How To Learn History

Everyone has a theory of history which forms the lens through which they see the world. For most people this is based on half-remembered high school classes, pop culture like Saving Private Ryan or Game of Thrones, and other sources they encounter passively. For some it’s based on a deliberate study of what happened in the past and why. Since the Renaissance, the deliberate study of history has been considered a core part of the liberal arts—that is, the craft of living as a free man. This is because, if your understanding of history is received from some other person or group, then they also control your politics and your civic participation.

Over the years I’ve given many people advice on how to study history. Here I’ve collected a bunch of the things I keep saying. 

1. To understand history, you need to know a big pile of facts about what happened, and you also need models and theories about how societies work and why events happen the way they do. Spend about four times as much effort learning about specific historical events as you spend learning abstract theory.

The ultimate goal is to understand why things happen. But before this, you must first learn what happened. The raw events form the foundation for the theory. Some dilettantes will spend a lot of time on the theory without ever looking deeply into specific events. This lets them sparkle at parties but leaves them ungrounded, unable to tell the true theories from the theories which merely sound good superficially, like the social science equivalent of hippies who bring up pop quantum physics when they talk about spirituality. Witty sophists can cite a string of historical cases that almost but don’t quite support their point, and it takes serious knowledge to tell this apart from good explanations that use historical citations which can bear weight. If I tell you something like “manorial-style agriculture arises after periods of anarchy and warlordism”, or whatever, then you can’t really evaluate my claim unless you happen to already know the history of several eras with anarchy and warlordism, and several eras of manorial-style agriculture, and all that in enough detail to tell whether the connection actually matters.

On the other hand, some people like to learn lots of events without going deep into the theories and explanations for why they happened. There’s nothing wrong with this, so far as it goes. Unlike studying only the theory, studying only the events won’t mislead or confuse you. A lot of excellent scholars have spent their lives cataloging facts about some hyperspecific niche like Roman grain mills without doing much interpretation, and the debt I owe to their kind is immense. That said, if your goal is to understand how society works, then this won’t get you there on its own.

A ratio of about four to one in favor of events seems about right in my experience, but this isn’t a strict law. If you find yourself reading at a ratio of three to one, or five to one, then you’ll do alright.

2. There’s far more history than anyone can learn in a lifetime. What’s worse, they keep making more. You have to prioritize. Choose topics that help answer live questions in your mind.

If you’re thinking about how to restore journalistic integrity and a trusted transpartisan media ecosystem, maybe you should read about the history of American journalism, or the Dreyfus affair, or the early days of the printing press. If you’re fired up about corruption in your city government, maybe you should read about Tammany Hall, or Roman tax farmers, or the British East India Company’s attempts to check its clerks’ corruption from half a world away. If you’re worried about an upcoming American civil war (don’t), then maybe you should look into our earlier civil war, or Rome’s Crisis of the Third Century, or the fall of the Han dynasty. And so on.

Historical case studies are much more effective when they’re driven by a specific question. This gives your research a focus and clarity that’s absent if you’re just like “Today I feel like learning about where railroads came from” or something. It also gives you a natural stopping point: once you’ve mined out the accessible insight for the motivating question, you can move on to a new vein.

3. Follow a pyramid structure for how deeply you research things. Do lots of shallow investigations, a decent number of medium-depth investigations, and a few very deep investigations.

There’s a lot of value in knowing only the rough outlines of a major event. It also gives you trailheads that you can investigate more deeply if you ever need to. These are quick to do, and you should do a lot of them. 

Going deep takes tons of time, and mathematically you can only fit in so many nine-month research projects before you die. However, you do need to go deep into some topics. This gets you the best understanding of the material, of course. It also lets you see how the sausage is made. You’ll read the primary sources, and see how they feed into the secondary sources and the summaries, and drill down into a disagreement between two different scholars. You’ll see how arguments and viewpoints spread and mutate from scholar to scholar across time. Then the next time you’re reading a quick summary, you’ll have an intuitive sense of how they know what they claim to know, which of their claims are likely to be solid, which might be well-justified inference, and which might be politically-motivated bullshit or ideological commitments.

One of the greatest benefits of historical study is that it lets you partially step outside of the assumptions of your own time and place. Everyone will intellectually agree that our society is constructed and arbitrary in many ways, but once you really understand the views and thought processes of a few alien societies, you will viscerally feel that ours is as insane and contingent as all the others. You can’t get that from summaries and secondary sources, or even from reading just a couple of primary sources. It takes serious study and reflection. There isn’t enough time to do this for everything, but it’s critical to do it for some things.

If you’re doing any novel theory of your own, then of course deep study becomes a necessity, because you’re looking for insights that the summaries and secondary sources have missed.

4. Find good sources. Use first-pass methods to get oriented—internet search, recommendations from friends, ask on social media, the book you heard some guy mention on a podcast last month, whatever. Sometimes this will lead you to great sources right off the bat, and sometimes it won’t. 

Even bad sources are useful for bibliography crawling. You look through the references and citations and dig up anything that sounds promising, looking for either something that’s worth reading in earnest, or at least for another bibliography to crawl through. Repeat as necessary.

If you find that a few different sources are all citing the same guy for critical arguments, then he’s probably a giant in the field and you should check it out. Also, pay attention to when each work is published. Some people assume that newer stuff is always better, but this isn’t necessarily true. There is a constant rise and fall of very small traditions of knowledge on historical microsubjects. For a given topic, there’s often a period of 40ish years where the scholarship was better than what came before or after. For medieval European military history, it was about 1890-1940 or so. For premodern Chinese economic history, it was about 1940-1980. And so on. If you’re going deep, you can probably identify the golden period, if there was one. When I’m not looking that closely, by default I gravitate towards 1890-1970, the high point of English-language academia on most subjects. For individual events or biographies I’ll gravitate towards something written 20-40 years after the event in question; this is often the best time to write because the subject is no longer an urgent political issue but the participants and witnesses are still alive.

Of course, if there’s a primary source which directly bears on your question, use that before any secondary source. Your high school teacher was just right about this one.

Once you know what you’re looking for, you can usually find it on Libgen or other pirate sites. Look, I’m not gonna do the whole “here’s an illegal site so that you can avoid it” song and dance. Libgen is illegal. The laws which prohibit it are absurd and evil, and I break them six times on a good day. The works of the great scholars have been passed down through the centuries as the birthright of all mankind. Only a monster would keep our civilization’s collected wisdom out of the hands of the greatest President or the humblest child.

If it’s a niche book that no one has uploaded, you can usually find it on Abebooks. When I do this, I’ll mail it to a service like this one to get it scanned, and upload the scan to Libgen for the next guy. You can also use Worldcat to find if there’s a copy at a library in your area. Personally I’m lazy and I’d rather get my own copy shipped to me if possible, but on rare occasions I’ll have to go to U.C. Berkeley or whatever and hit the stacks. One time I had to use Worldcat to track a rare manuscript to an Australian university library, then paid a student there to scan it and send it to me. So it goes.

5. Authors can be biased, or ignorant, or dishonest, or just wrong. For everything you read, consider the full causal chain that produced the artifact you’re reading. This one goes for any subject, not just history.

To infer from “here are some words in a book” to “here is what actually happened 700 years ago in Cairo” you have to think about how the author knows what she’s talking about. If she’s basing it on another author then you have to think about how that guy knows what he’s talking about, and so on.

Almost any author will be good at some things and bad at some things. Alice might have a firm grasp of factional politics but be hopelessly confused about economics—and worse, she might not realize this. Bob might have witnessed the events firsthand, yet be a terrible judge of character. Pay attention to how the author thinks, what he’s good at, and if there are any gaps in his process. What does he understand best? Does he see his writing as part of a larger social project? What will he say even if it’s unsupported or irrelevant? What will he leave out even if it’s true? What sort of evidence is the author paying attention to?

Be especially wary of modern books which rely heavily on secondary sources published after the author’s own career started, while neglecting older scholarship and primary sources. It’s a red flag for works that are more concerned with discourse and narratives than with actual historical events, a reflection of a reflection of a reflection.