Age of Invention: The Weight of Air

Welcome to my weekly newsletter, Age of Invention, on the causes of the British Industrial Revolution and the history of innovation. You can subscribe here:

Why was the steam engine invented in England? An awful lot hinges on this question, because the answer often depends on our broader theories of what caused the British Industrial Revolution as a whole. And while I never tire of saying that Britain’s acceleration of innovation was about much, much more than just the “poster boy” industries of cotton, iron, and coal, the economy’s transition to burning fossil fuels was still an unprecedented and remarkable event. Before the rise of coal, land traditionally had to be devoted to either fuel, food, or clothing: typically forest for firewood, fields for grain, and pastures for wool-bearing sheep. By 1800, however, English coal was providing fuel each year equivalent to 11 million acres of forest — an area that would have taken up a third of the country’s entire surface area, and which was many times larger than its actual forest. By digging downward for coal, Britain effectively increased its breadth.

And coal found new uses, too. It had traditionally just been one among many different fuels that could be used to heat homes, alongside turf, gorse, firewood, charcoal, and even cow dung. When such fuels were used for industry, they were generally confined to the direct application of heat, such as in baking bricks, evaporating seawater to extract salt, firing the forges for blacksmiths, and heating the furnaces for glass-makers. Over the course of the seventeenth century, however, coal had increasingly become the fuel of choice for both heating homes and for industry. Despite its drawbacks — it was sooty, smelly, and unhealthy — in places like London it remained cheap while the price of other fuels like firewood steadily increased. More and more industries were adapted to burning it. It took decades of tinkering and experimentation, for example, to reliably use coal in the smelting of iron.

Yet with the invention of the steam engine, the industrial uses of coal multiplied further. Although the earliest steam engines generally just sucked the water out of flooded mines, by the 1780s they were turning machinery too. By the 1830s, steam engines were having a noticeable impact on British economic growth, and had been applied to locomotion. Steam boats, steam carriages, steam trains, and steam ships proliferated and began to shrink the world. Rather than just a source of heat, coal became a substitute for the motive power of water, wind, and muscle.

So where did this revolutionary invention come from? There were, of course, ancient forms of steam-powered devices, such as the “aeolipile”. Described by Hero of Alexandria in the 1st century, the aeolipile consisted of a hollow ball with nozzles, configured in such a way that the steam passing into the ball and exiting through the nozzles would cause the ball to spin. But this was more like a steam turbine than a steam engine. It could not do a whole lot of lifting. The key breakthroughs came later, in the late seventeenth and early eighteenth centuries, and instead exploited vacuums. In a steam engine the main force was applied, not by the steam itself pushing a piston, but by the steam within the cylinder being doused in cold water, causing it to rapidly condense. The resulting partial vacuum meant that the weight of the air — the atmospheric pressure — did the real lifting work. The steam was not there to push, but to be condensed and thus pull. It saw its first practical applications in the 1700s thanks to the work of a Devon ironmonger, Thomas Newcomen.

Science was important here. Newcomen’s engine could never have been conceived had it not been for the basic and not at all obvious observation that the air weighed something. It then required decades of experimentation with air pumps, barometers, and even gunpowder, before it was realised that a vacuum could rapidly be created through the condensation of steam rather than by trying to suck the air out with a pump. And it was still more decades before this observation was reliably applied to exerting force. An important factor in the creation of the steam engine was thus that there was a sufficiently large and well-organised group of people experimenting with the very nature of air, sharing their observations with one another and publishing — a group of people who, in England, formalised their socialising and correspondence in the early 1660s with the creation of the Royal Society.

Yet many of the early experimenters with the nature of air, such as Evangelista Torricelli, Otto von Guericke, Denis Papin, and Christiaan Huygens, were Italian, German, French, and Dutch. The likes of Huygens and Papin may have been involved with England’s Royal Society, but the institution was just a branch of a much broader network of European scientists. Why, then, did the scientific observations only find their practical application in England? One influential answer is to do with England’s resource costs, and especially the presence of coal mines. The economic historian Robert C. Allen, for example, while fully recognising the key role played by seventeenth-century science, argues that “had the British coal industry not existed, there would have been no point going to the expense of developing the steam engine”.

Allen explains that the early engines were so energy-hungry that they were only cost-effective if applied directly to the source of their own fuel: the coal mines. It was where the fuel was cheapest. Thus, Allen argues that as a result of Britain having lots of coal mines, the nascent steam engine was kept alive as a technology for long enough that it could be developed and improved by various other inventors, until such time as it could be made cost-effective to apply it to other kinds of mine and then to rotary motion too. As Britain simply had more coal mines than anywhere else, Allen argues, British-based inventors were thus responsible for more of the steam engine’s improvements. Coal was the cradle for the infant industry.

But I don’t think this is quite right. The devil here is in the detail. Allen himself notes that although the first engine was made public in 1712 at a coal mine, Thomas Newcomen’s earliest experiments with steam engines, c.1700-10, occurred in Cornwall. The problem for Allen’s narrative here is that the earliest engines would thus have been used for pumping mines of tin, not coal. Indeed, I can hardly think of a more expensive place than Cornwall to have tried to invent a coal-fuelled steam engine. I say this because just last week I noticed some interesting details about Cornish fuel supplies in the travel accounts of Celia Fiennes, who meticulously recorded her visit to Cornwall in 1698 — just a few years before Newcomen’s very first engines were erected. Fiennes noted the same problem that Newcomen specifically set out to solve, which was that the tin mines near Cornwall’s southern coast had recently become especially prone to flooding. But she also commented on severe shortages of two of the most necessary resources: Cornwall, Fiennes noted, was almost entirely devoid of wood, which was a crucial material in the construction of any kind of pumping engine, regardless of whether it was powered by horse, water, or steam. This startup cost was likely offset for Newcomen by the fact that the government in the 1700s had begun to subsidise tin production. Investment was thus forthcoming. (It helped that one of the most powerful British ministers was the Earl of Godolphin, a major Cornish owner of tin mines.) But the other noticeably lacking resource, at least in Cornwall’s south, was coal.

Cornwall’s coal was largely shipped from Wales or Bristol, on Britain’s western coast. Yet Fiennes noted that during wartime these ships found it much riskier to sail around Land’s End and into the English Channel, due to the threat from French pirates that might be lying in wait. Cornwall’s northern coast was thus able to get Bristol coal cheaply, but during wartime the southern coast was hardly supplied with any kind of fuel at all, be it coal or even firewood. Fiennes described how her supper could be boiled by burning the abundant local gorse, but that roasting a meal was reserved only for special occasions because it required scarce firewood. When Fiennes was writing, the Nine Years’ War had just ended and the southern coast of Cornwall was beginning to receive its regular coal and wood supplies again. But an even more extensive conflict with France — the War of the Spanish Succession, lasting 1702-13 — was to cover almost exactly the period of Newcomen’s early experiments. French piracy in the English Channel was especially bad during this war, so the southern coast of Cornwall was likely highly dependent on burning local fuels. Coal, as well as timber, would have been both rare and expensive.

Now, this might not have been a problem for Allen’s narrative if Newcomen’s early experiments with the steam engine had been conducted on the well-supplied northern coast. But they weren’t. Again, the devil is in the detail. Thanks to the careful work of Newcomen expert James Greener, who has painstakingly reconstructed a timeline for the early engines, the first was likely erected in c.1703 at Balcoath — one of the most landlocked mines in the entire county. Both of its closest ports were on the ill-supplied southern coast, and to transport coal overland from the nearest northern port would have been so expensive as to be unthinkable. So how did Newcomen fuel this engine? He probably didn’t use coal at all, as Balcoath’s surrounding moors yielded abundant turf. Indeed, a couple of nineteenth-century sources positively state that this very first engine was powered by turf or peat, though the original source for the claim was unfortunately not recorded (a Victorian lapse in referencing that is frustratingly common).

So the earliest Newcomen engine — the one in which he ironed out all the kinks and got the machine to work smoothly and reliably — was likely fuelled, not by coal, but by turf. Interesting, it probably wasn’t the only one. The very first Newcomen engines to be exported were brought to Italy by the Venetian diplomat Nicolò Tron in 1717, where they were used to drain swamps. These would almost certainly have also been powered by the marshland’s turf — Tron’s main problem was nothing to do with profitability, it was that his British engineer accidentally drowned while bathing in a river soon after arrival. (Tron is said to have later applied his engines to draining mines, too, but his Scottish mine manager, James Stirling, apparently soon had to flee home after stealing the secrets of Venetian glass-making.)

And then there were Newcomen’s next two applications of the steam engine in Cornwall. Although they used coal for fuel, they were not at coal mines. Just like the first engine, they were used to pump tin mines, this time with easy northern coastal access to the Wales and Bristol coal supplies. Now, Allen might contend that neither of these additional mines were commercially successful, necessitating the move to the coal mines. But Greener points out a political reason for those failures: in 1710, following a Tory landslide, seaborne coal became a target for increased taxation. A mild increase in the taxes on coal that year were likely what prompted Newcomen’s move in search of new applications for his engine, especially to the unaffected inland coal mines of the Midlands. The engines at the Cornish tin mines persisted, however, until the government in 1713 increased the coal tax by a whopping 100%. The engines immediately became uneconomical, leading to the stoppage of both works. Even despite this setback, however, Newcomen seems to have persisted in his attempts to apply the steam engine to tin mines, setting one up there again in 1718. But in a stroke of extraordinarily bad luck, the following year the government drastically increased the duty on seaborne coals yet again. By 1721, there were no steam engines in Cornwall. The only ones left in Britain were those draining its coal mines.

I think these devilish details tell us a few important things. One is that coal was apparently not necessary to the invention and initial development of the steam engine, especially if the first practical application of the machine burnt turf. Not to mention the first application of the engine outside of Britain. The other important thing is that the presence of British coal mines does not appear to have been necessary to the further improvement of the Newcomen engine either. The technology could just have easily been improved if the engines had continued to pump the tin mines of Cornwall, with the fuel coming from anywhere, perhaps even the coal mines of nearby Belgium — if, that is, the government had not repeatedly intervened to raise the price of coal supplied by sea.

Overall, I think these details raise the relative importance of the air science. And although many of the key investigators on air pressure were not English by birth, the Royal Society as an institution successfully drew many of them into its direct orbit. Between Christiaan Huygens, Denis Papin, Robert Hooke, Robert Boyle, Henry Beighton, John Theophilus Desaguliers, and Thomas Savery, an enterprising ironmonger like Newcomen had a whole lot of expertise in England to draw upon for inspiration.

For subscribers to this newsletter, my new book Arts and Minds: How the Royal Society of Arts Changed a Nation is on sale for 25% off, with free worldwide shipping, when using the code AAM20 here.