Asante Empire

A stool from the sky

Drafted from Wilks, African Studies Series 13 (1975)formation

The Asante tradition has the priest Okomfo Anokye calling a golden stool down out of the sky, in a cloud of white dust, into the lap of Osei Tutu, and declaring that it held the soul of the nation and that the union of the Akan states was bound up in it.

Nobody sits on it. It is not a throne. The Asantehene is enstooled beside it and lowered towards it three times without touching it, and it has its own regalia, its own attendants and its own chair to rest on.

In 1701 the confederacy that had been assembled around it destroyed Denkyira at Feyiase and stopped paying tribute to anybody.

Gold

Gold, tolls and tribute, in that order, and the first of them is the reason Europeans called the country the Gold Coast.

The Akan goldfields had been supplying the trans-Saharan trade for centuries before Asante existed. Ghana, Mali and Songhai had all lived on that traffic at its northern end, taking a royalty on nuggets and a duty on transit; Asante sat at the southern end of the same system, on the ground the gold came out of.

Asante sat on the source, and by the eighteenth century the gold was going south to the Atlantic as well as north across the desert. Gold dust was the currency, weighed against brass weights — geometric ones and figurative ones showing animals, proverbs, tools and scenes, made in their thousands and now among the most collected African objects anywhere.

The state took a share of production, taxed the roads in both directions, and assessed tribute on the conquered states in gold, and Wilks's book reconstructs the accounting from the surviving correspondence.

And the other export

Asante's wealth in the eighteenth century came substantially from selling people to the Atlantic trade, and its wars produced the captives.

The same circuit had been running on the Saharan side for centuries — captives south to north, horses and salt north to south — and the argument now generally made is that categories of servility in West Africa were formed in that economy rather than by later European contact. Asante is the Atlantic-facing version and the largest single supplier on that coast for much of the century.

The state taxed and supplied that trade rather than operating it, which is a distinction worth keeping and not much of a defence. Asante did not sail ships or run barracoons. It sold people to the people who did.

Britain abolished its own slave trade in 1807, which removed Asante's largest market, and the nineteenth-century wars with Britain begin in the years immediately afterwards.

A state with a civil service

The reason so much can be said about how this state worked is Wilks, whose eight hundred pages are about exactly that.

A capital with wide swept streets and a stone palace. Great roads radiating from Kumasi with a courier service, stages, and fixed travel times the treasury used for accounting. A treasury with its own weights and its own officials. Provincial administration by appointed residents rather than only by tributary chiefs. A council at Kumasi with named offices and recorded factional politics — Wilks reconstructs an argument between an imperialist and an accommodationist party over policy towards the British that runs for decades.

That is a nineteenth-century African state described in the terms usually reserved for Byzantium or the Ottomans, and it is not a flattering reconstruction. It is what the records say.

Kumasi

Drafted from Wilks, African Studies Series 13 (1975)expansion

In 1817 Thomas Edward Bowdich came to Kumasi with a mission from the African Company of Merchants and published an account of it two years later. It is the first detailed European description of the Asante capital and it is not what its readers expected.

He describes streets — wide, straight, swept daily, some of them named — and houses of two storeys with polished red-clay walls, carved façades and open courtyards. He counts twenty-seven of the principal streets. He describes a market, a palace complex, and a cleanliness he compares favourably with London.

And he describes the reception, which had something over thirty thousand people in it: massed umbrellas the size of small tents, gold ornaments on a scale he found difficult to convey, horns and drums, and the king's own party arriving last after four hours.

Reading a witness who wanted to be impressed

Bowdich was negotiating a treaty and had reasons to make his hosts sound formidable, which is the opposite of the bias one normally has to correct for in European accounts of African states.

Ibn Battuta at the Malian court five centuries earlier shows that a hostile witness is worth more, because a traveller who objects to unveiled women and matrilineal inheritance is recording things his own assumptions made him notice. Bowdich is the other case: a man with an incentive to be dazzled, whose account has to be read down rather than up.

What survives that discount is checkable against other things. The street plan matches the archaeology and later descriptions. The scale of the gold matches the weights, the regalia and the indemnities Britain later demanded. The population estimates for the city — twenty to thirty thousand in the nineteenth century — are consistent across several independent visitors.

Odwira

The festival Bowdich saw part of is the Odwira, held annually, and it is the state's own account of itself performed for a week.

The stools of the dead Asantehenes were carried out and fed. The chiefs of the metropolitan and provincial divisions attended in person or sent representatives, and their attendance was the visible measure of whether they were still in the union. Tribute was presented. Cases were heard. The new yam could not be eaten until the ceremony released it.

The Odwira is where the two halves of the Asantehene's authority were operated at once. The Golden Stool held the soul of the nation and the blackened stools of the ancestors held the dynasty, and a chief who did not come was making a statement everybody could count.

It is also why the empire's collapse is legible in attendance records before it is legible anywhere else. A chief who stopped coming had seceded, and everybody present could see who was missing.

The Muslims at court

One detail from Bowdich that the general accounts leave out.

There were Muslim scholars resident at Kumasi — from the northern trading towns, serving as clerks, as writers of Arabic correspondence with the northern states, and as makers of protective amulets that Asante officers and the king himself wore.

Asante was not a Muslim state and did not become one. Ghana had run a comparable arrangement eight hundred years earlier and a thousand kilometres north — a royal town with the ancestral cult, and a Muslim merchant town six miles away with twelve mosques and salaried imams — and observes that the arrangement worked because the state did not propagate anything.

Kumasi is the same shape at smaller scale: a state that employed Muslims, used their literacy and their law, paid them, and carried nothing outward.

The great roads

Drafted from Wilks, African Studies Series 13 (1975)peak

Eight great roads ran out of Kumasi. They were numbered in the administrative usage, cleared and maintained, and the government kept a schedule of how many days each stage took.

That last is the detail Wilks's book turns on. A state that records travel times in days is a state that plans around them: the treasury used the figures for accounting, the council used them for deciding whether a report could have arrived, and the courier service — the nkwansrafo, road-wardens, and the runners who carried messages — worked to them.

Kumasi to the coast was about ten days. Kumasi to the northern trading towns about twenty.

A treasury with its own weights

Gold dust was the currency and it was weighed.

The brass weights are the objects most people have seen without knowing what they were for. Geometric ones in graded series for ordinary transactions; figurative ones showing animals, tools, birds, people, scenes from proverbs. They are made by lost-wax casting in their tens of thousands, they are one of the most collected classes of African object anywhere, and they are a monetary instrument.

The treasury — the Gyaasewa — kept the state's gold, its own standard weights, and officials to run both. Wilks reconstructs a working fiscal administration from the surviving correspondence: assessments on the provinces, arrears chased, death duties on officeholders, tolls on the roads in both directions, and the king's share of the goldfields.

Gold, tolls and tribute are all visible in that material, itemised.

What a nineteenth-century African state looks like when somebody has looked

Eight hundred pages by Wilks on how one state actually worked — the bureaucracy, the roads, the treasury, the succession, the named offices, the factions in council arguing about policy — is the fullest institutional account of any sub-Saharan African state before the colonial period.

Compare what is available for its neighbours. Ghana is one Andalusian geographer who never went. Mali is Arabic writers, one visitor, and an epic transcribed in the twentieth century. Great Zimbabwe has no text at all. Kanem-Bornu has a king list and one court historian's account of one reign.

Asante is not better documented because it was better organised. It is better documented because it survived into the nineteenth century, kept written correspondence, and was studied by a historian who spent a career on it.

That is worth saying plainly, because the temptation runs the other way. A state with a detailed institutional history looks more sophisticated than one without, and a great many African polities have no such history for reasons that have nothing to do with how they were run. What survives measures the survival at least as much as it measures the state.

And the fourth field

The succession is the best-documented part of the whole arrangement.

The entry into Kumasi

Drafted from Bowdich, Mission from Cape Coast Castle to Ashantee (1819)aside?What aside means

On 22 May 1817 an African Company mission from Cape Coast Castle walked into Kumasi, and its twenty-six-year-old secretary wrote down what happened with a precision nobody else managed.

They entered at two in the afternoon, passing under a dead sheep wrapped in red silk and slung between two poles. Upwards of five thousand people met them, most of them warriors, with what Bowdich calls awful bursts of martial music, discordant only in its mixture — horns, drums, rattles and gong-gongs going at once, deliberately, to subdue the visitors by first impression. The smoke of continuous musket fire closed the view down to the foreground.

They were halted while the captains performed a war dance inside a ring of their own men, waving English, Dutch and Danish flags together, firing blunderbusses so close that the flags occasionally caught fire.

What a captain wore

Bowdich itemised it, and the itemisation is why the passage is worth having.

A war cap with gilded ram's horns projecting in front and the sides extended out of all proportion with eagle feathers, fastened under the chin with bands of cowries. A red cloth vest covered in gold and silver amulets and embroidered cases of every colour that flapped as the man moved, mixed with small brass bells, horns and tails of animals, shells and knives. Leopards' tails hanging down the back over a small bow. Loose cotton trousers, and boots of dull red leather reaching half-way up the thigh, chained to the waist-belt and hung with bells, horses' tails and strings of amulets. A quiver of poisoned arrows at the right wrist. A long iron chain held between the teeth with a scrap of Arabic writing fixed to the end of it.

Bowdich's verdict on the assembled effect is a figure scarcely human, which is the reaction the dress was designed to produce.

The crowd

The exhibition lasted about half an hour, and then they were let through — the progress as gradual as if it had taken place in Cheapside, which is a young Englishman reaching for the only comparison he has.

The side streets were crammed. On the left the ground rose, so that rows of heads went up one above another. The open porches of the houses, like the fronts of stages in small theatres, were full of women and children waiting to see white men for the first time, and their voices were entirely drowned in the firing.

Then, at the palace, the caboceers passing with their retinues: bands of horns and flutes trained to play in concert, and enormous umbrellas made to rise and fall from the jerking of their bearers, which moved enough air to be a relief under a burning sun.

And what came next

Bowdich then describes something he would not soften, and it is set down in the same register as the umbrellas.

While they waited to be summoned, a man was paraded past them for some minutes. He was being tormented before sacrifice. His hands were tied behind him; a knife had been passed through his cheeks and his lips noosed on it in a figure of eight; one ear had been cut off and was carried in front of him and the other hung by a piece of skin; there were gashes in his back and a knife thrust under each shoulder blade. He was led by a cord through his nose, by men in tall caps of shaggy black skin, with drums beating ahead of him.

Reading both

Bowdich was there to negotiate on behalf of a trading company, and his account was published to argue that Asante was a serious power worth treating with rather than a nuisance to be crushed. That is a real bias and it runs the opposite way from the usual one.

What the book does, and what makes it valuable a century later, is refuse to choose. The same afternoon contains a court of extraordinary sophistication — concert bands, engineered umbrellas, a diplomatic reception staged with complete control — and a man being cut to pieces in the street as part of it.

Both are the same state. An account that keeps only the first is a brochure and one that keeps only the second is a justification for what Britain did afterwards, and Bowdich, who had every commercial reason to write the first, wrote both.

A map of the world

Drafted from Bowdich, Mission from Cape Coast Castle to Ashantee (1819)aside?What aside means

The most interesting thing the British mission found at Kumasi in 1817 was not the court. It was a Muslim scholar's house where men kept arriving from the other side of the Sahara.

Thomas Bowdich called on Baba three or four times a week. Baba wrote with a reed and vegetable ink, preferred them to the pens and pencils the Englishman offered, and had a manuscript full of intricate figures like horoscopes. He put his finger on one and explained what it was for:

if you have any hard palaver, this can make me settle it for you when no other person can; or if you have any dear friend in England you wish to see, tell me the name, and this shall bring him to you.

Bowdich answered that when Englishmen knew their palaver was right they left it to God, and that England was too good a place for him to wish anyone he regarded to leave it.

The business

Baba's pupils wrote on wooden boards. When somebody applied for a charm, one of the older ones wrote the body of it and passed it to Baba, who added a cabalistic mark and a mysterious fold; the customer took it eagerly, paid gold, and hurried off to have it put in the richest case he could afford.

That is one end of a very long supply chain, and Bowdich saw the other end of it at the same house. Every time he called there were strange Moors just arrived from different parts of the interior, staying with Baba. They always affected to deplore the ignorance of the Ashantees, and assumed the Englishman must find it as irksome as they did. One tested him by addressing him in several languages, none of which he recognised except one the man called Hindi, and named great cities Bowdich had never heard of — until he said Room, and told him that if he did not know Rome he was not a Christian.

Bowdich once asked Baba to draw him a map of the world. He drew one large continent encircled by a sea, and the sea bounded by a girdle of rocks.

The man from Timbuktu

On the second visit Baba sent hurriedly for a Moor he said was very learned and had just come from Timbuktu. The man showed no surprise at the sight of a European, and Baba explained why: he had seen three white men before, at Bussa.

Bowdich asked for the particulars and they were repeated through Baba:

that some years ago, a vessel with masts, suddenly appeared on the Quolla or Niger near Boussa, with three white men, and some black. The natives encouraged by these strange men, took off provisions for sale, were well paid and received presents besides … The next day, perceiving the vessel going on, the natives hurried after her … but the white men mistaking, and thinking they pursued for a bad purpose, deterred them. The vessel soon after struck, the men jumped into the water and tried to swim, but could not, for the current, and were drowned.

He thought some of their clothes were at Wawa, and did not believe there were any books or papers.

That is the death of Mungo Park, who had gone down the Niger in 1805 and never been heard of again. It reached a European for the first time in a scholar's house in Kumasi, from a man who had walked there from Timbuktu, told spontaneously and without being asked, in the course of explaining why he was not startled by the colour of Bowdich's face. Bowdich says the artlessness of it made a powerful impression on him. He heard the same account later from another Moor who had not been there.

What the house was

The point is not the anecdote, remarkable as it is. It is where the anecdote was standing.

Kumasi in 1817 is usually described as the capital of a forest kingdom facing the Atlantic: the gold went south to the coast, the guns came north from the Europeans, and the whole structure of Asante power is normally explained by that axis. Baba's house is the other axis, and it was working at the same time. The Asantehene kept Muslim scribes who wrote Arabic; charms made in Kumasi were sold for Akan gold; scholars arrived from Jenne and Timbuktu and stayed; and news travelled that road faster than it travelled to London.

Bowdich made his round of calls in the afternoons and found strange Moors in the house every time. He got a chart from one of them and sent certificates back to a British officer by his hand. And when he asked the resident scholar to draw the world, he was given one continent in a ring of sea inside a wall of rock — a cosmology as confident and as wrong as anything in the mission's own baggage.

The queen mother nominates

Drafted from Wilks, African Studies Series 13 (1975)peak

The Asante succession is the best-documented matrilineal succession anywhere, and it worked like this.

The Asantehene belonged to the royal abusua, the matriclan, and eligibility ran through the female line: a candidate had to be descended from the royal women, so the reigning king's own sons were not eligible and his sister's sons were.

Within that pool, the Asantehemaa — the queen mother, herself a senior woman of the matrilineage rather than necessarily the king's mother — nominated. Then the council of chiefs confirmed or refused. If they refused, she nominated again. In principle she had three attempts, after which the choice passed to the council.

So the office had two gates: descent through women decided who could be considered, and a council of men decided which of them it would be.

Why that is a good design

It solves both halves of the succession problem at once, which very few systems manage.

The first half is legitimacy: who has a claim. Descent answers that, and matrilineal descent answers it with something demonstrable — a man's sister's son is certainly of his blood and his wife's son is a matter of trust. Al-Bakri had been given the same reasoning in eleventh-century Ghana by people who practised it.

The second half is competence: which of the eligible candidates. A rule of descent alone cannot answer that, and every strictly hereditary monarchy eventually gets a child, an incompetent or a lunatic and has no lawful way to avoid him. Asante's council could simply decline, and did.

The cost is the one every elective system pays. When the council could not agree, there was no mechanism at all — and the civil war of the 1880s was fought in exactly that gap.

Where else the female line ran

Five times, on two continents, across four thousand years, with no possibility of contact between the earliest and the rest.

Elam, in the third and second millennia BC, where the throne passed from brother to brother and then to a sister's son.

Ghana, before 1068, where al-Bakri states the rule and supplies the reasoning.

Makuria on the middle Nile, from the seventh century to the fourteenth, where the king's mother and sister held formal office and appear in the cathedral paintings at Faras beside the kings.

Mali, in 1352, where Ibn Battuta found inheritance running to the sister's son and was scandalised by it.

Asante, into the twentieth century.

The convergence is not transmission. It is several societies arriving separately at the view that maternity is a fact and paternity is a claim.

What it did not mean

It did not mean women ruled, and the distinction gets blurred constantly.

The Asantehene was a man. The Asantehemaa was not a co-ruler; she was the person who named the candidate and who held her own court, her own stool and her own jurisdiction, particularly over cases involving women. Descent through the female line and rule by women are different things, and Asante had the first without the second.

What it did mean is that the most consequential political act in the state — the naming of a candidate for the highest office — was performed by a woman, in public, and could be refused by a council, in public, and everybody knew whose turn it was to do what.

Two parties

Drafted from Wilks, African Studies Series 13 (1975)contraction

After Wolseley burned Kumasi in 1874 and left, the Asante council spent twenty years arguing about what to do, and the argument is recoverable.

Wilks reconstructs two standing positions from the correspondence and the recorded debates. One — he calls it the imperialist party — held that the empire should be reassembled by force, the seceded provinces brought back, and Britain faced down. The other, the peace or accommodationist party, held that the tributaries were gone, that the coast trade mattered more than the northern one, and that an accommodation with Britain was the only way to keep anything.

Neither is a faction in the sense of a family or a region. They are positions, argued in council, by named men, across changes of ruler.

Why that is unusual

Because most states before the modern period are visible only as outcomes.

Songhai's kings were deposed at a rate of about one a decade and nobody can say what the factions were arguing about. Mali had two usurpations and no account of the reasoning. First Bulgaria's internal politics survives as a Byzantine chronicler's summary.

Here the reasoning survives. The Asante state wrote letters, kept them, and had a literate staff — the Muslim clerks at Kumasi, and later Africans educated at the coast — and enough of that correspondence was preserved on both sides of the frontier for a historian to reconstruct a policy debate.

It makes Asante one of very few pre-colonial African states where the question what did they think they were doing has an answer that is not an inference.

The twenty years

They were bad. The indemnity Wolseley imposed was larger than the state could pay, and paying any of it meant squeezing provinces that were already leaving.

Dwaben seceded and was reduced by force. The northern tributaries stopped sending. The Brong states went. Gyaman and Nkoranza went. By the mid-1880s the empire had been reduced to its metropolitan core, and the argument in council had become about whether there was anything left to accommodate with.

There was also a civil war over the succession, fought through the machinery the next chapter describes: the Asantehemaa nominating from the royal matrilineage, the council of chiefs confirming or refusing, and — when the council could not agree — no mechanism at all.

Prempeh

Prempeh I was enstooled in 1888 at about sixteen, after that war, and spent seven years doing what the accommodationist party had argued for: recovering the seceded provinces by negotiation where he could, sending an embassy to London in 1895 to argue Asante's case directly, and declining to accept a British protectorate.

The embassy was refused an audience. In 1896 a British expedition came to Kumasi without a battle, and Prempeh — who had been advised to fight and did not — submitted, was arrested along with his mother and much of the court, and was deported to the Seychelles.

He was there for twenty-eight years.

The judgement the record does not make

Whether the accommodationists were right is not answerable, and it is worth saying why rather than leaving the silence to look like neutrality.

The imperialist party's programme required beating Britain, and Asante had done that once, in 1824, and killed a British governor doing it. The accommodationist programme required Britain to accept a client state, and Prempeh's embassy establishes that it would not.

Both parties were arguing from a correct premise and both programmes were unavailable. That is the ordinary situation of a state in the path of an industrial empire.

1901

Drafted from Wilks, African Studies Series 13 (1975)end

Britain and Asante fought five wars between 1824 and 1900.

Asante won the first outright: at Nsamankow in 1824 an Asante army destroyed a British force and killed the governor of the Gold Coast, Sir Charles MacCarthy, whose skull was taken to Kumasi and kept as a royal object. It is one of very few occasions in the nineteenth century on which an African state beat a European force in a pitched engagement, and the only one on these pages besides Ethiopia's Adwa seventy-two years later.

The last is the War of the Golden Stool in 1900, and the country was annexed the following year.

What happened in between

1874 is the hinge. Wolseley's expedition reached Kumasi, found it evacuated, looted the palace and burned the city, and imposed an indemnity larger than the state could pay. What followed was twenty years of secessions, and the argument in council between the party that wanted accommodation with Britain and the party that wanted the empire reassembled by force.

Prempeh I was arrested in 1896 and deported to the Seychelles with his mother and much of the court. He was allowed home in 1924 as a private person and recognised as head of the Kumasi division in 1926; his successor was recognised as Asantehene in 1935, and the office exists.

The stool

In 1900 the governor, Frederick Hodgson, came to Kumasi, demanded the Golden Stool, and announced that he intended to sit on it.

The country rose. The rising is remembered under the leadership of Yaa Asantewaa, the queen mother of Ejisu — which is not incidental, since the office she held was one of the two that controlled the Asante succession. The British were besieged in the Kumasi fort for months and relieved; the rising was put down; Yaa Asantewaa was deported to the Seychelles and died there.

The stool was not found. It had been hidden in the forest and stayed hidden for twenty years. In 1920 African road-builders came across it by accident and stripped the gold ornaments; they were tried by the Asante and would have been executed had the colonial administration not commuted it to exile.

It exists. It is not on display.

The Golden Stool

Drafted from Wilks, African Studies Series 13 (1975)afterlife

In March 1900 the British governor of the Gold Coast, Frederick Hodgson, came to Kumasi, assembled the chiefs, and asked where the Golden Stool was. He wanted to sit on it.

The Asantehene Prempeh I had been deported four years earlier — arrested in his own capital, taken to the coast, and eventually exiled to the Seychelles with members of his family. The Asante state had been dismantled. What Hodgson had apparently not been told, or had not believed, was what the object he was asking about actually was.

The Golden Stool was not a throne. No Asantehene had ever sat on it. It was understood to contain the sunsum — the soul — of the Asante nation, it had its own attendants and its own regalia, it was rested on its own side-cushion so that it never touched the ground, and it was brought out at the Odwira and set beside the king rather than under him.

A governor demanding to sit on it was demanding to sit on the nation.

The war of the Golden Stool

The chiefs said nothing at the meeting and went home, and the country rose.

Yaa Asantewaa, the queen mother of Ejisu — an Asantehemaa of her own district, holding the office that nominated candidates for a vacant stool — is remembered as the one who put the question to the assembled chiefs and shamed them into fighting. The tradition has her saying that if the men of Asante would not go forward, the women would.

The rising besieged the British in the fort at Kumasi for some months. It was put down with troops brought from elsewhere in the empire, Yaa Asantewaa was captured and exiled to the Seychelles herself, and Asante was formally annexed in 1901.

The Golden Stool was not found. It had been buried in the forest, and it stayed buried for twenty years.

What was fought over

That is the part worth dwelling on, because the war has an obvious reading and the obvious reading is wrong.

Britain had already won. Asante had been beaten in the field repeatedly, its king was on an island in the Indian Ocean, its empire had gone, its capital was occupied and its trade was controlled. There was nothing left to defend militarily and the rising had no prospect of succeeding.

What the demand for the stool did was make submission mean something it had not meant before. The Asante had accepted defeat, occupation, deportation and annexation. What they would not accept was the physical transfer of the object their nationhood was held in, to a man who thought it was furniture.

Afterwards

Prempeh came back in 1924, after twenty-eight years, and was permitted to return as private citizen and then recognised as head of the Kumasi division. The Asante Confederacy was formally restored in 1935, under his successor, with the Golden Stool at the centre of it.

The stool itself was accidentally rediscovered by road workers in 1920 and partly stripped of its gold ornaments before it was recovered; the men responsible were tried by the Asante themselves and would have been executed had the British not intervened.

It is still there. The Asantehene remains one of the most substantial traditional authorities in Ghana, the Odwira is still kept, and the stool is still not sat on.

The name, again

One last irony outlasts Asante itself.

When the Gold Coast became independent in 1957 it took the name Ghana — after an empire eight hundred kilometres to the north-west that had ended seven hundred years earlier and had never held any of this ground. Asante, which had actually governed most of the new country's territory within living memory, did not supply the name.

A country named itself after the medieval empire it had no connection to, rather than after the one it was largely made of, because the first had been destroyed long enough ago to belong to everybody and the second was still a going concern with a king in it.