Why Is Harry Potter Popular—A Rebuttal

Ozy FranzFN 1 recently had a post titled Why Is Harry Potter So Popular? In that post, Ozy argued that Harry Potter isn’t very good, but was popular only “because something had to be.”

The basic argument is that popularity can feed on itself and become self-reinforcing. Thus, Harry Potter may have caught on at first more or less by coincidence and then snowballed its way to global prominence. As people recommended it to each other and enjoyed discussing it with each other, the popularity became self-reinforcing.

The post cites a famous paper that exposed different groups of people to the same set of music—and discovered that different songs became popular in each group. If one song happened to get popular at first in a particular group, that caused people to recommend it to others; as it gained traction/rose to prominence, it would crowd out other similar songs that could have gotten popular instead. In a different group, some other song might get the initial burst of popularity. Then, that song would be the one that snowballed into dominance in that iteration of the experiment.

Ozy’s post argues that this is exactly what happened with Harry Potter: that it was one of several mediocre (“good-but-not-great”) children’s fantasy books published around the same time, and that any one of them could have randomly turned out to be the mega hit that Harry Potter would be. The post mentions The Animorphs, A Series of Unfortunate Events, and the Time QuintetFN 2 as other series that could have taken off instead of Harry Potter.

This argument struck me as pretty odd—not least because I know that the author is much more of a fan of Harry Potter than I am. I’m going to argue against this point on the object level: I think there were a bunch of reasons Harry Potter became more popular than other, similarly well-written novels published around the same time.

In the next post, I’ll move to the meta level, and talk about why this sort of debate matters for more than just Harry Potter. I think this argument about Harry Potter exposes a more fundamental challenge we face whenever we try to answer complex questions. But first, the object level—here’s why Ozy is wrong:

The Model Is Too Simple

The basic problem with Ozy’s argument is that it focuses only on the quality of the writing in Harry Potter and on random chance. It implicitly argues that any of Harry Potter’s popularity that cannot be explained by writing quality must come down to chance.

Put more simply, Ozy’s post builds an informal model of Harry Potter’s success as a factor of exactly two inputs: (1) The literary quality of the text, and (2) random chance. Thus, an argument that shows that Harry Potter’s success was not attributable to the first input also show that it succeeded based on the second input.

What this ignores—I argue—is that Harry Potter has a bunch of features that were key to its success that had nothing to do with its literary quality—in simple terms, we could add a lot of other inputs to the model. To list off a few rapid-fire examples:

  • It had an everyman protagonist who was boring/non-distinct enough that most people could imagine themselves in his shoes.
  • It featured a wish-fulfilment opening plot where a regular-ish kid found out that he was secretly special, a common daydream.
  • In addition to Harry, the core cast was gender balanced and represented enough personality types that most readers could find someone to identify with even if they didn’t like Harry Potter himself.
  • The books were long enough—and published infrequently enough—that readers could get invested in a single book and could count on other fans having read the same books. In contrast, something like The Animorphs was published so frequently that even serious fans might have read a different number of books. And no one would ever look forward to the next one that much, because the last one had just come out not long ago.FN 3
  • Relatedly, the length of the books made them much easier to adapt into pretty good films, which changed the franchise from extremely popular-for-a-book to a global cultural phenomenon.
  • The books were set in England, giving them an appeal to Anglophiles in America.FN 4
  • The tone/style was generic enough to avoid polarizing readers, unlike A Series of Unfortunate Events.
  • The books had truly excellent world-building for a kids series—probably on the order of Star Wars, and well above the competition.
This last point is worth elaborating, because I am convinced that it happened entirely by accident. I believe that the way fictional universes have great world building is by having a lot of useless details. I think the useless part is key: a world where every detail is relevant to the plot won’t have quite the same effect, no matter how detailed it is.

This is related to the idea of Chekhov’s Gun, the idea that authors should “remove everything that has no relevance to the story. If you say in the first chapter that there is a rifle hanging on the wall, in the second or third chapter it absolutely must go off. If it’s not going to be fired, it shouldn’t be hanging there.” This may (or may not) be a good idea for many books, but it is definitely something that we, as readers/viewers are very used to. Thus, any time all the details of a setting end up used, it feels just like a book—exactly what we’re used to.

Conversely, when a setting has details that aren’t important—guns that are displayed but never fired—it makes the setting feel like a real place that we’re getting a window into, rather than an entirely artificial story. So the cantina scene in Star Wars is great for world building precisely because we meet so many irrelevant aliens. If each of those species later became important to the story, the effect would be greatly diminished, because the audience would see that they were only introduced to set up a later payoff.

And the first Harry Potter books have a ton of this sort of useless detail—or at least appear to. All the details about goblins and Gringotts, the Ministry of Magic, the Daily Prophet, Ollivander’s weird focus on wand cores; these all seems like exactly the sort of guns-that-aren’t-fired that make world building great. But, by the end of the series, all these guns end up getting fired—each of these “inessential” elements from the earlier books turns out to have been laying essential groundwork for a later plot development.

Thus, looking at the series as a whole, I’m convinced that the excellent world building was nearly accidental. It felt like a real, lived-in world, but only because readers experienced a lot of details as extraneous—when those details were actually key to developments five or six books later.

I believe that Harry Potter is so popular because it benefits from all these factors, and maybe more. More simply, our model should include way more factors than just “literary quality” and “random chance.” Next post, on to the meta level!


16 thoughts on “Why Is Harry Potter Popular—A Rebuttal”

  1. I think the world building was definitely big part of the appeal (or at least it was for me). The books created a world that felt imaginative and fun to occupy. It also had the effect of making the real world we live in seem full of secret potential and mystery. If the world of Harry Potter was real, what might secretly be happening at any of the mundane places we see or visit everyday?

    1. That’s a good point. TV Tropes has a page devoted to The Masquerade that makes the point that being set in a superficially real world “makes it easier for the fans to imagine what it feels like to live as a ‘normal’ person in the setting.” And that’s got to be part of Harry Potter’s appeal too.

  2. Interesting point about the value of having that ‘useless detail’. I’ve never read Harry Potter, but I think it helps explain why I enjoy Charlie Stross’s novels, but why they don’t leave much of an impression afterwards; they are very obviously novels, the result of deliberate craftsmanship, so much so that it almost distracts when reading them.

    1. Yeah, I think a lot of very enjoyable writing follows the Chekhov’s Gun dictat. I haven’t read Charlie Stross, but I think of mysteries as typically following that path. It would likely annoy readers in that genre if there were a bunch of “useless detail”/worldbuilding—a big part of the fun is being able to put together the puzzle as the mystery unfolds, and that would be basically impossible if all the important info were buried under useless detail. (Of course, mysteries can and do include red herrings, but a red herring is different from truly useless detail that doesn’t ever become relevant, even as a red herring.)

      In a lot of ways, I think Harry Potter turned out to be a mystery series—that just wasn’t obvious until book 6 or so.

  3. I definitely agree that the best of Harry Potter is the worldbuilding. I disliked the fifth book, and so hated the sixth that I never read the seventh. Yet, I still follow HP fanfiction communities and love essays analyzing various details. Without the strength of the setting, Harry Potter would just be yet another series that started out decently and then got sucked into a black hole of plot contrivances and gratuitous grimdark.

  4. I definitely think that you’re onto something with regards to the worldbuilding. If you look at the franchises that have spawned especially large and long-lasting fandoms, most have very large and detailed settings. Star Wars, Star Trek, Lord of the Rings, the Buffyverse, Firefly (which has an impressively large fandom considering how short-lived and financially unsuccessful the original show was), the DC and Marvel superhero universes…

    Incidentally, speaking as someone who did read the entire Animorphs series, I think that it actually does have a decent amount of this sort of worldbuilding. Scattered throughout the books are all sorts of references to alien species we never meet (or only meet once, like the various alien monsters that the main villain shapeshifts into during fight scenes), planets we never visit, events that happen offscreen, alien words we’re never told the meaning to, etc. However, most of this doesn’t show up until the later books. Especially before the token alien member of the team is introduced in book 4, the early books mostly just take place in a present day + secret alien conspiracy setting not that different from, say, The X-Files. So a reader who only buys the first one, two, or even three books isn’t going to encounter much in the way of worldbuilding.

    1. Yeah, I agree about Animorphs. Once Ax shows up—and especially once they start getting into the politics of the different alien races—it starts to have some really good world building of exactly the sort that made Harry Potter great. And it even does a fair amount of interesting work with the philosophical/free will issues at play and manages to make its (initially very one dimensional) villains surprisingly sympathetic.

      I really think that the biggest thing holding the series back was the number of books. The series as a whole wasn’t that long or anything—a quick google search indicates that had about twice as many pages as Harry Potter with ~10 as many books. And I’d be willing to bet large amounts of money that the page-count difference overstates the word-count difference based on the ways the different books were formatted and typeset.

      As a result, each book was incredibly short. This meant that, as you note, people had to read half a dozen or so books into the series before it even started to develop the more interesting bits. I bet if the full first “book” of Harry Potter had been time at the Dursleys, that series wouldn’t have taken off as well either.

      Really, it’s even worse than that. Like I was saying the other day about the Hardy Boys, this structure got in the way of telling good stories. Everything had to be self-contained enough that it could fit into extremely episodic content. Given that constraint (and the pace they pumped them out) I’m really impressed with how many big ideas they were able to tackle—still, I’m really curious what they could have accomplished without that self-imposed handicap.

      (The short length also put me in an awkward position on more than one occasion.)

  5. I think I came across that study in The Black Swan where a similar argument is made.

    The qualities you list seem important to the success of Harry Potter, but not attributes of every wildly popular thing. Did Minecraft hit these same criteria? If we lack the benefit of hindsight, can we identify the maximum peak potential of a new thing with a fixed set of criteria?

    I think that these attributes can only be identified reliably in retrospect. A specific book might be bland enough that we know it will not be popular, but many books are good enough to be popular, an order of magnitude more books than actually are popular.

    With infinite time one could read all books and perhaps make better predictions about which should stand the test of time, but this problem is intractable, except perhaps if humanity as a whole takes on the task. I suppose it has.

    Even if we knew the potential of a book there is so much randomness as to which gets popular first. It is said by many that Betamax is superior to VHS, and said by a few that mercurial is better than git, but once something is popular enough then it is popular because it was popular, and it excludes other similar things.

    1. Even if we knew the potential of a book there is so much randomness as to >which gets popular first. It is said by many that Betamax is superior to VHS, >and said by a few that mercurial is better than git, but once something is >popular enough then it is popular because it was popular, and it excludes >other similar things.

      That’s all correct, but I think we should be careful to distinguish between the network effect and the sort of snowballing popularity I was talking about. (Looking back, though, I’m not sure my post or Ozy’s post did a great job of making this distinction.)

      Specifically, with the network effect, something really does become better because more people use it. I don’t know enough about the technical details to have an opinion on mercurial vs. git, but GitHub is a classic example of network effects—even if it weren’t technically any better, the fact that so many people use it makes it much more useful on a practical level, because it means you’re more likely to find a useful project/a supportive community.

      In contrast, the in the music study the songs weren’t getting any better when they were more popular. It’s just that they caught people’s eye more, and became more popular than their quality would predict.

      Harry Potter’s success might be a bit of both, now that I’m making the distinction. To the extent that fans enjoyed participating in some of the cultural ephemera around the books (launch parties, discussions with friends, fan fiction) Harry Potter may actually have become better in the relevant sense because of how popular it was. That would be a network effect.

      On the other hand, Ozy is arguing (I think) that much of Harry Potter’s popularity is traceable to the non-network-effect snowballing that happened to the songs. I.e., that Harry Potter was popular because it got lucky and caught on, which directly made it popular—not that it got lucky, which made it better though network effects, which made it even more popular, etc.

      I also have much more to say about the rest of your comment, especially the the Black Swan stuff. (Which is a good book, even if Taleb comes off as obnoxious/arrogant half the time.) But I think I’ll save that for the follow up post.

    2. The thing is, Harry Potter wasn’t just popular. It had a level of success that was totally unprecedented for a children’s book series, and one that hasn’t been matched by anything that has come along since. Like “veronica d” wrote in a comment on the original post, the alternative scenarios include “no children’s book series achieves HP’s level of success” as well as “some other children’s book series is equally as successful as HP was.”

      The former scenario seems much more plausible to me from my research – looking at sales figures, no children’s book series with less than 10 installments even comes close to Harry Potter. The closest recent (IE first installment was published after 1980) series in terms of both sales figures and cultural impact, Twilight, has less than a quarter of Harry Potter‘s total sales and less than half as many sales per installment (if you divide total sales by number of primary installments, you get 72 million for HP, 30 million for Twilight, ~22 million for The Hunger Games and 5 million for ASOUE).

      It’s been twenty years since Harry Potter and the Sorcerer’s Stone was published. If its mega-success was primarily due to random chance, then surely something would have come along by now that at least approached that level of success. But nothing has.

      And I have no idea why you’re bringing up Minecraft, which is a product of a totally different medium and is probably closer to Lego than any book.

      1. It’s been twenty years since Harry Potter and the Sorcerer’s Stone was published. If its mega-success was primarily due to random chance, then surely something would have come along by now that at least approached its level of success. But nothing has.

        To play devil’s advocate, though, couldn’t you run that argument just as well in reverse? Like this:

        It’s been twenty years since Harry Potter and the Sorcerer’s Stone was published. If its mega-success was primarily due to [any set of features of the series], then surely something would have come along by now [and successfully copied those features at least well enough to have] approached its level of success. But nothing has [which indicates that Harry Potter’s success was due to random chance rather than anything duplicable].

        1. That is a good question. But I think you can chalk it up to the fact that novelty is an important part of the success of any creative product, as demonstrated by the existence of the pejoratives “rip off,” “rehash,” and so on. Simply copying a successful product won’t work because in this field being a direct copy is itself considered a bad thing. To put it simply, other book series can’t succeed in the exact same way that Harry Potter did because Harry Potter got there first. But if they had published a series with the relevant features before JK Rowling did, then yes, they might have enjoyed a similar level of success.

          Whereas if random chance was the primary factor, the above shouldn’t matter. If it’s all up to a roll of the dice, then you should eventually get a similar result if you roll the dice again enough times.

          Of course, the subject is much more complicated than this. We also have to factor in things like how the changing cultural zeitgeist affects which things can be successful when (to give an extreme example, it is highly unlikely that the TV series 24 would have been anywhere near as successful as it was if 9/11 hadn’t happened less than 2 months before the first season was scheduled to premiere). And I don’t deny that dumb luck can be a significant factor. But I think that pinning everything on dumb luck leaves too many things unexplained.

  6. I agree – IMHO, Harry Potter was a mega-success because (1) everybody else was reading it, which functioned as a recommendation and provided network effects because you could talk to other people about it, (2) reading the books is unusually enjoyable for children’s fiction, and (3) whether by design or chance, the movies were unusually good for literary adaptations.

    (1) Was definitely a snowball effect – first a bunch of kids liked it (compare Eragon, which pretty much stopped here), then there were stories about how many kids liked it, then there were stories about how many adults liked it, then the stories were “everybody’s reading it.”

    (2) I do think that Harry Potter is much more enjoyable for most people than average children’s fiction, and I agree with you that it’s primarily the world building and the reliability of the characters. The sheer delight when you walk around the corner at Universal and see Diagon Alley and Hogwarts realized is hard to describe.

    I love your point about worldbuilding, but want to argue with one detail. I haven’t read much background, but my guess as a reader is that the worldbuilding was intended and the Chekov’s guns came later. My guess is that Rowling set up the world by inventing a bunch of charming weird details – Dumbledore’s “few words” and Ron’s reaction to them, the chocolate frogs, quiddich, the monetary system, etc., and then as she wrote the books she picked details and added significance to them.

    (3) Shouldn’t be overlooked, either. Imagine if Avatar: The Last Airbender had been a cinema hit – I expect Nickolodeon would have gotten tens of thousands of new viewers for the animated series and its sequel.

    1. I agree that the worldbuilding elements were mostly turned into plot points retroactively. Rowling isn’t a natural plotter, and the difference in quality between the early ones (where she had time to think) and the later ones using those elements (where she was under the gun with her publishers) is very striking.

Leave a Reply

Your email address will not be published. Required fields are marked *