Pawns on the Gold Coast: The Rise of Asante and Shifts in Security for Debt 1680-1750

Judith Spicksley. Journal of African History. Volume 54, Issue 2. July 2013.

Employing the human body as a form of security was common in preindustrial societies. In western Europe, as in Africa, people historically had provided security for debt, but by the seventeenth century, if not before, it was unusual for people in Europe to be given as security for loans, although as pledges, people were used as a guarantee of performance. Indeed, the legal premise of security had itself undergone change. The rising demand for credit facilities in medieval Europe had, according to John H. Wigmore, increased the variety of obligations and forms of action linked to non-payment of debt, so that by the early modern period, the older idea of a pledge—essentially a forfeit—had been largely replaced by that of collateral security, in an attempt to guarantee the payment of a specific amount of debt.

However, this transition does not appear to have occurred simultaneously in Africa, where it remained customary for debts to be secured with human pawns, and may have been for some time. In their 1999 article, David Richardson and Paul Lovejoy revealed in detail the use of human pawns as a form of debt security in the Atlantic slave trade at Old Calabar in the later eighteenth century. They argued, moreover, that use of the body as a form of debt security was a key element in the success of that business—the adaptation by British and Old Calabar merchants of the local institution of debt bondage, or “pawnship,” proved crucial as a way of securing credit until Britain abolished its slave trade in 1807. The use of the body to secure commercial loans did not appear as a generalised feature of Afro-European relations. At Ouidah in the same period, security appears in general to have been unnecessary; overdue debts could be recovered through sale of the debtor and/or his family into slavery after successful prosecution in the king’s courts. The situation was similar at Bonny, where traders relied on the authority of the king to enforce credit arrangements.

On the Gold Coast, the securitisation of debt was a constant concern. Human pawns were occasionally in evidence from at least the seventeenth century, but records of the European trading companies suggest that gold, jewellery, and highly valued beads were more common as items of security, along with ceremonial items and other more mundane pieces. Such objects appear, at least in the second half of the seventeenth century—and even at the beginning of the eighteenth—to have been preferred. After that point, however, the relative importance of gold objects appears to have waned, and references to human pawns begin to dominate. This shift towards accepting people as security for debt is best evidenced in the records of the Danish West India and Guinea Company (WIGC), but additional materials from the Dutch West India and Guinea Company (WIC) and the Royal African Company (RAC) are utilised here to help explore the generality and complexity of the shift. On the Gold Coast, as at Old Calabar, there was no local political or legal agency active in the protection of credit, creating a situation in which traders had little option but to forge their own security arrangements. Why then did they choose to move away from the use of pawned objects and adopt the native institution of human pawn? The answer appears to lie in the parallel presence of two connected developments in Afro-European commercial relations: the declining availability of gold as an export commodity; and the rise to power of the Asante nation.

The Shift Toward the Use of Human Pawns

The provision of credit was a difficult issue for all the European companies trading off the west coast of Africa. Advice from company directors to European factors on the Gold Coast regularly cautioned against the extension of any form of credit to African traders, even though accounts offer ample evidence its use. Given the high level of European competition, company employees always ran the risk of upsetting the delicate trade balance, and recognised that failing to extend credit would risk loss of business to rivals. Offering credit was not always perceived as problematic. In order to function as permanent trading settlements, European forts had been required to develop a series of ongoing financial relationships with local peoples. Credit extended to Africans in coastal towns was therefore at some level redeemable by withholding the payment of rental agreements, customs duties, customary dues, and other perquisites from local allies, employees, and agents. And where debtors and their followers were living in the vicinity, the perception of risk was greatly reduced. Caution nevertheless remained the watchword, especially where personal wealth was at stake. According to William Bosman, chief factor for the WIC at the end of the seventeenth century,

The Negroes come daily to our Castle, or Fort, with their Gold; for which, after it is weighed, essayed and purified, they receive our Commodities; none of which ever go out of our Ware-houses before they are paid for; and if the Factor will give any Credit, it is on his own account, and he is consequently liable for the value to the Company.

Although Bosman gives no details of any security he might have required, the records of the European trading companies make it clear that gold pawn was a familiar and frequent form of security for commercial debt in the seventeenth century. An inventory of the Swedish African Company from 1658, for example, lists the pledges that had been left with the Company since the early 1650s, the bulk of which were in gold jewellery or highly valued agori beads; we find similar entries in the Dutch and Danish records. Nor was this purely a fort-based practice. Thomas Phillips, employed by the RAC as commander of the Hannibal, claimed he was often asked for credit to the value of two or three marks in gold:

promising payment in a certain number of days, when their servants or boys … return from the inland countries … Commonly they will leave some pledge in our hands till payment, as great collars of gold, and other large fatishes of exquisite workmanship, which their great men wear, and which they will be sure to redeem.

A review for the RAC of the ‘great causes of decay of [its] trade to Africa in 1705 not only echoed Dutch thinking on the responsibility for credit extension but reinforced the need for security in gold:

whosoever should trust a blackman should be liable for the summ, which would render the chiefs at Cape Coast Castle allwayes so cautious as never to trust but on gold pawns, which would have no consequence in the world on your trade.

If we turn now to the records of the WIGC, the first reference to pawned goods appears in 1674 as part of the inventoried effects of the late Governor Bartholomaeus von Groenenstein at Frederiksborg. This includes a long list of his pawns, many of them in gold, with their monetary value recorded in marks, ounces, and engels as determined by their weight. Items in von Groenenstein’s possession ranged from large neck rings, bracelets, and gilded chains to lumps of gold and ‘false gold,’ gilded snail shells and fetishes, and a gold-trimmed sword. Other items in pawn included various beads—some of them ‘aggrey’ (agori) beads—basins and old clothes, and one reference to a slave. Seven years later in 1681, the list of pawns being held at Frederiksborg—if considerably shorter—was not significantly different; again it reflected a preponderance of gold items, listed by weight and according to value. Where it does differ, however, is in the amount of additional information it provides on the nature of the debt: included in the list of pawns are details of the length of time for which the pawns had been held. It has been noted elsewhere that credit in African society was rarely time-limited, often being recalled on the basis of need. This is certainly supported by the Danish evidence. Debts had often been live for eight or nine years, and on one occasion for no less than thirteen. To each loan had been added interest at a rate of 25 per cent per annum, which though unlikely to be paid, nevertheless signalled to the Company directors at least that the governor was serious about the problem.

There are no further significant listings of pawns in the document collection for another forty years, although loans were still being issued and pledges taken. The next major piece of evidence appears in January 1722, when an itemisation was made of pawns and goods found in the writing desk and chest of Peder Østrup, who had assumed the office of interim governor at Christiansborg on the death of Knud Rost in 1720. The listing suggests that he, like von Groenenstein, was in the habit of securing credit on gold, gold jewellery, and other precious items. But an additional debit ledger drawn up in February 1722 reveals that he had also lent out some monies without security and others against human pawn. His lending policy was to come under considerable criticism from his successor. As David Herrn reported to the Sekret Council on his arrival in 1722, Østrup, ‘through his diverse, great and illegal extravagance … has brought everything into shameful confusion.’ His list of 73 debts included one clear example of the use of human security and three other possible cases.

Writing to the directors in Copenhagen in February 1722, Herrn promised to try and remedy the situation. But problems were evident as Herrn tried in vain to account for all the pawns being held on behalf of the Company.

The goods pawned against the Company’s receivables can be seen in the enclosed itemization, Letter J. There are however several which have no note, and we do not know to whom they belong, and how much has been borrowed on them. Of other pawns, half have been taken away and others again, which the owners wanted to redeem, are quite gone. I fear that this will cause some discord and inconvenience.

By July 1722, it was clear to Herrn that some of the difficulties he was encountering in attempting to recover the Company’s debts were a result of the fraudulent behaviour of Østrup.

The pawned goods have also been treated knavishly by P. Østrup, since many Negroes have come and asked for their pawns, which I swear to God I have never seen, and yet debt is claimed to be in pawn.

Østrup, it seems, had helped himself to a number of pawned items before being dismissed from his post. The passage of time did little to aid Herrn in his pursuit of the Company’s debts. Those listed on the Company’s books in March 1724 after his death as being owed ‘in the time of the late Knud Rost, Peder Østrup and the late David Herrn’ numbered 68, only five fewer than had been on the books two years earlier. Moreover, the bulk of the debtors that can be identified on both lists had significantly increased the amount of their borrowing. The ledgers provide continuing evidence of the use of pawned property, and a small number of examples of the use of human pawns. 18 rdl. had been lent against ‘2 Qvambus in pawn,’ for example and 54 rdl. 24 sk against ‘Cophij Zay, pawn Remido r[oarsman].’ Christian Syndermann, appointed governor in 1723, promised—as his predecessor had—to try and recover the Company’s outstanding receivables.

It therefore appears to us that the above-mentioned three governors [Rost, Peder Østrup and Herrn] have acted very badly, since they have lent out such a large and substantial capital to the Negroes, since these people are not minded or by nature like Whites, such that they will themselves voluntarily pay their debts. But a number of them think, when they are dunned, fair though it may be, that they are being done a great injustice. It will therefore create great trouble and difficulty for us, but we must not neglect, to the best of our ability and in accordance with our sworn duty, to remind them to pay their debts. If this should fail to bear fruit among them, then we will use such incentives as may be strong enough to make them pay their debts.

The veracity of Syndermann’s intention with regard to lending and his more proactive approach gains some confirmation from the list of his debtors drawn up in June 1725, which recorded only 42 names (including one Nune, ‘pawn[ed] fisherman’). Henrik von Suhm, his successor was even more determined. Though he retained Syndermann’s services ‘because he is thoroughly familiar with the trade,’ a commission was established to recover Syndermann’s outstanding debts. Von Suhm reported that as a result, ‘Negroes were pawned, then clapped in irons against payment of the debt, by which means some ofit has even been collected.’ However, this tactic backfired spectacularly, as African traders avoided the fort in fear or being panyarred for debt. Von Suhm was forced to admit the mistake:

I acquired such a bad reputation throughout the country that even the king sent a message to me asking if it was true, as people were claiming, that I had all trading Negroes who came here clapped in irons and then sold them, both for their own and for others’ debts.

Yet this was only part of what appears to have been a broader, more rationalised approach by von Suhm to the recovery and protection of debt. Sixty-one debtors appear on the Company debt books in both 1724 and 1726, but of those only 7 had increased their debt, and, in each case, the amount was relatively small (between around 3 and 5 marks). Moreover, as the debt book ledger in September 1726 reveals, a number of debts had been partly or fully redeemed through the forfeiture of gold pawns, and some others had been paid for in cowries. Other security, where recorded, appears to have been human, with remidors (oarsmen) frequently standing themselves in pawn for relatively small amounts of debt, usually less than 100 rdl. How much this is a function of recording practice is difficult to determine. Some of the remidors that appear as self-pawning in the 1726 debtors’ list also appear in the 1724 list with similar levels of debt but no accompanying reference to their pawned status. Was this information simply omitted earlier, or had they been asked to provide pawn subsequently as evidence of their intent to pay under von Suhm’s more pro-active debt recovery programme? Certainly new debt that can be identified after 1726 was predominantly secured on human pawn. Qvacun, a remidor in Orsu, had increased his debt from 166 rdl. in 1726 to 202 rdl. in 1729. While in 1726 he had offered himself in pawn, by 1729 it was clear that additional security was required: his son was also listed as a pawn for the debt. Likewise Old Cophi, who had repaid his outstanding debt in 1726, had by 1729 asked for further credit of 130 rdl.; the Danish insisted on holding two remidors, Afadu and Ando, for the bulk of the debt (114 rdl.).

Debtor lists in September 1727 and January 1729 reveal that the policy of accepting human pawns as collateral had continued under later Governors Frederik Pahl and Anders Wærøe, although a number of pawned objects were discovered in the Company’s gold chest in 1729. Agori beads were most in evidence, but the chest did contain two small gold fetishes and a gold bracelet. Later debt books say nothing about gold pawn, but do provide examples of human pawn; it is at this point that the shift towards the greater relative use of the body as security is apparent. In addition, increasing levels of indebtedness—or at the very least, demands for more security—are suggested by entries in Governor Christian Glob Dorph’s debit ledger of February 1744, in which, on at least two occasions, debtors had been required to offer their whole families in pawn as security for their debt.

The incomplete nature of the source material makes any attempt at a detailed calculation of the number of references to human pawns in the text unhelpful, but there is a rise in the number of incidental references to such pawns after the 1720s that becomes more insistent towards the middle of the century. Von Suhm had recognised that credit authorised by Company officers with African traders down at the coast, where the Company had lodges, was unlikely to be recovered, and that he needed to make use of existing indigenous debt recovery systems. In 1725, he informed the directors of his change of course—’all private [outstanding] debt will be allocated to the Company, so that no one must deal with the Negroes on his own account’; it would be left to the king or local caboceers ‘to collect it for us against reasonable payment.’ Perhaps there was, in addition, a move to recognise the value of local forms of security, and adapt their practice accordingly. Certainly there were problems for Europeans in relation to the use of gold. As already noted, it was relatively easy to embezzle and without it, African debtors could not be dunned for their debt. Sometimes items appear to have been misplaced or lost. In 1686, Ando Bonishee—a leading trader of Anomabu—had attempted to redeem his gold pawn, which he claimed consisted of both a neck ring and an arm ring, for which he had made satisfaction. Ralph Hassell, then RAC factor at Anomabu, who could only lay his hands on the arm ring, was advised by Bonishee that he would take fetish (swear on oath) before witnesses to prove his debt, and if his pawn was not returned in full, he would desire ‘the slave and five angles againe’ at its full value.

Unfortunately for Europeans, the use of human pawns did not provide a satisfactory solution to the problem of debt security either. Although creditors advised borrowers that unredeemed pawns would be sold upon expiry of their credit agreement, in reality, few European governors on the Gold Coast were willing to follow through on such threats. While WIGC directors were keen to recover outstanding debts, they had given express instruction ‘in the standing orders, Item 14, and in the governor’s special instructions, Item 11 … to treat the Negroes with leniency and kindness.’ Reporting as interim governor in 1752, Carl Engman echoed the thoughts of many of his predecessors—the only way to recover Company debt was by force, ‘such that the Negroes’ children or family are taken by the head and sent as payment to the West Indies.’ In adopting such a strategy, however, they risked losing trade to other forts, as disgruntled debtors moved their allegiances elsewhere. Moreover, African authorities on the coast were increasingly careful to regulate against the unlawful removal of their security. In 1747, Ludwig Rømer attempted to recover an overdue debt through the pretended sale of a pawn to an interloper. The debtor paid up, but the interloper refused to free the pawn unless he received a female as well as a male slave. Rømer had little option but to comply, since ‘according to the custom of the country,’ the debtor could demand four slaves for his pawn, if the pawn was not returned. By the 1750s, the penalty appears to have increased significantly. According to Marcus Svane, bookkeeper during Engman’s governorship, the latter had sold a number of pawned children belonging to the Ada Negroes to a Portuguese vessel for gold. As Svane informed the directors in July 1753, the cost of that premature sale could have been considerable.

For if the Ada Negroes got slaves to give and pay back what they had borrowed, and then demanded their children, and they were not to be had, they could according to their own and the country’s custom demand whatever they wanted for them, so that nothing else could settle this than whatever agreement could be reached. And then it would not have been cheaper than ten men slaves apiece, which was 30 men slaves for three impias and according to a simple calculation 3,000 rdl.

Yet keeping the pawns over significant periods of time was also costly for the Company. As Rømer was keen to explain,

when they [Africans] borrow from the Europeans, they do not pay even a low interest on the principal, but we have to feed the security they give, which usually means their own or slave children, sometimes feeding them for several years. Should we wish to sell the security, the debtor’s friends and any caboceers and fetish priests with whom they are acquainted come and plague us, begging us to have patience for yet a few more months—so [sometimes] we have fed their children until they became fully grown.

The French made similar complaints about the cost of maintaining Fante pawns in 1752, and the British were keeping pawns at their own expense in the latter part of the eighteenth century. It was also the responsibility of European companies to care for injured pawns.

There is evidence that on some occasions European traders on the Gold Coast refused to take human pawns. In July 1713, the WIC was looking for ‘as many securities as possible’ to guarantee a loan to Caboceer Caffo, chief of the ‘tJufferse, who had requested funds for the payment of allies in his war against the Akan. He already had some debt outstanding, but the precarious political situation made it necessary, in the view of the Dutch, that his request be granted, if with some care:

we know that if he had golden impias he would give them as a pawn or even as full payment for his needs, but that we cannot let the Company’s money go out of our hands without any proper surities; yet, he should try to produce sufficient impias, either golden ones or Conte de Terra, to the greatest value he can find, even if it were up to 100 bendas, but we cannot take Negroes as pawns or impias; firstly because we do not consider them sufficient, and secondly because that would cause too many rumours, and the Fantyns might hear of it, whilst we want to do this business in secrecy.

Nevertheless, human pawns were being offered as security for commercial debt in the 1710s by the state that had taken control of the bulk of the gold producing areas—Asante. And by the 1740s, this practice was also a feature of the other significant polity in charge of Afro-European commerce on the Gold Coast littoral—the Fante Coastal Coalition—who, in addition, had an interest in gold.

We should also re-examine the multiple forms, meanings, and terminologies used in discussions of security in the early modern period before trying to evidence change. Words used to describe security in the European trading companies’ records include (or have been translated as) ‘pawn,’ ‘pledge,’ ‘surety,’ and ‘hostage.’ In modern discussion, we generally understand security to refer to specific items of property deposited or handed over in order to guarantee the payment of a debt, because the concept of a surety—a pledge (either an object or a person) for the fulfilment of a promise or the performance of an agreement—is less common. In the seventeenth and eighteenth centuries, Europeans may rarely have thought of human pledges as securities for debt, but it remained common practice for people as pledges to be offered as security for the completion of contracts, the performance of a promise or agreement, or the safety of individual actors, especially in relation to trade. Nor was this practice unique to Europe. In the highly contested commercial context of the Gold Coast such pledges were vital, as Rømer noted in his advice to newly arrived personnel as late as April 1749:

Here at the fort eight caboceers’ sons have been and are still supported, namely three from Fante, two from the hills and two from the Rio Volta. These are all pawns from the most prominent caboceers. Those from Fante we have so that we can safely send goods ashore at these places to get provisions; those from the hills and the river we have here so that our factors at Ningo, Ada and Quitta [=Keta] can safely send goods to the most necessary places that belong under theses caboceers, to procure trade …

A brief outline of the various uses of human security in the records of the English, Dutch, and Danish trading companies will reveal the considerable flexibility of the practice. First, in the RAC records, pawned bodies served a variety of purposes: to secure the safety of factories and their goods; to act as hostages to ensure the safe return of negotiators and intermediaries; to secure the future payment of a court fine or compensation (rather than the payment itself), and as proof of an intention to comply with legal agreements; and to bind individuals to good behaviour. The Danish and Dutch records provide comparable instances and reveal in addition that the use of human pawns—as pledges for the performance of some current or future action—remained part of African and European practice throughout the period under investigation.

Some of the complexity of meaning is recovered by the comparison of two references to the same ‘pawn.’ In 1729, interim Governor Andreas Wellemsen, in handing over the books, factory, and warehouse, along with the son of the caboceer of Orsue town, to his successor, informed him that the boy was ‘in pawn for the town’s debt.’ A more expansive explanation at the end of the same letter, however, stated that the caboceer had ‘left his son in pawn in the fort as surety that the Negroes would not run away from the town, and that they would, as reported, pay off their debt as soon as possible.’ A decade or so earlier, an entry in the RAC document collection suggests a similar duality of purpose. In 1717, John Tau’s pawns—which appear to have been horns—were being held as ‘Security for his good behaviour to the Company & the payment of his debt amounting at first to ninety two Bendys which he has not at all complyed with in either case’ (he had recently ‘deserted the Companys interest’). If security could have a dual purpose, it could also take on a mixed format. Pawns received as a result of agreements with the Fetu for their good behaviour ‘and for Securing & reimbursements of the Company Charges in bringing them to Obedience’ included gold fetishes and a number of children, including ‘5 Boys 2 Girles Pawned by the Queen & Countery of Fetue for a debt.’

The use of free bodies as collateral for debt may not have been a desirable part of European practice in the late seventeenth century. In 1687, for example, the RAC bought slaves on a temporary basis to avoid the use of human pawn, expecting them to be redeemed later. That there was an early recognition of the commercial value of human pledges nevertheless is clear. En route for Copenhagen, Governor Busch recounted the offer that the King of Accra had made to assist in the retaking of Christiansborg, at the time being held by the Portuguese. The promised sureties—which include an agreement to forfeit in caseof failure—extended to:

his mother, sons and wives with all his gold, which he had taken with him to Friedriksberg. He would also quickly procure the slaves we wanted, and if this did not happen, then we had all his relatives as sureties and could then do with them as we liked. Mr Prand and I considered this, and we believed then that it could not go ill, for even if the King of Accra lost the battle, there would still be slaves in it for us. So we tried to maintain good friendly relations with him. He also asked us if, when the ship was at Lampe and he needed more powder or more muskets, we would help him with this—in return, however, for gold pawn.

The final comment, nevertheless, reveals the contemporary preference for gold. Busch had asked for gold pawn for the weaponry, and the English too appear to have been able to demand gold pawn in some instances, especially when credit was petitioned for guns and gunpowder. While Europeans seem to have preferred property—including slaves—as security, African traders may well have been pushing for the use of human security. There are further complications, however, in attempting to understand the nature and choice of security in the early eighteenth century.

Translators of the late seventeenth-century Danish records chose to refer to human security where it operated to provide a guarantee of performance rather than collateral for a loan as ‘pledges,’ ‘hostages,’ or ‘sureties.’ From the 1720s, however, there is a clear shift to use instead the more generic term ‘pawn’ for any form of security. The fact that this was indicative of shifts in contemporary usage is suggested by a parallel shift in the records of the Dutch company, although the RAC were already describing all forms of security—whether object or person, and regardless of their function—as ‘pawn’ in the 1680s. Certainly until the early eighteenth century, the human security that was offered by African traders in exchange for goods looks more like pledge than collateral. Although Ahenesa (Ansa), King of Akwami, was allowed to pawn his wife at James Fort, Accra, for 100 muskets in 1681, her sale would have been unlikely to provide sufficient collateral for the loan: the following February, Ralph Hassell paid 9 muskets each for 2 female and 1 male slave, which he believed was ‘extravigant rates for my Masters proffitt.’ Yet nine years later, the English refused to provide the King of Eguafo with the ten barrels of gunpowder he had requested because his gold pawn was insufficient: ‘it not being all good gold,’ he received only four. Instances in which there was disagreement over loan redemption also suggest that the indigenous premise of security on the Gold Coast was more akin to the concept of pledge than collateral. In 1722, the Danish were holding a number of gold items belonging to the King of Akwamu, weighing in at over 15 marks, upon which he had borrowed 30 benda of gold as well as some goods in c. 1718-19. In the 1722 ledger, he is listed as owing 640 rdl. at which point the security held by the Company exceeded the value of his debt. But sometime between 1722 and 1725, the king increased his loan to 3,974 rdl. He managed to pay back 57 rdl. on 31 July 1725 in cowries, but the Danish—keen to recoup as much as possible—decided to cash in their pawn, receiving 912 rdl. for part of the pawned gold on 30 September of the same year and a further 848 rdl. for the remaining gold on 8 October; this reduced the king’s liability to 2,157 rdl. Unfortunately for the Danish, the problem of the outstanding debt had to be taken up with the king’s successor. He was not averse to making good on his predecessor’s debt, but equated the total liquidation of the security with the cancellation of the loan. In 1727, he refused to pay the outstanding balance demanding the return of the gold or the annulment of the debt as well as compensation for the dishonour he perceived he had suffered.

For Africans, gold pawn does not appear to have been the norm. Instead, the provision of human pawns was what characterised indigenous practice, even though it was used to secure the performance of a wide range of obligations, including the payment of specific debts. In the Danish material, the 1720s marks the decade in which the instances of human pawn begin to supplant those of gold. But if factors and governors appear to have been adapting to African terminology as well as African practice, they were not prepared to risk capital on the legal premise of security as pledge or forfeit. In a parallel shift, therefore, human pawns appeared more frequently as guarantees of specific amounts of debt from the 1720s as debts and their security were calculated in relation to slave value. In 1744, for example, Jancon and Qvacu both owed 96 rdl.—the cost of a male slave at 100 per cent markup—for which they stood themselves in pawn. There were still negotiations to be made, as creditworthiness was related to status and bargaining power: where the relatives of caboceers and free Negroes were offered as security, the Danish seem to have accepted a balance of one pawn to the value of two slaves. Like the gold they replaced, Europeans wanted their human pawns to provide collateral security, but what had happened to switch the balance of preferences, and who was driving this change?

The European Demand for Gold

The extension of Portuguese medieval exploration to the Gold Coast of Africa in the late fifteenth century heralded the beginning of what could be described as the earliest documented gold rush, as European nations competed for access to the rich gold deposits of the area. Though significant in terms of its contribution to the early modern European economy, this gold rush has been rightly overshadowed by the broader and more devastating commercial endeavour of which it was part—the transatlantic slave trade. This too was European in its origins, but had global consequences that went immeasurably beyond its initial formulations.

‘Ghana, the land of gold’ is referenced in the writings of the early Arab geographer al-Fazari, and Arabic sources confirm the existence of a substantial trans-Saharan trade in gold from at least the tenth century. One major goal of Portuguese exploration then appears to have been to locate and secure sources of African gold in order to bypass Arabian merchants, and allow the Portuguese crown direct access to the precious metal. Only with the discovery of the Gold Coast in the 1470s did this begin to look feasible. As the gold trade was increasingly reoriented away from its northern focus towards the southern coast, there were opportunities for political change as traditional authorities in coastal kingdoms were undermined. The most relevant of these for the discussion at hand was the rise of the Asante nation in the early eighteenth century—whose identity and power was intricately intertwined with gold—and the parallel emergence of the Fante Coastal Coalition. Their control over a broad swathe of the coastal littoral allowed them to monitor and direct commercial operations between the Asante traders inland and the Europeans on the coast. The emergence of these new polities in the early eighteenth century coincided with the period in which supplies of gold as an item of commerce were dwindling.

Initially, the Portuguese had imported slaves into the Gold Coast as an exchange good but by the seventeenth this trade had been reversed, although until the eighteenth century slaving remained relatively small scale. Richard Bean’s 1974 note in the The Journal of African History sparked what has become a long-running debate on the timing of the relative preference of Europeans for gold and slaves, as indicated by their commercial value. For Bean, gold remained the most coveted good until the later seventeenth century. After that point higher export prices for slaves, and the depletion of the most accessible gold deposits, in his view, appear to have driven up labour costs in mining. Ernst van den Boogaart’s challenge almost twenty years later, incorporating evidence from a revision of the number of slave imports into Spanish America before 1650, argued for a much earlier swing: ‘from 1600 onwards the value of slave exports surpassed that of gold exports. This has been demonstrated … for 1623-32 and it is also valid for 1600-22.’ The more recent work of David Eltis has sought to reclaim the turn of the eighteenth century as the fulcrum of change. Only then did slaves become more important than gold ’as gold deposits were mined out and, more important, as both slave prices and numbers of slaves carried increased sharply.’

Price increases are often the most visible signal of shifts in the demand/supply equilibrium, and as such form a substantial part of existing historical analyses. John Fynn, for example, has argued that a rise in prices encouraged ‘the Africans to abandon the tedious job of working in the gold pits for the comparatively easy task of raiding for slaves.’ Eltis too claims that ‘the slave-gold mix shifted sharply at the end of the century in response to a change in the relative price of gold to slaves.’ The documents of the monopoly trading companies offer some qualitative support for this increase. In 1704, Willem de la Palma, director general for the WIC, claimed that since 1700-1 the price of slaves had risen ‘by more than a half.’ Johannes Postma’s more definitive Dutch figures support his claim, indicating a rise from 30 guilders in 1680 to 45 in 1703; by 1721, the price demanded had risen to 80 guilders. The price of slaves at the Danish forts had risen too. Starting from 10 rdl. in 1680, the prime cost of a slave rose to 16 rdl in 1700, reaching 32-38 rdl. in the period 1700-24. Price data for the British also reveal a rise in the cost of slaves toward the end of the seventeenth century, with Bean’s figures, for example, suggested a near doubling of prices from £5•43 in the four years between 1698 and 1702 to £10•24 in the following four.

Explanations for the causes of price rises at the end of the seventeenth century have focused on increasing demand for African labourers in the Americas, both as plantation slaves and as miners. Certainly the cost of slaves in the Caribbean appears to have risen, and with the discovery of gold in the region of Minas Gerais in Brazil at the end of the century, the Portuguese were looking for labourers who were fitter and better suited to work in the mines than those from Angola or Kongo. The entry of English interlopers into the trade as a result of the end of the RAC’s monopoly in 1698 has also been cited as a causal factor in the sudden racking up of competition. As Johan Thrane, governor at the Danish fort, reported in 1700, ‘there are so many ships here at the Coast that they might have rained down from the clouds.’ There was certainly a significant racking up of slave exports. The estimated number of slaves leaving the Gold Coast rose from 75,377 in the period 1676-1700 to reach 229,239 in the following 25 years, with the biggest single yearly increase occurring between 1699 and 1700 (4,817 to 11,203). The annual total then fell back, and apart from odd years when slave numbers were high, did not reach and maintain similar levels until the late 1760s. Given the fact that the number of slaves leaving the Gold Coast appears to have trebled between the later decades of the seventeenth century and the early decades of the eighteenth, the supply of slaves must have had considerable elasticity. Moreover, African sellers appear to have been pushing the terms of trade. Despite the fact that men were in greater demand than women, male slaves as a proportion of all slaves fell from a figure of 52 per cent in 1662-97 to 47 per cent in the period 1698-1713.

The change in commercial practice could not have failed to have an impact on contemporary discussion. In August 1704, de la Palma reported to his directors in Amsterdam that increased competition was causing the inhabitants ‘to concentrate on the whole much more on the slave than on the gold trade, finding the former more profitable’; he now regarded commerce in slaves—rather than gold—’as the unique cornerstone of Y.H’s interest.’ This appears to have affected his procurement strategy, and did not go unnoticed elsewhere. In March 1706, the Danish Governor Erich Lygaard, reporting to his superiors in Copenhagen, argued for a similar change of direction:

Even if capital in gold cannot be traded to be sent home with every ship (which would require that the trade got better than it is now) one can regularly get slaves, with which the Company (in my view) must first and foremost be maintained. Not a year goes by but the Dutch send out four, five or more slave ships, and if they did not keep up this trade, they would long since have been ruined.

Yet the policies of the Dutch, Danish, and English trading companies show considerable vacillation, driven largely by the level of returns from the slave trade. In the last quarter of the seventeenth century, the Dutch, according to Postma, had ‘exported nearly half of all the gold mined in West Africa.’ Their much poorer performance in the first decade of the eighteenth century may have encouraged a broadening of trade, but interest in gold did not wane. The high cost of Gold Coast slaves in 1710, for example, occasioned a reversal of existing practice with the recommendation that instead of buying slaves, traders should ‘rather barter the cargoes which arrive for gold and other current trade goods.’ Similar sentiments appear in the RAC material. In 1711, bemoaning the consequences of the trade ‘being laid open’ to interlopers and attempting to regain its monopoly status, the Company proposed a scheme ‘by which, not only a sufficient Number of Negroes will be bought for supply of all the Plantations, but very large Returns will be made Annually in Gold, and other Merchandize, to the great encrease of Bullion, and the Wealth of this Kingdom.’

For the Danish, complying with conflicting Company directives was often tricky. In July 1720, a disappointed Governor Rost reported that trade at Accra was worsening:

Now neither gold nor cowries comes here to the fort, but always slaves. We are obliged to buy some of the best to cover the current costs of the fort. Although your Lordships the Directors tell us in your orders that we must not take goods for goods, we still think that it is better (since the trade is now so inordinately poor) to engage in some trade in slaves rather than refrain altogether.

Governor von Suhm offered a suitable compromise. In requesting more coastal transport and ‘kill-devil’—young rum or brandy from St Thomas—to improve his trading prospects, he explained his motives and his strategy:

To shed further light on the above, I postulate that the true advantage of the Company in the trade lies in purchasing slaves for the goods, and then attempting to get them out at a profit, either with our own ships or to interlopers. For what I reported in my humble letter of 20th September 1725—that the slaves could be sold to the Portuguese for gold at Fida [=Ouidah]—was, as Your Lordships can see from the same letter, on the condition that the fort was always so abundantly supplied with goods that we would have no need to seek them from the interlopers.

The Danish had made earlier attempts to trade for gold with the Portuguese, allowing one captain in 1702 ‘to anchor his ship for a few days under the protection of the fort to sell tobacco and rum’; he, in return, had promised to ‘purchase from the Company’s cargo as much as he had sold for gold.’ By 1730, if not before, it was Portuguese demand that was informing procurement strategy. As Wærøe reported in December of that year:

Finally in this item, Your Lordships the Directors are pleased to recommend us not to purchase other than young Portuguese slaves. We can assure you that the poorest man slave we purchase, and at the lowest price, we can sell again to English and French ships here at the coast for at least 112 rdl., and the poorest woman slave for 80 rdl., and this for good wares, so that for the goods we can get the same money back in gold, and sometimes even a profit on it. Your Lordships can thus see for yourselves that if the gold trade with the Akenists would come in here, we could use all the slaves we purchase: selling them for goods and the goods for gold, which we believe gives us a better profit; and sending the gold home with our ships. And no slaves will die on the voyage to the West Indies.

This policy had also been adopted by the Dutch, who the English claimed had been trading for gold as coastal middle men since at least the 1680s. The WIC directors eventually called a halt to the practice, believing it reduced the supply of slaves for the Dutch West Indies.

Other policies confirm the continuing importance of gold for the European trading companies. Throughout the first half of the eighteenth century, for example, the Danish and Dutch companies discussed the building of new lodges and the acquisition of new forts, and continued to entertain the possibility of finding and working their own mines, despite a number of unsuccessful attempts. The Dutch finally decided to open up slave trading to anyone willing to pay them a ‘recognition’ fee in ‘liquid gold, or otherwise the same value in ivory.’ The British too, were still dreaming of gold. The RAC shifted its focus towards the export of gold and ivory after 1731, and its successor company in 1750 was encouraged to do the same. James Hippisley, who was employed at a number of forts on the Gold Coast in the 1750s, pleaded for the erection of a fort at Cape Appolonia in 1764, at least in part on account of that place having ‘always been remarkable for its plenty of gold.’ Indeed it was the only place in Hippisley’s opinion, where the English could ‘depend on getting gold, on (what is now so improperly called) the Gold Coast.’

The African Demand for Gold

Hippisley was one of the few commentators to explain the decline in gold availability in Lower Guinea in the eighteenth century in relation to changes in African demand. Gold, he revealed, had now ‘become as precious among them as with us, for this plain reason, that every necessary and indulgence are as readily purchased with it in Guinea, as in our parts of the world.’ This view contrasts with the existing historiography, which sets declining availability within the context of rising levels of hostility and the impact of this on production and exchange. The understanding that wars ‘made gold scarce but negroes plenty’ was well known among European factors. This trade-off had significant implications for business at the very end of the seventeenth century with the onset of what seemed to Willem Bosman at least to be a situation of perpetual warfare. It not only meant that trade routes from the areas of gold production were closed, but that sources of the metal itself were left unexploited as miners were called up to fight. In addition, wars required large amounts of funding, all which had to be paid for in gold, the local currency. Much of this was obtained through the imposition of a war tax—apeatoo—and other special levies, but gold was also a key element in the negotiation of alliances. Some groups resorted to bribery—the Akyem reputedly offered the Agona a large brass pan full of pure gold if they would assist them in defeating their enemies; others were ‘oblig’d to buy a Peace at the price of a large Sum of Gold.’ For the victors, however, war could also be a source of gold; defeated peoples had to pay indemnities in gold as well as slaves, and the amounts could be substantial. Since gold functioned both as local currency and an item of commerce, any fall in production could have significant implications for the money supply, as well as the export market. Certainly the value of gold appears to have risen. From as early as 1727, Danish factors reported that African sellers were demanding 20 per cent more European goods for gold than for slaves, a situation that appears to have continued until at least the 1750s. The impact on Danish stocks was significant—the small amounts of gold being obtained were barely enough to cover fort costs, and from 1728, Company directors ordered that traded gold was not to be used for local purchases or the payment of expenses.

There was no complete withdrawal of African sellers from the gold market. While demonstrating an increased preference for payment in gold, merchants were nevertheless prepared to expend gold if the deal was sufficiently tempting. According to John Snow’s letter to the RAC of 31 July 1705, access to high quality goods allowed the Dutch to ‘catch the gold, & slaves being little of their trade they then dismiss them to us.’ That gold was available is also clear from Dutch sources in 1705—guns and powder could elicit the elusive precious metal, as could other quality items. And around ten years later, James Phipps recorded that he had had to barter ‘with Capt Maulthus an Interloper from London for some goods to Dispatch a gang of Ashantees that brought about one thousand weight of Elephants Teeth besides some gold.’ Danish governors too made repeated requests to Copenhagen for regular deliveries of good quality commodities, but were frequently disappointed; they turned to sourcing ‘the proper goldgoods,’ after 1739 at least, from George Fryer, a London merchant.

Gold had a number of functions in Gold Coast societies that can help to account for changes in export preferences. Clearly, since it was ‘the principal medium of exchange in the process of buying and selling,’ we would expect demand for gold to rise alongside the expanding volume of both internal and external trade. As well as money for exchange, customs duties had to be paid in gold, for example, and the rising levels of criminal and civil offences that tend to accompany population growth and urbanisation would have led to a greater number of judicial fines. Gold in addition served to grease the wheels of the justice system. But there were other social factors at play. That the peoples of the Gold Coast, as Pieter de Marees recorded in his early travelogue, were ‘very keen on their Gold,’ was a function of the fact that status, wealth, and the ownership of gold were closely entwined in Gold Coast culture and practice. Nobility could be achieved through birth, the performance of meritable action on behalf of the state, or the accumulation of money, which allowed the individual in question to undergo a public ceremony of ennoblement. The lure of aristocratic status therefore provides part of the context for Bosman’s critique of the Gold Coast peoples, who ‘very greedily heap up Money and Goods; to which their Minds are strongly enclined.’ But he also recognised that gold was equated with the control of people: ‘Kings are obliged in this Country to preserve their Power by dint of force; wherefore the richer they are in Gold and Slaves, the more they are honoured and esteemed; and without those they have not the least command over their Subjects.’

For one group of people in particular, however, gold appears to have wielded an extraordinary significance, and it may not be accidental that their rise to power coincided with the large-scale retraction of gold as a commodity from the export market. That group is the Asante. Gold was not the only source of wealth in precolonial Asante—other sources included people, food, goods, and land. Yamao Ponko, whose name in Asante became ‘a byword for the accumulation of wealth’ in the eighteenth century, appears to have made his fortune through trade and the buying up of land and people in debt. But gold was the ultimate performative symbol of wealth—when approached by others claiming to be his financial equal, Ponko demanded they ‘show him their assets in gold.’ Gold had been used as a demonstration of wealth on the Gold Coast from the sixteenth century at least, but in Asante, the metal appears to have been gaining considerable additional value. Much of this has been attributed to its symbolic relationship with power through the concept of the Golden Stool. Drawn down from the heavens at the founding moment of Asante at the end of the seventeenth century, the Stool was thought to contain, in the words of John Fage, ‘the spirit of the whole Ashanti nation,’ of which the king was to be the eternal guardian. Gold was thus linked symbolically to national power, identity, strength, and continuity, ensuring it a singular place in Asante culture and practice.

The manifestations of this link are not difficult to see. From its central part in the retelling of the creation story, to its role in the defeat of national enemies, gold was bound to the life and beliefs of the GoldCoast peoples. If the Golden Stool marked the foundation of the Asante nation, it was the potential loss of large quantities of gold that inspired the need for it. Gold makes a regular appearance in folk tales and proverbs, for example, and was called upon to mark important lifecycle events.’A Negro of Fortune,’ according to James Wyatt, had pieces of gold laid in his coffin, and gold was a significant element in traditional wedding celebrations. As a source of mystical power, gold could be offered up by priests in advance of major events such as war, trade, and travel, and for the resolution of existing problems, often related to illness. Finally, gold ornaments, as a demonstration of military prowess, were worn by the Asante in battle, soldiers being ‘frequently so loaded with Gold and Conte de Terra, that they can scarce march.’ In addition, the source of Asante power was protected through the concept of the sacred. Asante military activity secured control over all the major auriferous areas in the early eighteenth century, and while there were sound economic reasons for preventing European access to gold, Bosman claimed that the Asante held the gold mines to be sacred, ‘and consequently take all possible care to keep us from them.’

What may be most significant for the discussion at hand, however, is Thomas McCaskie’s claim that after 1700, the Asante state sought to control not only the sources of gold, but access to them. Such action had a number of facets. Fines in gold for adultery paid to the king had already been part of the judicial system on the Gold Coast in the early seventeenth century. But McCaskie claims that the Asante state extended this principle to the payment of all fines, tributes, taxes, and other levies, all of which had to be paid in gold. In the case of the introduction of death duties, attributed to the reign of Opuku Ware in the second quarter of the eighteenth century (and without specific limit), the state ‘customarily took the overwhelming bulk of the gold dust in any estate.’ Moreover, as Ivor Wilks has noted, the payment of death duties was cleverly linked to the notion of the good citizen—the okaniba:

the individual who by the exercise of his, or her, own skills and industry made money and therefore at death became, through the system of death duties, a benefactor of the nation. It is implicit in this view of society that wealthy citizens will take every opportunity conspicuously to display their riches and to acquire formal recognition of their achievements.

Since the state also determined who should be awarded noble status for the accumulation of wealth, from the very early eighteenth century, the introduction of measures to maintain control of entry saw the death penalty decreed for both the circulation of counterfeit gold and the making of regalia from it. If, by 1750, the Asante state had secured, in Fynn’s words, ‘complete control of the gold, ivory and slave resources in the interior of the Gold Coast,’ it had also become the main repository for, and redistributor of, gold.


The European gold rush that was precipitated by the discovery of substantial quantities of the precious metal on the Gold Coast in the fifteenth century led to the export of huge amounts of gold in the two centuries that followed. European demand does not appear to have been curtailed in the eighteenth century, but there was a clear reduction in the amount of gold available for exchange, which in the historiography has been linked to falling levels of production as a result of political unrest among emergent African states vying for commercial supremacy. Consequently, from around 1700, the European trading companies turned increasingly to slave trading as demand for slaves in the Americas expanded. Moreover, it has been suggested that slave trading on the Gold Coast emerged as ‘a second-best alternative for both sides,’ as Africans attempted to ‘maintain the flow of Atlantic imports threatened by diminishing gold production.’ From a European perspective this is not difficult to evidence. Gold as a commodity item continued to preoccupy the minds of company directors throughout the period under discussion, even if supply was limited. Evidence is not so clear from an African perspective, however. The value attributed to gold on the Gold Coast—both materially and symbolically—suggests that gold rather than slaves may have been the ‘second-best alternative’ for African traders during the early eighteenth century. Given the dual function of gold on the Gold Coast—as both currency and commodity—extracting the nature of preferences from export levels is problematic. Certainly war reduced the amount of gold available for export as production fell and demand for money to fund the prosecution of war increased. But for the Asante especially, gold became more than a material possession and a route to personal glory. As a symbol of national power, unity, and strength, its level had to be monitored carefully by the fledgling state, and losses kept to a minimum. War also meant that slaves became more plentiful in the eighteenth century, in direct opposition to supplies of gold, and were more easily accumulated. In addition, they needed to be disposed of fairly rapidly; they were costly to keep and a threat to security. How fortunate for traders then, that the very end of the seventeenth century brought an expanding number of Europeans who were willing to buy them, as commercial endeavours multiplied substantially in response to rising prices. Gold was still available for exchange in the early eighteenth century, but at a higher price and only for those ‘gold goods’ that were in high demand among African traders.

This renegotiation of commercial preferences appears to have spilled over into the related area of credit preferences in the early eighteenth century. During the seventeenth century, Europeans—quite familiar with the use of pawned objects as security for lending—had asked their African debtors for gold, jewellery, ceremonial items, and slaves when making advances. Not until the early eighteenth century did human pawns as security for commercial loans appear to become part of customary practice. At this point, Europeans were experiencing problems with the use of gold pawn, and the opportunity to gain additional cargo in the case of default may well have worked initially to make the use of human pawns seem attractive. Yet it was soon evident that such pawns could be as much if not more problematic—they required food and support and were difficult to sell off the coast without losing local goodwill. The Danish, at least, appear to have remained reluctant to accept human pawns as security, preferring to provide funds instead through the purchase of young boys as slaves. Gold pawn was still in evidence, but like its counterpart in trade, appears increasingly to have been obtained only for the ‘right’ sort of goods.

Problems of interpretation remain. The shift to human pawns may be a function of the source selection, and the WIGC was a relatively small concern. Secondly, the printed records of the WIGC, which serve as the primary source for this investigation, are not comprehensive, taking the form of an edited selection chosen for interest and relevance after a process of consultation. The Dutch records too were originally collected as research notes for a thesis. In addition, neither of these collections were consulted in their original language, raising frustrating problems of interpretation and translation; meanings can change across cultural and linguistic boundaries, as well as between periods of time. Europeans were familiar with the use of pawned objects to secure debts, and with pledges and hostages as surety for the performance of a contract or an agreement. Africans, on the other hand, used the concept of security in relation to a broad range of obligations, but do not appear to have recognised any difference between the idea of a pledge or forfeit, and that of collateral for debt. Mixed portfolios of security suggest there may have been a process of transition in which African debtors offered human pawns that Europeans accepted as what we might call sureties while Europeans demanded pawned objects as collateral for any monetary debt. Gradually, however, and in response to a withdrawal of gold pawn, Europeans moulded the broader African concept of human ‘surety’ to include a more specific idea of collateral security that they could utilise. For African debtors, the end result was often the same—the forfeit of their pledge—but the supra-local nature of that forfeit demanded institutional protections. By the mid-eighteenth century, customary fines for the unauthorised removal of pawns had been introduced in an attempt to restrain impatient European traders.

The lower preference amongst African debtors for gold pawn was likely to have been a consequence of its dual function as commodity and currency. Given the importance of gold in social mobility, growing levels of demand alongside a declining supply would have tended to encourage hoarding. Moreover, during a period of considerable political and social dislocation, the material and symbolic value of gold as a source of wealth and power, and as visible evidence of status, would surely have discouraged even its temporary loss. The need to maintain currency levels would also have cautioned against removing large amounts of gold from the monetary system to function as pawn. Indeed, the increased use of human pawning appears most marked in the Danish material in the later 1720s, after an aggressive period of debt recovery involving the liquidation of much of the gold pawn that had been held at Christiansborg. Other European trading companies may have taken a less parsimonious approach, but in any case the decision to shift away from the use of gold pawn may not have been in European hands.

Since the Asante were in control of the main auriferous regions by the early eighteenth century, understanding their behaviour is crucial to explanations of change. According to McCaskie, we have to consider that fact that ‘the accumulation of wealth as imperative and as a yardstick, and the deeply resonant meaning of wealth as symbol and as mnemonic, were abiding and central features of Asante life, history and self-knowledge.’ Their rise to power would no doubt have been costly, but from the creation of the nation with the calling down of the Golden Stool in 1698, gold had constituted the primary determinant of power, status, identity, and wealth; as early as 1702, Asante traders were offering human pawns for commercial credit. The Fante Coastal Coalition too emerged in the wake of high levels of inter-state conflict, and though we know less about their relationship with gold, by the 1740s at least, they were also offering human pawns in commercial credit dealings. Extracting the idea of an expansion in the use of humans as a form of debt security in the early eighteenth century from the available documents has required a measure of informed interpretation. Yet historicised within the particular context of early eighteenth century change, it does not seem unreasonable to argue that it reflected a genuine shift in practice that was instigated by a rising preference for, and control of, gold in Asante, and an increased demand for slaves in the Americas. Such developments were greatly aided by the high levels of European competition that existed on the Gold Coast, allowing African kings, chiefs, and merchants more opportunities to influence the terms of trade. The absence of a single political authority prepared to safeguard credit also played a part, but political strength in this instance appears to have been less significant than strategy. The sociocultural, economic, and symbolic value attached to gold by the Asante people was audaciously attached to an ambitious programme of nationalist expansion that also drew on gold for its success. But that was not all. Slavery exists wherever repayment of debt can be attached directly to the sale of the body, ensuring that the level of enslavement is determined ultimately by the form and nature of debt recovery. The culture of ’rights-in-persons,’ so clearly outlined by Suzanne Miers and Igor Kopytoff in the 1970s, offered new opportunities to enterprising people with bodies at their command. That the prevailing indigenous system of debt recovery allowed those bodies to be used in both the securitisation and liquidation of debt was of crucial importance to European traders, who largely adopted the institutional practices of their trading partners in order to support and extend their commercial ambitions.