Fossil Fuels and Industry

Foundations

What you are actually burning

Take a lump of coal. About 300 million years ago, in a hot swamp forest, a plant used sunlight to pull carbon dioxide out of the air and build itself. The plant fell into water, did not fully rot, was buried, compressed, and heated for a very long time.

The energy released when you burn it is that sunlight. Not a metaphor for it. The actual photons, captured by photosynthesis, stored as chemical bonds, sitting underground since before the dinosaurs.

Oil and gas are the same story with different organisms, mostly marine plankton, and different burial conditions.

So the sentence that opens this lesson is literal. Industrial civilisation runs on ancient sunlight.

The ceiling that had always existed

Here is the structural point, and it is the one worth carrying out of this lesson.

Every society before about 1750 lived on a flow. Sunlight arrives at a fixed rate. It grows crops, grows trees, drives the wind and the water cycle. Muscle power, human and animal, is stored sunlight from this year’s food. Firewood is stored sunlight from the last few decades.

You could use that flow more cleverly — watermills, sailing ships, better ploughs — but you could not exceed it. Wanting more energy meant wanting more land, and land was finite and already occupied. That is a hard ceiling, and every agrarian society in lesson 7.6 pressed against it.

Fossil fuels are a stock. Three hundred million years of accumulated flow, sitting there, and available all at once to anyone who could dig.

Switching from a flow to a stock removes the ceiling. That is what happened, and almost everything else in the modern world follows from it.

Why it started where it did

Now the hard question, and the one where the popular answer is worst.

The popular answer is that Britain was more inventive. This does not survive contact with the evidence: China had invented printing, gunpowder, the compass and cast iron, and India was the world’s largest manufacturer of cotton textiles. Inventiveness was not the scarce input.

Several factors are documented, and their relative weight is genuinely disputed among economic historians. The honest version is to lay them out.

Coal, where it was useful. Britain had coal at shallow depths, near navigable water, close to where people lived. China had enormous coal reserves too, but concentrated in the north-west, far from the economic centre in the Yangzi delta. Transport was the binding constraint. Having the resource is not the same as having it cheaply.

The self-fuelling loop. Thomas Newcomen’s engine of 1712 was built to pump water out of coal mines. It was hopelessly inefficient, which did not matter, because at a coal mine the fuel was effectively free. Coal financed the machine that got more coal. Every improvement afterwards had a market waiting.

Wages and prices. Robert Allen’s argument is that British wages were high and British energy cheap, so a machine replacing workers paid for itself in Britain and did not pay for itself in Delhi or Suzhou, where labour was cheaper. On this account the machines were built where they were profitable, not where people were smart.

This is an elegant argument and it is contested. The wage comparisons rest on assumptions about what a consumption basket contained in each place, and reasonable changes to those assumptions move the numbers.

Colonial resources and markets. Kenneth Pomeranz argues that the advanced regions of Europe and China were broadly comparable in the eighteenth century, and that two things broke the symmetry: coal, and the New World. American land supplied cotton, sugar and timber that Britain would otherwise have had to grow on land it did not have — what he calls ghost acres. And colonies supplied markets that could be closed to competitors.

Institutions, finance, knowledge networks. Patent systems, joint-stock companies, insurance, and the publishing culture of lesson 8.2. All real, all hard to weigh.

The situation is that we have a list of documented factors and no agreed weighting. Anyone who gives you one clean cause is selling something, and usually a politics.

What it did to India

This is where the story stops being about machines.

In the middle of the eighteenth century, India was the largest producer of cotton textiles in the world. Bengal muslin, Gujarati and Coromandel cloth, dyed with techniques Europe could not match. It was exported across Asia, Africa and Europe, and English producers lobbied for tariffs against it precisely because it was better and cheaper.

By 1900 that industry was gone, and India was importing cloth from Manchester made partly from Indian cotton.

The careful economic history is more interesting than either the crude nationalist or the crude apologist version, and it comes in two phases.

Phase one, before British factories mattered. Clingingsmith and Williamson find that Indian deindustrialisation begins in the eighteenth century, before Lancashire could out-compete anyone. Their mechanism: the collapse of Mughal authority and the disorder that followed disrupted agriculture, pushed grain prices up, and so pushed up the nominal wages Indian weavers had to be paid. Indian cloth got more expensive to make for reasons internal to India.

Phase two, after about 1810. Now British factory production, with fossil-fuelled machinery, undercuts Indian cloth on price in Indian markets themselves. And Indian producers face this without the tariff protection that an independent government would have used — the trade policy was set in London, and it was not set in India’s interest.

So the collapse has an internal component and a colonial component, and the colonial component is real and documented without needing to be exaggerated.

This matters for how the whole chapter is read. The Industrial Revolution was not a rising tide. For a large part of the world it was a period of absolute decline in manufacturing capacity, and the divergence in the second half of the nineteenth century is partly a story of one region industrialising and another being deindustrialised.

The bill nobody was reading

One last thing, and it sets up the next two lessons.

Carbon that took the Carboniferous hundreds of millions of years to remove from the atmosphere and bury is being returned to the atmosphere over about two hundred years.

That is a rate difference of roughly a factor of a million.

In 1800 nobody knew this mattered. The physics that says it matters — that carbon dioxide absorbs infrared radiation — was worked out during the nineteenth century, and the connection to global temperature was made explicit by the century’s end. But the machinery was already built, the cities were already growing, and the ceiling was already gone.

Lesson 4.7 described the Earth as a coupled system in which the carbon cycle regulates temperature over geological time. This lesson is about the moment one species started running that cycle backwards at speed.

Deeper

Why the Carboniferous made so much coal

Two conditions coincided. Extensive tropical wetland forests, dominated by lycopsids with high lignin content, grew in basins that were subsiding, so plant material was repeatedly buried below water where decay was slow. The once-popular idea that coal accumulated because fungi capable of degrading lignin had not yet evolved has been substantially weakened by molecular clock and palaeobotanical work; the tectonic and climatic explanation, in which subsiding equatorial basins and a wet climate did the work, carries more of the load than it once did. Either way, roughly 90 per cent of the world’s coal-bearing strata date from the Carboniferous and Permian.

Stock versus flow, quantified

Wrigley’s argument is that the English economy before 1800 was an organic economy whose energy came from the annual photosynthetic flow, and that the constraint was land. Producing enough charcoal to smelt iron at nineteenth-century volumes would have required forest area on a scale England could not spare from food production. Coke smelting broke that link. This is why he treats the transition as a change in kind rather than degree: an organic economy has a computable maximum, and a mineral-energy economy does not have one in the same sense.

How the wage comparison works, and where it wobbles

Allen constructs real wages by pricing a fixed basket of goods in silver in each city and dividing the nominal wage by its cost. The difficulty is that a basket appropriate to London — bread, beer, meat, coal, cloth — is not what a Delhi labourer bought, and substituting a locally appropriate basket changes the ratio. Broadberry and Gupta approach the comparison from a different direction, using grain wages and silver wages, and reach a picture in which advanced Asian regions had high productivity in some sectors but lower silver wages. The two literatures agree that the divergence was substantial by 1800; they disagree about how much of it existed in 1700, and the disagreement is largely methodological.

What the Indian data can and cannot support

Roy’s caution is worth repeating: colonial-era Indian statistics were collected for revenue and administration, not for measuring living standards, and the coverage is uneven across regions and periods. Aggregate national estimates for pre-1870 India carry wide uncertainty. This means quantitative claims about Indian decline should be stated as ranges, and that arguments resting on precise figures for eighteenth-century Indian GDP are weaker than they look. The qualitative fact of textile-sector collapse is not in doubt; its exact magnitude is.

Frontier

Weighting the causes of divergence

The debate is active and quantitative work continues, using new wage series, height data as a proxy for nutrition, and reconstructed national accounts. It is unlikely to produce a single answer, partly because the factors interact: coal made machines worth building, machines made colonial markets more valuable, colonial markets financed further mechanisation.

Counterfactual India

Whether India would have industrialised under an independent government with tariff autonomy cannot be settled, because the counterfactual has no data. What can be examined is the record of the places that retained policy autonomy, and that comparison is an active area of research rather than a closed case in either direction.

Transcript and shot list

Target runtime about 9 minutes. Built as follow the energy: where it came from, why it was found here first, who paid for it, what it did to the air. Shot types: ANIM is rendered from code with Manim, AI is a generated shot from Higgsfield with the shared style reference, REAL is agency imagery with credit, CAM is the narrator.

[00:00] S01 · AI · A Carboniferous swamp forest, giant lycopsids, humid, no animals larger than an insect in frame. Slow push in.

Three hundred million years ago, a plant in a hot swamp used sunlight to pull carbon dioxide out of the air and build itself. Then it fell into water, did not fully rot, and got buried.

[00:20] S02 · ANIM · The plant compressing through burial into a coal seam; then a hand holding a lump of coal; then the stored photons releasing as it burns.

That is coal. And the energy that comes out when you burn it is that sunlight. Not a metaphor for sunlight. The actual solar energy, captured by photosynthesis, locked into chemical bonds, sitting underground since before the dinosaurs existed. Oil and gas are the same story with different organisms, mostly marine plankton. So this is literal: industrial civilisation runs on ancient sunlight.

[00:55] S03 · ANIM · A flow diagram: sunlight arriving at a fixed annual rate, driving crops, trees, wind, water, muscle. A hard ceiling drawn across the top, labelled LAND.

Here is the structural point, and it is the one to carry out of this lesson. Every society before about 1750 lived on a flow. Sunlight arrives at a fixed rate. It grows crops, grows trees, drives the wind. Muscle power is stored sunlight from this year’s food. Firewood is stored sunlight from the last few decades. You could use the flow more cleverly, with watermills and sailing ships and better ploughs, but you could not exceed it. Wanting more energy meant wanting more land, and land was finite and already occupied.

[01:35] S04 · ANIM · The ceiling shattering as a stock appears below ground: 300 million years of accumulated flow, all available at once.

Fossil fuels are not a flow. They are a stock. Three hundred million years of accumulated sunlight, sitting there, available all at once to anyone who could dig. Switch from a flow to a stock and the ceiling disappears. Almost everything else in the modern world follows from that one sentence.

[02:05] S05 · ANIM · Newcomen’s 1712 engine pumping water from a coal mine; free fuel at the pithead; the loop closing as more coal comes up.

So why did it start where it did? Start with a machine. Thomas Newcomen’s engine, 1712, built to pump water out of coal mines. It was hopelessly inefficient, which did not matter at all, because at the pithead the fuel was free. Coal financed the machine that got more coal. Every improvement afterwards had a market already waiting for it.

[02:35] S06 · ANIM · Britain’s coal at shallow depth near navigable water; then China’s coal in the north-west, far from the Yangzi delta, with the transport cost drawn as distance.

But that only works if the coal is where you need it. Britain had coal at shallow depths near navigable water, close to where people lived. China had enormous reserves too, and they were in the north-west, a long way from the economic centre in the Yangzi delta. Transport was the binding constraint. Having a resource is not the same as having it cheaply.

[03:05] S07 · ANIM · Allen’s argument: high British wages, cheap British energy, a machine paying for itself in Britain and not in Delhi or Suzhou. Then the consumption-basket assumption highlighted and marked contested.

Then wages. Robert Allen argues that British wages were high and British energy was cheap, so a machine that replaced workers paid for itself in Britain and did not pay for itself in Delhi or Suzhou, where labour was cheaper. On that account the machines got built where they were profitable, not where people were clever. It is an elegant argument and it is contested, because those wage comparisons depend on what you assume a person’s shopping basket contained in each place, and reasonable changes to that assumption move the numbers.

[03:45] S08 · ANIM · Pomeranz’s ghost acres: American land supplying cotton, sugar and timber, drawn as extra territory attached to Britain; colonial markets shown as closable to competitors.

Kenneth Pomeranz adds the other half. He argues the advanced regions of Europe and China were broadly comparable in the eighteenth century, and two things broke the symmetry: coal, and the New World. American land supplied cotton, sugar and timber that Britain would otherwise have had to grow on land it did not have. He calls it ghost acres. And colonies supplied markets that could be closed to competitors.

[04:15] S09 · CAM · Narrator direct to camera.

So where does that leave us? With a list of documented factors and no agreed weighting between them. Coal geography, wages, colonial resources, institutions, finance. All real. Anyone who hands you one clean cause for the Industrial Revolution is selling you something, and usually a politics.

[04:40] S10 · REAL into ANIM · Bengal muslin and Coromandel painted cotton (museum imagery; confirm credit). Then trade routes carrying Indian cloth across Asia, Africa and Europe, and English producers lobbying for tariffs against it.

Now the part that stops being about machines. In the middle of the eighteenth century, India was the largest producer of cotton textiles in the world. Bengal muslin, Gujarati and Coromandel cloth, dyed with techniques Europe could not match. Exported across Asia, Africa and Europe. English producers lobbied their own government for tariffs against Indian cloth, precisely because it was better and cheaper. By 1900 that industry was gone, and India was importing cloth from Manchester, made partly from Indian cotton.

[05:25] S11 · ANIM · Phase one: Mughal authority collapsing, agricultural disruption, grain prices rising, weavers’ nominal wages rising, Indian cloth becoming more expensive to make — all dated before Lancashire mattered.

The careful economic history is more interesting than either the crude nationalist version or the crude apologist version, and it comes in two phases. Phase one starts in the eighteenth century, before Lancashire could out-compete anybody. Clingingsmith and Williamson find that the collapse of Mughal authority disrupted agriculture, pushed grain prices up, and so pushed up the wages Indian weavers had to be paid. Indian cloth got more expensive to make for reasons internal to India.

[06:00] S12 · ANIM · Phase two after 1810: British factory cloth undercutting Indian cloth inside India; the tariff asymmetry shown as policy set in London.

Phase two runs from about 1810. Now British factory production, powered by the fossil fuel we have been talking about, undercuts Indian cloth on price inside Indian markets. And Indian producers face that without the tariff protection an independent government would have used, because trade policy was set in London and it was not set in India’s interest. So the collapse has an internal component and a colonial component, and the colonial component is real and documented without anybody needing to exaggerate it.

[06:40] S13 · ANIM · Two curves diverging after 1800: one region’s manufacturing output rising, another’s falling in absolute terms.

Which changes how you read this whole chapter. The Industrial Revolution was not a rising tide. For a large part of the world it was a period of absolute decline in manufacturing capacity. The divergence of the nineteenth century is partly one region industrialising and partly another being deindustrialised, at the same time, connected.

[07:15] S14 · ANIM · Carbon buried over hundreds of millions of years; the arrow reversing and running back to the atmosphere over two hundred years; the ratio stated as roughly a million to one.

And there is a bill nobody was reading. Carbon that took the Carboniferous hundreds of millions of years to pull out of the atmosphere and bury is being put back over about two hundred years. That is a rate difference of something like a factor of a million.

[07:50] S15 · ANIM · Nineteenth-century radiative physics: carbon dioxide absorbing infrared; the connection to surface temperature made by the century’s end; factories already built, cities already growing.

In 1800 nobody knew this mattered. The physics that says it matters, that carbon dioxide absorbs infrared radiation, was worked out across the nineteenth century, and the link to global temperature was explicit by the end of it. By then the machinery was built, the cities were growing, and the ceiling was already gone.

[08:20] S16 · ANIM · Callback to lesson 4.7: the Earth’s carbon cycle as a regulator on geological timescales; then one species running it backwards at speed. Calendar returns.

Lesson 4.7 described the Earth as a coupled system, where the carbon cycle regulates temperature over geological time. This lesson is about the moment one species started running that cycle backwards, fast. Next lesson: what happens when the whole thing accelerates at once.

Shot list

One AI shot, fourteen code-rendered shots, one museum image, one narrator shot. Estimated Higgsfield use: 2 to 4 generations including retries.