa (Figs. 10, 11, 15; also see Cole & Ross 1977: 64–69; Rattray 1969b: 313–315; Ross 1983). In addition to being cast, a major difference is that the nkuduo lids typically fit inside the vessel rims, while mforowa lids generally fit on the outside. As will be discussed, the latter was likely a later innovation in the stylistic evolution of mforowa.

Figure 10 – A kuduo from a probable mortuary context disturbed by small-scale gold mining at Eguafo

Figure 10 – A kuduo from a probable mortuary context disturbed by small-scale gold mining at Eguafo

The open grill work on the base is stylistically similar to the chased and repoussé decoration found on the bases of mforowa. The grid in the background is 1 cm squares.

© C.R. DeCorse

Figure 11 – Drawings of late 17th or early 18th century forowa fragments from Elmina showing the lid, the lid’s rim, body, bottom and base

Figure 11 – Drawings of late 17th or early 18th century forowa fragments from Elmina showing the lid, the lid’s rim, body, bottom and base

The repoussé decoration on the lid features a crocodile or lizard. The truncated oval representations on the top center of the body may represent stylized treasury bags or locks, both of which suggest security. The angular, recurring motif to the right of the flower-like motif may be a stylized key. The repoussé ribbed decoration on the base is reminiscent of the open grill work found on the bases of some nkuduo. The bottom of the forowa is curved.

© C.R. DeCorse

  • 34 Nana Kodjo Nquandoh III, July 14, 1990; Badu Prah, July 14, 1990; Kofi Obusu, August 9, 1990 at Elm (...)

21There is also some functional overlap between nkuduo and mforowa. Both were high status articles often used to contain valuables, which often functioned in mortuary contexts. However, nkuduo traditionally served in more prestigious shrine and royal contexts (Cole & Ross 1977: 67; Silverman 1983b). Christensen (1954: 71) recorded that mforowa were used “For keeping gold dust, shea butter, and mixture of shea butter, cowrie shells and herbs for magical powers to cure. Owned by an important man and buried with the owner.” Informants at Elmina indicated that mforowa were traditionally placed under the head of the deceased at burial.34 While mforowa served in these specialized contexts, they are ethnographically documented in more utilitarian settings; commonly described as containers for shea butter, a fat extracted from the nut of the shea tree still commonly used as a moisturizer and pomade (Ross 1974). Doran Ross (1974: 42) suggested that the scarcity of public references to the ritual use of mforowa may be a consequence of the limited access outside observers had to these contexts.

  • 35 Ross (1974: 45) initially suggested mforowa production began in the 19th century, but revised this (...)
  • 36 See following discussion of imported brass vessels associated with Elmina mforowa. For other exampl (...)

22Substantive study of mforowa was undertaken by Ross, who suggested they were produced from circa 1780 to 1930 and that their stylistic origins could be traced to the earlier cast brass nkuduo.35 Ross also suggested that the evolution from flatter to dome shaped lids occurred later. Ross’ assessments of the age and origin of mforowa were based on the study of examples in museum collections and limited European sources. The earliest documentary references to the containers may date to the early 17th century. De Marees (1987: 52), referring to the diversity of metal basins traded on the coast stated: “These Basins they use for various purposes: they use the small Neptunes to immure in tombs on the graves of the dead, and also to carry something or other in” and “in chased Basins they put their ornaments and trinkets.” Writing at the start of the 18th century Bosman (1967: 128) is possibly referring to mforowa manufacture when he notes the “making of Copper Ointment Boxes” among the manual arts of the coast. While these descriptions may refer to mforowa, the ambiguity in description make it difficult to be sure that the distinctive mforowa are represented. Both documentary and archaeological data indicate that European brass vessels and mforowa often served in similar functional contexts.36

Archaeological Insights into Mforowa Origins

  • 37 In addition to the examples discussed here, mforowa and mforowa fragments have been recovered from (...)

23Mforowa recovered archaeologically from the site of Elmina allows for the refinement of their origins, stylistic development, and manufacture (DeCorse 2021a).37 A striking aspect of the Elmina site was the close chronological control afforded by an array of European ceramics, glass, tobacco pipes and other trade materials. The production ranges of some of these materials can sometimes be dated within a few decades (e.g., DeCorse 2021a: 145–174; Deagan 1987; Noel Hume 2001). These data support the nkuduo-stylistic origins of mforowa, the beginnings of mforowa production in the 17th century, and increasing stylistic elaboration of mforowa during the late 19th and 20th centuries.

An Early 18th Century Forowa from Elmina

  • 38 Heidi Miller, personal communication 3-30-22.
  • 39 The disturbed, disarticulated nature of the burial is typical of the Elmina site. The sub-floor bur (...)

24The oldest forowa recovered thus far was excavated from a subfloor burial context at the African settlement Elmina, in a stone walled structure likely built in the late 17th or early 18th century on earlier 16th century midden deposits (Figs. 11, 12). This area is located in front of, and relatively close to, Elmina Castle and was the location of some of the African settlements most impressive structures. A late 17th to early–18th century age of the burial is based on clear stratigraphic context and associated European tobacco pipes, glass, and ceramics (DeCorse 2021a: 98-100). The forowa was associated with the post cranial bones of a robust adult, probably a 35 to 50-year-old male;38 an individual who would have lived and worked in the late 17th and early 18th centuries. The burial was fragmentary and unarticulated, being one of several interments intruding into earlier deposits beneath the structure.39 The associated artifacts were grouped together below the fragmentary leg bones, which were stained green through long exposure to the associated copper-alloy artifacts. The forowa was found nested with two European sheet brass vessels and a large, partially reconstructed, red slipped African tobacco pipe. The largest of the brass vessels appeared to have been approximately 30 centimeters in diameter. The regular shape and folded rim indicate it was imported vessel, and may correspond to the Neptune basins described by de Marees (1987: 52, 182). It was, however, very poorly preserved and was little more than a green, copper oxide stain in the soil.

Figure 12 – Schematic diagrams and detail of the bottom of the forowa shown in Figure 11

Figure 12 – Schematic diagrams and detail of the bottom of the forowa shown in Figure 11

The forowa’s pieces were attached using a combination of crimping, rivets, and staples.

© C.R. DeCorse

25The other associated European vessel was a better-preserved brass tobacco box (Fig. 13). It’s shape and engraved lid are similar to Dutch examples of 17th to 18th century age (e.g., Kisluk-Grosheide 1988). The lid is decorated with inscribed words and pictures that form a riddle. The riddle has not been deciphered but the script suggests a Dutch origin (Jan van Doesburg, personal communication). Engravings of the sun, krug, and crown can be discerned.

Figure 13 – A Dutch riddle tobacco box from a late 17th early 18th century context at Elmina

Figure 13 – A Dutch riddle tobacco box from a late 17th early 18th century context at Elmina

Historical and archaeological data indicate that mforowa and European metal vessels often functioned in similar contexts.

© C.R. DeCorse

  • 40 The forowa, associated brass pan, and surrounding soil were stabilized, removed as an unexcavated b (...)
  • 41 The curved base of the forowa is smooth without the pronounced battering marks characteristic in th (...)

26The forowa itself was severely degraded and some sections had deteriorated to no more than a copper oxide stains.40 However, portions of the lid, sides, and bottom survived and these provided a clear indication of the container’s construction and appearance. The forowa is similar to examples Ross (1983) considers early transitional mforowa forms with nkuduo features. The overall shape of the Elmina forowa is squat and wider relative to its height than is seen in later examples. The lid is flat and fits inside of the vessel analogous to nkuduo lids. This contrasts with ethnographically and historically documented mforowa of late 19th and 20th century age, which are typically domed with lids that fit outside of the vessel (Ross 1983). The bottom of the forowa is curved; it is possible that the Fante metal workers utilized curved portions of imported brassware.41

  • 42 The bottoms of mforowa are typically made from a single piece of brass, but pieces with bottoms con (...)
  • 43 The sheet brass staples may have been easier to fabricate than copper wire rivets. A copperware fra (...)
  • 44 Tchakirides (1999: 18–20) analyzed the samples using a Micromass Sector 54 multi-collector mass spe (...)

27At least seven pieces of brass were used in the forowa’s construction, possibly representing different sources of brass. The majority of extant mforowa are typically constructed of five pieces of brass—the lid, the lid’s rim, body, bottom and foot—fastened by crimping and riveting (Ross 1974: 45). The bottom of the Elmina forowa was secured to the body by crimping, while the base was riveted to the bottom. The bottom was formed using at least two separate pieces that were overlapped and joined with copper alloy or sheet brass staples (Fig. 12).42 The staples were made of strips of copper alloy/brass sheets with pointed ends, which were inserted through the pieces of metal to be joined and the ends bent over to make the attachment. One of the bottom fragments is also thicker and may represent a patch.43 The lead isotope ratios in the sheet brass used in the lid’s rim also suggest two different sources may be represented, though macroscopically the pieces of brass are indistinguishable (Tchakirides 1999: 12).44

28Decorations on the vessel are comparable to those seen on other mforowa and in other examples of Akan art. The base of the vessel had a ribbed decoration, interspersed with a “X” pattern reminiscent of nkuduo decorations and which are also common on later mforowa. The decorative elements on the sides and top are both geometric and figurative. A flower-like pattern is found on the side (Ross 1974: 46). In addition, the oval decorations on the upper edge of the sides may be stylized representations of leather treasury bags or, perhaps, locks. Although uncertain, the design between these figures may be an oversize key. Symbolically the treasury bag and the lock convey the same measure of security in this context (Doran Ross 1989: pers. com.). The final figurative motif is a lizard or crocodile found on the lid. These decorations, often prominent on mforowa, are among the earliest decorative motifs in Akan art (Cole & Ross 1977: 63–69; Ross 1974; 1983; Silverman 1983a: 183–187).

29Another example of an early transitional forowa form with nkuduo features is provided by a sheet brass forowa lid recovered from a spoil heap left by small scale gold miners at the Atofosie locus, Eguafo. The lid is flat, 15.5 centimeters in diameter, with a hinge and hasp of sheet brass, and cast brass handle (Fig. 14). These features, while often found on nkuduo are not found on mforowa (Ross 1983). The hinge and hasp of the Eguafo lid were attached with rivets, while the handle was attached with sheet brass staples. The handle and the lid are decorated with incised and stamped patterns, the lid has a small repoussé circle in the center.

Figure 14 – An early forowa lid with a hinge, hasp, and cast brass handle from a disturbed context at Eguafo

Figure 14 – An early forowa lid with a hinge, hasp, and cast brass handle from a disturbed context at Eguafo

a. top; b. bottom. The grid in the background is 1 cm squares.

© C.R. DeCorse

30While the Elmina and the Eguafo mforowa can be seen as early transitional forms, they represent almost the full range of construction methods seen in later mforowa and other Akan sheet brass objects, including crimping, riveting, stamping, and repoussé decoration. Only cutout decorations and the domed lids seen in later examples are lacking. Indeed, in some respects they are more sophisticated than the examples from 19th century contexts discussed below. This suggests that they are not prototypes, but part of an established metal working tradition extending further back in time. In a similar vein, the decorative inventory, including zones of patterned decoration, flower-like patterns, locks or treasury bags, are similar to those found on earlier Akan art, including nkuduo. This is further indication that the origins of the mforowa production lie well within the 17th century.

19th Century Mforowa

31Five 19th century mforowa were recovered from discrete archaeological contexts at Elmina. They are distinct in their size, style and method of manufacture from the late 17th–early 18th century Elmina example. Their construction is comparable with pieces having late 19th and 20th century provenances, consisting of the lid, the lid’s rim, body, bottom, and foot, fastened by crimping and copper riveting. All also have lids that fit outside of the body. However, the lids are flat, or only slightly convex, something that again points to domed mforowa lids being a later innovation.

  • 45 The sub-floor contexts in this area were complicated, consisting of shallow deposits beneath 1873 f (...)
  • 46 Only single samples from the lid’s rim and body were analyzed, so it is possible that different sou (...)

32Two mforowa were excavated from probable 19th century contexts in a room adjoining the late 17th or 18th century stone walled structure where the early 18th century forowa was located (Fig. 15). The broken, but largely intact mforowa were not in situ but had likely been associated with a sub-floor burial, or burials disturbed by construction and a circa 1873–1880 interment. The two mforowa likely date circa 1820–1850, based on their stratigraphic contexts and the trade materials represented.45 The mforowa were constructed from five separate pieces of sheet brass making up the lid, the lid’s rim, body, bottom, and basal foot crimped and riveted together. In both cases the base was riveted to the bottom. The body of the smaller forowa is 3.9 centimeter high and 6.9 centimeters in diameter. The flat bottom was cut from a single piece of sheet brass. Lead isotopes analyses of this forowa suggests that a similar source of brass was used to construct the top and bottom of the vessel (Tchakirides 1999: 21).46 A repoussé band, vertical lines, and a stamped scalloped pattern decorate the body and lid. The base of the forowa is undecorated. The body of the second forowa recovered at this location measures 3.8 centimeters high and 7.5 centimeters in diameter. The body is decorated with simple repoussé and stamped incised geometric patterns, while at least a portion of the base was decorated with faux-grill work.

Figure 15 – 19th century mforowa from Elmina

Figure 15 – 19th century mforowa from Elmina

© C.R. DeCorse

  • 47 There were four partially overlapping burials in this location. The burial data are reviewed by Mil (...)

33Three additional mforowa were excavated from discrete 19th century burial contexts in Fisherkrom (the fishermen’s village), the name historically used to refer to the area of the old Elmina settlement located on the point of land extending south-southeast of Elmina Castle (Figs. 16, 17; DeCorse 2021a: 52–53, 97–98). Archaeological and documentary sources indicate that the African settlement did not extend into this area until after the Dutch takeover of Elmina in 1637. The large, stone walled structure excavated was built during the 18th century on water-washed bedrock and sandy deposits of 15th–17th century Portuguese material. The burials that contained the three mforowa likely date circa 1850–1873.47

Figure 16 – A 19th century forowa from Fisherkrom, Elmina

Figure 16 – A 19th century forowa from Fisherkrom, Elmina

© C.R. DeCorse

Figure 17 – A 19th century forowa from Fisherkrom, Elmina, side view (left) and top view (right)

Figure 17 – A 19th century forowa from Fisherkrom, Elmina, side view (left) and top view (right)

The radiating lines and twelve-point repoussé star in low relief on the top of the lid are barely visible. The forowa’s base is missing. The photograph also illustrates the poor condition of many of the sheet brass artifacts recovered archaeologically. The grid in the background is 1 cm squares.

© C.R. DeCorse

34The Fisherkrom mforowa were each constructed of five separate pieces of sheet brass making up the lid, the lid’s rim, body, bottom, and base. The body diameters are 8.5, 7.4, and 7.5 centimeters. The base of the larger forowa was attached to the bottom with copper rivets (Fig. 16). It is undecorated except for a four-point repoussé star on the lid. The two smaller mforowa may have both been associated with the burial of a middle age adult female (35–50 years old). The two smaller mforowa are distinct in decorative style and construction and might easily be the product of the same crafts person (Fig. 17). The base, bottom, and body were all joined together by crimping the sheet brass of the vessels’ bodies over the edges of the bottom and the top of the base, with no riveting. The bodies and the slightly convex lids are decorated with stamped lines. A twelve-point repoussé star in very low relief, accents the lid of one of the mforowa.

35The five19th century mforowa suggest an increasing refinement in manufacturing technique, including the attachment of the base to the bottom and body with a single crimped joint. The bodies also become less squat; the diameter less than the height. The absence for any archaeologically or historically documented dome-shaped mforowa prior to the late 19th and early 20th centuries suggests this form is a later innovation. The comparative simplicity of the 19th century mforowa discussed, and possible association with female mortuary contexts illustrates their function in varied socioeconomic contexts.

36A late-19th through early 20th century age for domed mforowa suggests a similar age for related artifact classes. A mis-matched pair of sheet brass rattles was viewed in Ghana for sale by a Hausa trader in the 1980s, one of which is illustrated in Figure 18. The construction of the rattle is comparable to the methods used in mforowa construction, essentially consisting two domed mforowa lids joined together and attached to a handle using copper rivets. The rattle’s repoussé and stamped decoration consists of bands and zones of patterned decoration with no figurative images. Both rattles reputedly had coastal Fante provenances. Their construction and decoration were very similar, possibly the products of the same maker. In the 1970s, Daniel Crowley documented two similar rattles from southern Ghanaian contexts (Ross 1974: 49). Ross describes the rattles as having the “typical applied decoration” found on mforowa. Given the similarity in construction to the dome shaped-mforowa, these objects likely date to the late 19th or early 20th century.

Figure 18 – Rattle constructed of sheet brass, comparable to the method used in the domed lids of late 19th–early 20th century mforowa

Figure 18 – Rattle constructed of sheet brass, comparable to the method used in the domed lids of late 19th–early 20th century mforowa

Collected in coastal Ghana with purported Fante provenance.

© C.R. DeCorse

The Late 19th Century and Decline

37Mforowa production may have started to decline in the late 19th century, a period that witnessed dramatic socioeconomic changes both globally and locally. The disruptions caused by the Anglo-Asante wars of 1873–1875, 1895–1896, and 1900 impacted trade, possibly limited access to brass goods, and may have contributed to the decline of local metal working (e.g., Garrard 1979: 43). Yet, the decline of mforowa production more likely lies with global political and economic changes than localized disturbances (e.g., DeCorse 2021b; Frankema et al. 2018; Hobsbawn 1989: 34–83).

38With the passage of abolition legislation, European nations sought alternative ways to profit from their imperial outposts on the African coast. The expanding European spheres of influence and the nascent colonies of the 19th century were equally seen as sources of raw materials and markets for European manufactures. European colonial interests reached their apogee with the partition of Africa at the Berlin Conference of 1885. In Ghana prior to the 19th century, private merchant companies of various European nations had maintained trade forts along much of the coast. These small emporia existed with the permission of local polities and exerted little territorial control over the adjacent hinterlands. However, during the 19th century Britain consolidated its interests (DeCorse 2016, 2021b; Newbury 1971). In 1821, the British government withdrew the charters of the private companies and created the British Gold Coast Colony, and gradually extended control over the coast from Cape Three Points to the Volta River. Britain was granted the rights to acquire territorial sovereignty over the Ghanaian hinterland by the General Act of the Berlin Conference. Asante and the territories of northern Ghana formally became British protectorates in 1902 (Pakenham 1991: 504–523).

  • 48 In particular, there seems to have been a marked increase in the range of items traded on the coast (...)
  • 49 The production of objects specifically for export to certain areas was an early and important aspec (...)

39The colonial partition of Africa was concurrent with increasingly sophisticated organization of production, technology for brass production, and mechanization, especially in Britain. This had important consequences for African brass working. Progressively sophisticated production yielded mass produced products that competed with, and soon replaced, African industries (Mokyr 1990: 113–150). European merchants had long tailored their shipments to meet demands on particular parts of the West African coast (e.g., DeCorse 2021a: 142–151).48 However, during the 19th and early 20th centuries European manufacturers made an increasing array of implements, ornaments, and products for the African trade, many aimed at specific markets. There were hoes of varied styles comparable to local examples, clay pipes that copied local designs, and glass cowries, teeth and imitations African beads made specifically for the West African trade (O’Leary & DeCorse 2022). 49

40The products marketed in Ghana, included a variety of machine-made anklet and armlet rings made from brass tubing. Examples of the last reportedly had an internal diameter of from “2¾ up to 3½ and 4 inches” (approximately 5.7 to 10 cm), the opening left unsoldered “in order to admit of their being opened the more readily, to allow of their being placed on the arm and leg of the wearer” (Aitken 1866: 320). They were decorated by milling tools, and finished by burnishing and lacquering. A bracelet corresponding to this description was found in a late 19th century context at Elmina. It is made from a brass tube and is decorated with machine engraved bands (Fig. 7). There were also cylindrical, sheet brass match or strike-a-light containers with simple machine-engraved decorations and, in some cases, pressed decorations of European locomotives (Fig. 19).

Figure 19 – European machine-made match or strike-a-light container, probably British colonial period, 20th century

Figure 19 – European machine-made match or strike-a-light container, probably British colonial period, 20th century

The lid measures 27 mm in diameter.

© C.R. DeCorse

Figure 20 – Machine made forowa, probably British colonial period, late 19th century or early 20th century

Figure 20 – Machine made forowa, probably British colonial period, late 19th century or early 20th century

The grid in the background is 1 cm squares.

© C.R. DeCorse

41European-made mforowa afford an even more striking example of production specifically for a localized African market (Fig. 20). Given the distinctive form and limited distribution of mforowa to southern Ghana, such objects could only have been fabricated with the Akan coast in mind. In form, they are identical to the locally made domed mforowa. However, the European mforowa bodies, bases, and lids are seamless, with machine-engraved decorations. Notably, these decorations are comparable to those found on the machine-made bracelet of brass tubing recovered from a late 19th century context at Elmina (Fig. 7). The bottoms of the machine-made mforowa observed form a separate container, the main body of the forowa serving as the lid. Although generally unmarked, these imported mforowa were likely all colonial era British manufactures. Ross (1974: 45) noted a unique silver forowa marked ‘Birmingham 1926’ in the collections of the British Museum.

42Writing in the 1950s J. Sylvanus Wartemberg, the local historian of Elmina, explained the demise of local production in the face of European imports. He observed that the Elmina pottery industry

“...suffered a set-back after the British Empire Exhibition of 1925 to which specimens of local production were sent; the immediate effect was a diminution of interest in the local industry as a result of the importation of cheap enamelware imitation of the local specimens, which eclipsed and paralized [sic] the industry to some appreciable extent” (Wartemberg 1951: 72).

43Trade goods such as these, coupled with regular steamship service (Lynn 1989), led to a general collapse of local craft industries, including mforowa production. Brass casting and brass working became minor, small-scale industries. This limited niche only expanded in the 20th century, facilitated by the tourist trade and the availability of scrap brass rather than the importation of greater amounts of raw material.

  • 50 This time range somewhat corresponds to Garrard’s suggestion that the majority of all mforowa produ (...)

44The vast majority of mforowa in museum collections have high domed lids and likely date to the second half of the 19th and early 20th centuries.50 Made of sheet brass, mforowa are less likely to survive than the cast brass nkuduo. Mforowa of more recent manufacture can still be found in Ghanaian markets, as can Akan Akua’ba figures, gold weights, and other articles, many suitably ‘antiqued’ to attract collectors. These modern mforowa are readily distinguished by their lack of patina, distinct manufacture, crude decoration and, often, the use of heavier gauge sheet brass. Akan mortuary rituals are elaborate and still include placement of grave goods with the deceased. It may be that mforowa continue to play a role similar to the past, but this has not been documented.

Conclusion

45Archaeology offers singular information into the origins and development of artistic traditions not assessable through other sources. On one hand, the comparatively close chronological control afforded by the known production ranges of many imported trade materials allows more refined dating than is often possible on African artifacts of the early modern world. On the other, archaeological contexts also furnish more detailed information on behaviors and technologies than can sometimes be found in written and oral accounts.

46The Elmina mforowa and other metal objects described are among the best documented examples of these materials recovered archaeologically. They provide new insight into both the early European trade and the origins of local craft production. However, they do not provide the full picture of Akan metal working. Although the relative paucity of sheet brass objects from earlier contexts and from other areas support the development of mforowa manufacture on the Fante coast during the 17th century, more excavation and careful documentation of materials is needed. It is likely that the origins of Akan metal working and earlier prototypes of mforowa will be pushed further back in time. Indeed, it would be surprising if imported brass vessels were not being readily reworked soon after the initial arrival of quantities of European brassware on the coast in the 16th century.

Acknowledgements

47I acknowledge with gratitude the late Doran Ross and Timothy Gerrard, pioneers in the study of Akan metal working, who both provided comments on some of the archaeological finds discussed here. Tobias Skowronek offered important guidance on the interpretation of the lead isotope ratios in the sheet brass. I am also grateful to Tobias and Jan van Doesburg for providing comments on the Dutch riddle tobacco box inscription. I also thank Gérard Chouin, Sean Reid, Matthew O’Leary, and the anonymous reviewers of this article for their very helpful input.

Haut de page

Bibliographie

Agorsah E.K. (1975) – Unique Discoveries at Efutu, Ghana. Sankofa, 1: 88.

Aitken W.C. (1866) – Brass and brass manufactures. In: S. Timmins (ed.), The Resources, Products and Industrial History of Birmingham and the Midland Hardware District, London, Robert Hardwicke: 225–380.

Alpern S.B. (1995) – What Africans got for their slaves: A master list of European trade goods. History in Africa, 22: 5–43.

Amartey S. (2021a) – Archaeology and Settlement Histories along the Pra River Southern Ghana, circa 500 B.C.–AD 1970. PhD thesis, Department of Anthropology, Syracuse NY, Syracuse University.

Amartey S. (2021b) – A Multi-Component settlement site in the Shama Hinterland, southern Ghana. Nyame Akuma, 96: 20–25.

Barbot J. (1992) – Barbot on Guinea: The writings of Jean Barbot on West Africa 1678–1712. P.E.H Hair, A. Jones & R. Law (eds.), London, Hakluyt Society.

Bellis J.O. (1972) – Archaeology and the Culture History of the Akan of Ghana, A Case Study. Ph.D. Dissertation, Bloomington, University of Indiana.

Blake J.W. (1967) – Europeans in West Africa, 1450–1560. Nendeln, Krauss Reprint.

Bosman W. (1967) – A New and Accurate Description of the Coast of Guinea. London, Frank Cass.

Calvocoressi D. (1977) – Excavations at Bantama, near Elmina, Ghana. West African Journal of Archaeology, 7: 117–141.

Chouin G.L. (1995) – A Brass Basin at Abakrampa. Ghana Studies Council Newsletter, 8: 67.

Chouin G.L. (2009) – Forests of power and memory: An archaeology of sacred groves in the Eguafo polity, Southern Ghana (c. 5001900 A.D.). PhD thesis, Department of Anthropology, Syracuse University, Syracuse, NY.

Christensen J.B. (1954) – Double Descent Among the Fanti. New Haven, Human Relation Area Files.

Collette Q., Wouters I. & Lauriks L. (2011) – Evolution of historical riveted connections: joining typologies, installation techniques and calculation methods. Structural Repairs and Maintenance of Heritage Architecture, 12: 295–306.

Cole H.M. (1979) – Art studies in Ghana. African Arts, 13(1): 26–27.

Cole H.M. & Ross D.H. (1977) – The Arts of Ghana. University of California, Los Angeles.

Cook G. D. (2012) – The Maritime Archaeology of West Africa in the Atlantic World: Investigations at Elmina, Ghana. PhD thesis, Department of Anthropology, Syracuse University, Syracuse, NY.

Cook G.D., Horlings R. & Pietruszka A. (2016) – Maritime Archaeology and the Early Atlantic Trade: Research at Elmina, Ghana. International Journal of Nautical Archaeology, 45(2): 370–387.

Craddock P.T. & Hook D.R. (1995) – Copper to Africa: Evidence for the International Trade in Metal with Africa. In: D.R. Hook & D.M. Gaimster (eds.), Trade and Discovery: The Scientific Study of Artefacts from Post-Medieval Europe and Beyond. London, British Museum Press: 181–193.

Daniell W.F. (1852) – On the Ethnography of Akkrah and Adampe, Gold Coast Western Africa. Journal of the Ethnological Society, 4: 1–32.

Davies O. (1961) – Archaeology in Ghana. London, Thomas Nelson.

Day J. (1973) – Bristol Brass: A History of the Industry. Newton Abbot, David and Charles.

Deagan K. (1987) – Artifacts of Spanish Florida and the Caribbean, 15001800. Washington D.C., Smithsonian Institution Press.

DeCorse C.R. (2005) – Coastal Ghana in the first and second millennia AD. Change in settlement patterns, subsistence and technology. Journal des Africanistes, 75 : 43-54.

DeCorse C.R. (2016) – Tools of Empire: Trade, Slaves, and the British Forts of West Africa. In: D. Maudlin & B.L. Herman (eds.), Building the British Atlantic World: Spaces, Places, and Material Culture, 1600-1850, Chapel Hill, University of North Carolina Press: 165–187.

DeCorse C.R. (2021a) – An Archaeology of Elmina: Africans and Europeans on the Gold Coast 14001900. Washington D.C., Smithsonian Institution Press.

DeCorse C.R. (2021b) – Atlantic Slavery, African Landscapes: Change and Transformation in the Era of the Atlantic World, In: D.W. Tomich & P.E. Lovejoy (eds.), The Atlantic and Africa: The Second Slavery and Beyond, Albany, SUNY Press: 131–158.

De Marees P. (1987) – Description and Historical Account of the Gold Kingdom of Guinea. (Translated and annotated by Albert Van Dantzig and Adam Jones). Oxford, Oxford University Press.

Denk R. (2020) – West African Manilla Currency: Research and Securing of Evidence from 14392019. Hamburg, tradition.

Ehrlich M.J. (1989) – Early Akan Gold from the Wreck of the Whydah. African Arts, 22(4): 52–57.

Frankema E., Williamson J. & Woltjer P. (2018) – An Economic Rationale for the West African Scramble? The Commercial Transition and the Commodity Price Boom of 1835–1885. Journal of Economic History, 78(1): 231–67.

Fox C. (1986) – Asante Brass Casting. African Arts, 19(4): 66–71.

Fox C. (1988) – Asante Brass Casting: Lost-wax casting of gold weights, ritual vessels and sculptures, with handmade equipment. Cambridge African Monographs 11, African Studies Centre.

Garrard T.F. (1979) – Akan metal arts. African Arts, 13(1): 36–43, 100.s.

Garrard T.F. (1980a) – Akan Weights and the Gold Trade. New York, Longman.

Garrard T.F. (1980b) – Brass in Akan Society to the Nineteenth Century: A Survey of the Archaeological, Ethnographic and Historical Evidence. MA Thesis, Department of Archaeology, University of Ghana.

Garrard T.F. (1986) – Brass casting among the Frafra of northern Ghana. Unpublished Ph.D. dissertation, Department of History, University of California, Los Angeles.

Garrard T.F. (1988) – Introduction: The historical background to Akan gold-weights. In: C. Fox, Asante Brass Casting: Lost-wax casting of gold weights, ritual vessels and sculptures, with handmade equipment, Cambridge African Monographs 11, African Studies Centre: 1–13.

Garrard T.F. (1989) – Gold of Africa; Jewelry and Ornaments from Ghana, Côte d’Ivoire, Mali and Senegal in the Collection of the Barbier-Mueller Museum. Munich, Prestel.

Hamann N. (2007) – Forging an Atlantic World: An Historical Archaeological Investigation of African-European Trade in Metalwares. M.A. Thesis. Department of Anthropology, University of West Florida, Pensacola.

Hamilton H. (1967) – The English Brass and Copper Industries to 1800. London, Frank Cass.

Herbert E.W. (1984) – Red Gold of Africa: Copper in Precolonial History and Culture. Madison, University of Wisconsin Press.

Hobsbawn E. (1989) – The Age of Empire, 18751914. New York, Vintage Books.

Jones A. (1983) – German Sources for West African History 1599–1669. Studien zur Kulturekunde 66, Weisbaden, Franz Steiner Verlag.

Jones A. (1985) – Brandenburg Sources for West African History 1680–1700. Studien zur Kulturekunde 77, Weisbaden, Franz Steiner Verlag.

Jones A. (1995) – West Africa in the Mid-Seventeenth Century: An Anonymous Dutch Manuscript. Emory University, Atlanta, Georgia, African Studies Association Press.

Jones A. (2005) – Written sources for the Material Culture of the Gold Coast before 1800. Journal des Africanistes 75 (2) : 99.

Kankpeyeng B.W. & DeCorse C.R. (2004) – Ghana’s Vanishing Past: Antiquities, Development and the Destruction of the Archaeological Record. African Archaeological Review 2(2): 89–128.

Kea R.A. (1982) – Settlements, Trade, and Polities n the SeventeenthCentury Gold Coast. Baltimore, Johns Hopkins.

Kisluk-Grosheide D.O. (1988) – Dutch Tobacco Boxes in the Metropolitan Museum of Art: A Catalogue. Metropolitan Museum Journal, 23: 201–231.

Kiyaga-Mulindwa D. (1980) – The “Akan” Problem. Current Anthropology, 21: 503–506.

Lynn M. (1989) – From Sail to Steam: The Impact of the Steamship Services on the British Palm Oil Trade with West Africa, 1850–1890. Journal of African History, 30(2): 227–245.

Miller H. (2021) – A Biocultural Analysis of the Impacts of Interactions Between West Africans and Europeans, During the Trans-Atlantic Trade at Elmina, Ghana. Ph.D. thesis, Department of Anthropology, University of South Florida.

Mokyr J. (1990) – The Lever of Riches: Technological Creativity and Economic Progress. New York, Oxford University Press.

Morton V. (2019) – Brass from the Past: Brass made, used and traded from prehistoric times to 1800. Oxford, Archaeopress.

Murdock G.P. (1959) – Africa: Its Peoples and their Culture History. New York, McGraw Hill.

Newbury C.W. (1971) – The Tariff Factor in Anglo-French West African Partition. In: P. Gifford & W. R. Louis, France and Britain in Africa: Imperial Rivalry and Colonial Rule, New Haven, Yale University Press: 221–259.

Nguyen T.W. (2020) – Scandal and Imprisonment: Gold Spinners of 17th Century England. In: Hidden Stories/Human Lives: Proceedings of the Textile Society of America,17th Biennial Symposium, October 15–17, 2020. https://digitalcommons.unl.edu/tsaconf/

Noel Hume I. (2001) – A Guide to the Artifacts of Colonial America. Philadelphia, University of Pennsylvania Press.

Nunoo R.B. (1957) – Excavations at Asebu in the Gold Coast. Journal of the West African Science Association, 3(1): 12–44.

O’Leary M.D. & DeCorse C.R. (2022) – Marketing Colonialism: Smoking, Pipes and Cultural Entanglement in Colonial Senegal. Paper presented at the Society for American Archaeology meetings, April 2022.

Oppong C.E., Dompre H.O., Asante E.A. & Kissi S.B. (2019) – An Examination of the Production Processes of Brass Casting Among the Asantes: The Case of Krofofrom in Ashanti Region of Ghana. American Journal of Art and Design, 4(4): 58–67. doi: 10.11648/j.ajad.20190404.13

Ozanne P. (1962) – Excavation at Dawu, Transactions of the Historical Society of Ghana, 6: 119–123.

Pakenham T. (1991) – The Scramble for Africa: The White Man’s Conquest of the Dark Continent from 1876 to 1912. New York, Random House.

Penfold D.A. (1971) – Excavation of an Iron Smelting Site at Cape Coast. Transactions of the Historical Society of Ghana, 12: 1–15.

Pietruszka A.T. (2011) – Artifacts of Exchange: A Multiscalar Approach to Maritime Archaeology at Elmina, Ghana. PhD thesis, Department of Anthropology, Syracuse University, Syracuse, NY.

Pole L.M. & Posnansky M. (1973) – “Excavation of and Iron-Smelting Site at Cape Coast: Two Comments. Transactions of the Historical Society of Ghana, 14(2): 263–266.

Quarcoopome N.O. (1997) – “Art of the Akan.” Art Institute of Chicago Museum Studies 23(2): 134–147, 197.

Rattray R.S. (1969a) – Religion and art in Ashanti. Oxford University Press, London.

Rattray R. S. (1969b) – Ashanti. Oxford, Clarendon Press.

Reid S.H. (2022a) – Fragmented Landscapes: An Archaeology of Transformations in the Pra River Basin, Southern Ghana. PhD thesis, Department of Anthropology, Syracuse University, Syracuse, NY.

Reid S.H. (2022b) – Fragmented Landscapes: An Archaeology of Transformations in the Pra River Basin, Southern Ghana. Afrique : Archéologie & Arts, 18: 99–100.

Renschler E.S. & DeCorse C.R. (2016) – The Bioarcheology of Entanglement: Burials from Elmina, Ghana. Nyame Akuma, 86: 34–42.

Ross D.H. (1974) – Ghanaian forowa. African Arts, 8(1): 40–49.

Ross D.H. (1983) – Four unusual forowa from the Museum of Cultural History. In: D.H. Ross & T. Garrard (eds.), Akan Transformations, Museum of Cultural History Monograph Series, 30: 54–59.

Ross D. H. (2002) – Gold of the Akan from the Glassell Collection. Houston, Museum of Fine Arts.

Ross D.H. (2009) – Royal Arts of the Akan West African Gold in Museum Liaunig. HL Museumsverwaltung GmbH.

Ross D.H. (2014) – Akan Leadership Arts. In: Art and Life in Africa. Iowa, University of Iowa.

Shaw T. (1961) – Excavation at Dawu. New York, Thomas Nelson and Sons.

Shinnie P.L. & Kense F.J. (1989) – Archaeology of Gonja, Ghana: Excavations at Daboya. Calgary, University of Calgary Press.

Silverman R. (1983a) – History, Art and Assimilation: The Impact of Islam on Akan Material Culture. Dissertation in Art History, University of Washington, Seattle.

Silverman R. (1983b) – Akan Kuduo: Form and Function. In: D.H. Ross & T.F. Garrard (eds.), Akan Transformations, Los Angeles, Museum of Cultural History Monograph Series, 30: 10–29.

Silverman R. (2014) – Akan Brass Casting. In: Art and Life in Africa, University of Iowa. https://africa.uima.uiowa.edu/topic-essays/show/27?start=7 Accessed December 17, 2021

Silverman R. (2019) – Red Gold: Things Made of Copper, Bronze and Brass. In: K.B. Berzock (ed.), Caravans of Gold, Fragments in Time: Art, Culture, and Exchange Across Medieval Saharan Africa, Princeton, NJ, Princeton University Press: 256–267.

Spiers S. (2007) – The Eguafo Kingdom: Investigating complexity in southern Ghana. PhD thesis, Department of Anthropology, Syracuse University, Syracuse, NY.

Stahl A. B. (1994) – Innovation, Diffusion, and Culture Contact: The Holocene Archaeology of Ghana. Journal of World Prehistory, 8: 51–112.

Terray E. (1995) – Une histoire du royaume abron du Gyaman : Des origines à la conquête coloniale. Hommes et Sociétés. Paris, Karthala.

Tchakirides T.F. (1999) – Isotopic Composition of Lead in Sheet Brass as an Indicator of the Provenance of Ghanaian Ceremonial Objects. Honors thesis in Geology, Syracuse University.

Van Dantzig A. (1976) – English Bosman and Dutch Bosman: A Comparison of Texts, II. History in Africa, 3: 92–126.

Volper J. (2021) – Three Silver Cimpaaba: Afro-European Exchanges on the Kongo Coast in the 18th and 19th Centuries. Tribal Art, 25(4): 82–95.

Wallerstein I. (1980) – The Modern World System II: Mercantilism and the Consolidation of the European World Economy, 16001750. New York, Academic Press.

Wartemberg J.S. (1951) – Sao Jorge d’El Mina, Premier West African European Settlement: Its Traditions and Customs. Ilfracombe, England: Stockwell Ltd.

Haut de page

Notes

1 Akan refers to an ethnolinguistic group that includes the Akwapim, Akyem, Anyi, Aowin, Asante, Bono, Denkyira, Fante, Kwahu, Nzima, and Sefwi in the southern portions of Ghana and Ivory Coast. Although a linguistic classification, the Akan share a great deal cultural similarity. For discussion and review see DeCorse 2021a: 19; Kiyaga-Mulindwa 1980; Murdock 1959: 253; Terray 1995: 21–31.

2 Syracuse University’s Central Region Project has surveyed 100s of sites in central and western coastal Ghana and excavated a number (DeCorse 2021a, viixviii). Although evidence for iron production in coastal Ghana can be said to be widespread during the 1st millennium AD, indications of large scale production are limited. The earliest dated furnace is from the Komenda Asaba site where slag and iron artifacts were recovered from a context provisionally dated to AD 230 to 630 (Chouin 2009: 727–728). A poorly dated smelting furnace associated with a small slag scatter was also excavated at Cape Coast (Penfold 1971; Pole & Posnansky 1973). Further west in the southern Pra River basin, a slag mound near Adiembra dated to the early 1st and 2nd millennium CE represents comparatively small-scale production (Reid 2022a: 60). Many other sites in the central and western Ghanaian coast likely dating to the 1st millennium AD include small scatters and isolated fragments of iron slag. Many of these contexts also produced quartz flakes, stone beads, and ground stone celts (e.g., Amartey 2021a: 172–178, 2021b; DeCorse 2005).

3 Aside from the isotopic characterizations of the Elmina and Eguafo brass vessels discussed, the elemental composition of the cast artifacts, crucible residues, and other metal artifacts considered, have not been assessed and the term ‘copper alloys’ is used to indicate this uncertainty. However, with the expansion of European trade, it is very likely that local production relied on European brass brought to the coast. In the case of mforowa, the material used was European sheet brass. Insights into the composition, origins and ages of metal artifacts recovered archaeologically in Africa continue to be limited by the lack of detailed information (see Craddock & Hook 1995; Herbert 1984: 99).

4 This observation was made by Timothy Garrard (1980a: 104), who also noted that early castings may have been in gold because of the limited availably of copper alloys. Also see review in Craddock & Hook 1995.

5 Stylistically, the crotula are similar to ethnographic pieces associated with the Frafra in northern Ghana (Timothy Garrard personal communication; also see Garrard 1986: 571; Shinnie & Kense 1989: 212). De Marees (1987: 175) may illustrate similar ornaments in use on the Ghanaian coast during the early 17th century, possibly associated with slaves from the interior. Pieces such as these hint at the long-distance connections that existed between coastal emporia, such as Elmina, and the northern forests and savannah.

6 The more complete of the Elmina crucibles has an exterior diameter of 5.5 cm and was at least 6 cm in height (Fig. 3). The vessel was made by pinching; the surface is somewhat irregular and roughly smoothed. The paste is highly vitrified, with coarse temper, uniform grey in color, with larger dark grey or black inclusions. The paste is very distinct from other local pottery. No other casting material was recovered from this context.

7 De Marees referred to kacraws or kakaras, a type of faux gold dust used as currency at Elmina and some other portions of the coast by the early 17th century (DeCorse 2021a: 127–128, 230 note 88; de Marees 1987: 65). There are numerous references to the adulteration of gold dust with other metals on the coast through the 19th century (e.g., Garrard 1980a: 66).

8 Similar crucibles, mold fragments and casting debris were recovered from the sites of Asebu in the Central Region and Dawu in the Eastern Region, both likely also representing 17th century contexts (Garrard 1980b: 105–108; Nunoo 1957: 15–17; Shaw 1961: 59–62; Ozanne 1962: 121). The Dawu excavation produced substantial amounts of casting debris, including 288 crucibles and 318 mold fragments. Garrard (1980b: 88–105) suggested that the Dawu site chronology is longer than proposed by the excavator and that the earliest casting evidence might date to the 16th century. He further suggested that the casting represented at Dawu was in its formative stages at that time, with the workers somewhat unfamiliar with the technique (Garrard 1980b: 100–101, 104). However, the European trade beads from Dawu include type-varieties possibly dating to the 18th and 19th centuries, thus post-dating any of the chronologies proposed. A 17th century age for the early occupation of the site is more likely, but the site chronology remains uncertain. Some of the crucibles from Begho and examples documented ethnographically in northern Ghana are larger than those documented at Elmina, Asebu and Dawu, and the molding more elaborate (see Garrard 1980b: 115–124; 1986).

9 This observation is based on the examination of museum pieces attributed to the Akan, as well as observations of modern brass casting in Ghana and Ivory Coast. The casting processes are well documented ethnographically (e.g., Fox 1986, 1988; Oppong et al. 2019; Rattray 1969a: 309–316; Ross 2009: 164–190). There is some reference in documentary sources to Akan goldsmiths producing wire, a process reliant on heating and drawing primarily recorded in metal working traditions in eastern, east-central and southern Africa (Herbert 1984: 78–82). Describing coastal Akan metal workers in the late 17th century, Jean Barbot (1992: 527, 541 note 37, 542–543 note 41) noted their ability to make “drawn gold thread which they make fine as a hair.” William Bosman also refers to the fine gold and silver hat bands “the Thread and Contexture of which is so fine, that I question whether our Goldsmiths would not be put to it to imitate them” (Bosman 1967: 28–129; cf. Van Dantzig 1976: 117). Notably, however, the manufacture of gold thread is not documented elsewhere and does not seem to appear in later sources. It is possible that Barbot was observing gold thread recycled and rewoven from European clothing. During the 17th century, gold and silver thread in clothing was both an expression of wealth and a means of movable currency, particularly in England (e.g., Nguyen 2020). There is some reference to the trading of European hats and clothing on the West African coast (Alpern 1995) and gold thread may be a poorly documented aspect of European trade.

10 There is occasional confusion in the classification of nkuduo and mforowa in museum collections, some mforowa described as nkuduo and vis a versa. The stylistic differences and similarities of the two forms are discussed below. However, the major distinguishing features are that nkuduo are cast and the form originated earlier, while the mforowa are made of sheet brass and date later.

11 Forming alluvial gold such as that found in coastal Ghana does not require smelting or melting. Large, naturally occurring pieces are malleable and can be readily shaped using cold-forming processes and hammered into thin sheets. Akan chiefly regalia such as staffs, stools, beads, bracelets, swords, and sandals made from wood, leather and other materials are still covered with gold or gold-colored foil (e.g., Cole & Ross 1977: 134–169). Stools and wooden thrones are also sometimes covered with sheet brass (e.g., Quarcoopome 1997: 139). Garrard (1980a: 38, 100) suggests that the Akan learned the method of covering objects with thin gold sheets through the northern gold trade. However, the 1st millennium AD bead from Coconut Grove indicates that the method of beating gold into thin sheets—and possibly covering wooden objects—could have existed much earlier.

12 Garrard (1980a: 105–107) discusses this and other 17th century documentary references to gold objects on the coast in his review of the early gold trade, but the type of manufacture represented in these sources is unclear.

13 Barbot (1992: 527) states that there were many goldsmiths at “Boutrou [Butre], Comendo [Komenda], Mine [Elmina], Berku [Senya Bereku] and elsewhere” who produced a wide array of objects and that “They particularly excel in filigree of the kind that is cast and run through a mould.”

14 Alpern (1995: 11–18) provides a compilation of the varied finished metal goods and unprocessed metals that Europeans traded on the coast.

15 The cargo of the Elmina shipwreck, provisionally identified as the Groeningen, included rolls of lead sheathing (Cook et al. 2016) and a gold weight made of lead was recovered from a 17th–19th century disturbed context at Eguafo.

16 Two iron molds for lead musket balls and lead spew were recovered from late 19th century contexts at the Elmina town site (DeCorse 2021a: 172).

17 Iron smelting, casting and the cold working of brass sheet metals seem to have been traditionally regarded as distinct occupations by the Akan, and in documentary, oral and ethnographic sources they are referred to as distinct activities. Yet, collectively they represent the range of metal working activities that the Akan would have been familiar with. Garrard (1980a: 3), for example, suggests that the initial working of gold may have been done by blacksmiths who shaped lumps of gold into simple ornaments using iron working techniques.

18 Even during the following centuries, brass casting on the coast may have remained limited compared to production in the interior. Garrard (1980b: 67, personal communication 1987) posited that the total number of casters that operated throughout the entire Akan area between the 15th and the 19th centuries may have been no more than a few hundred, with only a small portion located in the coastal settlements. On the other hand, descriptions by Jean Barbot (1992: 527) suggest that artisans casting gold were common in the coastal settlements associated with European trade posts by the late 17th century.

19 See examples of trade lists and reviews of the brass trade in Alpern 1995: 12–18; Garrard 1979, 1980a: 105; Herbert 1984: 125–132; Jones 1983: 4, 1985: 60–74, 1995: 12–13, 2005; Morton 2019.

20 Chouin (1995, 2009: 272–282) has documented large copper or copper alloy basins of likely 16th or 17th century European origin still in use in ritual contexts. The cargo of the Elmina Wreck, provisionally identified as the Groeningen, included stacks of nested brass basins, pewter bowls and manillas (Cook et al. 2016). The Groeningen was a Dutch West India Company ship that burned and sank when one of its cannons exploded while saluting the Castle in 1647. The ship, just arrived on the coast and not yet unloaded, was laden with what was described as a “goet cust cargesoen” (good coast cargo). Divers identified 34 stacks of nested brass basins on the site, some in stacks several meters long, and additional quantities are likely present (Cook 2012: 182).

21 See Cole & Ross 1977: 65, 85 note 23; Garrard 1979: 43, 1980a: 186, 1983: 54; Ross 1974: 45, 1983: 54.

22 See discussions and references in Aitken 1866: 231–232, 310; Day 1973; Cook 2012: 182–191; Hamilton 1967: 343–344; Hamman 2007; Morton 2019: 88-95; 189-200; Pietruszka 2011: 105–113. More efficient brass rolling mills were in use in Britain by the late 17th century but battery mills continued to predominate for the next 100 years (Day 1973: 35–36; Hamilton 1967: 343–344; Morton 2019: 193–199, 229–235).

23 Jones (1983: 4) observes that the majority of metal goods the Portuguese and Dutch traded in West Africa during the 16th and 17th centuries were produced in southern and western Germany, and were often referred to in trade lists as “Nuremberg goods”.

24 Aitken is confusing in his references to West Africa. While he refers to the Gold Coast (modern day Ghana), he then mentions “Old Calabar”, a historically important trading town in what is now Nigeria. The coasts of both Ghana and Nigeria were both important areas of British trade by the mid–19th century.

25 The rods recovered archaeologically at Elmina were approximately a foot in length (30.5 cm). Although Aitken suggests brass rods were commonly buried in graves, none have been recovered from mortuary contexts in coastal Ghana.

26 For example, while tons of manillas were sent to Elmina, only two fragments were found in archaeological excavations (DeCorse 2021a: 124).

27 Notably, de Marees (1987: 52) made a similar observation in the early 17th century, “Although these [brass] Basins are brought there [coastal Ghana] in such quantities and are not as perishable commodity as Linen, one does not see much old brass-ware there; so, there must be a huge population in the Interior which uses and employs such quantities of imperishable goods.”

28 Local manufacture in or near Elmina is perhaps suggested by the prevalence of these items at Elmina compared to other sites. However, this may also be representative of the volume of trade at Elmina, as well as the excavation contexts represented.

29 For other examples of sheet brass objects related to the gold trade see Bellis 1972: 71–73; DeCorse 2021a: 130–131; Garrard 1980b: 90, 108–111). Based on the available information, all of these artifacts date to the 17th century or later.

30 Riveting, a procedure already common in antiquity, was widely used in the manufacture of European brass wares traded in West Africa (e.g., Collette et al. 2011). Riveting was also a common European method of repairing vessels. Modern rivets are hollow or solid metal cylinders with a head on one end. The rivets are inserted through holes drilled in the pieces of metal to be joined and the end without a head pounded flat making a permanent fastening. Rivets in examples of mforowa examined at the Smithsonian National Museum of African Art, the Fowler Museum at UCLA, and Ghanaian antiquities markets appear to all be solid metal cylinders, likely segments of copper wire. The rivets do not appear to have pre-formed heads; rather both ends are pounded flat, something that could be done by cold-working. Riveting is not present on pre-European contact Akan vessels and the technique is likely associated with the European trade and the availability of sheet brass.

31 Surveying mforowa from documented provenances in the collections of the Ghana National Museum, Crowley found that specimens clustered in the Central Region and extended no more than 60 miles inland, with no examples documented from Kumasi and vicinity (Ross 1974: 43). Also see Cole & Ross 1977: 64–66; Ross 1983. Documented archaeological examples are exclusive to the Fante coast.

32 It is possible that mforowa were inspired by European forms. However, as there are no obvious parallels or antecedents in European brass wares this requires more examination (Cole & Ross 1977: 85 note 27).

33 The only kuduo from a coastal archaeological site is the example from Eguafo (Fig. 10).

34 Nana Kodjo Nquandoh III, July 14, 1990; Badu Prah, July 14, 1990; Kofi Obusu, August 9, 1990 at Elmina, personal communication.

35 Ross (1974: 45) initially suggested mforowa production began in the 19th century, but revised this estimate and pushed back the origins to the 17th century (Cole & Ross 1977: 65, 85 note 23). However, Ross further revised this date in favor of a circa 1780 to 1930 production range (Ross 1983: 54). He was influenced by Garrard, who argued that Akan sheet brass working was associated with the increasing availability of rolled sheet brass in the late 18th century, with the majority of mforowa production occurring between 1830 and 1930 (Garrard 1979: 43, 1980a: 186).

36 See following discussion of imported brass vessels associated with Elmina mforowa. For other examples of European brassware vessels used in mortuary and ritual contexts see Bellis 1972: 61–62; Daniell 1852: 16.

37 In addition to the examples discussed here, mforowa and mforowa fragments have been recovered from other archaeological contexts at Elmina and other sites. However, these are too fragmentary to determine their form and/or lack clear archaeological context (e.g., Davies 1961: 34; Garrard 1980b: 85–155; Nunoo 1957: 15). Two fragments of mforowa were excavated from late 17th to 19th century disturbed burial contexts at the Atofosie locus, Eguafo (Spiers 2007: 201–202, 256). The forms of these mforowa could not be determined, but one fragment has traces of repoussé decoration and was likely part of a domed forowa lid. The second piece is a complete bottom 8.7 centimeters in diameter. The bottom was roughly hammered into a convex shape. The body and bottom had been attached by crimping and the base riveted to the bottom. A third fragment Spiers identified as a forowa base has a number of tacks used for attachment and is more likely brass boss. Fragments of more than two dozen additional mforowa were also found at Eguafo in spoil heaps from small scale gold mining.

38 Heidi Miller, personal communication 3-30-22.

39 The disturbed, disarticulated nature of the burial is typical of the Elmina site. The sub-floor burials were often impacted by later interments and construction, as well as the intentional removal of some of the mortuary contents (DeCorse 2021a: 187–191; Miller 2021; Renschler & DeCorse 2016).

40 The forowa, associated brass pan, and surrounding soil were stabilized, removed as an unexcavated block, and returned to the University of Ghana for study. Although the artifacts were in situ, they were in an advanced state of decay and unsalvageable. With the permission of the Ghana Museums and Monuments Board, the forowa fragments were brought to the United States for conservation and study in 1993. However, conservators at the Smithsonian National Museum of African Art considered conservation impossible given the fragments poor condition.

41 The curved base of the forowa is smooth without the pronounced battering marks characteristic in the locally formed domed lids of later mforowa. A fragment of a similarly curved piece of brass, likely the bottom of a forowa, was found in a 19th century fill context at Elmina (Fig. 6). The vessel would have been approximately 11 centimeters in diameter. The surface of the brass is even and shows concentric circular grooves and a central dimple, attributes indicative of European metal vessels produced by spinning, a method seen in brass wares from the Elmina Wreck (Cook 2012: 187-191; Pietruszka 2011: 105–113).

42 The bottoms of mforowa are typically made from a single piece of brass, but pieces with bottoms constructed using two pieces can be seen in the collections of the Fowler Museum at UCLA (Doran Ross 1989: pers. com.).

43 The sheet brass staples may have been easier to fabricate than copper wire rivets. A copperware fragment from a 16th–17th century context at Bosomtwi includes both staples and rivets (Chouin 2009: 729). The small, irregular shape of the metal suggests this was a repair, as opposed fastenings used in the vessel’s manufacture. Other sheet brass artifacts recovered archaeologically, as well as many mforowa examples in museum collections (Ross 1974: 43), also bear evidence of repairs. The evidence of patching on mforowa is an indicator of local mending. However, it should also be noted that European metal workers also patched vessels, and used and patched European brassware was traded to the West African Coast (e.g., de Marees 1987: 55; Jones 1995: 164, 173).

44 Tchakirides (1999: 18–20) analyzed the samples using a Micromass Sector 54 multi-collector mass spectrometer at Syracuse University. The Variscan lead in the brass from the early 18th century forowa, the Dutch riddle tobacco box, and the early 19th century forowa discussed below may indicate geologic sources in Central and Western Europe, which were used extensively in brass workshops in those regions. The isotopically distinct samples from the early 18th century forowa have pre-Variscan lead, which is rare in Europe but occurs in Swedish copper and this would result in a mixture of lead isotope ratios (Tobias Skowronek, personal communication April 6, 2022). For general discussion of the industry see Morton 2019: 77-98.

45 The sub-floor contexts in this area were complicated, consisting of shallow deposits beneath 1873 floor levels resting on bedrock that were disturbed by later construction and/or a circa 1873–1880 burial.

46 Only single samples from the lid’s rim and body were analyzed, so it is possible that different sources are represented in other parts of the vessel. The Variscan lead in the brass may indicate geologic sources in Central Europe (Tobias Skowronek, personal communication April 6, 2022).

47 There were four partially overlapping burials in this location. The burial data are reviewed by Miller 2021. The associated mortuary finds are still being studied, but all of the burials clearly pre-date the 1873 destruction of the town.

48 In particular, there seems to have been a marked increase in the range of items traded on the coast by the Dutch during the late 17th and 18th centuries.

49 The production of objects specifically for export to certain areas was an early and important aspect of European trade. The earliest trade items brought to West Africa by the Portuguese included manillas (copper alloy bracelets) made to specification in Flanders and Germany, and various forms of manillas were produced in England and elsewhere through the 19th century (Denk 2020). Bristol produced ‘Guinea Kettles’ for the African market: large, deep-sided pans used for salt production (Day 1973: 169 note 42). France and Portugal made Kongo cimpaaba (ceremonial swords) for chiefs in Cabinda during the 18th and 19th centuries (Volper 2021), while Joseph Banks made brass patu clubs for the Māori in Polynesia (Coote 2008). Some of these items were ‘high status’ articles that served in specialized contexts. However, hoes, machetes and metal pans of specific forms were more utilitarian; improved versions of local items and implements.

50 This time range somewhat corresponds to Garrard’s suggestion that the majority of all mforowa production was between 1830 and 1930 (Garrard 1979: 43; 1980a: 186). Garrard’s assessment was based on his study of museum and antiquity markets, including some pieces with ethnographic context. The vast majority of these examples are dome shaped and likely represent more recent production.

Haut de page

Table des illustrations

URL http://journals.openedition.org/aaa/docannexe/image/3707/img-1.jpg
Fichier image/jpeg, 252k
Titre Figure 1 – Map of coastal Ghana showing locations mentioned in the text
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-2.jpg
Fichier image/jpeg, 129k
Titre Figure 2 – Two crotula (paired bells) excavated at Elmina from 18th and 19th century mixed fill contexts
Légende Artifacts such as these were likely produced in northern Ghana and hint at the wider trade networks with which the coastal emporia were connected
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-3.jpg
Fichier image/jpeg, 150k
Titre Figure 3 – Crucible fragment from a pre–1873 mixed fill context at Elmina
Légende The crucible’s paste is highly vitrified and distinct in appearance from other local ceramics
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-4.jpg
Fichier image/jpeg, 37k
Titre Figure 4 – Illustration from Pieter de Marees (1602) depicting how men dress on the Gold Coast
Légende - A (at left) is described as “a Nobelman or great Monsieur, whom they call Brenipono, as he goes through the streets every day, with a cap on his head, like a Bonnet; his mantle or wrapper is of Linen.” - B. “shows a Merchant, whom they call Batafou, coming from distant Lands to trade on the seashore, wearing a hat made of Dog-skin on his head and a roll of Cotton or Linen around his body; in one hand he has assegai and in his right hand a copper Basin.” - C. depicts “an Interpreter, who comes with the Merchants or Peasants to buy things on the Ships, having on his head a little Hat made of Sugar-cane.” - D. (in the background) “shows how a Merchant, having done his trade with the Dutch, returns home with his Catious or Slaves, loaded with his Merchandize.” De Marees 1987: 33, from de Marees 1602: Plate No. 2, reproduced courtesy of Leiden University Libraries, Digital Collections https://digitalcollections.universiteitleiden.nl/​search/​Pieter%20de%20Marees?type=edismax
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-5.jpg
Fichier image/jpeg, 244k
Titre Figure 5 – Brassware from the Elmina Wreck, provisionally identified as the Groeningen, a Dutch West India Company ship that burned and sank in front of Elmina Castle in 1647
Légende The ship was loaded with cargo arriving on the coast, including stacks of nested brass basins, pewter bowls, and manillas
Crédits © Nicole Hamann
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-6.jpg
Fichier image/jpeg, 134k
Titre Figure 6 – Artifacts possibly locally made or modified in coastal Ghana from 19th century contexts at Elmina
Légende Rim of imported brass basin with engraved decoration (left); a copper wire bracelet (upper right); brass ring and bracelets; a forowa bottom made from a European vessel (lower right)
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-7.jpg
Fichier image/jpeg, 61k
Titre Figure 7 – European bracelet made from brass tubing with machine engraved decoration
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-8.jpg
Fichier image/jpeg, 94k
Titre Figure 8 – Artifacts related to the gold trade made from imported sheet brass: a gold dust shovel and a spoon of heavy sheet brass, both from 19th century contexts at Elmina; sheet brass gold dust box and spoon with engraved and repoussé decoration
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-9.jpg
Fichier image/jpeg, 186k
Titre Figure 9 – A forowa with a domed lid
Légende This is the most common form seen in historically and ethnographically documented examples and likely dates to the late 19th and early 20th centuries. The grid in the background is 1 cm squares.
Crédits C.R. DeCorse collection
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-10.jpg
Fichier image/jpeg, 38k
Titre Figure 10 – A kuduo from a probable mortuary context disturbed by small-scale gold mining at Eguafo
Légende The open grill work on the base is stylistically similar to the chased and repoussé decoration found on the bases of mforowa. The grid in the background is 1 cm squares.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-11.jpg
Fichier image/jpeg, 162k
Titre Figure 11 – Drawings of late 17th or early 18th century forowa fragments from Elmina showing the lid, the lid’s rim, body, bottom and base
Légende The repoussé decoration on the lid features a crocodile or lizard. The truncated oval representations on the top center of the body may represent stylized treasury bags or locks, both of which suggest security. The angular, recurring motif to the right of the flower-like motif may be a stylized key. The repoussé ribbed decoration on the base is reminiscent of the open grill work found on the bases of some nkuduo. The bottom of the forowa is curved.
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-12.jpg
Fichier image/jpeg, 126k
Titre Figure 12 – Schematic diagrams and detail of the bottom of the forowa shown in Figure 11
Légende The forowa’s pieces were attached using a combination of crimping, rivets, and staples.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-13.jpg
Fichier image/jpeg, 129k
Titre Figure 13 – A Dutch riddle tobacco box from a late 17th early 18th century context at Elmina
Légende Historical and archaeological data indicate that mforowa and European metal vessels often functioned in similar contexts.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-14.jpg
Fichier image/jpeg, 255k
Titre Figure 14 – An early forowa lid with a hinge, hasp, and cast brass handle from a disturbed context at Eguafo
Légende a. top; b. bottom. The grid in the background is 1 cm squares.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-15.jpg
Fichier image/jpeg, 496k
Titre Figure 15 – 19th century mforowa from Elmina
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-16.jpg
Fichier image/jpeg, 274k
Titre Figure 16 – A 19th century forowa from Fisherkrom, Elmina
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-17.jpg
Fichier image/jpeg, 28k
Titre Figure 17 – A 19th century forowa from Fisherkrom, Elmina, side view (left) and top view (right)
Légende The radiating lines and twelve-point repoussé star in low relief on the top of the lid are barely visible. The forowa’s base is missing. The photograph also illustrates the poor condition of many of the sheet brass artifacts recovered archaeologically. The grid in the background is 1 cm squares.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-18.jpg
Fichier image/jpeg, 407k
Titre Figure 18 – Rattle constructed of sheet brass, comparable to the method used in the domed lids of late 19th–early 20th century mforowa
Légende Collected in coastal Ghana with purported Fante provenance.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-19.jpg
Fichier image/jpeg, 46k
Titre Figure 19 – European machine-made match or strike-a-light container, probably British colonial period, 20th century
Légende The lid measures 27 mm in diameter.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-20.jpg
Fichier image/jpeg, 189k
Titre Figure 20 – Machine made forowa, probably British colonial period, late 19th century or early 20th century
Légende The grid in the background is 1 cm squares.
Crédits © C.R. DeCorse
URL http://journals.openedition.org/aaa/docannexe/image/3707/img-21.jpg
Fichier image/jpeg, 42k
Haut de page

Pour citer cet article

Référence papier

Christopher R. DeCorse, « Brass Working and Mforowa Manufacture among the Akan of Coastal Ghana during the 17th–20th centuries »Afrique : Archéologie & Arts, 18 | 2022, 11-38.

Référence électronique

Christopher R. DeCorse, « Brass Working and Mforowa Manufacture among the Akan of Coastal Ghana during the 17th–20th centuries »Afrique : Archéologie & Arts [En ligne], 18 | 2022, mis en ligne le 02 novembre 2022, consulté le 27 avril 2025. URL : http://journals.openedition.org/aaa/3707 ; DOI : https://doi.org/10.4000/aaa.3707

Haut de page

Auteur

Christopher R. DeCorse

crdecors@maxwell.syr.edu — Department of Anthropology, Maxwell School of Citizenship and Public Affairs, Syracuse University, 209 Maxwell Hall, Syracuse, New York 13244, USA

Haut de page

Droits d’auteur

CC-BY-NC-4.0

Le texte seul est utilisable sous licence CC BY-NC 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search