There’s a particular type of alternate history whose premise is: what if [fill in the blank past society] industrialized? (Rome is a particular magnet for this.)

The challenge of such speculation is that we have precisely one data point for what de novo industrialization looks like. Many parts of the world have industrialized, but they’ve done it by adopting the concepts and technologies developed elsewhere. As a result, our explanations for how it happens run the risk of being just-so stories, with no way to test them and see if they’re correct. Those being the only explanations we have, though, we pretty much have to go with them whenever we attempt to depict either an alternate historical industrialization, or this process happening in a secondary world.

But before we ask what it takes to industrialize, we should first look at what industrialization is.

I’m going to give a simple answer to this. An industrial society is one that’s figured out mechanized methods of production, rather than everything having to be done by hand. In order make that mechanization work, we had to harness new sources of energy — specifically, fossil fuels — and then reorganize labor around creating and operating the machines. As a consequence of such changes, a society of this type develops more specialized division of labor, and also tends to support higher, denser populations.

So: how do you get there from an agrarian society where muscles provide most of the power?

Obviously this is in large part a technological question. A Bronze Age society can’t industrialize for the simple reason that their metallurgy can’t support the kinds of technology necessary for powerful steam engines; hunter-gatherers, even less so. Even an iron-working society can’t necessarily manage it, because a boiler capable of surviving useful levels of pressure isn’t something any old blacksmith can bang together. But technology is only one side of the equation, and if all you’re looking at is the metallurgy, it’s easy to think that surely any place with good blacksmiths could figure it out — that it’s pure chance no other time period industrialized. In reality, you also have to ask yourself, what are we making these machines for?

Yes, aeolipiles — primitive steam turbines — existed nearly two thousand years before the Industrial Revolution got rolling. But they were essentially toys, producing very little power and using up tons of fuel to do it. They had no practical function. It took a completely different design to arrive at a steam engine that could do anything useful . . . and the odds that anybody was going to put in the work for that design were low, because what purpose would it serve?

When your vision of the Industrial Revolution is that change at its height, with massive engines driving locomotives or machines that fill whole rooms, you miss how inefficient, ineffective, and unreliable early steam engines were. Even if some Greek inventor tinkered around with the aeolipile or asked “I wonder if there’s a better approach?”, he would wind up spending tons of money and effort on making a device that still wasn’t worth it. The argument I’ve seen — the best just-so story we have for the Industrial Revolution — is that it started where it did and when it did because eighteenth-century Britain found itself in a situation where even a kind of crappy steam engine was better than no engine at all: coal was needed for heating purposes, their coal mines had gotten deep enough that they were flooding with water, and oh look, the fuel you need for the engine is right there where you’ll be using it. No need to pay for transporting it anywhere. The economics worked out to make that a problem worth solving with a new technological development.

Coal has been used for a long time in cooking and heating, but we’ve tended to go for the easy surface deposits first, and to switch away from it when those become less accessible. The roots of Britain’s industrialization probably lie in deforestation and the more intensive mining of coal in the century or two leading up to the development of actual steam engines — a set of circumstances that didn’t prevail in, say, Rome. They handled their mechanical problems with slave labor and had much less need for coal, living where they did; as near as I can tell, peninsular Italy had very little coal anyway (compared to Britain). So trying to invent a steam engine there would be a solution in search of a problem to solve: not a situation that favors the kind of technological development that has to pass through multiple not-very-effective stages before it gets to the good stuff.

And the good stuff, as you all probably learned in school, is steam engines that are smooth and efficient enough to be useful in textile production. Once you have those, it’s worth the cost to build them in places other than on top of coal mines and transport coal to them. Other uses, too, but after the water-pumping prologue, textile industrialization really is Act I of the Industrial Revolution, because it’s an easy place for a better (but still not amazing) engine to make a difference. So here, again, the just-so story says Britain was the right place at the right time: they had huge industries in both wool and (thanks to colonialism) cotton, meaning that productivity gains in something as basic as the spinning of thread could produce absolutely explosive growth. Everything after that — trains and steamships and cool steampunk gadgets — is flying on the momentum created by coal mining and thread.

Of course, all of this is the mundane path to industrialization. In a speculative world, it’s entirely possible to change the starting conditions and create a different trajectory; so long as it still follows the general pattern of “non-muscle energy source allows for new, mechanized, mass production,” it will feel industrial. If that energy source is the discovery of a vein of some mineral which, when a small quantity is placed into a device, becomes an abundant form of power, maybe nobody has to slowly iterate through crappy devices to reach a point where it makes economic sense to transport the stuff elsewhere. Or it’s a method of channeling magical power from the sky, recently discovered by an innovative sorcerer, which turns out to be useful for some productive task. (Quite possibly it’s still textiles: as noted in the previous essay, those are, alongside food, one of the basic survival requirements that have historically demanded the most time and labor.)

I’ll admit to ambivalent feelings about that latter example, because of what kind of magic I like in my stories. An industrialized form of magic is one that, by definition, can be depersonalized. At that point, no matter what words you attach to it, I no longer find it very magical: it’s just technology by a different name. I can still enjoy stories in such a setting; I’ll just enjoy them for reasons other than the magic. And I freely admit this is a personal opinion, not one shared by every reader. For worldbuilding purposes, it’s entirely fine to create a speculative twist on the process of industrialization — and then it helps to understand what does and does not make sense!