The past decade has seen much ink spilled on the relationship between European and Asian economies between 1500 and 1800. This has focused in part on how the European re-export of American and Japanese silver to China forged a global currency relationship. This debate has yet to include Africa, but this article shows however that West and West-Central Africa were part of the global price revolution of the sixteenth and seventeenth centuries.

Classic discussions of the price revolution have pitted Earl Hamilton's thesis on the impact of New World silver imports in the sixteenth century against historians such as John Elliott and Keith Wrightson, who claim that a focus on currency imports ignores the role of demographic increases and the rise of local demand in Europe.Footnote ¹ These discussions of England and Spain tend to suggest that economic changes in these European economies arose from internal forces and the creation of demand. This has also long been a feature of historiography on Dutch economic growth, as Kwame Nimako and Glen Willemsen note.Footnote ² However, the validity of this internally-driven growth thesis has been challenged by the more recent linkage of the price revolution in Europe to transformations in China. Once this material is considered, the plausibility of the internal thesis for inflation and economic growth diminishes: and it is here, this article argues, that the evidence from West and West-Central Africa is vital.

Dennis Flynn and Arturo Giraldez have driven this new work so far.Footnote ³ Flynn and Giraldez show that silver imports to China in the sixteenth century provoked rapid silver depreciation, reducing the profits of the middlemen Portuguese and Dutch; whereas in 1635 it took 13 ounces of silver to buy an ounce of gold in China, fifty years previously it had taken only six ounces.Footnote ⁴ Moreover, they argue that this process diminished the profits from silver mining in the New World, contributing to the crisis of the Spanish empire.Footnote ⁵

This interpretation holds that the place of currency imports in the rising inflation of silver prices trumped that of internal demand growth. The evidence from West and West-Central Africa discussed in this article supports this view, revealing a significant expansion of local currency supplies, inflation, and yet some areas of negative demographic growth. Though there were increases in demand, these were not driven by demography: demand was increased by pushing markets of exchange inland, as scholars such as J. E. Inikori and Paul E. Lovejoy have noted.Footnote ⁶ Transformations in the structure of demand do not reveal a ‘special case’ among European economies. Nor is the nature of European demand increases universalisable into a general economic model. The evidence considered here thus argues for the price revolution's importance in Africa, and seeks to decentre and modulate its analysis as seen from Europe and Asia.

Much evidence exists as to the interlinkage of African and global economic histories through currency. First, there is the place of West African gold, as shown classically by Vitorino Magalhães Godinho.Footnote ⁷ But beyond the Gold Coast, the question of Africa's mines was also important. In West-Central Africa, a principal reason for European interest in the sixteenth and seventeenth centuries was the perceived existence of rich seams of copper, gold, and silver. As Beatrix Heintze notes, hardly any document from this period ignores these mines.Footnote ⁸ With a silver-copper alloy known as vellón in Castile, the frequent citations of copper mines influenced West-Central Africa's interconnections to the world economy.Footnote ⁹

In spite of this importance of Africa as a source of global currency, the continent's economies have never been considered as part of the price revolution; even though, as Flynn and Giraldez write, the export of contraband silver from Buenos Aires in exchange for contraband slaves from Angola formed part of this global picture.Footnote ¹⁰ The importance of the silver/slave trade nexus in this period is underlined by Portuguese royal policies on Luanda of the 1650s, during Portugal's war of independence from Spain: these permitted a direct slave trade from Buenos Aires because of the associated inflow of silver, but prohibited a triangular slave trade originating from Spain.Footnote ¹¹ Given that most of this silver was destined for the Chinese economy, such evidence shows, as Nimako and Willemsen argue, that the functioning of the Atlantic world, including the slave trade system from West and West-Central Africa, should be considered at least in part within a global economic system.Footnote ¹²

It is important to grasp why African economic changes have not traditionally been seen as part of the price revolution. As the foregoing historiographical summary shows, discussion of the price revolution has hitherto been restricted to the role of bullion, and in particular of silver imports in currency depreciation. This particular process of currency change did not happen in Africa, which has therefore been cut out of this discussion. Nevertheless, new research shows that there are problems with this approach. Akinobu Kuroda's work on the complementarity of monies shows that until recent times the history of currency is one of plurality involving specie and non-specie,Footnote ¹³ and moreover that such pluralism represents normal market mechanisms rather than ‘primitive forms’ of financial operation. As Kuroda shows, ‘commodity currencies’ (that is, not specie/bullion) had specific uses determined by local preferences in many parts of the world.Footnote ¹⁴ Hence, incorporating different types of currency beyond specie into an understanding of the price revolution – such as those in use in West and West-Central Africa – is a sensible approach.

That being so, moreover, Asian and European economic changes may only be fully understood if their relationship with African economic trends is also delineated. Yet this process has not been completed by historians. The two major synthetic works on longue durée African economic history were written many years ago by A. G. Hopkins and Ralph Austen.Footnote ¹⁵ One of the main causes of this is what Richard Reid called the ‘presentism’ of Africanist studies;Footnote ¹⁶ and yet, as Jane Guyer has shown, to study the historical creation of currency systems in West and West-Central Africa is also to study the systematic construction of Atlantic Africa as a purveyor of raw materials with an extractive base, and to understand the deep roots of the current unequal currency system.Footnote ¹⁷

What are now called ‘hard’ and ‘soft’ currencies originate in some of the features analysed here, on this account. As Guyer has recently noted, hard currencies are those which retain value over time, where soft currencies depreciate.Footnote ¹⁸ Though these terms arose to describe the patterns of currency exchange since the Second World War, this article shows how this material difference encapsulates the trends in West and West-Central African currencies in the seventeenth century, where African currencies depreciated and lost value in relative global terms when compared to other currencies in widespread use. The pattern that emerges is one between currencies used in Africa losing relative global value compared to those such as silver and gold used elsewhere.

This may seem strange to some readers, since gold-dust and gold nuggets were currency in areas of the Gold Coast in the seventeenth century. However, as Ray Kea has noted, it was in these areas a commodity currency; that is, dependent on the production of gold as a commodity traded to the outside, and moreover coexistent with the cowrie currency whose relative value declined consistently.Footnote ¹⁹ Currency history in West and West-Central Africa is therefore vital not only to understanding the price revolution as a whole, but also to conceptualising longstanding structural economic relations between the region and the world, something resonating with recent debates on historical inequalities in global development.Footnote ²⁰

The history of currency is also very important in establishing the extent of markets and credit institutions in precolonial Africa.Footnote ²¹ As Inikori has noted, currencies constituted the dominant imports into Atlantic Africa until the middle of the seventeenth century, and may have helped expand market mechanisms.Footnote ²² And yet there has been meagre sustained research on the subject of currencies for the period to 1700, and the most important overall summary, by Philip D. Curtin, is unfortunate in suggesting that African monetary systems were isolated from worldwide metallic monetary systems, and that cowries were not thought of as money by Europeans – where Portuguese regularly shipped nzimbu shells from Brazil to Angola and were happy to receive payments in nzimbu in the sixteenth and seventeenth centuries.Footnote ²³ Nevertheless, some historians have recognised the importance of imports through Atlantic ports on monetary structures in the region before the eighteenth century, albeit without extensive data to drive the point home.Footnote ²⁴

The analysis proposed here is therefore new in bringing primary data to the pre-eighteenth-century period, and drawing theoretical connections with models developed for later periods. Some scholars will feel that including large parts of West and West-Central Africa offers too general a perspective, without sufficient awareness of local variations and factors in demand. This is in part dictated by the limitations of the sources available for the study of this question during the seventeenth century. Instead of a detailed corpus of economic data from one or two archival bases, analysis of this subject requires the integration of fragments from multiple archival records, written in a variety of languages. This approach speaks to the nature of the sources themselves, and the aggregation of the different materials consulted slowly builds a strong evidence-based picture. Moreover, what emerges from the article are the ready comparisons which can be made between regions as different as Angola, the Bight of Benin, the Gold Coast, and Upper Guinea; comparisons which suggest that one of the factors in the gradual convergence of social and human experiences across the region – the process contributing to what V. Y. Mudimbe called ‘the invention of Africa’ – was precisely its comparative position in world economies.Footnote ²⁵

This article therefore ranges widely in order to build this picture. Much material is drawn from the region between the Gold Coast and the Bight of Benin, with further comparative examples from both Angola and Upper Guinea. The article argues that changes in currency use in these varied regions indeed formed part of the global price revolution. We see that, as elsewhere, one important consequence of the widespread import of currencies was inflationary. Inflation itself did not provoke negative consequences for economic productivity, but what mattered for future economic trajectories was the relative value of currencies used in the region.

The place of gold exports is central to this. These gold exports depressed the relative value (and purchasing power) of the multiple currencies used within West and West-Central Africa, accelerating the process of the ‘softening’ of African currencies. This in turn increased the volume of African trade goods needed to purchase items from Atlantic traders, and the area reached by the internal market for exchange and trade in these currencies for slaves and other commodities.Footnote ²⁶ Thus demand was increased not through demographic growth, but through the expansion of the market area and of exchange mechanisms, as processes of inflation and relative currency depreciation (or ‘softening’) took hold. Importantly, this process was not just restricted to areas where gold dust was a principal legal form of exchange (for example, Asante): it was the relative global value and weight of currency imports to and exports from Africa, which determined the relative position of the plural monies used within the continent, and their relative weight within global trade exchanges.

In part, the argument relating to inflation here therefore follows a monetarist approach and emphasises currency imports in the price revolution in West and West-Central Africa. Nevertheless, the article does not take a strictly monetarist stance because it does not adopt a narrow monetarist view of currency uses. As the examples studied show, currency held both a role in quantitative accumulation, and a mutually constituted social meaning connected to social reproduction and changing relationships to Atlantic trade. Indeed, as Akinwumi Ogundiran has shown so clearly in the case of Oyo, it was through the social and ritual uses of the cowrie currency used there that the power and meaning of Atlantic transformations was most powerfully expressed.Footnote ²⁷ Such social meanings were however lost to exchange value in the wider Atlantic world, as the rise of merchant capital made the production of surplus-value the key component of global currency regimes: hence while such parallel use values help in understanding the immediate impact of currency imports, on a global scale they contributed to the creation of soft currencies in Atlantic Africa.

Wherein lay the origin of these parallel use-values of currencies? As Joseph Miller observed for West-Central Africa, currencies in this period were not solely valued for their exchangeability for material possessions.Footnote ²⁸ There are two factors that need to be taken into account to understand this point. In the first place, in common with many commodity currencies, those used in West and West-Central Africa often had an intrinsic value beyond their inherent exchangeability. Hence, cloth could be worn, copper and iron melted down and used to make functional and decorative objects, and cowries could – and were – used extensively for decorative purposes, for instance as a form of decoration on bags and as jewellery, alongside many other non-monetary functions.Footnote ²⁹ Using Georg Simmel's theories on the philosophy of money, we can recognise that, whereas standardised currencies presuppose a general hierarchy of value in relation to the material world, multiple currencies such as existed in most (if not all) of the region studied here suggest plural and less standardised relations between individuals, currencies, and the objects of trade.Footnote ³⁰

In the second place, moreover, imported currencies very often acquired an important ritual function which was valued as much or more than their monetary function. Indeed, Mervyn Hiskett noted this for the cowrie currencies of the western Sudan, arguing that the Hausa used cowries as a means of divination as well as a store of value;Footnote ³¹ while, as Lars Sündstrom argued, cloth was often used in ritual contexts relating to brideprice and other formal duties.Footnote ³² In the case of Oyo, meanwhile, Ogundiran's detailed analysis based on archaeological and oral records shows the importance of cowries in burial, divination, household Orí shrines, and public religious architecture.Footnote ³³

Hence the key to grasping the importance of this social meaning of currency is to diversify our understanding of the word ‘value’ beyond a reductive meaning linked to exchangeability, by seeing how value also inhered to growing stores of religious and spiritual power through the use of cloth and cowries in apparently ‘non-productive’ ways. This view builds on Gareth Austin's critique of traditional institutional economics, namely that it attempts to make normative European beliefs in contract-enforcement, rational-choice political economy, and – we might add here – currency use.Footnote ³⁴ Moreover, this discussion of multiple currency use is not a matter of simply re-engaging with the old debate between formalists and substantivists, now long overcome.Footnote ³⁵ Rather, just as Kuroda has encouraged scholars to move beyond a rigid sense of singular ‘money’, it recognises the importance of grasping the diversity of ‘value(ations)’ which contributed to the use of currencies at this time and the development of market mechanisms; thereby following Jane Guyer's imprecation to move beyond the analyses which ‘rest on the qualitative invariance of money’.Footnote ³⁶

Thus the evidence considered here places the growing ‘softening’ of West African currencies in global context. The relationship of this process to the export slave trade is important. As West and West-Central Africa's currencies softened through the seventeenth century, the export slave trade grew. Fundamentally, the material of this article moves Paul Lovejoy's model linking inflation and imperialist expansion in the late eighteenth century back to the seventeenth century, expanding on Sündstrom's awareness that such inflation was also relevant to the seventeenth century, and that inflation in the West and West-Central African context tended to follow large imports of currency.Footnote ³⁷ Where successive stages of nineteenth-century inflation corresponded to new phases of European imperialism and capitalist development, it is argued here that the prior inflation of the seventeenth century is also symptomatic of an earlier such shift, towards the Atlantic slave trade and away from a more mixed economy involving cloth exports.Footnote ³⁸ The argument pursued here is that this seventeenth-century phase linked inflation to the expansion of value extraction through labour, and the creation of a global currency system in which West and West-Central Africa were increasingly disadvantaged.

In order to make these suggestions, this article is divided into three further parts: in the first part, we look at commercial changes in West Africa in the seventeenth century, with a particular focus on the Gold Coast and the Bight of Benin and comparative material from West-Central Africa; in the second part we look at new evidence for currency imports and their effects; and in the final part we see how this relates to the different uses of currencies and the creation of hard and soft currencies.

*

Prior to 1700, the region stretching from Elmina to the Cross River and Elem Kalabari in the Niger Delta was a partially integrated trading area with a manufacturing base in cloth production. The export slave trade was already intensive, but manufacturing, together with the export of raw materials such as ivory (and gold on the Gold Coast), was also important. However, by the latter seventeenth-century exports of enslaved Africans began to predominate.

A significant trading station here was Allada, where trade grew from the late sixteenth century onwards.Footnote ³⁹ Nicolás Ngou-Mve suggested that after 1610, Allada's slave trade to Mexico expanded significantly.Footnote ⁴⁰ Similar conclusions were made by David Wheat for the trade to Cartagena, in what is now Colombia, for by the late 1610s enslaved Africans from Allada were appearing in significant numbers there.Footnote ⁴¹ Certainly by this time there was an active trade here, with Damião Ramires shipping two vessels to Allada and one to the Niger Delta region – which he then called ‘Calabar’ – annually between 1619 and 1625, all trading for slaves in return for cowries.Footnote ⁴² By the early seventeenth century, therefore, Allada's Atlantic trade was becoming heavy. An anonymous Dutch text dating from 1602 described how the lagoon access to Allada ‘is much used to be entered into by the Portugals, and is well known … because of the great number of slaves which are bought there’.Footnote ⁴³

These references make clear that the trade in enslaved Africans was pivotal to Allada, as Robin Law emphasises for the early seventeenth-century trade. We know that 400 enslaved Africans were sent annually to São Tomé to work on the sugar plantations in the 1640s, and that there was a regular trade in enslaved people from here to the Americas as Ngou-Mve points out.Footnote ⁴⁴ By the 1640s, enslaved Africans named as ‘Alladas’ were regularly mentioned in Brazil.Footnote ⁴⁵ However, slavery was still complemented by other export trades from Allada. This is important since it suggests that the expansion of slave trading (c. 1600) probably coincided initially with an expansion of Allada's other export trades. The Dutch sailor Dirck Ruiters described Allada as a place where much ivory was traded.Footnote ⁴⁶ Meanwhile, an anonymous account, composed in 1620, describes a flourishing trade in ivory, cotton cloths woven in Allada, palm oil, and provisions for ships.Footnote ⁴⁷ This mixed trade continued, since a navigational aide published in Madrid in 1635 also described a mixed trade in Allada, where peppers, skins, and gold were exported as well as enslaved Africans.Footnote ⁴⁸

This export of cotton cloths tied Allada into a broader regional manufacturing base. By this time there were regular trading relations between the Portuguese of São Tomé and the peoples living further east of Allada, in Benin, Elem Kalabari in the Niger Delta, and on the Forçados river. The Portuguese bought cotton cloths in all these areas as well as slaves, although these cloths may have been manufactured further inland and then transported to the coast. The manufacturing of loom-patterned cotton was well-established, confirmed in the savannah regions of West Africa by the eleventh century, and there was also extremely skilled weaving in the northern part of Igbo country in present-day Nigeria.Footnote ⁴⁹ The importance of this regional cloth manufacture in trade of the time is shown by an anonymous late seventeenth-century document, which describes how in ‘past eras of governance’ (governos passados) small ships used to come to São Tomé regularly from the coast to disembark slaves and cotton cloths.Footnote ⁵⁰

How widespread, then, was this Atlantic trade in West African cloth? An inventory of goods sent by the Dutch from Olinda in Brazil to Luanda in Angola, in 1641 included thirty pieces of ‘ordinary cloth’, thirty pieces of red cloth, and 2,200 bolts of ‘cloth from Guinea’.Footnote ⁵¹ Writing in 1642, the Dutch factor in Luanda, Cornelis Hendrickx Ouman, wrote that ‘cloths from São Tomé, Benin and Ardra [sic] are in high demand along this coast and are sold at a good profit’.Footnote ⁵² His predecessor Pieter Moortamer confirmed the production of cloths in São Tomé in 1641, revealing also how this trade touched not only the Gold Coast region and the Bight of Benin, but also areas as far as Angola.Footnote ⁵³ An undated document, probably from the 1650s, describes the ship De Pietas arriving on the Caribbean island of Curação with cloths from Guinea, which were then to be re-exported to New Netherland (New York).Footnote ⁵⁴ In 1670, agents of the First Dutch West India Company (hereafter: OWIC) reported that the only trade in Benin was for the purchase of small pieces of Benin cloth;Footnote ⁵⁵ and as late as 1681, cloth traded from São Tomé was still used to make shirts worn in Brazil.Footnote ⁵⁶

This evidence on West and West-Central Africa's textile manufacturing base in the seventeenth century ties it into broader trends in global economic history. In return for these cloths, for the slaves, gold, ivory, and provisions, traders in Allada bought cloths woven in India and cowries from the Maldives; both of these items were used as currencies in the Bight of Benin, and as both were imported from Asia, both belonged to the global currency trade.Footnote ⁵⁷ Nevertheless, as the seventeenth century continued, the slave trade began to predominate. By the 1630s, enslaved Africans from Allada were found in the sugar plantations of Brazil, as well as in the ports of Cartagena and Mexico already discussed.Footnote ⁵⁸ Slave trading soon accelerated rapidly for the Brazilian provinces that the Dutch had occupied, with a frequent passage of ships from Allada to Elmina and thence to Dutch Brazil.Footnote ⁵⁹ The English too were trading slaves from Allada, adding to the traffic.Footnote ⁶⁰ By 1669, Sieur Delbée, a French naval officer, described Allada as the centre of the slave trade in the whole region.Footnote ⁶¹

This discussion shows that Asian and European Atlantic trades intersected in Atlantic Africa in the seventeenth century. Connections stretched even beyond those considered here, with a persistent cowrie trade in Yunnan into the seventeenth century; the strength of the cowrie currency system in Yunnan reveals a certain shared global currency system tying West and West-Central Africa and the Maldives to China.Footnote ⁶² The Gold Coast, Bight of Benin, and Angola were nodal points for global intersections, where currencies from Asia (cloths and cowries) were transhipped and became part of a different process of exchange linking the important gold exports and the increasing export of human beings to the import of some of those currencies. From the Asian perspective, this process was part of the accelerating competitiveness of Indian cloth exports.Footnote ⁶³ Thus, West and West-Central Africa were central places in this process of expanding global economic exchanges.

This material already gives us much to relate Africa to the price revolution. The role of imported Asian currencies emphasises that this expansion of trade required a rapid growth in exchange, and hence expanding imports of locally-used currency as a medium of exchange. Moreover, as we shall see in the next section, it matters that this trade in cowries imported from Asia had characterised the pre-Atlantic age, via the trans-Saharan trade. Hence, the rapidly increasing imports of this currency under the Portuguese led to expansion of a pre-existing currency base, with the creation of new Atlantic markets requiring more currency to be imported for exchange. This expansion of the use of currencies within Africa is critical when considering Inikori's view as mentioned earlier on the relationship between currency imports and internal market development, for it does suggest that demand was created by expanding the market, and that the expansion of currency imports facilitated the market's growth inland. Coincidentally with the expansion of the currency base and the market, however, a substitution of the export slave trade for other exports has also been shown. The connection is not immediately apparent, but as the remainder of the article elucidates, this was one of a range of major social impacts connected to the developments examined here.

*

Before looking at the evidence for currency imports in this era, it is worth discussing the existing state of the historiography of currency in Atlantic Africa. The pioneer of currency history for West Africa was Marion Johnson, whose works on cowries and the nineteenth-century mithqal remain standard references, relating cowrie and gold monetary zones, as well as West Africa's currency history to the wider world.Footnote ⁶⁴ Subsequent works were published, but even some of Johnson's excellent work depended on secondary sources, in the case of the Portuguese, meaning that although the subject's significance was thus established work remains to be done – especially for this early period analysed here where Portuguese sources are so important.Footnote ⁶⁵

Meanwhile, the historiography of money has ignored Africa's place in the history of the price revolution. Historians of currency have tended to perceive a graduated progression from ‘primitive’ forms of money to a universal standard of tender.Footnote ⁶⁶ Both Karl Polanyi and Orlando Patterson, who wrote on the history of various parts of Africa and the diaspora, referred to ‘archaic’ forms of economy where parallel relative currencies were in used.Footnote ⁶⁷ As Tzvetan Todorov noted in his analysis of Columbus, for Columbus different systems of exchange equalled the absence of a system.Footnote ⁶⁸

This article follows Todorov's approach, as well as those of Guyer and Kuroda noted in the introduction. As Flynn and Giraldez note, there were four main monetary substances in this period – gold, silver, copper, and cowries – and they never flowed in tandem anywhere at once: it was not because of a European trade deficit that American silver was re-exported to China, but rather because Europe and China were part of a global currency system in which differing values facilitated arbitrage profits, as well as resolving large trade deficits between European and Asian economies.Footnote ⁶⁹ Similarly, West Africa's role as an exporter and re-exporter of gold should not be seen narrowly in terms of this gold's destination in Europe, but rather in the context of the way in which these exports formed part of a wider global trend to develop differential exchange systems, forerunners of today's hard and soft currencies. Here, Africa's place as a destination for some of this global currency flow is fundamental; moreover, the fact that both before and after the era of Atlantic trade gold exports flowed continuously north across the Sahara underpins the region's longstanding importance in global currency flows.Footnote ⁷⁰

As already mentioned, the overwhelming majority of goods imported for trade to Africa in the sixteenth and seventeenth centuries were used as currency.Footnote ⁷¹ In Angola, the predominant imported currencies were the libongo cloths woven north of the Zaire River in Loango and – as we have seen – shells called nzimbu, imported from Brazil by 1600.Footnote ⁷² Further north, in Upper Guinea, iron was being imported from the late fifteenth century onwards; it was the product in highest demand, and by the early seventeenth century at the latest goods were exchanged in units measured in iron bars, while small hoe-shaped iron coins were being manufactured further inland.Footnote ⁷³ Meanwhile, Eugenia Herbert shows that copper was a unit of value in much of Africa for many centuries, and that copperwares imported from the late fifteenth century onwards, much of it as manilla arm-rings, constituted a currency import.Footnote ⁷⁴ There was, too, a mass import of cowries from the early sixteenth century onwards, especially to Elmina and the Bight of Benin, and this may have been linked to a subregional system that placed cowries as part of the system of gold exchange stretching from the forests of the Gold Coast north to the Sahel.Footnote ⁷⁵

What, then, was the role of imported currencies in driving socioeconomic changes, and in connecting West Africa to the price revolution? With a greater stock of currency in circulation, there was more scope for exchange. Expansion of the currency base beyond traditional sources for the cowrie, copper, and iron imports also precipitated inflationary pressures with social impacts. Fundamental to these imports was increased global mobility from the sixteenth century onwards.

The expanding place of the gold exports is significant in this picture. Initially, the Portuguese presence on the Gold Coast was related to gold production, and indeed as Ivor Wilks showed, the early Portuguese gold exports were closely linked to the existing trade conducted by Wangara merchants from the goldfields to Djenné and the empire of Mali.Footnote ⁷⁶ Subsequently, after a lull in the later sixteenth century, the latter seventeenth century saw the re-emergence of West Africa as the major producer of gold for the world market.Footnote ⁷⁷ Indeed gold exports grew rapidly in the period studied in this article, and are thus directly related to the processes under discussion here: as Kea showed, gold exports quadrupled from the Gold Coast between 1618 and the second half of the seventeenth century, and indeed in 1679, Dutch factors were able to purchase in two weeks more than twice the volume of gold that they had purchased in the whole of 1645.Footnote ⁷⁸ Significantly, this did not only affect the old heartlands of gold production in the southern forest zones of the Gold Coast; new mines were opened up also in Senegambia in the 1680s, trading predominantly with trans-Saharan merchants, and producing a rapid spike in prices of ordinary goods according to the English trader Cornelius Hodges.Footnote ⁷⁹

The impact of the revived export gold trade on West African economies in the later seventeenth century was significant. As Nehemiah Levtzion once noted, it was after 1700 that kola nuts took over from gold as the principal export from the forest zone to the Sahel.Footnote ⁸⁰ These changes were closely linked to the rise of Asante, which by the eighteenth century operated what Lovejoy described as a bimetallic monetary system, which connected Atlantic cowrie imports to its gold exports and the growing trade between Asante and Hausaland in northern Nigeria. Gold was mined and used in the Asante heartland, as well as being exported through the coastal ports; while in the northern provinces it coexisted with cowries through the kola export trade, with the kola-producing forests of Abron and Bono Mansu under Asante control by 1750.Footnote ⁸¹ Moreover, this prominence of the region in the world gold trade continued in the eighteenth century through the Gold Coast's role in the re-export of Brazilian gold.Footnote ⁸² This export of gold, it is argued here, had a key role in the emergence of the hard/soft currency system, depressing the value of currencies circulating internally in Africa vis-à-vis global values as gold's stock rose; and as we saw above, even where gold was used as a currency within West Africa, it was primarily a commodity currency.Footnote ⁸³

This development makes historical processes of currency in- and outflows from these regions important to the price revolution. In West Africa, as with the new fluxes of global trades linking Asia and Europe, expanded local currency supplies derived from imports were related to the increasing currency exports, and the need to increase imports of other goods to balance these exports as the relative price of gold grew. This increased the supply of local media of exchange, and thereby facilitated increasing volumes of exchanges and the effects of these exchanges on societies. The need to expand the market and demand accelerated, and it is this global context to the expansion of the currency supply which explains the extraordinary rise in commercial trade on what had hitherto been a commercial backwater, the sea coast; by 1665, Michiel de Ruyter described 200–300 canoes coming to trade with his ship at Elmina.Footnote ⁸⁴

It is significant, moreover, that in the cowrie trade the Portuguese rapidly expanded the exchangeability inherent in an existing economic system, as noted above. Writing in the fourteenth century, Ibn Battuta noted that Maldive cowries were sold in Gao and Timbuktu; from here, they were transported down the Niger river and were thus an existing medium of exchange by the early sixteenth century, when the Portuguese expanded the base by using cowries as ballast on their return voyages from Goa to Portugal.Footnote ⁸⁵ Writing in 1529, João Lobato noted that there was a heavy cowrie trade operating through São Tomé, where smuggling was rife.Footnote ⁸⁶ This confirms Robert Garfield's suggestion that one to two tons of cowries were imported annually through São Tomé in the 1510s in order to finance the slave trade through Elmina.Footnote ⁸⁷

What, then, was the impact of these currency imports? Inikori sees them as positive, and as stimulating production and inter-market trading. Moreover, they integrated the regional economy, connecting the economies of the coast with those of the Central Sudan and Songhay through the exchange of gold and cowries; furthermore, they acted as a way of integrating West and West-Central African economies through the trading exchanges facilitated by the São Tomé traders. A further consequence was inflationary. In Atlantic Africa, as in China and Europe, there was a strong connection between these imports and price inflation, or the price revolution. In the case of the cowrie imports, Johnson estimates five-fold inflation between c. 1500 and c. 1775.Footnote ⁸⁸ For copper, Herbert cites evidence that from 1630 onwards, the Dutch imported between 545 and 763 tons per year, ten times what the Portuguese had imported the previous year.Footnote ⁸⁹ Hence, although the weight of the various currency ‘products’ – iron and copper bars, manillas – varied significantly, volumes of imports certainly increased. Large amounts of copper were traded at Elmina, much of it brought from the copper mines east of Loango.Footnote ⁹⁰

Eventually, copper's value depreciated. A. F. C. Ryder points out that the inflationary impact of copper imports had already been noted by Portuguese sources further to the east, at Benin, in the sixteenth century; whereas in c. 1500 an enslaved African was sold for 12–15 copper manillas according to Duarte Pacheco Pereira; by 1517 this had risen to 57 manillas.Footnote ⁹¹ By the early seventeenth century, iron emerged as a replacement currency; perhaps hardly surprising where iron was already being manufactured by Malinke smiths for use as a local currency.Footnote ⁹²

Turning to the Gold Coast and Allada, new evidence also supports the view that currency imports led to inflation and currency proliferation. The accounts of the Dutch ships Halve Maen from Amsterdam and Eendracht from Zeeland, trading in Allada in 1636, list the prices for which enslaved Africans were purchased in various media of exchange (Tables 1 and 2).

Table 1. The Halve Maen. Source: NA OWIC, Inventarisnummer 52, no. 46, 6 Apr. 1636.

Table 2. The Eendracht. Source: NA OWIC, Inventarisnummer 52, no. 46, 6 Apr. 1636.

Copper manillas were no longer nearly as valued as other currencies traded at Allada in the mid-1630s: between Pacheco Pereira's report in Benin in 1500 and this report from nearby Allada in 1636, inflation in manillas ran at something like 700%; and even copper bars were only worth one third of iron bars.

The fact that iron bars were heavily traded in the Gold Coast emerges in a 1646 inventory of goods traded there, which shows that the value in (Dutch United Province) florins of iron bar imports was only exceeded by cloth imports in that year.Footnote ⁹³ In sum, the most valued items in this region appear to be iron and cloth, at least in terms of the volume required to trade. Iron had taken over from copper, and by the 1640s heavy iron rings weighing 14 pounds were being used further along the coast in Elem Calabari as a measure of the worth of an enslaved African, and frequent mention was made of iron bars as a form of currency in Allada and near the English fort at Koromantyn.Footnote ⁹⁴

When this evidence is compared to information for the Gold Coast, it suggests that even between 1636 and 1652, there was rapid inflation in the value of these imported currencies. In 1652, a Dutch ship purchased 600 enslaved Africans at Accra for 40,000 pieces of light linen cloth known as lijwaet, at approximately 66 pieces per person.Footnote ⁹⁵ In the face of this, it is hardly surprising that, at the Gold Coast, Louis Dammel of the ship Prins Willem – who made this observation – was told that the exchange value of cloth had gone down. Moreover this was a process of currency depreciation at work elsewhere, as in both Angola and Upper Guinea slave ship captains noted that the value of their currencies increased markedly at this time.Footnote ⁹⁶ In Upper Guinea, the price of an iron bar in panos – the standard cloth currency unit in the early seventeenth century – rose from five panos per bar in 1616 to five and a half in 1618.Footnote ⁹⁷ In West-Central Africa, there was at the same time a widely noted inflation in the value of the nzimbu, which as we have seen was being imported in bulk from Brazil.Footnote ⁹⁸

We can also, finally, compare this evidence for early seventeenth-century inflation of copper bars and their substitution with iron with evidence from later in the century. In 1681 at the port of Ouidah – not far from Allada – the English factor noted that slaves cost 12 iron bars, an increase of 33 per cent from the 9 bars noted in this Dutch source for 1636;Footnote ⁹⁹ a 1686 report placed the exchange as at 14 iron bars per slave, representing a further increase, and by 1705 the rate was of 18 bars per slave.Footnote ¹⁰⁰ Meanwhile the inflation of manillas had been even stronger, and was then at 220 per slave, up from 100 in 1636, an increase of 120 per cent.Footnote ¹⁰¹ For cowries, meanwhile, whose use was burgeoning at Ouidah at this time, the price for slave purchase rose from eighty Dutch pounds in weight in 1686 to one hundred in 1705.Footnote ¹⁰² Mass currency importation and the growth of the market thus clearly tied economic and social processes in this part of Africa to those in the broader Atlantic world.

In sum, the evidence of the sixteenth and seventeenth centuries reveals both a significant expansion of the currency supply as part of a global system of exchange, and the rise of the sort of inflation in prices commonly associated with the price revolution (see Table 3, below). This is a convincing argument for taking Africa's place in this global process seriously. Gathering mobility increased the import of currencies already in use, with these increasing imports balanced by gold exports and expanding areas of market exchanges and demand within and between West and West-Central Africa. In order to understand how this global process accelerated the hard/soft currency system, however, it is vital to consider the question of currency use, and it is to this that this article now turns.

Table 3. Overview of inflation of currencies in the Bight of Benin, sixteenth–seventeenth centuries.

*

One important correlation of the foregoing material is as already noted that as currency imports increased, the export slave trade also increased in many parts of West and West-Central Africa. Meanwhile, the export and re-export of African manufactures declined. Clearly this was not a uniform process: slave exports fluctuated from different regions throughout the era of the Atlantic slave trade, and it would be wrong to propose a standardised system. Moreover, as Inikori notes, during the eighteenth century there were also heavy imports of goods such as alcohol and firearms that were not used as media of exchange.Footnote ¹⁰³ Other factors must also be considered in the growing slave trade from West Africa, not least Gareth Austin's argument that, owing to a relative abundance of land and the lack of economies of scale for production, there was little economic incentive in West Africa to contract for land or free labour.Footnote ¹⁰⁴

Therefore it is not the case that the relationship between currency imports and slave exports was a universal and permanent one. Rather, the argument here is that when the Atlantic slave trade accelerated in the seventeeth century, these local currency imports and the relation of local currencies to global currencies accelerated this process by creating a need for growing exports to match the increasing currency imports. Thus, while the growth of the slave trade and the rise of slave prices in this period is usually explained through the expansion of demand in the New World, this explanation fails to take account of the rising demand for imported currencies within Africa which also was an important factor. Inflation as shown in the rising prices of enslaved Africans was caused by a balance of local and global demands, and not just by rising demand in the New World and an unmediated impact of the external economy in shaping West and West-Central African societies.

Comparative examples from across Atlantic Africa illuminate the point. The interconnection of currency imports, inflationary consequences and a rising export slave trade has already been shown for Allada. But Allada was not alone. In West-Central Africa, the cloth currency system of libongos used in Luanda from the early seventeenth century onwards, and imported from the Loango coast, saw rapid depreciation from the 1640s, largely because of the increasing import of Dutch and Indian-produced cloth.Footnote ¹⁰⁵ As Loango's cloth exports disappeared in favour of imports, after 1670 Loango became a key site for slave exports to the Dutch.Footnote ¹⁰⁶ By the second half of the eighteenth century, it was estimated that 16,000 slaves departed Loango annually.Footnote ¹⁰⁷

What then was the connection between the seventeenth-century growth of labour exports and currency imports? As noted in the introduction to this article, key was the way in which increases in accumulated value related to control of the labour supply. The export slave trade meant that much ‘surplus labour’ and consequent global capital accumulation was concentrated outside Atlantic Africa. Meanwhile currency imports contributed to inflation of local currencies, and to the expansion of the market and demand inland; in this context, the tendency towards surplus-value accumulation in the Atlantic, and the relationship this had to labour supply, encouraged a shift towards slave exports from West and West-Central Africa to match these growing imports.

The expansion of the market thus of course promoted trade and internal market development, but it was also part of a growing disjunction between African economies and the rest of the Atlantic world owing to the export of labour used to accumulate value elsewhere. In sum, as Guyer suggests, there is no contradiction between unequal Atlantic exchanges on the one hand, and the growth of commerce and production in West and West-Central Africa on the other.Footnote ¹⁰⁸ The relationship of labour to capital accumulation encouraged the growing divergence between hard currencies in global use and soft currencies used in these parts of Africa, which moreover could not be converted outside the continent and therefore encouraged export of that which was in most demand, namely enslaved people. Here too, however, we should recall the significance of the parallel currency use related to social meaning mentioned in the introduction to this article. This was an important part of this process, for in many parts of Africa social reproduction became connected to the multiple uses of these currency imports.

The relationship between currency imports and the social meaning and use of currencies in West Africa is clear in the case of copper. Klaas Ratelband's evidence shows that by the 1640s, a huge range of finished copperware was being imported into the Gold Coast, including 1,804 pounds (note: the measure of the pound at this time varied according to which part of the Dutch republic the writer was from), of large pans, 912 pounds of small pans, and 474 pounds of kettles, and 1,848¼ pounds of barbers’ basins.Footnote ¹⁰⁹ Indeed, there seems to have been a preponderance of finished products over copper bars in the Gold Coast trade; the cargo of Jacob Ruychaver for trade at Allada, the Bight of Benin and the Gold Coast in 1645 contained only finished copper products, and no copper bars, though some iron bars were laded.Footnote ¹¹⁰

Thus there were important changes in how copper products were circulated and in the exchange value that they held along the Gold Coast by the 1640s. Transferable items of exchange value, such as manillas and bars – which archaeological evidence suggests were originally melted down for local manufacturers to use – began to coexist with finished products.Footnote ¹¹¹ Whereas previous generations of historians might have looked to industrial practices in the Dutch Republic to explain this, the causes of these changes were equally dependent on the uses to which copper currencies were put. As Herbert stresses, the ritual power of copper in many Atlantic African societies meant that copper-made objects held a transformative power which is what gave them value beyond their use-value as an end-product.Footnote ¹¹² As she noted, copper is for instance ubiquitous in ‘virtually all pre-Islamic and non-Islamic burials thus far examined’.Footnote ¹¹³ Copper had an exchange and a ritual value, just as tacula redwood did in West-Central Africa, and the kola nuts used as a form of currency did in Upper Guinea.Footnote ¹¹⁴ Moreover, as Ogundiran suggests for Oyo, these uses expanded and accelerated along with the growth of the Atlantic economy, making the seventeenth century a pivotal moment in these social transformations and the varying uses to which currencies were put.Footnote ¹¹⁵

This multiple use of copper imports is confirmed by seventeenth-century Dutch accounts. While noting the many practical uses to which copperware products were put, such as using basins for drinking and pans for cooking, such sources also note a strong ritual use: small barbers' basins were used to place offerings in the graves of the dead, just as tacula was used in funerary ceremonies in Ndongo, in West-Central Africa.Footnote ¹¹⁶ Such evidence adds to the previously mentioned research of Ogundiran on Oyo, and underscores Guyer's insight that ‘calculative rationality’ was not the zero-sum game of all currency in West and West-Central Africa, where rather diversity of ‘value’ was an intrinsic quality of currency. The evidence is thus especially clear for cowries and copper; currency, following Guyer's analysis, was not valued only for its quantitative accumulation, but as a store of spiritual and political power.Footnote ¹¹⁷

It is important not to generalise this insight to its applicability to all currencies, since it is certainly not the case that this was the only way in which cowries – to take one example – were used. Here, we need to bring chronology to our understanding of economic use. Evidence from pre-Islamic societies suggests that cowries originally had a strong ritual use as well as an economic function, being used by diviners among the Bamana of Segu (Mali) and the Soninké of Kaabu (Upper Guinea).Footnote ¹¹⁸ Nevertheless over time cowries came to function more strictly as money in the Central Sudan, and oral accounts of Segu describe how the later eighteenth-century king Ngolo presided over ‘the time of cowries. Rows of granaries full of cowries were the legacy of Ngolo’.Footnote ¹¹⁹ Hence, the increasingly strict monetisation of cowries on this account dates to a later period, when the increasing imports via the Atlantic made ever more cowries available as currency; nonetheless, even here, as Ogundiran's evidence shows, such monetised use went with a persistent diversity of value and the growing use of cowries in divination and other stores of ritual value.

Moreover, as the examples from Kaabu and Segu show, cowries certainly held non-monetised value at the same time. The fact that the multiple currencies in use in the region had – at various times – differing functions in monetary, social and spiritual terms, is significant for understanding transformations in social history. It means that the vast expansion in copper imports (and of other currencies) had implications not only for ordinary inflation, but also for the ways in which many symbols of power and ritual increased their social value in West Africa, and were deployed by political and religious leaders and within households as expressions of the meaning of Atlantic exchanges.Footnote ¹²⁰ Thus as these imports increased dramatically in the middle part of the seventeenth century, the deployment of these imports for both practical purposes and as a store of ritual power and social status increased.

It is worth, therefore, underlining the extent of currency imports in the eighteenth century. Cloth imports, for instance, formed by far the bulk of the goods traded to West Africa in the mid-seventeenth century. A list of goods sent to Elmina in 1645, tabulated by Klaas Ratelband, makes this point well. The dominance of light cloth (lijwaet) in exports is clear, with 28,2695 ¼ ells of lijwaet sent from the Dutch republic in that year, 268 cannekins, 1,440 slaeplakens (pieces of linen), and various other types of cloth including 177 large annebaes (blue-and-white striped linen) and 2,671 small annebaes, 2,140 pieces of tijckt (a cloth with blue stripes and red designs), 1,037 pieces of cleden (a checked cloth from Haarlem), and over 1,000 ells of Indian satin.Footnote ¹²¹ Although not all cloth was by any means used as currency, this is still indicative of substantial growth in this area, and of the ways in which – as noted above – the use of such imports to accumulate power and social value may have been expanding at this time. As for cowries, meanwhile, many documents from the 1620s reveal a large cowrie trade to Allada in particular, with the use of cowries shipped from São Tomé to buy slaves in Allada being noted in 1625, and huge shipments of cowries from India to São Tomé authorised by the Portuguese authorities.Footnote ¹²² The Dutch trading to West Africa in 1624 in Philips Van Zuylen's fleet noted the value of 41,838 pounds of cowries laden for Allada.Footnote ¹²³

There were, in sum, large imports of many of the currencies which held a variety of uses in West and West-Central Africa in the seventeenth century. A relationship has been posited here between this development, the price revolution, and the growing emergence of an early hard and soft currency system, where currencies in use in these regions of Africa did not hold their relative value on a global level. The apparent relationship of labour to accumulated capital value, and the growing export of enslaved African labour in exchange for these currency stocks, added to the forces weakening West and West-Central African economies on a comparative global level. Finally, with the importance of gold exports continuing, the growing strength of gold in the world economy inevitably weakened the cowrie's exchange-value.Footnote ¹²⁴ Exchangeability within a system of currencies that gave value to gold declined, just as, as we have seen, volumes of gold exports increased. Therefore, the attraction of other currencies and goods in Africa rose, needed to balance gold and slave exports now of such importance to the production of global value. These labour exports dwarfed and eventually extinguished the global cloth exports, which, as we saw in the first part of the article, were important to the mixed economy of parts of West and West-Central Africa during the first half of the seventeenth century. By 1700, a disjunction had emerged between regional and global exchanges.

Many historians posited that these imports did not have a negative impact on industry in Africa.Footnote ¹²⁵ Indeed, in the case of cloth the expansion of the market through currency imports and global trade did stimulate production and regional trade. Cloth production grew in many regions, in particular in Kano where extensive regulations governed the cloth trade – and where cloth was also traded from Borno (and after 1810 Sokoto) – as well as among the Asante and Bambara.Footnote ¹²⁶ Widespread cotton cloth imports, moreover, did not crush cotton production in the Niger delta according to David Northrup.Footnote ¹²⁷ The development of a weaving industry in Cape Verde also enhanced draw loom weaving techniques across West Africa, according to Venice Lamb.Footnote ¹²⁸ The emerging economic pattern was of the production and consumption of increasing amount of cloth, both imported and locally produced.

Nevertheless, it is not universally the case that African-made cottons ‘were holding their own against Indian cottons’ in the seventeenth century.Footnote ¹²⁹ Loango-produced cloths became much less important in the trade to Angola, while the export of cloths from Allada and other parts of West and West-Central Africa into the Atlantic seems to have declined sharply by 1700. A disjunction was emerging between Africa's global exports of labour and gold – both of which contributed to the growing emergence of the hard/soft currency system – and the regional trading system, where the expansion of the market encouraged expansion of cloth production and trade. This is in line with the apparent paradox noted above, of an increase in African trade coinciding with a relative increase in inequality in the global system; in local contexts, trade expanded, but in global terms the economic weight of West and West-Central Africa was in decline, and thus expansion of trade alone did not bring prosperity.

*

This article has suggested that large imports of currencies in the sixteenth and seventeenth centuries had consequences for trade, currency use, and socially constituted power and wealth in West and West-Central Africa. The study shows how interconnected these features were to the global economy, and sees these changes as integral to globalising the understanding of the price revolution. The inflation which characterised some of this process in the seventeenth century was an early phase of the linkage between cowrie inflation and European imperial expansion which appears to characterise some aspects of the late eighteenth- and nineteenth-century economic history of West Africa; here, as Lovejoy has shown, the increase of cowrie imports from East Africa, as part of the rise of palm oil production and European trade expansion, was a major cause of inflation.Footnote ¹³⁰

Several conclusions follow from this for both African and global histories. While the historiography on the interconnections of European and Asian economic histories has expanded hugely, it has yet to integrate an African perspective. And yet this article has shown how the demand for labour in the Atlantic regions accelerated the inflation of the cowrie currency, which was itself imported from Asia. It is thus by integrating Africa into the global price revolution that a fuller understanding of the interconnections of mobility that forged that revolution emerges. Meanwhile, contextualising African economic histories without reference to world economic trajectories is dangerous: local systems of value and production were indeed fundamental, but they intersected with global changes in demand, and it was indeed in the balance of these demands – of local and global agencies – that many of the features identified in this article took place.

A second consequence relates to the price revolution. The material here suggests that demographic growth and the local rise of demand is an insufficient explanation of the European price revolution; demand increased in West and West-Central Africa too, not through demographic growth but rather an expansion of the market. There is thus no universalisable conclusion predicated on the special features of European growth. While the mass import of currencies had huge inflationary consequences in West and West-Central Africa in the sixteenth and seventeenth centuries, these were not accompanied by a universal demographic rise or by an increase in productivity. Though scholars continue to dispute the demographic impact of the slave trade, it is known that in several instances there was a correlation between the trade and demographic decline. In Luanda in the second half of the seventeenth century, as the inflationary spiral of libongos continued unchecked, the region became hugely depopulated through warfare and the slave trade. Writing in the 1640s, the Capuchin missionary Giovanni Francesca di Roma noted the depopulation of Kongo caused by successive civil wars;Footnote ¹³¹ while the size of the army of the manibamba, to the north of Luanda, had halved by the 1650s.Footnote ¹³² Meanwhile, in the Gold Coast during the eighteenth century, Fante traders cited a large fall in population as caused by the slave trade.Footnote ¹³³ Archaeological analyses of the networks of earth walls around Benin City also suggests a larger pre-Atlantic population than was the case by 1700.Footnote ¹³⁴

Thus though no scholarly consensus is likely to emerge on this point, there was certainly no direct correlation between inflation and demographic growth, as has been argued for the European price revolution. Looking at the price revolution in a more global context is thus important. When Africa is included in the picture, the interconnection of the export slave trade to economic trajectories becomes explicit. It is not that the rise in the slave trade was alone a sufficient condition for the region's relative global impoverishment, but rather that the boom in the export slave trade went together with other economic and social impacts of the global price revolution there.Footnote ¹³⁵ These impacts connected to world economies through the rising labour demand in the New World, and the impact which that demand had on the relative value of currencies used in West and West-Central Africa, and in the longer term on the region's relative economic position of in the world sphere. In the end, it should perhaps be little surprise that the growing economic and political disadvantages faced by West and West-Central African societies came down to money as much as to anything else.