History Vs. Future

I’ve long puzzled over differing interest in history and the future, in both fiction and non-fiction. And I’ve finally collected some numbers.

Amazon.com says it has 37 million books on offer. Here are the fraction of those books it says are in these named categories:


Note that Amazon has no “future studies” category, so I listed the two future-themed categories I found. Here are the fraction of books associated with related keyword phrases:


Why the far larger interest in real history, relative to all the other combinations of future/history and real/fictional? It can’t just be a simple history vs. future effect, nor a real vs fiction effect – it is some sort of combination effect.

GD Star Rating
Trackback URL:
  • Vulture

    Even in my nice, updated firefox, those TIFF files won’t load at all. Any chance of converting them into something more friendly?

  • Jess Riedel

    Past non-fiction is the preferred quadrant selected by wanting to know truth with relatively high confidence. Non-fiction is obviously truthier than fiction, and the arrow of time means we can find out about the past much more reliably than about the future.

  • > It can’t just be a simple history vs. future effect, nor a real vs fiction effect – it is some sort of combination effect.

    Doesn’t seem like a combination effect to me; you have one real category (history) and 3 fictional categories (historical fiction, SF, and future).

    • The greater popularity of fiction in general belies the hypothesis that truth is more popular. [It’s not a matter of organizing the immediately presented data to get the simplest structure, but rather of finding a plausible explanation.]

      • Sigivald

        True as it goes.

        On the other hand, historical fiction, just like modern fiction, can provide a lot of truth.

        The best sort – with impeccable research – combines a fictional story with a wealth of absolutely accurate historical detail.

        (See e.g. O’Brian’s naval novels.)

        Likewise, modern fiction can, around the fictional narrative, provide any amount of perfectly truthful detail about the background.

        (The difficulty there being, if it’s an area unknown to one directly, telling the accurate ones from the horribly inaccurate…)

      • Fiction which is constrained to pretend it’s not fiction is probably not going to be very good fiction and have the worst of both worlds. Not that it makes much sense to be parsing fine distinctions in as noisy a procedure as this with a single datapoint.

      • truth_machine

        Fiction in general is a huge category that swamps the tiny speculative categories given in the chart … gee, what a surprise that they have small percentages.

      • The comparison I intended for contrast with fiction is the broad category nonfiction. Best sellers are (much?) more likely to be fiction.

      • truth_machine

        Did I imply otherwise? The point is that it isn’t surprising that history is much larger chunk of nonfiction than science fiction is a chunk of fiction.

      • I was responding to the claim (gwern’s) that a real-versus-fiction effect explains history’s popularity versus futorology. This isn’t the correct explanation. We don’t prefer the real to fiction in books–as proven by the very popularity of general fiction compared to general nonfiction (not shown in the dataset).

        In various ways, it’s obvious that history is a richer field than futurology, and I too am struck by Robin’s not seeing any of them. I suspect the reason is that he doesn’t make direct use of much historical evidence (rather, using putative general principles, which were perhaps derived from history by others) in his futurology.

  • TheBrett

    With history, you can go into nearly infinite amount of details about any particular segment in time. With futurology, though, it’s much harder and more speculative – and you can find yourself embarrassingly out of date in only a few years, so there’s less incentive to write such books.

  • Sieben

    Speculating about the future associates you with the geeky sci-fi crowd that attends comic conventions and doesn’t bathe.

    Writing about History associates you with all sorts of prestigious groups, liberal/conservative/academic/etc.

    History fits a lot more of people’s self image than futurism.

    Then there’s also the fear of being wrong. You can be wrong about the future but you can always argue that you’re right about the past.

  • endril

    I’ve always thought it was strange in the first place just how much writing there is about history. You could read books about World War 2 for much longer than the length of the actual war.

  • Bafflement about this baffles me. (Am I missing something?) Of course there’s more written about the past: to make even limited predictions about the future, you must analyze a great deal of data about the past.

    Anyone interested in the future must be interested in the past, but not the reverse.

    • oldoddjobs

      “Who controls the past controls the future; …”

  • James Andrix

    History is easier to write.

  • advancedatheist

    Just wait awhile. Kurt Vonnegut in one of his late interviews said that if you live long enough, you will know “the future” of the people you knew earlier in life, assuming you have kept in touch with them.

    For example, I now know the future of Ralph Whelan, based on what I remember of him from his days as an Alcor employee and cryonics activist in the early 1990’s: He died back in September at age 46, but he let his funding lapse, and his parents stepped in for conventional body disposal.

    • Tragic. I wonder if they will ever learn enough to regret doing so in the future. http://www.brainpreservation.org might have been able to prevent such a casual dismissal of his morality and his life’s purpose. Then again, maybe not. Most people are really no smart enough to avoid voting for Hitler. He never loses his popularity, he just takes new forms, even though he’s one of the easiest kinds of people to create.

  • Kevin

    History + nonfiction is truth. The other combinations are speculation. Just like this blog seeks truth, so too do readers, it seems.

    • truth_machine

      Yeah, but they’re more likely to find it.

      • oldoddjobs

        Tell it, sister!

  • Doug

    While I don’t disagree with the idea or conclusions of this post, I think it’s probably better to compare books published in a given year, rather than total books in a library. A book written about the Civil War in 1970 is probably still relevant today. A book written about the future in 1970 is very likely not, and is more than likely out of print.

    Even if history and futurology garner equal academic interest (not saying they do), one would expect the back catalog in a library to heavily skew towards history. I’d say look at just books published over some recent time period (which I’m sure still favors history).

    • I used the data I could find; happy to use annual data if I can find it.

      • truth_machine

        Like the drunk under the lamppost. Please tell me this isn’t how most economists operate.

  • Sigivald

    Why the far larger interest in real history, relative to all the other combinations of future/history and real/fictional?

    … because it is real?

    I mean, you say it can’t be just a simple history vs. future effect, but … why not? It sure seems on its face that it can be that simple.

    The future is known and interesting. The future is unknown, and interesting … in a different way, less amenable to useful writing.

    Futurism is a niche taste, in other words. I think at least partially because the track record is so abysmally bad in terms of accuracy of prediction.

    Many people prefer sure knowledge [or something easily confused with it, admittedly] to what rounds to speculation*.

    (Also, what Doug said; it’s more meaningful to compared current output than historical sum output. Remember that for one thing plenty of history is put out essentially as residue of a doctoral thesis, while there are, ahem, far fewer doctorates in “future history”.)

    (* Science fiction being a subset of “speculative fiction”, but I don’t include it in the future speculation category because for the most part if doesn’t even pretend to be a prediction of the actual future, more a mental exercise, if done right, about a notionally-possible one.

    That, of course, excludes space opera and other things called science fiction that are not properly so at all.

    Exclude those and I doubt you get .5%, though I don’t know how you’d determine the number short of manual sorting.)

  • david condon


    I’m surprised I’m the first to mention this.

    • truth_machine

      What? Your school didn’t have a required science fiction curriculum?

  • truth_machine

    As Santayana noted, history *matters*. And there’s only one. Science fiction is make believe, it’s generally implausible or frankly impossible, and it’s a niche subcategory of fiction, so talking about the *general* appeal of fiction is just stupid.

    • Books about history are not directly reporting truth. They are trying to *infer* truth from our many scattered noisy clues. Just like our books about the future are trying to infer that from many clues.

      • They are trying to *infer* truth

        The same logic applies to all empirical truth claims; that is, we have only inferential knowledge of the present.

        Our knowledge of the past is more like knowledge of the present than is knowledge of the future. This is because there is nothing comparable pertaining to the future like memory (or the written historical record) is to the past. (Notice that there is little writing about the prehistorical period, which might be, with the absence of an historical record, analogized to writing about the future.)

      • truth_machine

        Yes, clearly. It’s hard to wrap one’s head around the immense intellectual dishonesty of Hanson’s response to my criticisms.

      • truth_machine

        That’s your response? Really?

        You have an awful lot to overcome.

  • Jim

    Because “history” is not only an interest in the past but more importantly a genre of writing that was well established 2500 years ago. “Future” has no such connotation. You would find the reverse if you were to compare amazon searches of “novel” to searches of “old” or “ordinary.”

  • lemmycaution

    There also is not that much you can say about the future, It hasn’t happened yet. Most people can provide the bulk of their speculations on the future in an essay.

    • lemmycaution

      Lots of history and science books end with speculative chapters on the future. They often seem unfocused and not really worth reading.

      • …Because most people are dull, immoral, and unimaginative.

      • Yet elsewhere you call for political organizing. Who with? Who for?

  • Axa

    Perhaps Amazon is not the best place to look for the future. I wonder what are the percentages in scientific literature. In the other extreme: patents are 100% about the future.

  • Friendly-Hi

    It’s probably quite simple: We know or at least can reconstruct a lot about historical details, while trying to picture details about the future is futile. It’s hard enough to even picture and predict broad trends especially considering that they are largely interacting with each other. Overall there can simply be much less said in terms of volume about the broad trends we might expect in the future than about all the details and broad strokes of history.

    I would imagine that’s the main reason but there there are probably also a slew of lesser reasons contributing.

    1) By speculating about the future people will point and laugh if you get it wrong, thus you incur a potentially massive hit in social status if you choose to write about the future. It seems much more risky to write about the future than about the past.

    2) Any halfway well-researched picture of how the future might look like is too crazy to swallow for most people, thus they are not interested in reading those “crazy speculations”. And by extension the author is crazy as well and to be avoided.

    3) My impression is that higher education talks a lot more about the past and little (if at all) about the future. Thus people in scientific fields simply know more about the past than what to expect from the future. Also overall there are a lot more fields of study concerned about the past (especially if we include natural history) than fields mainly concerned with the future.

  • Due to the use of knowledge in society, in the sense written about by Hayek, specialists in any era can find out useful things about the past. Also, since humanity has not outgrown control by sociopaths, and the technology of law/politics/government is how progress has been made or lost with respect to dealing with them, it makes sense to improve our knowledge of past events, especially the events that led to the industrial revolution. It also makes sense to collate into a consistent narrative exactly what has transpired with banking, for-profit investment in war and destruction, and other high-hierarchical-level human systems.

    Most “futurists” are not very familiar with the information contained in G. Edward Griffin’s “The Creature From Jekyll Island,” or Rothbard’s less conspiracy-minded (but not better) “The Case Against the Fed.” Few science fiction books care to examine a likely future, focusing instead on entertaining possibilities. A notable exception are the two sci-fi books by James Halperin, and “The Moon is a Harsh Mistress,” by Heinlein. Those contain a lot of likely-to-happen individual ideas. (Not necessarily the Moon v. Earth thing, but more the “brainlike” processors, technology helps outmode enslavement thing.) In any case, I found the focus on history to be entertaining in The Moon Is a Harsh Mistress, because it became relevant again, in the future.

    I also thought Dune’s interpretation of Hitler’s historical record in the far future was interesting, as was Norstrilia’s conception of Drexlerian nanotech (written before Drexler’s “Engines,” I think).

    Most people lack any moral framework at all, simply accepting what is handed to them by authority figures and taking it as valid. History contradicts this bland conformist acceptance, by teaching us that Stalin and Hitler were wrong, and their policies were wrong. Well, Bloomberg is attempting to take people’s right to own firearms away in Nevada, as we speak, so that’s one policy that is similar to the totalitarian position on guns (only police should have them).

    History can be analyzed not just as discrete facts and occurrences, but as evidence that human networks of a certain level of advancement were able to universally advance to a certain level of comprehension, to the point where X% of a network was able to support, communicate, and modify ideas necessary to survive.

    That is interesting to me. I think it will also be interesting to powerful computers in the future, when they are trying to decide how to interact with humans.