If you prefer to listen rather than read, this blog is available as a podcast here. Or if you want to listen to just this post:

Or download the MP3


Recently I’ve been working my way through the Iliad (the Richard Lattimore translation as recommended by Harold Bloom). And as I’ve been doing so, I noticed something I hadn’t on my previous read-through, something having to do with…horses.

Horses are mentioned a lot in the Iliad. One of Homer’s frequently used epithets is “breaker of horses”, but he also talks about “horse tamers”, and “horse drivers”. And of course everyone knows about the Trojan Horse, though, to be clear, that doesn’t make an appearance in the Iliad. But it wasn’t the fact that there were horses which caught my attention, it was the fact that no one rode those horses, the horses are used exclusively to pull chariots. On some level I may have already known this, but it was only on this recent read-through that I paused to consider the implications of this fact.

I’m no expert on the history of horses, but I did some digging and this is what I found. While horses were domesticated no later than 2000 BC and probably as early as 3500 BC, the earliest evidence for them being ridden in warfare is from around 850 or so BC. (The Trojan War, or at least the event that we think of as the Trojan War is generally considered to have happened around the 11th or 12th century BC.) And evidence that horses were ridden even outside of warfare before then is suggestive, but mostly inconclusive. I find this interesting for a whole host of reasons, but primarily for what it says about how long it can take for the full utility of a new technology to become apparent, even in the realm of war, where it’s a matter of life and death.

Those that have spent any time studying the Mongols (and if you haven’t I could not recommend the Hardcore History series on the Mongols enough) will know that before the harnessing of gunpowder the ultimate military unit was the mounted archer, particularly once the stirrup was invented. And yet by the time the Mongols used them to such devastating effect, horses had been domesticated for thousands of years. Bows which could be used from horseback had existed for almost as long, and while stirrups were a recent invention there is nothing about their construction which kept them from being invented much, much earlier. So why did it take so long? And is there anything which is currently just sitting around waiting for a slight improvement before it entirely changes the world?

As for the first question, while riding horses is baked into our psyche now, I’m sure that the first time someone suggested getting on the back of something alive, which could buck you off without any warning, seemed pretty crazy. Also if you’re mostly using horses to facilitate archery it would seem logical that a stable platform like the chariot, with a 360 degree field of fire was much better than trying to do the same thing from horseback. Also even something which seems obvious can take a long while to develop. The classic example of this is wheeled luggage.

I’m not sure when it would first have been useful to have wheeled luggage. You obviously want a reasonably flat surface that’s free of mud and horse crap. But once that’s in place, and perhaps even before then, you could imagine something like a traditional steamer trunks with wheels built into one corner being easily wheeled from place to place. But even if flat surfaces were uncommon and horse crap all too common, it certainly would have come in handy by the time of World War II, and particularly in the immediate aftermath as commercial air travel took off. And yet the first time someone actually put wheels on luggage was 1972 or, as has been frequently pointed out, after we landed on the Moon. Further, this was not the wheeled luggage most of us are familiar with this was the typical old style suitcase with a strap you could pull, and which tipped over all the time. The rollaboard style didn’t come along until 1991! This is despite all the necessary technology existing for decades before hand.

There are certainly other examples of this sort of thing, but you can see that even over very long periods (centuries in the case of horses, decades in the case of wheeled luggage) obviously innovative improvements can elude us. And it doesn’t seem to be important that it’s a matter of life or death (horses) or if it’s something that millions of people could notice (luggage). It can still be overlooked for a disconcertingly long time.

To return to my two questions, I’m not sure why certain innovations take so long, there are obviously lots of potential theories, but the fact is that they do. And as far as the second, if I knew of a slight improvement that would change the world I should be out there raising capital rather than writing this blog. That said I do have what I think is a good (or perhaps awful) candidate, though unfortunately it’s not the kind of thing you raise money for. That candidate is… nuclear weapons.

I’ve blogged before about nuclear weapons, and they’re definitely something I frequently worry about, but a combination of my observation about the Iliad and a recent Bloomberg article by Tyler Cowen brought the subject back to the top of my list. Combining all three of these sources together we come up with the following list of reasons why I think nuclear weapons should still be humanities number one concern (which is Cowen’s position as well.)

1- Just as it took a long time to figure out the “killer app” (literally) for horses, it could be that we haven’t yet figured out the most effective way to use nukes. For example, I personally think there are probably methods of just threatening to use nukes that could be horribly effective in the hands of an aggressive nation. (Look how far Hitler got on nothing but aggression and confidence with nations who had no appetite for violence.)

2- If it potentially takes many man hours before a novel way of doing things is uncovered, as in the case with luggage, we don’t actually have all that many “nuke man hours”, however you choose to define that term. This is good in general, but may be bad if it leads us to believe we’ve ruled out or accounted for all of the possible scenarios for nuclear weapons to be used.

3- Cowen points out that every era has their recency bias. In the immediate aftermath of World War II and during the Cold War people were very worried about war because we had just one. Now that we’re experiencing the Long Peace, people don’t worry about that anymore. Now they worry about banks failing because of bad mortgages. And in the immediate aftermath of 9/11 we worried about terrorism. In each of those cases our worries were based more on what sprang to mind than on what the true probabilities were.

4- Following from the above Cowen goes after Steven Pinker, and makes an argument I’ve also made. Those who paint a rosy picture of a future without nuclear war have to be right every single year, while those who worry about it only have to be right once for all their worry to be justified and for all the optimism to seem fantastically naive. Or as Cowen says it:

Yes, the arguments for optimism often appear stronger than the arguments for pessimism, and indeed they are. When it comes to nuclear weapons, however, the arguments for pessimism only have to be true once — and that is likely to happen sooner or later, no matter how positive the general trends

5- Cowen’s final point is that, as with all technology, nukes are becoming “easier and cheaper to build” and beyond that other improvements are being worked on, like hypersonic delivery systems. These are also things I’ve talked about before but it sounds more impressive coming from Cowen.

6- Related to the above, ICBMs, particularly solid fuel MIRVs (Multiple Independently targetable Reentry Vehicles) are basically impossible to defend against, and if there is any wiggle room, hypersonic missiles will definitely get rid of it. Absent the ability to mount a successful defense we can only deter through the promise of retaliation, but that means we need to be willing to retaliate, and that we can’t make any mistakes in this area, which if history is any indication is harder than you think.

7- Additionally, as nukes become easier and cheaper to build more countries are likely to acquire them, which means that the bipolar game of deterrence which got us through the Cold War is going to increasingly become a much more difficult multipolar game. And this doesn’t even take terrorist nukes into account, which is an entirely separate massive threat.

8- Despite all of the above, people appear to have stopped worrying about nuclear weapons or nuclear war. In every way that matters nuclear weapons are at least as dangerous as they have always been, and in many ways much more dangerous. But the amount of attention and worry they attract is lower than at any point since their invention. Consequently, vigilance is much lower than in the past, and there’s much less effort being put towards related diplomacy, treaties and other activism.

I think, as always, the key problem is the people have a very narrow time horizon. Is there going to be a nuclear war in 2019? I’m 99% sure there won’t be. But it’s not just a question of 2019, or 2020, or 2025. Nuclear weapons cannot be uninvented. They’re not a danger for the rest of the decade, or the rest of the century, they’re a danger for the rest of forever.


The Iliad goes into a surprising amount of detail when describing death. On the other hand, I have assiduously avoided describing someone getting “struck in the chest near the nipple, and the point fixed itself in his lungs”. If you appreciate that, consider donating.