Empire, slavery & unequal exchange

Thesis

The Industrial Revolution was not a European miracle that happened to coincide with Europe’s four-continent empire. The empire — and specifically the Atlantic slave-sugar-cotton complex — was a constitutive part of the IR. Enslaved African labor grew the cotton that fed the Lancashire mills. Caribbean sugar plantations generated the profits that financed the shipping, insurance, banking, and industrial investment of Bristol, Liverpool, London, and Glasgow. Colonial markets absorbed the output of the factories. Extracted Indian and American raw materials substituted for scarce European ones. Without the Atlantic system, the capital would have been smaller, the demand would have been smaller, the raw-material supply would have been inelastic, and the IR would have been delayed or been a much narrower phenomenon.

The thesis comes in several strengths. Eric Williams’s 1944 Capitalism and Slavery made the strongest form: slave-trade profits were a major source of British industrial capital. Mid-20th-century quantitative economic history (Engerman, Solow, O’Brien) argued Williams had vastly overstated the numerical contribution. A 21st-century revival — Inikori, Beckert, Berg — has resuscitated the thesis in a broader form, shifting the emphasis from slave-trade profits to the full ecosystem of demand, shipping, finance, and raw-material flows that the Atlantic complex generated.

The Great-Divergence-scale extension of this argument — which is more robust than the IR-scale version because empire varies between Europe and Asia — is the empire-and-coerced-extraction position.

Lead proponents

• Eric Williams — Capitalism and Slavery (1944), originally an Oxford DPhil thesis. The classic statement and the benchmark the later literature argues against and with.
• Joseph Inikori — Africans and the Industrial Revolution in England (2002). The central modern revivalist argument: measures the full contribution of the Atlantic system (shipping, finance, exports, demand effects, processing industries) to British industrial development, not just slave-trade profits per se.
• Sven Beckert — Empire of Cotton (2014). A global history of cotton capitalism from the slave plantation to the Lancashire mill. Beckert’s organizing concept is war capitalism — the regime of slavery, expropriation, and state-backed violence that preceded and underwrote the later “industrial capitalism” of factory and wage labour. The claim is not that the two were sequential but that the industrial system rested permanently on the coerced one: as late as 1860 the Lancashire mills ran on slave-grown fibre, and the fibre supply collapsed when the slave labour did, in the “cotton famine” of 1861–65.
• Maxine Berg — has tied British industrial development to the colonial complex from the demand and product-design side. In Luxury and Pleasure in Eighteenth-Century Britain (2005) and a series of papers, Berg argues that Asian and colonial goods — Indian calicoes, Chinese porcelain, West Indian sugar — created the consumer appetites and the import-substituting product imitations (Wedgwood’s china, the printed-cotton industry) that drove early British manufacturing innovation. The mechanism is taste and product competition, not capital transfer.
• Pat Hudson — with Berg and in her own regional work, has connected the Atlantic and Asian trades to British industrial finance and to the West Riding and Lancashire production complexes, insisting that the older national-accounting tests measured the wrong quantity (slave-trade profit shares) and missed the systemic integration.
• Kenneth Pomeranz, Ronald Findlay & Kevin O’Rourke — not partisans of the slavery thesis as such, but suppliers of its modern quantitative scaffolding. Pomeranz’s ghost-acres accounting (in The Great Divergence, 2000) estimates that the sugar, cotton, and timber imported from the New World by ~1830 substituted for tens of millions of acres of British land that the home island did not possess; Findlay & O’Rourke (Power and Plenty, 2007) reconstruct the scale of Atlantic trade flows and frame the slave-plantation complex as a central node of the early-modern world economy.

The cotton complex

The empirically tightest version of the thesis runs entirely through cotton, because cotton was both the IR’s pace-setting sector and an unambiguously slave-grown crop. The raw-material magnitudes are stark. British raw-cotton imports rose from roughly 2.5 million lb in 1760 to about 56 million lb in 1800 to roughly 592 million lb by 1840 — a more-than-200-fold increase across the heart of the IR, almost entirely absorbing the rising output of the American slave South. The US share of British cotton supply climbed from negligible in the 1790s (when the West Indies and Brazil dominated) to roughly three-quarters by 1860; the enabling technology was Eli Whitney’s gin (1793), which made short-staple upland cotton — the crop the lower South could grow on newly cleared Indian land worked by an enslaved labour force that doubled from ~700,000 in 1790 to nearly 4 million by 1860 — commercially viable.

Raw cotton imports into Britain (million lb, log scale), c. 1700–1860. Hover for year-by-year values. Sources: Mitchell, British Historical Statistics; Deane & Cole.

Data table

Year	Imports (million lb)
1700	1.99
1710	0.72
1720	1.97
1730	1.55
1740	1.65
1750	2.98
1760	3.87
1770	4.76
1780	6.77
1790	31.45
1800	56.01
1810	132.49
1820	151.67
1830	263.96
1840	592.49
1850	663.68
1860	1,390.94

Beckert’s contribution is to render this as a single integrated system rather than a supply curve. The Mississippi planter, the New Orleans cotton factor who advanced him credit against the next crop, the Liverpool broker who financed the Atlantic leg on London acceptances, and the Manchester spinner who bought the fibre on the same credit chain were nodes of one circuit of capital. The system’s dependence on coercion was demonstrated, not theorized, in 1861: when the Union blockade cut off Southern cotton, Lancashire’s mills went short, ~500,000 operatives were thrown into distress, and the British cotton industry scrambled — only partially successfully — for Indian, Egyptian, and Brazilian substitutes. The “cotton famine” is the natural experiment the thesis points to: remove the slave-grown input and the lead sector seizes.

This is the part of the empire thesis that even sceptics have trouble dismissing. The dispute is less about whether slave-grown cotton was load-bearing for the cotton industry — it plainly was — than about whether the cotton industry’s specific dependence translates into a claim about the IR as a whole, given cotton’s modest share of aggregate GDP and employment (see the rehabilitation debate about how to weight a small but explosively growing sector).

Key arguments

1. Cotton was the IR’s lead sector, and it was a slave-grown crop. By 1860, ~75% of Britain’s cotton imports came from the US South, grown by enslaved labour. Raw imports rose from ~2.5M lb in 1760 to ~592M lb in 1840. The Lancashire mill complex could not have existed at scale on European or Indian cotton alone.
2. The Atlantic generated export markets and capital flows. West Indian plantations bought British manufactured goods in quantity; profits were remitted to Britain and invested (partly) in industrial and financial infrastructure. Modern estimates of the plantation complex’s size (Findlay & O’Rourke, Inikori) are much larger than the old slave-trade-only calculations.
3. Shipping, insurance, and finance were co-constitutive. The Atlantic trade built the maritime and financial capacity (Lloyds of London, marine insurance, bills of exchange, triangular-trade shipping infrastructure) that underpinned both imperial commerce and domestic industry. These are hard to disentangle from the “industrial” economy.
4. India fits the same pattern. Indian cotton textiles were the world’s dominant manufacture into the 18th century, and the early British cotton industry began as an import-substituting imitation of Indian calicoes (so threatening that the Calico Acts of 1700 and 1721 banned their import to protect home producers). After the East India Company’s conquest of Bengal (Plassey 1757, diwani 1765), the flow reversed: Bengal’s weaving export trade collapsed, British machine-spun yarn and cloth captured first the Indian and then the world market, and India shifted from manufactured-textile exporter to raw-cotton and raw-material supplier and captive market for Lancashire cloth. The reciprocal of British industrial rise shows in the Maddison series cited in the debate index — Indian per-capita GDP fell across the colonial 19th century. Utsa Patnaik and the older “drain theory” (Dadabhai Naoroji, R. C. Dutt) quantify the net transfer from India to Britain; the magnitudes are contested but the directional pattern of deindustrialization is not.
5. Ghost acres, revisited. Pomeranz’s coal-and-geography story already includes the New World in the explanation — but he frames the ghost acres as land, not labor. The empire-thesis makes coerced labor explicit: ghost acres only mattered because enslaved labor worked them.

The demand and consumption channel

The most distinctively modern limb of the revival shifts attention from capital supplied by empire to demand created by it — the channel Inikori and Berg emphasize precisely because it sidesteps the Engerman profit-share objection. Two mechanisms operate. The first is colonial export markets: the protected markets of the West Indies, North America, and (later) India absorbed a large and rising share of British manufactured exports across the 18th century, and the goods they bought — coarse woollens, metalwares, guns, and increasingly cottons — were disproportionately the products of the dynamic, mechanizing sectors. A plantation economy of enslaved people who had to be clothed and equipped, and of planters consuming British finished goods, was a captive demand sink that pulled British manufacturing output upward. The second is the consumption revolution at home: Berg’s work argues that the inflow of colonial groceries (sugar, tobacco, tea, coffee) and Asian manufactures (calicoes, porcelain) reshaped British consumer demand and provoked the import-substituting product innovations — printed cottons imitating Indian calico, Wedgwood’s creamware imitating Chinese porcelain — that were among the IR’s commercial drivers. Sugar consumption per head rose roughly twenty-fold across the 18th century; the sweetened-tea complex that sugar and tea built was the mass-consumption habit around which much of the Atlantic trade turned.

The demand argument is harder to dismiss with a profit-share calculation but also harder to pin down causally: export demand and domestic consumer demand were real and empire-linked, but whether they were necessary (rather than substitutable by home or European markets) is exactly what the counterfactual debate cannot settle.

The Engerman accounting and the Inikori response

The decisive blow to the strong Williams thesis was an accounting exercise, and the modern revival is an argument about what that accounting should count. Stanley Engerman’s 1972 Business History Review comment took Williams’s own claim — that slave-trade profits were a major source of British industrial capital — and gave it the most generous numerical reading he could. He estimated slave-trade profits at perhaps £300,000–£500,000 a year in the peak late-18th-century period, against a British national income of order £100 million and gross investment of order £10 million. On these figures slave-trade profits were on the order of 1% or less of national income and could have financed only a small fraction of total investment even if every penny were reinvested in industry — which it manifestly was not. Robert Paul Thomas, Barbara Solow, and others extended the exercise to the whole West Indian colonial sector and reached compatible conclusions: the plantation economy was profitable but not large enough, relative to a £100-million economy, to have been the pump of industrialization. By the 1980s the strong Williams thesis was, within economic history, dead.

Inikori’s revival concedes the Engerman arithmetic and rejects the question. Slave-trade profits narrowly defined were indeed small; but the relevant magnitude, Inikori argues, is the whole Atlantic system’s contribution to British growth — the export demand that pulled the manufacturing sectors, the shipping and shipbuilding the trade required, the marine insurance and financial services it generated, the raw materials it supplied, and the processing industries (sugar refining, tobacco, rum, cotton spinning) that grew up around it. By Inikori’s accounting, trade with the Atlantic basin accounted for a large and rising share of British manufactured exports through the 18th century, and the export sector was where the most dynamic, most mechanizing industries sat. The dispute is therefore not about the data but about the unit of analysis: Engerman measures a profit stream against national income; Inikori measures a demand-and-services system against the growing, tradable part of the economy. Findlay & O’Rourke’s reconstruction of Atlantic trade volumes is broadly congenial to Inikori’s framing; the national-accounting tradition (O’Brien’s “periphery was peripheral” argument) is not.

Key evidence

• Williams’s original accounting — slave-trade and plantation profit shares in British capital formation. The numbers were contested and substantially revised downward in the 1970s–80s Engerman/Solow tradition.
• Inikori’s broader accounting — extending to shipping, processing, and intra-empire trade, rather than just slave-trade profits. Produces numbers several times larger than Engerman’s.
• Cotton supply-chain reconstruction (Beckert) — documenting the physical and financial integration of the Mississippi Delta plantation, New Orleans factor, Liverpool broker, and Lancashire mill as a single system.
• Counterfactual exercises — what if cotton had to be grown by wage labor at world wages? Estimates vary wildly, but the higher estimates suggest British cotton output would have been a fraction of observed levels, and industrialization would have taken a substantially different shape.

Major critiques

— ENGERMAN: the slave-trade was too small to be the engine. Stanley Engerman’s upper-bound accounting in the Business History Review (1972) put British slave-trade profits at perhaps 1% of national income in the peak late-18th-century years. Even generously construed and fully reinvested, these profits cannot plausibly have been the main source of industrial investment. Williams’s strong numerical claim is largely conceded to be wrong; the modern Inikori/Beckert revival accepts this and responds by widening the scope of accounting — which Engerman’s successors regard as moving the goalposts.

— O’BRIEN: the periphery was peripheral. Patrick O’Brien’s influential 1982 Economic History Review paper (“European Economic Development: The Contribution of the Periphery”) argued that the entire trade with Asia, Africa, and the Americas was too small a share of European economic activity to have driven the transformation — gross profits from the periphery were on the order of a few percent of European GDP. The IR, on this reading and on that of the coal/geography and institutions schools, was a transformation of productivity through coal, machinery, and institutions, not of capital scale. Capital was not the scarce factor — British interest rates and savings behaviour suggest investment was demand- and opportunity-limited, not savings-limited. (O’Brien has since partly softened this in light of the cotton and fiscal-military arguments.)

— Counterfactual comparators: Belgium, the Netherlands, and parts of Germany industrialized with minimal or no Atlantic slave empire, while Spain and Portugal held enormous Atlantic empires for centuries and did not industrialize. The cross-country correlation between empire and industrialization is far messier than the strong thesis implies, which pushes the burden onto specifically British mechanisms.

— Selection on the outcome: Britain industrialized and had an empire, so the empire must have mattered — a classic over-reading of a one-case correlation. The revivalist literature tries to discipline this with explicit counterfactuals (what would British cotton output have been at world wage rates for the fibre?), but the estimates vary so widely as to be only suggestive.

— Disaggregation: “the empire” is not one thing. The slave-plantation complex, the East India Company conquest, free-trade informal empire, and colonial settler economies operated on different logics and timescales; lumping them together makes clean tests difficult and lets the strongest sub-case (cotton) carry rhetorical weight for the weaker ones.

Status

Contested, in live revival. The strong Williams claim (slave-trade profits funded industrialization) is generally rejected. The broader claim (the Atlantic complex was a significant and underappreciated component of British industrial growth, and cotton capitalism cannot be understood without slavery) is actively contested and is currently gaining ground in the scholarly literature, particularly in global and imperial history as distinct from narrow economic history. The disagreement between the two approaches is partly methodological (national-accounting vs. systems-level) and partly political, which keeps the debate heated.