Abstract: A shirt occupies half a square metre around one body, and the world that had to be coordinated to produce it is enormous. This article follows the materials of clothing across world history through the concept of the multimaterial hinterland: all the relations through which the materials of a garment are generated, obtained, transported and transformed, across species, places and durations. It begins before textiles, with the recruitment of surfaces that other living beings had already made, and shows how wool and silk drew sheep and silkworms into new dependencies on humans, how fur made territory in northern North America and Siberia, how cotton bound plantation slavery, Atlantic shipping, credit and Lancashire mills into one trajectory and helped to make industrial capitalism, and how synthetic fibres replaced the harvest of living producers with hydrocarbons that ancient organisms buried, extending the hinterland to oil fields, refineries, sewing floors and the places where garments end. Materials differ in properties such as the ratio of value to mass, dependence on living producers and capacity to be mechanised, and these differences help to explain why each produced a different kind of world-historical trajectory. The materials of clothing accumulate rather than replace one another, so that a single coat is an archaeological site worn on the body. The article distinguishes this genealogical claim from commodity-chain analysis. A supply chain describes where the parts of a garment come from now, whereas a hinterland describes what had to become possible before such a garment could be made.
I. Before the Shirt
Consider an ordinary cotton shirt, taken from a drawer and put on without thought. Hardly anyone asks, while getting dressed, what had to exist before this object could exist. Once the question is asked, the apparently singular object explodes. The cotton required plants, land, water, sun, harvesting and ginning, and before that the domestication of the plant and its movement between continents. The thread may be polyester, which required oil or gas, refineries and a chemical industry. The buttons may be plastic, shell or horn. The dye required chemistry, and before the chemistry of the nineteenth century it required indigo plants, madder roots or insects. The shirt was cut and sewn by people working machines in factories connected by ships and trucks to growers, spinners, weavers, dyers and buyers, and it was packed, shipped, stored and sold before it reached the drawer.
The garment is therefore the visible endpoint of a vast multimaterial hinterland. The concept can be defined simply. A garment's multimaterial hinterland consists of all those relations through which the materials necessary for its production are generated, obtained, transported and transformed. The hinterland includes living beings, cotton plants, sheep, silkworms, beavers, and the human beings who grow, raise, hunt, spin, weave and sew; it includes found and extracted matter, water, soil, minerals and hydrocarbons; and it includes the made infrastructures, mills, ships, railways and refineries, through which these materials are brought together.
The hinterland is enormous relative to its product. A shirt occupies perhaps half a square metre around one body. A garment is a mobile made environment surrounding that body, a portable remediation of the conditions to which the body is exposed, and its hinterland extends across species, continents and centuries. This article turns the garment inside out, asking how much of the world must be coordinated to produce that small mobile environment.
The hinterland is easy to overlook for a reason that applies to successful coordination generally. A garment that works withdraws from attention, and the world that produced it withdraws with it. The hinterland becomes visible mostly when it is interrupted or when it fails: a blockade of ports, the collapse of a factory, a shipment of clothes that nobody can sell. The discussion below returns repeatedly to such moments, because misfit leaves an archive and fit leaves only a shirt.
The framework used here treats multimateriality as one of five mediations through which living beings remain coordinated with a world, alongside multisensorial embodiment, being-with, multiversal dwelling and multisymbolization. Multimateriality names the relations through which living beings recruit, produce, select, displace, transform, combine and eventually anticipate material forms. It has a layered evolutionary history, resting on two preconditions, material dependence and metabolic externalisation, and unfolding through six partially ordered layers: recruited self-produced matter, recruited found matter, selected and displaced matter, transformed matter, combined matter and projectively fabricated matter. Its criterion is recursive uptake, since a material difference becomes multimaterial only when it enters a living being's further coordination. Later layers accumulate without abolishing earlier ones, and found and made name relations to aspects of a configuration rather than classes of object. Later layers also remain answerable to the older conditions on which their viability depends, a principle that can be called downward answerability. Dwelling is used strictly here for non-human-made conditions such as terrain, water, climate and light, so that garments, mills and ships belong to multimateriality even when their purpose is to modify the dwelling conditions to which bodies are exposed.
The asymmetry between the intimacy of the need and the scale of the hinterland gives the article its world-historical question: what happens when increasingly large human populations require increasingly large quantities of increasingly elaborate bodily coverings? The answer is extraordinary. Landscapes change. Species change. Plants are carried between continents. Animals are killed in enormous numbers. Human beings are enslaved. Rivers are harnessed to turn mills. Factories are built and cities grow around them. Railways connect ports with manufacturing centres. Empires seek control of the territories where materials are found. Chemistry creates substances that had never existed.
Clothing is an especially revealing case because the need it answers is so mundane. Human beings need and want to cover their bodies, every day, everywhere. The need is recurrent and nearly universal, and so the demand for the materials of clothing has been constant, cumulative and immense. The history of clothing accordingly becomes a history of making the world capable of producing clothes.
II. Before Textile Production: Recruiting Living Surfaces
The first distinction is crucial. Human beings did not initially need to manufacture fibres, because other living beings had already manufactured coverings. Skin, hide, fur, feathers and wool are products of animal bodies. Bast fibres in the stems of flax, hemp and nettle, the fibres around cotton seeds, the inner bark of trees and the leaves of grasses and palms all possess properties that humans could recruit. The earliest history of clothing begins by recruiting surfaces that other living processes had already made.
This connects the history of clothing with the deepest layers of multimateriality. The oldest form of making, recruited self-produced matter, is exemplified by the mollusc that secretes its shell and the spider that draws its web from its own glands, and the next layer, recruited found matter, is exemplified by the hermit crab that occupies a shell another animal made. The human wearing of skins belongs to the second of these relations. A fur is no raw material from the perspective of the animal that grew it. It is an extraordinarily elaborate zoetic production, the outcome of a long evolutionary history in which the animal's lineage came to regulate its temperature, shed water, and display itself to others through the structure of its hair. Human beings appropriate that production into another recursive trajectory, in which it will regulate a different body's temperature and display a different body to others.
The deep history of clothing therefore runs less from nature to culture than from one living being's embodied formation, through its recruitment by another living being, to a transformed covering. The formulation matters because it keeps the other species in view. The beaver, the sheep, the silkworm and the cotton plant are no mere sources of stuff. Each is a living process that produced, for its own purposes, the material humans came to wear, and each was drawn by that fact into a relationship with human beings that transformed its life, its population and in several cases its biology.
Plant fibres enter the history very early. Twisted and dyed fibres of wild flax around 30,000 years old were found at Dzudzuana Cave in Georgia (Kvavadze et al. 2009), and cordage, netting and woven textiles, perishable as they are, appear to have been far more important in Palaeolithic life than their scarce remains suggest. Plant fibres required a more elaborate transformation than skins. The stems must be retted, soaked until the soft tissue rots away from the fibres, then beaten, combed, spun and woven. Yet the fibres themselves were still produced by plants. The transformation began from material a living process had already made.
Some traditions developed transformations of living surfaces that required neither spinning nor weaving. Across Polynesia and much of the tropical Pacific, the inner bark of the paper mulberry and related trees was soaked and beaten into broad sheets of barkcloth, tapa, which could be layered, decorated and exchanged in great quantities at ceremonies. Hides were softened by rubbing in animal brains or fats, or smoked over fires, so that they would neither rot nor stiffen. Wool and hair could be matted into felt by moisture, heat and pressure, as the pastoral peoples of Central Asia did for their clothing and for the walls of their tents. Each of these techniques worked with a property that the living source had supplied, the fibre structure of bark, the collagen of skin, the scales of hair, and turned it into a covering the source had never been.
This relation remained fundamental for almost the whole of human history. Until the twentieth century, very nearly every garment worn anywhere began as the product of some other living being. What changed over the millennia was the depth of human intervention in the living processes that produced it.
III. Domestication: Making Animals Produce Clothing Materials
Clothing became implicated in the transformation of other species when human beings began not only to find suitable coverings but to shape the living beings that produced them. Wool is the great case.
Sheep were domesticated in southwestern Asia roughly ten thousand years ago from wild mouflon, whose coats consisted mainly of coarse hair over a short, fine underwool that was shed each spring. Early domestic sheep were kept largely for meat. The woolly fleece of later breeds, in which the fine underwool grows long and continuous and the coarse hair is reduced or absent, was the result of selection over several millennia, and its spread has been associated with what Sherratt (1981) called the secondary products revolution, the growing use of animals for products they yield while alive, such as milk, traction and wool.
Once human beings bred animals partly for the properties of their fleece, something profound changed. Humans no longer merely found suitable material. Through selective breeding, husbandry, feeding, seasonal management and protection, they recursively altered the living processes that produced the material they wanted. The sheep became part of a system of production that spanned two species, and its body was reshaped toward the needs of other bodies. Most modern wool breeds no longer shed their fleece naturally. They must be shorn, and an unshorn sheep can suffer from heat, immobility and infestation. The animal has been made dependent on the human intervention that its transformation required. Godfrey-Smith (2024) argues that judgements about the use of animals should attend to their whole lives and not only to the moment of harvest or slaughter, and the woolly sheep is an instructive case, since the whole of its life has been shaped to the requirements of shearing.
This is an important intermediate relation between found material and projectively fabricated material. The material is still produced by another living organism, but humans have progressively altered the conditions of its production, and eventually the organism itself, so that particular material properties recur. It is the recursive modification of another species according to anticipated multimaterial properties, a relation that goes well beyond domestication in the general sense. Its mechanism repays a closer look. Selective breeding works through what Godfrey-Smith (2024), following Wright (1976), calls feedback from past cases: fleeces that served their keepers well were kept and bred from, so that the past successes of the flock explain the qualities of its descendants. Nobody needs to have represented the merino in advance. The contrast with the synthetic fibres discussed below, in which a property is specified before any material exists, marks the difference between two routes by which forms come about because of their effects.
Wool then reorganised landscapes and polities. In medieval Castile, the Mesta, an association of sheep owners chartered in 1273, drove flocks of merino sheep hundreds of kilometres each year between summer and winter pastures along protected routes, and the Crown for centuries forbade the export of living merinos in order to keep its monopoly of the finest wool. In England, wool was the country's principal export in the later Middle Ages, and the Lord Speaker of the House of Lords still sits on the Woolsack, a cushion stuffed with wool, as a reminder of it. The conversion of arable land to sheep pasture through enclosure drove people from the land, and Thomas More's Utopia (1516) described sheep so greedy that they devoured men. When merino sheep reached Australia at the end of the eighteenth century, wool became the foundation of a pastoral economy that expanded across Aboriginal lands. The fleece on a sheep's back thus mapped pastoral landscapes, dispossessions and colonial frontiers across several continents.
In the Andes, the domesticated llama and alpaca and the wild vicuña supplied fibres that became the basis of a political economy of cloth. Murra (1962) showed that in the Inca state cloth functioned as a primary form of wealth and obligation. Households owed textiles to the state, specialised weavers produced the finest cloth, called cumbi, for the ruler and for gifts, soldiers were issued clothing from state storehouses, and the burning or bestowal of cloth marked the most important political and ritual occasions. The fine fibre of the vicuña, captured in periodic royal hunts, shorn and released, was reserved for the highest ranks. Here the materials of clothing did more than reorganise landscapes. They became the medium through which a state recorded and enforced the relations between rulers and subjects, and the herds and hunts that produced the fibre were organised accordingly.
IV. Silk: When a Fibre Reorganises Eurasia
Silk gives an even stranger relation between species. The domestic silkworm, Bombyx mori, descends from a wild moth of eastern Asia, and it has been so transformed by thousands of years of rearing that it can no longer live without human beings. Its larvae feed only on mulberry leaves supplied by their keepers, and its adults cannot fly. Each larva spins a cocoon from a single continuous filament of protein that can run to several hundred metres or more. To obtain the filament unbroken, the cocoon must be heated to kill the pupa before the moth emerges and cuts its way out, and the filament must then be softened and reeled.
Silk thus required the cultivation of mulberry trees, the management of silkworm populations through their delicate life cycle, the development of techniques for reeling, twisting, weaving and dyeing, and, for long periods, the guarding of production knowledge. Silk fabrics were being made in China by the third millennium BCE, and for centuries China was the sole source. The Byzantine historian Procopius records that around 552 CE monks brought silkworm eggs to Constantinople concealed in hollow canes, allowing the Byzantine Empire to begin its own production (Procopius 1928). Sericulture was also intensely laborious. Silkworms must be fed fresh mulberry leaves many times a day and kept at the right temperature and humidity, and reeling requires constant skilled attention. In China, Japan and elsewhere, this work fell overwhelmingly to women and girls in rural households, and later in reeling mills. In the late nineteenth century raw silk became Japan's leading export, and the earnings of silk, produced largely by young women in rural filatures, helped to finance the country's industrialisation.
The multimaterial hinterland of silk extended spatially on a continental scale. The network of routes across which silk and many other goods travelled was named the Silk Road, Seidenstraße, by the geographer Ferdinand von Richthofen only in 1877, and the routes were never built for silk alone. They carried horses, spices, metals, paper, glass, religions, technologies and diseases, and much of their trade was regional rather than end to end. Silk was nonetheless consequential enough to give the whole network its retrospective name. Han dynasty China paid tribute in silk to the nomadic Xiongnu, and later dynasties used bolts of silk as a medium of taxation and payment. Pliny the Elder complained that Rome's appetite for eastern luxuries, silk among them, drained the empire of silver.
The material itself matters here. Silk combines high value with low weight, durability and ease of packing. A pack animal carrying silk carries a great deal of value in a small load, and the load survives months of travel without spoiling. Compare timber. A kilogram of fine silk and a tonne of ordinary timber pose radically different problems of transport, and the distance over which each can profitably be moved differs accordingly. Clothing thus reveals another dimension of multimateriality. The ratio of value to mass helps determine the possible spatial extension of the networks that move a material. The Silk Road is partly a consequence of what silk physically is.
The principle generalises. Materials with a high value-to-mass ratio, silk, spices, precious metals and dyes such as cochineal, sustained long-distance trade across premodern worlds in which transport was slow and expensive, while bulky materials of low value travelled short distances or not at all. The shape of premodern networks of exchange was therefore partly written in the physical properties of the materials they carried. Changes in transport, from the sailing ship to the steamship and the railway, altered this relation and allowed bulkier clothing materials, wool and above all cotton, to travel the world in quantities that silk never approached.
V. Fur: Clothing Makes Territory
European demand for fur, and above all for beaver, drove an extraordinary expansion into northern North America. The fashionable felt hats of seventeenth- and eighteenth-century Europe were made from the dense underfur of the beaver, whose microscopic barbs allowed it to mat into a strong, waterproof felt. Hatters preferred pelts that had already been worn for a season or two by their Indigenous owners, fur side inward, since wear loosened the long guard hairs and left the underfur ready for felting. The felting process itself came to use a solution of mercury nitrate, and the chronic mercury poisoning of hatters gave English the phrase mad as a hatter. The hat on a European head was thus connected through its material to animals, to Indigenous bodies and to the poisoned bodies of workers.
The trade that supplied the fur is often told as a story of Europeans arriving and collecting resources. The system depended fundamentally on Indigenous knowledge, hunting, preparation of pelts, transport, existing routes and trading relations. Cree, Ojibwe, Dene and many other peoples trapped, processed and carried furs, supplied the food, canoes and snowshoes on which traders depended, and set many of the terms of exchange. Indigenous women were central to the fur-trade society that grew up around the posts, as partners, interpreters, guides and makers of the clothing and provisions that made travel possible (Van Kirk 1980), and the Métis emerged as a people within this world.
The demand for fur also produced territorial infrastructures. In 1670 Charles II granted the Hudson's Bay Company a charter to the lands draining into Hudson Bay, a territory named Rupert's Land that covered much of what is now Canada. Posts such as York Factory became distribution and administrative centres linking inland fur supplies to European markets. The rivalry between the Hudson's Bay Company and the Montreal-based North West Company carried posts and routes westward to the Pacific. At Fort Langley on the Fraser River, founded as a fur trading post in 1827, the Colony of British Columbia was proclaimed in 1858. When the Hudson's Bay Company surrendered Rupert's Land in 1869–1870, it passed to the new Dominion of Canada. Harold Innis (1930) argued that Canada emerged along the lines of the fur trade, its boundaries following the drainage systems through which furs had moved, and that the country existed because of its geography and its staple rather than in spite of them.
The same pattern appears on the other side of the pole. Russian expansion across Siberia from the late sixteenth century was driven to a large degree by sable and other furs, collected as tribute, known as yasak, from Indigenous peoples, and fur was for a time among the most valuable sources of revenue for the Muscovite state. Two empires extended themselves across the northern hemisphere in pursuit of materials for clothing.
The argument can now be stated directly. A demand for bodily covering in one mesocosm can reorganise relations among species and among people thousands of kilometres away. Someone wanting a fashionable beaver hat in Paris or London became connected, through many intermediaries, to beavers, waterways, Indigenous trappers, voyageurs, trading posts, ships, merchants, chartered companies and imperial institutions. The connection went beyond the circulation of a commodity, because the multimaterial demand produced territory: posts, routes, charters, colonial proclamations and eventually national boundaries. The materials required to make clothes have helped make territories.
The beavers themselves were changed. Trapping reduced beaver populations across much of North America, and because beavers are ecosystem engineers whose dams create ponds and wetlands, their removal transformed rivers and landscapes. Life has always shaped rivers, and Godfrey-Smith (2024) notes that rivers with definite banks and bends owe part of their form to plants, whose roots hold banks in place, and that the shape of the land follows in turn from the shape of the water. The fur trade altered the dwelling conditions of entire watersheds by removing one of the animals that had shaped them. On the Pacific coast the pattern repeated with the sea otter, whose exceptionally dense fur was prized in China. Russian, British and American traders pursued it along the coasts from Alaska to California in the late eighteenth and early nineteenth centuries, the Russian-American Company establishing a chartered presence in Alaska in the process, and the animal was hunted almost to extinction. Once again, a fur demanded for clothing in one part of the world drew a territorial and commercial frontier across another.
VI. Cotton: When Clothing Reorganises Bodies, Land and Labour
Cotton is the historical centre of this article, because its fibre links several enormous transformations at once: cultivation, plantation agriculture, enslavement, Atlantic shipping, ports, credit, mechanised spinning and weaving, factories, urbanisation, railways, industrial labour and global markets.
Cotton was domesticated independently in South Asia, in the Americas and in Africa, and for millennia the most accomplished cotton textiles in the world were made in India. Indian calicoes and muslins were so popular in early modern Europe that the English Parliament passed Calico Acts in 1700 and 1721 restricting their import and use, in order to protect domestic woollen and silk manufacturers (Riello 2013). The later dominance of British cotton manufacturing was thus built in part on excluding the Indian textiles it would eventually replace, and the subsequent decline of Indian hand weaving under competition from Lancashire cloth gave the spinning wheel its later significance as a symbol of Indian self-rule.
The raw material for British mills came increasingly from the Americas, and above all from the slave plantations of the United States. After the cotton gin mechanised the separation of seeds from short-staple cotton in the 1790s, cotton cultivation spread rapidly across the American South, and the enslaved population of the United States grew from about seven hundred thousand in 1790 to almost four million in 1860. Enslaved people were forcibly moved in their hundreds of thousands from the older states of the upper South to the new cotton lands of the Deep South, and their labour was driven by violence to extract ever greater quantities of cotton per person. By the middle of the nineteenth century, most of the raw cotton spun in Britain was grown by enslaved people in the United States (Beckert 2014).
Multimaterial analysis prevents slavery from becoming an external social context of textile history. The coercive organisation of human bodies was part of the production system required to deliver the material. This has to be stated carefully. The cotton plant did not require slavery; it had been grown for millennia by peasants in India, Africa and the Americas, and it has been grown since by free labour, sharecroppers and colonial smallholders. Particular historical systems organised cotton production through racialised chattel slavery. Once that system was established, British textile manufacturing became deeply dependent on its output, and contemporaries in Parliament and elsewhere acknowledged the dependence openly. The dependence became visible when the supply was interrupted. During the American Civil War, the blockade of Southern ports produced the Lancashire cotton famine of 1861–1865, when mills closed and hundreds of thousands of workers were thrown into destitution, and many of them nonetheless supported the Union cause against the slaveholding South. The famine also sent British manufacturers and officials in search of other sources. Cotton cultivation expanded rapidly in Egypt and India during the war, railways were built into cotton-growing districts of western India to carry the crop to Bombay, and in the following decades colonial administrations promoted cotton growing across Africa and Asia, often through compulsion (Beckert 2014). The interruption of one material trajectory rerouted the whole hinterland, and new landscapes, labour regimes and infrastructures were drawn into the production of the same fibre.
The fibre can be followed. It moved from plantation to gin to bale, from bale to riverboat and ship, from ship to the docks of Liverpool, from Liverpool by canal and, from 1830, by the Liverpool and Manchester Railway to the mills of Manchester and the surrounding Lancashire towns, from mill to yarn, from yarn to cloth, and from cloth back through Liverpool to markets across the world, including India. Each stage was a transformation, a displacement or a combination in the multimaterial sense. Each also required credit, insurance, contracts, brokers and exchanges, and each connected the fibre to a different set of human bodies, enslaved, waged, seafaring, merchant or consuming.
This yields a methodological principle. When the material transformation is followed, apparently separate histories become one history. The history of slavery, colonial history, transport history, labour history, industrial history and the history of fashion appear as intersecting strands of a single multimaterial trajectory, the trajectory of cotton becoming clothing, rather than as separate contexts surrounding it. Beckert (2014) called the violent system through which cotton was secured war capitalism, emphasising expropriation, slavery and state power. The multimaterial account agrees on the violence and adds a question about the material itself: why this fibre, with these properties, could be grown in these places, moved over these distances, and transformed by these machines into cloth cheap enough to clothe much of the world.
Forced labour in cotton did not end with emancipation in the Americas. It persisted in colonial cotton schemes in Africa and in state-organised harvests elsewhere into the twenty-first century. The cotton plant's history remains entangled with the coercion of the bodies that grow and pick it.
VII. The Mill: When the Production of Cloth Produces Industrial Capitalism
Textile manufacture was among the central sites of industrialisation rather than one of its incidental beneficiaries. The spinning jenny of the 1760s, Arkwright's water frame and the mill he built at Cromford in 1771, Crompton's mule and eventually the power loom transformed the spinning and weaving of cotton from household work into factory production. The earliest factories in Britain grew up around the cotton industry of the later eighteenth century, first beside rivers that could turn waterwheels and later around steam engines burning coal. Godfrey-Smith (2024) reports Lovelock's proposal to date the Anthropocene from the first efficient coal-burning steam engine, Newcomen's of 1712, which was built to pump water out of a coal mine. The engine that would later drive the mills was first a device for reaching more of the fuel on which it ran.
Here the production of clothing began producing a new form of human temporal coordination. Machines required workers, and workers had to arrive at specified times and stay until the machines stopped. Production became synchronised by the clock and the bell. Children and adults were coordinated with machinery that set the pace of their work. E. P. Thompson (1967) described how industrial capitalism imposed a new time discipline on working people, replacing the rhythms of tasks and seasons with the uniform hours of the factory. Factory discipline became a problem for owners and a grievance for workers. Urban settlements grew around the mills, and Engels (1845) described the conditions of Manchester's working class in the 1840s with an intensity that made the city a symbol of the new industrial order. By the 1830s Parliament was already confronting the conditions of factory textile production, and the Factory Act of 1833 forbade the employment of children under nine in textile mills, limited the hours of older children, and created a factory inspectorate to enforce the rules.
The usual direction of explanation runs from capitalism to the textile factory: capitalist relations of production created factories, and textiles happened to be the first industry organised that way. The multimaterial account suggests a complementary direction. The problem of producing enormous quantities of standardised cloth from a fibre that could be spun and woven by machines generated pressures toward machine coordination, concentrated labour, capital investment, factories, transport networks, regulation and wage dependence. Textile production was one of the material domains through which industrial capitalist arrangements became historically realisable, and clothing is not thereby cast as the cause of capitalism. The particular properties of cotton, a fibre strong and uniform enough to be spun by machines, supplied from abroad in rapidly growing quantities, made it an especially suitable domain for the purpose.
The mill spread with the fibre. In the United States, the mills of Lowell, Massachusetts, founded in the 1820s, recruited young women from New England farms and housed them in company boarding houses. In India, the first cotton mill in Bombay opened in the 1850s, and Bombay and Ahmedabad became centres of an Indian textile industry that competed with Lancashire for the Indian market and eventually surpassed it. Later in the twentieth century, cotton spinning and garment making moved again, to Japan, then to Hong Kong, South Korea, Taiwan and China, and then to Bangladesh, Vietnam, Cambodia and elsewhere. In each place, the production of cloth and clothing was among the first forms of factory industry, and in each it drew rural people, very often young women, into urban wage labour.
The mill also displays the non-supersession principle at the scale of a building. The machines were iron and the power was coal, but the mills remained dependent on the river water that turned their first wheels and kept the air humid enough for spinning, on the bodies of workers whose hands pieced broken threads, and on a plant that had to be grown under the sun on the other side of an ocean.
VIII. From Harvesting Fibres to Designing Fibres
For almost the whole of human history, clothing began with fibres generated by living organisms. The cotton plant made cotton, the sheep made wool, the silkworm made silk and flax made linen, and human beings recruited and transformed what these living processes produced. The twentieth century altered this relation.
The transition passed through an intermediate stage. Artificial silk, later called rayon or viscose, was developed from the 1880s onward by chemically dissolving cellulose from wood pulp or cotton waste and extruding it as a filament. The fibre was made by chemistry, but its raw material was still produced by trees. With nylon, developed by Wallace Carothers and his colleagues at DuPont in the 1930s and sold as stockings from 1940, the relation changed more fundamentally. Polyester followed, developed in 1941 by John Rex Whinfield and James Tennant Dickson at the Calico Printers' Association in Manchester, in the city that cotton had built. Elastane and acrylic followed in their turn.
With synthetic fibres, the desired properties of the fibre increasingly became projectively specified before the fibre existed: strength, elasticity, water resistance, durability, texture, thermal behaviour, washability, retention of colour. Chemistry could then work backward toward a material capable of producing those properties. This is projective multimateriality in exceptionally pure form. The fibre is designed as a molecule before it is synthesised, and its properties are specified before any living process has been asked to supply them. In the terms noted earlier, the route from past feedback to represented outcome is complete: the sheep's fleece was bred by keeping what had worked, and the polymer is made to meet a specification that no earlier fibre satisfied.
Dyes had already travelled the same path. For millennia, colour came from plants, insects and molluscs: indigo from the plants of India and elsewhere, madder from roots, crimson from the cochineal insects of Mexico, whose trade enriched the Spanish Empire, purple from the glands of sea snails. In 1856 William Perkin, attempting to synthesise quinine, produced mauveine from coal tar, the first synthetic dye, and a chemical industry grew rapidly around the manufacture of colours. When synthetic indigo came onto the market in 1897, it destroyed within a few decades the Indian indigo industry, whose plantation system had already provoked the indigo revolt of 1859–1860 among Bengal cultivators forced to grow the crop.
Synthetic fibres promised an escape from sheep, silkworms and cotton plants, and the escape was real in one respect: the fibres are no longer grown. Their feedstocks nonetheless come from somewhere, and where they come from is itself part of the history of living producers. Photosynthesis stores the energy of sunlight in chemical bonds, and when the remains of plants and plankton are buried before they can break down, the carbon they hold is interred instead of returned to the air (Godfrey-Smith 2024). Coal, oil and gas are that interred matter. Most synthetic fibres are made from petroleum and natural gas, and polyester has become the most widely produced fibre in the world, accounting for more than half of global fibre production. The apparent escape from living producers therefore replaced the harvest of living plants and animals with the mining of the bodies of dead ones, laid down over millions of years. It also produced a new multimaterial hinterland of wells, pipelines, refineries, petrochemical complexes and global hydrocarbon infrastructures, reaching into the geology of the Middle East, the Gulf of Mexico and the North Sea. The hinterland extends downstream as well. Synthetic garments shed microscopic fibres when they are washed, and shoreline surveys have traced synthetic fibres of the kind released by laundering to coasts around the world (Browne et al. 2011). Such fibres now reach rivers, oceans and the bodies of marine animals. The synthetic garment did not abolish the material hinterland of clothing. It changed its geology, and it extended the hinterland into the waters that receive what the garment sheds.
The hinterland of contemporary clothing also includes its sewing and its disposal. Garments are still sewn largely by hand-guided machines, by workers concentrated in countries where wages are low, and the collapse of the Rana Plaza building near Dhaka in 2013, which killed more than eleven hundred garment workers, made visible how much of the cost of cheap clothing is carried by the bodies of those who assemble it. At the other end of the garment's life, clothes discarded in wealthy countries are exported in bales to markets such as Kantamanto in Accra, where traders buy them unseen and a large share proves unsaleable, and discarded garments accumulate in landfills and open dumps from West Africa to the Atacama Desert. The hinterland of a shirt therefore extends backward to the plant and the oil well, sideways to the sewing floor, and forward to the places where the shirt will end.
IX. A Stratigraphy That Never Disappears
The materials of clothing did not succeed one another in a simple sequence from fur to wool to silk to cotton to synthetics. All of them remain. Human beings still wear animal skins, as leather shoes, jackets and belts. Wool remains, from merino base layers to the tweed of old jackets. Silk remains, in saris, ties and linings. Cotton remains the most familiar fibre of everyday clothing across the world. Synthetic fibres proliferate. Regenerated cellulose fibres such as viscose and lyocell occupy an intermediate stratum between plant and chemical plant. Blends of natural and synthetic fibres have become normal.
A single contemporary garment can therefore contain several historical strata at once. Imagine a winter coat with a wool outer shell, a cotton lining, polyester insulation, nylon components, a metal zip, plastic buttons, synthetic dyes and perhaps a leather trim. Each material carries its own trajectory: the sheep bred over ten thousand years toward its fleece, the cotton plant with its history of slavery and industry, the oil formed from ancient organisms and extracted through modern infrastructures, the metal mined and smelted, the leather from an animal raised and killed. The coat is almost an archaeological site that one wears, its layers deposited by different histories and brought together only in its making.
This makes clothing an exceptionally powerful demonstration of cumulative layering rather than evolutionary replacement. Later relations to matter are added to earlier ones without abolishing them, and in clothing the layers are literally stitched together and worn next to the skin.
X. The Multimaterial Map of World History
The large claim of this article can now be made. Place the major materials of clothing on a world map, and something extraordinary happens. Silk draws routes across Eurasia. Wool maps pastoral landscapes, transhumance routes, enclosures and settler frontiers. Fur draws colonial networks across northern North America and Siberia. Cotton connects plantations in the Americas with Atlantic shipping, Liverpool, Lancashire, India and global markets for cloth. Synthetic fibres connect clothing to oil fields, refineries and chemical industries. Dyes add further lines, from Mexican cochineal to Bengali indigo to the coal-tar chemistry of Britain and Germany. The history of clothing begins to look remarkably like world history.
The map is organised around material trajectories rather than nations. World history conventionally takes civilisations, empires, states, religions, economic systems or oceans as its organising units. Multimaterial history can instead ask what worlds had to be connected and transformed so that particular materials could acquire particular forms. The units of analysis become the trajectories of materials, and the civilisations, empires and states appear as what those trajectories passed through, connected, enriched and sometimes created.
Materials differ along dimensions that help to explain why each drew a different trajectory. Silk has a very high ratio of value to mass and so could travel by camel and mule across a continent. Wool depends on living producers whose reproduction, movement and pasture must be managed. Cotton is a fibre strong and uniform enough to be mechanised at every stage and so could be scaled almost without limit. Fur depends on wild animals whose populations can be exhausted, so that its frontier keeps moving. Synthetic fibres depend on no living producer at all and on an infrastructure of extraction that the ancient dead supplied. Such properties belong to a multimaterial profile of each material, and comparing profiles allows the differences between trajectories to be explained and not merely narrated.
Clothing is an ideal demonstration because the final stake is so mundane. Human beings want and need to cover their bodies. Satisfying that recurring requirement eventually connected silkworms, sheep, beavers, cotton plants, enslaved people, Indigenous hunters, merchants, sailors, mill workers, chemists, oil wells, factories, railways, ports, states and empires. The asymmetry between the intimacy of the need and the scale of the world reorganised to meet it is the central finding of the article.
The approach has close predecessors. Mintz's (1985) history of sugar showed how a single substance connected Caribbean slavery to the diets of English workers, and Hopkins and Wallerstein (1986) proposed that commodity chains, followed from raw material to final product, could reveal the structure of the world economy. Histories of cotton by Riello (2013) and Beckert (2014) have followed one fibre across the globe. The multimaterial approach shares their commitment to following materials and adds two things. It places the materials in a longer evolutionary history of relations between living beings and matter, beginning before domestication, and it compares materials by their properties, their value-to-mass ratio, their dependence on living producers, their capacity to be mechanised, in order to explain why each material produced the kind of world-historical trajectory it did.
XI. The World in a Shirt
The ordinary garment with which the article began turns out to be the temporarily stabilised endpoint of an immense historical process. The argument can now be distinguished explicitly from commodity-chain analysis. The claim goes beyond the observation that every commodity has a geographically dispersed supply chain. The stronger claim is genealogical. Each material entering clothing carries a different evolutionary and historical trajectory of relations between living beings and matter, and the garment brings those trajectories together. A supply chain describes where things come from now. A multimaterial hinterland describes what had to become possible, over thousands of years and across many species, before anything like this garment could be made.
Human beings did not simply learn how to make clothes. In learning to clothe themselves in ever greater quantities and varieties, they progressively reorganised parts of the world around the production of clothing materials. Silkworms and mulberry trees were incorporated into specialised systems of cultivation. Sheep were bred toward desired fleeces. Beaver populations became entangled with continental trading networks and colonial territorial expansion. Cotton connected plantation slavery to mechanised mills and industrial cities. Synthetic fibres connected the clothed body to petroleum and chemical engineering. None of these layers simply replaced the previous ones. They accumulated. The history of clothing is therefore written across landscapes, species, oceans, factories and empires as well as on bodies. To follow the materials from which clothes are made is to discover that an apparently intimate technology of bodily covering has repeatedly participated in remaking the world.
Selected References
Beckert, Sven. 2014. Empire of Cotton: A Global History. New York: Alfred A. Knopf.
Browne, Mark Anthony, Phillip Crump, Stewart J. Niven, et al. 2011. "Accumulation of Microplastic on Shorelines Worldwide: Sources and Sinks." Environmental Science and Technology 45 (21): 9175–9179.
Engels, Friedrich. 1845. Die Lage der arbeitenden Klasse in England. Leipzig: Otto Wigand.
Godfrey-Smith, Peter. 2024. Living on Earth: Life, Consciousness and the Making of the Natural World. London: William Collins.
Hopkins, Terence K., and Immanuel Wallerstein. 1986. "Commodity Chains in the World-Economy Prior to 1800." Review (Fernand Braudel Center) 10 (1): 157–170.
Innis, Harold A. 1930. The Fur Trade in Canada: An Introduction to Canadian Economic History. New Haven: Yale University Press.
Kvavadze, Eliso, Ofer Bar-Yosef, Anna Belfer-Cohen, et al. 2009. "30,000-Year-Old Wild Flax Fibers." Science 325 (5946): 1359.
Mintz, Sidney W. 1985. Sweetness and Power: The Place of Sugar in Modern History. New York: Viking.
More, Thomas. 1516. Utopia. Leuven: Dirk Martens.
Murra, John V. 1962. "Cloth and Its Functions in the Inca State." American Anthropologist 64 (4): 710–728.
Procopius. 1928. History of the Wars, Books VII.36–VIII. Translated by H. B. Dewing. Loeb Classical Library. Cambridge, MA: Harvard University Press.
Riello, Giorgio. 2013. Cotton: The Fabric That Made the Modern World. Cambridge: Cambridge University Press.
Sherratt, Andrew. 1981. "Plough and Pastoralism: Aspects of the Secondary Products Revolution." In Pattern of the Past: Studies in Honour of David Clarke, edited by Ian Hodder, Glynn Isaac, and Norman Hammond, 261–305. Cambridge: Cambridge University Press.
Thompson, E. P. 1967. "Time, Work-Discipline, and Industrial Capitalism." Past and Present 38: 56–97.
Van Kirk, Sylvia. 1980. Many Tender Ties: Women in Fur-Trade Society, 1670–1870. Winnipeg: Watson and Dwyer.
Wright, Larry. 1976. Teleological Explanations: An Etiological Analysis of Goals and Functions. Berkeley: University of California Press.