Book review: ‘Fossil Capital: The Rise of Steam Power and the Roots of Global Warming’, by Andreas Malm.

When I bought this book I thought it was going to be a history of steam power in the textile industry in the eighteenth and nineteenth centuries. I was looking forward to reading about cross‑compound engines, Corliss valve gear, centrifugal governors and indicator diagrams.

When I opened it I was surprised to discover that it is, in fact, a sequel to Karl Marx’s Capital. It attempts to weave (no pun intended) the idea of fossil fuel into the fabric of Marx’s theory of class struggle. It aims to show that capitalism and fossil energy are so intimately connected that they cannot be separated, from which it follows, according to the author, that only the overthrow of capitalism can avert climate change.

The book is full of words and phrases that only Marxists use, such as ‘structural crisis’, ‘surplus value’, ‘historical process’, ‘commodity fetishism’, ‘primitive accumulation’, ‘subsumption of labour’, ‘property relations’ and even ‘bourgeois property relations’. The author appears to have a checklist of ideas from Das Kap and to be ticking them off one by one.

I should have guessed from its title that it would be something like this, in which case I wouldn’t have bought it. If you are tempted to buy it for the same reason as I was you now know everything you need to know about it and I will forgive you if you don’t read any further.

Assuming you do wish to know more I’ll continue. The book’s argument starts from the following premises:

  1. The earliest textile mills were powered by water.

  2. This required them to be located in the countryside far from centres of population.

  3. To attract workers to these sites, mill owners had to pay good wages and also build housing and amenities.

  4. Steam engines enabled capitalists to locate their mills in towns where labour was plentiful and consequently cheaper and where there was no need to build housing or amenities.

  5. A gradual switch from water to steam power occurred in the textile industry in the early 19th century. The proportion of the textile industry's motive power supplied by steam reached 50% by sometime in the 1830s.

  6. The move to steam power happened well before all the available water power sites were occupied, in fact they never were all occupied, implying that it was not a shortage of water power that caused the switch to steam.

  7. Steam power was in most cases more expensive than water power, per unit of energy delivered, meaning that the switch didn’t occur because of the relative costs of the two power sources.

These premises are true, at least in the early stages of the industrial revolution. From them the author draws the following conclusions:

  1. The switch from water power to steam power in the textile industry occurred because it shifted the balance of power away from workers towards capitalists. It was not driven by technological innovation but by ‘social relations’. Steam power was purely and simply a weapon in the class struggle. It existed solely to enable ‘Capital’ to suck more ‘surplus value’ out of ‘Labour’.

  2. If the country had been socialist at that point in history, or at least not capitalist, there would have been no reason to switch from water power to steam power and global warming would have been averted.

  3. If we overthrow capitalism our economy will automatically become low carbon.

Malm claims that the prevailing view among historians is that the switch to steam power in the textile industry occurred either because steam power was cheaper than water power or because the demand for power outstripped the available supply, or both—a view he, though nobody else, calls ‘the Ricardian-Malthusian paradigm’. He seems to think that refuting this view is sufficient to prove his hypothesis.

However, the book ‘Stronger than a hundred men: a history of the vertical water wheel’, by Terry S. Reynolds, published in 1983, devotes a whole chapter to the decline of the water wheel. This concludes that:

In large part it was due to three major deficiencies of water power: (1) irregularity and unreliability; (2) geographical inflexibility; and (3) natural power limitations. …

… Water power was often not close to established transportation routes, markets, raw materials, and labor supply, and usually it was not economically feasible to move water power to the proximity of these essentials. …

Further, as mass production, national markets, and a railroad network (made possible by the mobile steam engine) emerged, power became only one of many determinants of industrial location. Increasingly, proximity to a labor supply, to raw materials, to markets, and, especially to established transportation lines, became more important than power costs. The steam engine gave industry the freedom to locate close to these.

Going a bit further back, Sadi Carnot’s famous ‘Reflections on the motive power of fire and on machines fitted to develop that power’, published in 1824 and still in print today, says:

The heat engine is already at work in the exploitation of our mines, for driving our ships, digging out our ports and rivers, forging steel, fashioning wood, milling grain, spinning and weaving cloth, transporting the heaviest loads, etc. It seems that one day it must become a universal source of power and in this respect supplant animals, water and wind. Over the first of these sources of power the heat engine has the advantage of economy; over the other two, the invaluable advantage that it can be employed and remain in uninterrupted use irrespective of either time or place.

It has, therefore, always been well known, at least in some sections of academia, that, in the early 19th century, steam power’s advantages over water power were were regularity and geographical flexibility, not cost or scarcity of water power. And yet, before now, no one has suggested that its uptake in industry had anything to do with ‘class struggle’.

Not having read the historians Malm attacks I cannot judge whether or not their theories conform to his caricature of them. Assuming for the sake of argument that they do, it still doesn't follow that Malm's explanation is correct. His challenge is not to prove that his premises are true, it is to prove that his conclusions follow from them. He seems to think that there are only two possible explanations of the shift from water to steam power in the textile industry—his view and the ‘prevailing’ view—and that disproving one is sufficient to prove the other. However, as Reynolds points out in the above quote, power supply and labour supply are just two of many factors that influence the choice of industrial location. Others include the availability of raw materials, machinery, transportation and markets. Power is not the only or even the biggest cost so it would be surprising if it were the sole determinant of location.

Malm effectively acknowledges this in a section entitled ‘Steam and Agglomeration Economies’ in Chapter 7, where he discusses the large number of factors that tend to cause clustering of industries in urban centres. The water wheel, by tying a factory to a particular location, counteracts the pull of these urban agglomerations. When the steam engine appeared on the scene it cut that tie and the pull could no longer be resisted. The decision to locate in a town is based on a large number of factors and has nothing to do with class struggle.

Malm’s argument would have more force if he had shown that the total cost per unit output of an urban steam powered textile mill was higher than that of a rural water powered one but the book doesn’t attempt to do that. A capitalist’s decision to build a new factory in a particular location is driven by overall profitability, which depends on total cost, not the cost of one particular factor of production considered in isolation.

An important part of Malm’s argument is that, in addition to being cheaper, workers in urban steam powered textile mills were more thoroughly subjugated, compliant and easily exploited (or, to use Marxist terminology, ‘subsumed’) than those in rural mills. He makes much use of a phrase from an 1833 article in the Edinburgh Review by the Scottish Economist J. R. McCulloch, which says:

… the invention of the steam engine has relieved us from the necessity of building factories in inconvenient situations merely for the sake of a waterfall. It has allowed them to be placed in the centre of a population trained to industrious habits. (Malm’s emphasis)

Aha!, you can almost hear him saying: ‘Here is an admission from the bourgeois horse’s mouth that my thesis is correct!’. He then uses the phrase ‘trained to industrious habits’ over and over again throughout the book to emphasise the claim that workers in mill towns were easier to control than those in rural water powered mills.

A single phrase doesn’t constitute evidence. In an attempt to provide supporting evidence for this idea Malm discusses various strikes, riots, machine smashing and other disturbances. Although I have not rigorously counted all the references to such disturbances in the book, it seems to me that overwhelmingly more of them are in towns than in rural mills. The most detailed discussion of anti capitalist violence is in Chapter 10, “‘Go and Stop the Smoke!’: The Moment of Resistance against Steam”, which is devoted entirely to violence against steam powered mills. This doesn’t suggest that urban workers were any less feisty or more downtrodden than rural ones. Anyway, use of language changes over time and in 1833 ‘industrious habits’ may just have meant what today would be called ‘industry experience’.

He may be arguing that workers were ‘resisting’ steam because they saw it as a threat. This means that whichever set of facts is correct Malm's hypothesis would be proved. If it turns out that there were more disturbances in rural than in urban mills it proves that urban workers were more easily controlled. If the opposite turns out to be the case it demonstrates that ‘Labour’ recognised that steam gave ‘Capital’ an advantage in the class struggle and tried to resist it.

The fact that mill owners bought steam engines in order to locate in towns where, among other things, labour was cheap and plentiful doesn’t imply that workers in towns were easier to control. It just implies that they were easier to find, as were many other factors of production.

In Chapter 8 the author argues that the imposition of the various Factory Acts starting in 1833 hastened the demise of water power. If the overall thesis of the book were correct you would expect that factors tending to shift the balance of power back towards workers would have the opposite effect. This also applies to his arguments relating to the introduction of legislation to limit the length of the working day. Another example, by the way, of ticking off topics from The Holy Book. The evidence presented in this chapter actually shows that to operate a water powered textile mill profitably a capitalist needed to treat his workers more coercively than he would have needed to do in a steam powered mill.

It is pertinent at this point to ask what it means to say that the choice of steam was ‘motivated by power over labour’? Such a statement only makes sense when viewed through the distorting filter of Marxian theory, which interprets the whole of human history as a titanic battle between ‘Capital’ and ‘Labour’. To someone who doesn't view the world in these terms—which is most people—the idea isn't even meaningful.

Also like the original Capital this book focuses almost exclusively on the Lancashire cotton industry, with occasional excursions to Scotland. Other sectors of the economy are ignored. It has been remarked that if Friedrich Engels had lived in Birmingham instead of Manchester then Das Kap would have turned out very differently and the same could be said of this book. Sectors such as steel, non-ferrous metals, transport, mining, ceramics, cement, basic chemicals and dozens of smaller industries, as well as the use of coal for heating were, and are, also important. Though data are hard to find, I would be very surprised if the textile industry ever accounted for more than ten percent of UK coal consumption at any one time. Most of these other industries, at least until the late 20th century, were technologically unsuited to renewables. Even if textiles had continued to use water power these other industries would not have been able to develop without fossil energy—you couldn't have a water powered inter-city train, for example, or use water to reduce iron ore in a blast furnace—and would still have caused massive growth in coal consumption. Undoubtedly human ingenuity would have found alternative solutions, as it is now being required to do, but it would certainly have been more difficult and would have occurred later.

At some point in the 19th Century, improvements in the cost and efficiency of steam engines did result in steam power becoming cheaper than water power and the energy requirements of industry did outgrow the available hydropower resource. Malm says this is irrelevant and that it is only the initial cause of the switch that counts. In Chapter 12: ‘The myth of the human enterprise’, after acknowledging that:

… the Ricardian-Malthusians might object that the astounding growth of the global economy and population witnessed over the past two centuries would not have been possible without fossil fuels – and in this they are, of course, correct.

he then says:

But no functionalism can explain why an institution came into existence in the first place, since that would violate ‘one of the few simple and self-evident rules of causality: if event E happened before event C, it cannot have happened because of it.’ The original transition to fossil fuels cannot possibly have been caused by the subsequent function of opening up a subterranean continent twenty times the size of Europe. Saying so would be to indulge in teleology.
But surely the watercourses available on the British Isles could not have powered all their industries in, say, 1979? Granted, but any such hypothesised later scarcity cannot, again, explain the turn to steam, any more than the causes of the Second World War can be found after 1945.

If the purpose of the book were limited to showing that the initial uptake of the steam engine in the British textile industry in the early 19th Century was not caused by cost or scarcity then this would be a valid point. However, if that were all the book aimed to do it would have no contemporary relevance. To show that present day world wide fossil fuel consumption is driven solely by ‘certain social relations’ would require a discussion of what would have happened if the country had not been capitalist when that point was reached. Would people have shunned fossil fuels and forgone further economic growth? Or would they have turned to them anyway? The book doesn't address this question.

To answer it we need only look at some recent examples of socialist societies, such as the pre-collapse Soviet Bloc and its satellites. These were big inefficient consumers of fossil fuels. Furthermore, between the end of World War 2 and the late 1970s Britain was the most nearly socialist it has ever been—in fact its energy system was nationalised—yet its carbon emissions continued to rise, energy was used very inefficiently and there were virtually no renewables. Therefore socialism is not inherently low carbon. Malm would, I expect, argue that these examples are not truly socialist but such an argument would be an example of the ‘no true Scotsman’ fallacy.

The old Soviet Bloc was high carbon because, at that time, climate change wasn’t seen as an issue and fossil fuels were convenient and, certainly by the early 20th century, cheaper than renewables. In other words, the factors that drove fossil fuel consumption in the old Soviet Bloc were exactly the same as in the capitalist west. Ergo, your position on the political spectrum has no bearing on your carbon intensity. Fossil fuels are just as intimately woven into the fabric of socialist systems as they are in capitalist ones.

Anyway, if it’s only the very first use of steam that counts then that was in the Cornish metal mining industry and was driven by the need to pump water out of the mines—none of which were located in towns either before or after the introduction of steam.

By the way, an internet search for the phrase ‘the human enterprise’ comes up with many hits none of which correspond to the idea Malm is trying to debunk, so maybe this is just a straw man like ‘the Ricardian-Malthusian paradigm’?

The core of the book is Chapter 13, ‘Fossil Capital: The Energy Basis of Bourgeois Property Relations’. In this chapter, Malm sets out his ‘theory’. This is that fossil energy is a pre-accumulated stock of concentrated ‘value’. The processes that accumulated it saved capitalists the effort of accumulating it themselves. At least that’s what I think it means, but I may be wrong. Based on these ideas, he derives the ‘formula of fossil capital’, modelled on similar formulae in Das Kap, but much more complicated:

Figure 1: The formula of fossil capital.

If anybody out there understands this could they please explain it to me?

He then introduces the weird concepts of ‘abstract time and space’. At this point the argument descends even further into gobbledegook and my eyes began to glaze over.

Strangely, at the end of the book Malm says that the overthrow of capitalism can't be accomplished quickly enough to avert catastrophic climate change and urges international collaboration for a planned energy economy instead. Actually quite a sensible suggestion.

In the acknowledgements section the author mentions that several of the book’s core arguments were first outlined in a paper entitled ‘The Origins of Fossil Capital: From Water to Steam in the British Cotton Industry,’ Historical Materialism 21 (2013), 15–68. I find it astonishing that there should exist a serious—in the sense that it is produced by a reputable publisher and subscribed to by university libraries—academic journal called ‘Historical Materialism’. Coming across such a thing is like finding a journal devoted to astrology or the healing power of crystals on the shelves of a university library.

Far from being a ‘theoretical masterpiece’, as its cover blurb claims, this book is just a load of tendentious Marxist nonsense that uses selectively chosen facts and logical sleight of hand to shoe horn the story of the steam engine into a class struggle template. If you are not a Marxist you will find its description of historical facts interesting but its argument ridiculous. The book is just another example of the academic left talking to itself—a conversation that, fortunately, doesn’t impinge on the world outside university humanities departments. On the positive side, it’s very well written for someone whose native language isn't English.

Fossil Capital was published by Verso Books in 2016 (

© Copyright 2018 Howard J. Rudd all rights reserved.