A technology that raises output per worker by a factor of hundreds does not stay in one country, and what it did on its way out is the balance sheet of the whole subject.
The previous lessons have taken Britain from a flat wage series to an inspected, drained, partly reformed industrial economy. This last one asks what the change meant for everybody else: who copied it and why they could, whose manufacturing it destroyed, what it ran on, and what it left in the atmosphere. It is also the place to state plainly what the course as a whole has argued, since a reader who has followed the chain should now be able to account for the break.
How the technology left
Britain tried to keep it. Exporting textile machinery was illegal until 1843, and the emigration of skilled artisans was prohibited until 1825, with penalties on anyone who recruited them.
Neither ban worked, because the knowledge was in people and people walk. Samuel Slater, who had served an apprenticeship under a partner of Arkwright's, sailed for New York in 1789 having memorised the machinery, since he could carry no drawing past the customs officers, and built the first successful water-powered spinning mill in the United States at Pawtucket in 1793. Francis Cabot Lowell toured British mills as a gentleman visitor in 1810 to 1812, memorised the power loom, and built an integrated spinning and weaving mill at Waltham in 1814. William Cockerill, an English carpenter, built spinning machinery at Verviers from 1799 and his son founded the ironworks at Seraing in 1817 that made Belgium the first industrial power on the Continent.
The pattern is consistent. The bans slowed diffusion by a decade or two and did not prevent it, because tacit knowledge held in a skilled head is not something a customs house can search for. Every subsequent attempt at technological containment has run into the same problem.
Why some countries could copy it and others could not
The interesting question is not who wanted the machines, since everybody did, but who could use them.
Belgium had coal, iron ore, a dense population and proximity to Britain, and industrialised first. Germany had coal in the Ruhr and Silesia and, after the Zollverein customs union of 1834, a market large enough to be worth building for; its industrialisation ran through railways, banks and chemicals rather than through cotton, and in the first years of the twentieth century German pig iron output passed Britain's. The United States had land, coal, iron and, above all, a chronic scarcity of labour that made labour-saving machinery pay even more obviously than it had in Lancashire. Japan is the case that breaks any argument from geography or culture: after 1868 the Meiji state bought the technology deliberately, built model factories like the Tomioka silk mill of 1872, sent students abroad, hired foreign engineers on enormous salaries and dismissed them once their knowledge had been absorbed.
Alexander Gerschenkron's account of this, published in 1962, is the one that generalises. His argument is that late industrialisers do not repeat the leader's path: the later a country starts, the more the missing pieces have to be supplied by institutions rather than by private accumulation, so Britain industrialised on reinvested profits, Germany on investment banks, and Russia on the state itself. He also observed that late starters skip stages, importing the newest technique rather than working through the sequence, which is why Japan built its first mills around ring spinning while Lancashire was still running mules.
The other side of the ledger
For a smaller group of countries, British industrialisation was not an opportunity but a demolition, and India is the case everyone argues about.
Paul Bairoch's estimates of world manufacturing output shares are the standard frame. In 1750 India produced roughly a quarter of the world's manufactures and Britain about 2 per cent. By 1880 Britain was at nearly 23 per cent, its peak, and India at under 3. India, which had clothed much of the world, was importing British cloth by the 1820s and had lost its export trade entirely by mid century.
Example. Take India's share of world manufacturing as 24.5 per cent in 1750 and 2.8 per cent in 1880. If world manufacturing output rose sixfold over that period, what happened to India's output in absolute terms, and what would world output have had to do for India's absolute output to be unchanged?
Set world output at 1 in 1750, so India produces 0.245. In 1880 world output is 6 and India produces . So India's absolute output fell by per cent, not by the 89 per cent that the fall in share suggests. For the absolute level to be unchanged, world output would have had to multiply by . The distinction between a share and a level is the single most abused piece of arithmetic in this whole literature, in both directions: a share can collapse while the level rises, if the world grows fast enough, and a stable share can hide a collapse. It is worth adding that the sixfold figure is itself an estimate with wide error bars, and that Bairoch's shares are reconstructions rather than measurements, so this exercise establishes the shape of the correction rather than a number to quote.
Now you. How much of India's deindustrialisation should be attributed to British tariff policy rather than to the productivity gap?
Answer
Both mattered and the productivity gap did more, though the policy is not a footnote. Britain taxed imported Indian calicoes at rates that reached 70 or 80 per cent while British cloth entered India at a few per cent, and the East India Company's administration did nothing to protect Indian weavers, so the trade was not conducted between equals. But the earlier lesson on cotton gives the harder number: the self-acting mule spun a hundred pounds of yarn in about 135 hours against roughly 50,000 for an Indian hand spinner, a ratio of about 370 to one. No tariff schedule that any government could plausibly have set would have offset a gap of that size, and the collapse of Indian spinning also happened in markets Britain did not control. The most defensible summary is that Indian hand spinning was destroyed by machine productivity, that Indian weaving survived considerably longer and partly recovered, and that British policy determined that India would not build mills of its own until much later, which is the part properly laid at the door of colonial government rather than of technology.
What the mills ran on
There is one input the British cotton industry could not have replaced, and by the 1850s most of it was grown by enslaved people in the American South.
The chronology fits together uncomfortably well. Eli Whitney's gin of 1793 made short-staple upland cotton, which grows across the whole southern interior, cheap to clean, and the crop moved west into Alabama, Mississippi and Louisiana with the people forced to grow it. The enslaved population of the United States rose from 697,681 at the 1790 census to 3,953,760 in 1860, a factor of 5.7, much of that increase moved south and west by an internal slave trade that broke up families as a matter of routine.
Example. United States cotton output was about 1.5 million pounds in 1790 and about 4.5 million bales of roughly 450 pounds in 1860. What growth rate is that, and what does output per enslaved person do?
The 1860 crop is pounds, which is 1,350 times the 1790 figure, a growth rate of per cent a year sustained for seventy years. Per enslaved person, output goes from pounds to pounds, a factor of about 240. Two cautions are essential. The denominators include children, the elderly, and the majority of enslaved people who did not pick cotton at all, so this is a crude ratio for the whole enslaved population rather than a measure of anyone's work. And the rise has several causes competing for credit: the gin, the far better land of the southwest, improved cottonseed varieties documented by Alan Olmstead and Paul Rhode, and, in Edward Baptist's account, systematic escalation of violence to raise picking rates, a claim Olmstead and Rhode have disputed on the evidence of the plantation records.
Now you. Eric Williams argued in 1944 that profits from slavery and the slave trade financed British industrialisation. What is the strongest evidence against the strong version of that claim, and what survives it?
Answer
The strongest evidence against is arithmetic about magnitudes. Estimates of slave trade profits put them at a small fraction of British domestic investment, on the order of a per cent or two, and British industrialisation was financed overwhelmingly out of retained profits and local partnerships, as an earlier lesson noted when observing that the investment rate rose only from about 6 to about 12 per cent of national income. A mechanism that supplies one or two per cent of the funds cannot be the cause of the whole. What survives is the input argument, which is stronger and does not depend on Williams at all: by the 1850s the great majority of the raw cotton entering Britain came from the American South, the industry had no substitute of comparable price or quality, and when the American Civil War cut the supply the Lancashire cotton famine of 1861 to 1865 threw hundreds of thousands out of work. Britain's largest export industry ran on a fibre grown by enslaved people, while the same country had abolished its slave trade in 1807 and slavery in its colonies in 1833. Both facts are true and they belong in the same sentence.
The carbon
The last item on the balance sheet was invisible to everyone at the time and is the one that has outlasted the rest.
Burning coal turns its carbon into carbon dioxide. Take British coal as roughly 70 per cent carbon by mass; then a ton of coal yields 0.7 tons of carbon, and since carbon dioxide has a molecular mass of 44 against carbon's 12, that is tons of carbon dioxide per ton of coal.
Example. Britain produced about 62 million tons of coal in 1850 for a population of about 20.8 million. What was that in carbon dioxide per head, and how does it compare with Britain today, which emits roughly 250 million tons for about 68 million people?
The 1850 output gives million tons of carbon dioxide, which is tons per person. Present-day Britain gives tons per person. So Britain in 1850, with no cars, no aircraft, no electricity and no central heating, emitted about twice as much carbon dioxide per head as Britain does today. The result is not a trick, though it needs two honest qualifications. Some of that coal was exported or converted to coke and gas rather than burned domestically, so the figure overstates British consumption somewhat. And modern Britain imports a great deal of embodied carbon in manufactured goods, so its consumption-based figure is considerably higher than its territorial one. The core finding survives both: an economy running on nothing but coal is an extraordinarily carbon-intensive economy, and Britain's per-head emissions peaked long before living memory.
Now you. Estimate the total carbon dioxide from British coal between 1750 and 1900, taking output as roughly 5, 15, 30, 62, 110 and 225 million tons in 1750, 1800, 1830, 1850, 1870 and 1900. World emissions today run about 37 billion tons a year. What does the comparison suggest?
Answer
Joining the points with straight lines and summing the areas gives roughly 8,840 million tons of coal across the 150 years, which at 2.57 tons of carbon dioxide per ton is about 22.7 billion tons. Against present world emissions of 37 billion tons a year, Britain's entire coal-fired industrial revolution amounts to about seven months of what the world now emits. That comparison is the useful one, and it points the opposite way from the way it first reads. Britain's own contribution to the stock of atmospheric carbon is small, so the historical significance is not the quantity but the template: what Britain did was demonstrate that an economy could be run on a stock of fossil carbon rather than on the annual flow of sunlight falling on its own acreage, and every country that has escaped the Malthusian trap since has done it the same way. The atmospheric concentration of carbon dioxide has gone from about 277 parts per million in 1750 to about 425 today, and essentially all of that increase belongs to the process this course has described, spread across the countries that copied it.
The balance sheet
Set the whole thing out, since a reader who has come this far is entitled to the summary.
What was gained is the end of the trap the course opened with. For four centuries English real wages oscillated without trending, and the model that explained why had no exit. After about 1820 output per head rose and kept rising, at rates that compound into transformation, and the population rose with it, which is the one thing the Malthusian model forbids. Everything now taken for granted about modern life, including the expectation that a child born today will outlive their parents' generation, rests on that break.
What was paid is a list this course has tried not to soften. Two generations of workers produced 46 per cent more and were paid 12 per cent more. The people who moved into the towns lost fifteen years of life expectancy on Farr's tables. Children worked in mills until Parliament stopped them, and the stopping took forty years and a factory inspectorate. Indian spinning was destroyed by a productivity gap of several hundred to one and Indian industrialisation was postponed by policy. The largest British industry ran on cotton grown by enslaved people. And the fuel that made all of it possible has left a bill nobody knew was being run up.
Both halves are true, and holding them together without discounting either is what the subject actually requires. The change was worth having and it was not free, and it was not paid for by the same people who first received it. That is not a paradox: it is the ordinary shape of large historical changes, and it is the reason arguing about the Industrial Revolution has never stopped.
One question is left open, and it is the one a reader should take away. The escape happened once, in one place, under a specific set of conditions: dear labour, cheap coal, a state that would enforce a contract, and a culture of measurement. Every country that has escaped since has done it by imitation, with a working example in front of it. Whether the same escape can be made a third time, out of the carbon economy and into something else, with the same speed and without the same costs, is the practical form of the question this course has been asking about the past.