IELTS Academic · Reading

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

The persistence of cash

Forecasts of the cashless society have a long history and a poor record. Each new payment technology has been expected to displace notes and coins, and each has instead been added alongside them. Understanding why requires separating two things that are often confused: the share of transactions conducted in cash, and the quantity of cash in circulation. The first has fallen almost everywhere. The second, in most economies, has risen. The explanation lies in what cash is used for. As a medium for everyday purchase it is genuinely in retreat, displaced by cards and, in much of sub-Saharan Africa, by mobile money accounts operated over ordinary handsets. As a store of value it has if anything strengthened, particularly where confidence in banks is uneven or where interest rates make holding notes cost little. A household that pays for transport by phone and keeps its savings under the mattress is not behaving inconsistently; it is using two instruments for two purposes. Mobile money complicates the picture further, because it is frequently described as a substitute for cash when it functions in practice as a channel for moving it. A user typically converts notes to electronic value at an agent, transfers that value, and the recipient converts it back at another agent. Cash enters and leaves the system at both ends. What has been replaced is not the note but the journey — the bus fare and the day lost carrying money to a relative in another state. This matters for policy because the two readings imply different things. If mobile money substitutes for cash, a central bank might expect demand for notes to fall and plan its printing accordingly. If it instead extends the reach of cash, demand may hold steady or grow even as electronic transaction volumes climb. Several African central banks have found the second pattern, and have had to revise currency forecasts that assumed the first. There is a further consideration that the transactional framing tends to obscure. Cash is the only payment instrument that works without a network, without a charged battery and without an identity record. Those properties are unattractive to a regulator pursuing traceability, and they are exactly the properties valued by someone whose network is unreliable or who has reason to avoid being recorded. A design that treats those users as a problem to be engineered away misreads why the instrument has lasted.

A weed with a use

Water hyacinth is a floating plant native to the Amazon basin. It was carried to other continents in the nineteenth century as an ornamental, admired for a pale violet flower, and it has since become one of the most troublesome aquatic weeds in the world. It reached West Africa in the 1980s. On the lagoons around Lagos, and on stretches of the Niger and Benue, it now forms mats dense enough to stop a boat. Its success rests on how fast it reproduces. A single plant sends out horizontal stems that produce daughter plants, and under favourable conditions a population can double in under a fortnight. It does not depend on seed to spread, though it produces seed that can remain viable in sediment for many years. A mat that has been cleared can therefore return from material that was never visible. The damage is easiest to see in fishing communities. A mat blocks light, and the plants beneath it die. As that material decomposes it consumes dissolved oxygen, and fish either leave or suffocate. Canoes cannot be pushed through a thick mat, so landings that have been used for generations fall out of use. Where a mat sits over slow water it also provides breeding habitat for mosquitoes and for the snails that carry schistosomiasis. Control has proved expensive and rarely permanent. Mechanical harvesters lift the plants out, but a harvester is costly to run and the cleared water is recolonised within weeks unless the work is continuous. Herbicides work quickly and are cheap, but they are applied to water that people drink and fish in. The most durable results have come from introduced weevils that feed on the plant, although the effect builds over several seasons rather than in one, which makes it a difficult thing to fund. A different approach treats the mats as a crop. The dried stems can be woven, and a small industry making baskets and furniture has grown up around several Nigerian lagoons. The plant can also be digested to produce biogas, or composted. None of this removes the weed: the quantities harvested are trivial beside the quantities growing. What it does is give the people most affected a reason to keep harvesting, in places where no agency is funded to do it for them.

The terracottas of the Jos Plateau

In 1928 a tin miner working near the village of Nok, on Nigeria's Jos Plateau, turned up a small terracotta head in his sluice. It was set aside as a curiosity. Over the following two decades, as alluvial mining spread across the plateau, hundreds more fragments emerged from the gravel — heads, torsos, limbs, and occasionally a figure almost complete. By the 1940s the archaeologist Bernard Fagg had recognised that these objects shared a distinctive style and were unlikely to be recent. The Nok figures are modelled rather than moulded, built up from coarse clay containing visible grains of quartz. Most are hollow, with walls of even thickness, and were fired at a relatively low temperature. The heads are disproportionately large for the bodies, and the eyes are rendered as triangles or segments of a circle, with the pupil pierced straight through the clay. Hair is shown as elaborate buns and plaits. Many figures wear beaded ornaments at the neck, waist and ankle. Dating them has proved harder than describing them. Almost none has been recovered from an undisturbed archaeological layer. The mining that brought them to light also destroyed the stratigraphy that would have dated them, and the great majority of pieces in circulation have no recorded findspot at all. The dates most often quoted — roughly 900 BCE to 200 CE — rest on a small number of controlled excavations, principally those conducted from the 1990s onwards near Kaduna, together with thermoluminescence readings taken from the clay itself. What the figures were for remains unsettled. They are too numerous and too varied to be simple portraits. Some scholars have argued for a funerary purpose, noting that a few fragments appear to have been deliberately broken. Others point to the recurring gestures — a hand raised to the chin, a figure kneeling — and suggest a ritual repertoire whose meaning is now lost. A third position holds that the question is badly framed: a tradition lasting a thousand years is unlikely to have served one purpose throughout. The looting of sites has made all of this worse. Because Nok pieces command high prices, plateau farmland has been dug over systematically since the 1990s, and objects reach dealers stripped of every context that would make them informative. A figure without a findspot can still be admired. It can no longer be interrogated.