How to Invent Everything is a book by Ryan North, of Dinosaur Comics fame. Its premise - time travel is real! You’ve just used your time machine, but now it’s broken and you’re stuck in the past. You want to:

  1. Not die.
  2. Recreate your modern life of luxury.

How to Invent Everything doesn’t literally cover everything, but you can see it as a book about the highlights of civilization. It covers basic survival tips, building up to important technologies like the compass, then ending with fundamental science concepts and a basic intro to computing and logic gates. Its main thesis is that with the right reference text (itself), you could reproduce 12000 years of human progress in, say, 5 years of work.

Odds you’re ever in the exact same scenario as the premise? Vanishingly small. Still, it’s a fun read if you’re interested in exploring how humanity improved its quality of life from first principles.

* * *

The main reason I like this book a lot is that it’s very, very reductionist. That necessarily falls from the premise, but it makes it easy to trace the links between technological innovation. Steam engines relied on precision engineering to make airtight pistons. That required a standard unit of measure and knowing how to make steel. Steel requires knowing how to smelt iron, and that requires kilns. To get the kiln to high enough temperature, you want it to be made out of bricks, and bricks are easy if you know how to make charcoal. There are many details to fill in for getting the raw materials, and the book deliberately focuses on the technology rather than the politics of convincing people to help create your vision, but having the road map to go from nothing to the core of the Industrial Revolution is a big deal.

(It’s also fun to consider how different things would be if certain natural resources weren’t so plentiful on Earth. So much technology traces its lineage to the chemical and physical reactions that turn wood and mud into useful materials, and it’s not like wood has to exist on a planet with life.)

Over and over again, the book points out the difference between when inventions were invented, and when they could have been invented. The gap is usually decades, or centuries. It’d be easy to conclude that humans are dumb, and in some sense that’s true. If you assume that evolution increases intelligence very slowly, and civilization appears as soon as an organism is smart enough to construct it, then the natural conclusion is that humanity is the dumbest species that’s just smart enough to create civilization.

So yes, we’re dumb. On the other hand, when you read the “solution” to different technologies, it can be hard to fault humanity. The prerequisites for the bicycle were around for a long time, but considering their design, even minor changes make the bicycle just a novelty instead of a more efficient means of transportation. (See this post by Jason Crawford for more analysis.)

It’s also easy to see how environmental factors can hamper innovation. Take crop rotation for instance. The core idea of crop rotation is that if you grow the same crop every year, and haven’t invented fertilizer yet, your crops will use up nitrogen in the soil. Next year’s crop is bad, the one after that is worse, and eventually the soil is no longer fertile. The easiest way to fix this is to let fields lie fallow, which means growing no crops to give time for the land to recover. Alternatively, you can plant legumes like clover, which replenish nitrogen thanks to their symbiotic relationship with nitrogen-fixing bacteria.

Now, imagine you are a generic human around 10000 BC. You’ve just discovered the basics of farming, and have reached an uneasy level of food security. Some bozo is saying you should leave a perfectly good field alone for long term sustainability. Are you going to listen to them? Keep in mind that you don’t know the chemistry, you have no way of learning the chemistry because the right tools don’t exist yet, crop experiments have a long delay to payoff, and you have limited experimentation budget because messing up crops means people go hungry. It’s easy to see why we don’t have evidence of two-field crop rotation until 6000 BC, and four-field crop rotation wasn’t invented until around the 17th century. Discover the correct causal links seems very hard, especially given the other confounding factors.

I don’t want to give humanity too much credit. Hot air balloons took a really long time to invent, given that baskets, controlled fire, and fabric all existed for centuries. Incubators for premature babies were not used until 1857. All that needed was a warm box, and their introduction reduced premature infant mortality by 28 percent. The Pelton wheel is a more efficient waterwheel from the 1870s, about 1400 years after the earliest known waterwheel. Its primary differences are using pressurized water and a better paddle shape. It really doesn’t feel like discovering those two should have taken 1400 years, but it did. It’s made worse when the book cites a source that Pelton was inspired by spraying a cow in the face.

Newpaper articl about Pelton

I mean, cool story, but it’s a bit embarrassing no one thought about the physics earlier.

* * *

Does How to Invent Everything offer any greater truths about the universe? Well, maybe. Here are my takeaways.

Models don’t have to be correct to be useful. Many times, people had the wrong idea about why a technology worked, but that didn’t stop it from working. For thousands of years, no one knew where alcohol came from when brewing beer, but people brewed it anyways. Similarly, there is evidence the Egyptians knew that eating liver helped them see in the dark, but it would take many years to learn this was because livers are heavy in vitamin A.

Technology can be lost. Rongorongo is a set of glyphs found on Easter Island that might be an independent invention of writing, but it’s hard to tell because no one knows how to read it. In Rapa Nui culture, only the privileged few could be taught how to read and write, and after a smallpox epidemic went through the island, anyone literate was dead and the symbols were just that - symbols. As another example, the cure for scurvy was lost and rediscovered seven different times between 1400 and its final discovery in 1907. That’s crazy. You wouldn’t think “eat a fresh orange” would be so hard for a society to remember, and yet there it is. Arguably, this is an argument we should focus on better archiving of information.

Society advances when information spreads easily. One question How To Invent Everything toys with is what point in history would let you most influence the trajectory of humanity. It concludes that the answer is either introducing spoken language, or introducing writing. Spoken language allowed humanity to share complex concepts with each other, which let ideas live after death. Writing (and later, the printing press) allowed people to share those concepts further than their community. Even if an entire village disappeared, if their writing remained, their ideas could persist.

Both these technologies are so incredibly useful to society that it’s hard to see how progress was made before their invention, and they look a long time to invent. Anatomically modern humans are estimated at 200,000 years ago, spoken language is estimated at 50,000 years ago, and writing is estimated at around 3200 BC. Again, I don’t fault humanity for this. Getting a community to agree on the meanings of arbitrary sounds seems really hard when no one knows what language is. Getting that same group to agree on how those sounds map to written glyphs seems extra hard. It’s a horrible coordination problem.

If the engine of invention is powered by people sharing random ideas until good ones emerge, then I can’t help but wonder if the best inventions are ones that make sharing ideas easier. Social media’s gotten a beating ever since 2016, but I am coming around to the argument that social media increases the volume of both bad information and valuable information. In this view, you can argue for free speech both morally and practically. Morally, people should not have to self-censor. Practically, progress can come from ideas that sound heretical in the present day, and it can be hard to tell whether history will agree with our judgment.

Atomic gardening is insane. Not actually important, I just thought it was cool. Find a source of radiation. Put it in the center of a field. Plant a bunch of crops around the radiation source at varying distances. Hope you get useful mutations. You have basically no control over the results, it’s just random search in genetics space. It’s the kind of crazy idea that could only come from the 1950s, when atomic hype was high and fear about radiation was low. Some grapefruit varieties sold today trace their lineage to atomic gardening, and shooting X-rays at Penicillium molds led to a mutation that 5x-ed penicillin production. I suppose history has judged it as a net positive, but I almost feel personally offended that such a brute force method worked.