Making Fan Theories

So you've seen some Pokémon fan theories and think they're really cool. Suppose you want to create some of your own. Great! The more, the merrier.

What Is a Theory?

A theory, in a sentence, is a model of the universe: a set of assumptions about what we can't see that explains what we do see. The theory of gravity, for instance, proposes that all things with mass attract one another according to a certain mathematical formula, and it consistently explains the way that we observe objects falling towards the Earth, the way that the Earth orbits the sun, and all manner of other observations that we make every day. The mathematical formula itself is invisible - it's not written in the sky somewhere - but we know it, or at least can be pretty darned sure that we know it, because it's what fits consistently with all the observations we make.

Fanmade theories about something like Pokémon, of course, can't come close to the power, rigor or usefulness of proper scientific theories. Crucially, whereas scientific theories don't even get to be called that unless the predictions extrapolated from their assumptions about the universe have been rigorously tested and consistently turned out to be true, fans aren't in any position to test their theories about works of fiction created by people who may be halfway around the world, so fan theories make do with just being fitted to previous observations. Also crucially, whereas the ultimate goal of science is always to search for the actual truth, fan theories are more about entertaining and intellectually stimulating other fans, hence why that lack of rigor isn't a problem for most fan theorists.

But despite their general lack of scientific testability, fan theories nonetheless vary wildly in how seriously they're taken, both by their creators and other adherents. Very unserious theories are often referred to as "crack theories"; they may extrapolate wildly ridiculous things from the flimsiest of evidence with the intention of amusing the reader more than anything else. An example of a crack theory is the "theory" that Imprison works by suing the other Pokémon for copyright infringement, causing it to be imprisoned, which popped into my head once and caused me to write it into a stupid drabble. On the other hand, other theories are much more refined and make a point of being genuinely plausible. Some more serious theories are still only entertained as a theoretical possibility or a "what if?" thing; others permeate their adherents' interpretations of canon and are treated almost as if they were truth.

It's important here to note that whereas scientists are concerned only with the real universe that we live in, fans have two different universes to deal with: the fictional universe of their fandom, and the actual universe in which the canon was created. This gives rise to two different types of fan theories: in-universe theories and meta theories. An in-universe theory pretends that the world portrayed in canon is real and proposes a model of that world that explains observations made within the canon. Obviously, since the universe of the canon is fictional, in-universe theories can't actually be true or false so long as they don't actively contradict fundamentals of canon, though they can simply be lousy theories (more on that later). An example of a well-known in-universe theory is the theory that Ash has been in a coma since Pikachu fried those Spearow in the first episode of the anime, with the whole series since being his dream, and that this is the reason for various peculiarities such as the main characters' lack of aging, Team Rocket's wimpiness and so on. Meta theories, meanwhile, are theories about the real world, speculating on the creators' actual real-life reasons behind the creative decisions we observe in canon. Unlike in-universe theories, meta theories are definitively either true or false; we just don't know it for sure, hence the possibility of theorizing about it. One well-known meta theory is the theory that Butterfree was originally intended to evolve from Venonat, whereas Venomoth was supposed to evolve from Metapod, but that they were switched accidentally or deliberately after they were designed. In-universe theories and meta theories start to cross over a little when it comes to what in-universe interpretations the creators intended, but in-universe theories generally aren't concerned with the intentions of the creators, since an in-universe theory assumes the universe is real and has no creators.

As you can see by the examples, a "model of the universe" doesn't have to mean some big all-encompassing universal principle - it just means some hypothetical premises we propose within the universe we're working in (such as "Ash is actually in a coma" or "The creators actually designed Butterfree to evolve from Venonat") that, if taken to be true, provide a reason or mechanism that could explain the actual facts we observe in that universe (that the main characters of the anime don't appear to age, or that Butterfree and Venonat look suspiciously similar). That's what a theory fundamentally is.

What Makes a Good Theory?

A good fan theory needs three things:

  • A question. A theory needs to be about something: it must answer some sort of a "why?" or "how?" question one might pose about the canon, whether on the in-universe or meta level, such as "How come Butterfree and Venonat look so similar?" It isn't always a mystery per se - sometimes there is already an obvious explanation, but you thought of a different one - but it has to be some sort of a question calling for an explanation. The question itself isn't an important part of the theory when it's finished, exactly - in particular, the same theory can be arrived at from multiple different angles that start from different questions - but starting to make up a theory without a concrete idea of what it is you want to explain with it is likely to just end up in an aimless mess, so always have some question in mind.
  • An answer. Obviously, the theory needs to answer the question in a plausible way that makes sense and doesn't contradict the facts as we know them within the theory's universe (though less serious theories can be lax with the plausibility part). The answer is the "model of the universe" discussed above, and thus is also the speculative part - here's where you can simply make stuff up. In the Butterfree/Venonat theory's case, the answer posed is, "Because originally Venonat was supposed to evolve into Butterfree, not Venomoth."
  • Evidence. This is the key to making a theory that is actually worthwhile. Evidence here means any fact within the theory's universe - so a fact established within canon for in-universe theories, or a real-life fact about the canon and its creation for meta theories - that makes more sense in the theory's interpretation of the world than in the default interpretation, such as, "Butterfree and Venonat have identical eyes, very similar limbs, are both dark-colored and have noticeably similar poses in both the Japanese Red/Green and Yellow, despite not being related, whereas Caterpie and Metapod share almost no similarities with Butterfree, nor Venonat with Venomoth aside from being purple." It is evidence that makes one theory sound more plausible than another; without evidence, your theory has no power.

Keep in mind that it is very important that the theory have more explanatory power than the default for at least some piece of evidence: without positive evidence supporting the theory over the default assumption, your theory is meaningless. What grants a fan theory any kind of value is that "That explains everything!" feeling you get, whether it's the humourous "Pfft, well, that would explain it" variation or the awestruck "Oh, wow, it all makes so much sense now!" variation. You're not explaining anything if the stuff you purport to explain makes just as much sense without your explanation; then you're just needlessly overcomplicating things.

What is the default assumption? It is a 'null hypothesis', or, basically, the 'boring answer' to the question it poses, such as "The similarities between Venonat and Butterfree are just a coincidence". Null hypotheses vary in plausibility, completeness and general explanatory power depending on the question, and accordingly different theories need different amounts of evidence to surpass the default assumption. If you have a hard time figuring out what the default assumption for your question is, the default assumption is probably something like (for meta theories) "The creators just randomly felt like it" or "It's a coincidence", or (for in-universe theories) "It's just a legend" or "It's magic." These explanations are all extremely unsatisfying, of course, making them very easy to get the upper hand on, but you'd be surprised by the number of fan theories that don't manage even that.

A few common mistakes must be noted here:

  • Pointless cynicism does not automatically make for a good theory. Many interesting fan theories are somewhat "edgy", such as the Ash-in-a-coma theory and the theory that the R/B/Y rival's Raticate was killed in the player's battle with him on the S.S. Anne, and many people enjoy the way these theories present darker reinterpretations of a franchise as generally idealistic as Pokémon. However, frequently people just miss the point and make up a "theory" that has nothing over the default interpretation other than being edgier. "Maybe all the wild Pokémon you beat in the games really die rather than fainting!" or "Maybe healing items don't really heal the Pokémon, they just give them a boost that makes it take even more pain for them to faint!" are not proper theories unless you have facts from canon that appear to back them up better than the opposite. You can be as cynical as you like, and you can be so with perfect awareness that it's wholly opposed to the creators' intentions (so long as you're not advancing it as a meta theory), but without somewhat compelling evidence there is no theory, period.
  • Making something up wholesale is not a theory; it is fanfiction. I love fanfiction, but call it what it is. Can you write a piece of fanfiction presenting some mindbending reinterpretation of the Pokémon world? Sure. But that's not a theory unless it presents an explanation of a canonical fact that is supported by canonical evidence. You can add elements that are completely uncanonical, of course, provided they don't actively contradict canon - but those can only come into the answer part, not the question or the evidence. "Scyther hate Scizor because the latter's red coloration drives them mad" is not a theory, because Scyther don't canonically hate Scizor. A story about a trainer who trains on a route until wild Pokémon stop appearing and then realizes that he's killed all the Pokémon on the route is not a theory, because you made up the "evidence". Go ahead and write fanfiction, but remember that fanfiction is all that it is.
  • Observations are not in themselves theories. Pointing out that Ninetales' Yellow Pokédex entry says it's the reincarnation of nine saints is not a theory. Pointing out that there are ghosts in the Old Chateau is not a theory. Even pointing out simple truths that most people don't know or haven't thought about is not a theory. Remember, a theory is a model: it's a set of speculative ideas about the unseen that explain what we do see. You can make a theory about how the reason Ninetales' tails have different mystical powers (as per the Gold Pokédex entry) is that each tail contains the spirit of one of those nine saints, or a theory about who those people at the Old Chateau are and why their ghosts behave as they do, but the observations by themselves aren't theories - not even if you write an article posing dramatic questions about them. A theory has to actually answer those questions.

Creating a Theory

So when it comes to making a theory, the first thing to do is to get some proper inspiration. A lot of the time you get the idea for an explanation for something in a sudden burst of inspiration without having had any intention of making a theory beforehand, but otherwise, is there something about Pokémon that genuinely puzzles you? How about something that's just kind of strange that could have an alternative explanation? Remember to start with a question: start by figuring out the thing you're going to explain.

Once you know what you want to make a theory about, you need to actually make up the theory. The first thing to do here is to do research: find out everything there is to know about the subject, either on the in-world or meta level as appropriate. You don't want to awkwardly end up having a theory that contradicts the facts; that's just a useless theory. Know your stuff beforehand and avoid that. A good place to start finding out the facts is Bulbapedia, though not everything there is accurate, so be careful, look at the discussion pages for more information if some of the information seems suspect, and try to find independent verification. If the question is about a game mechanic, look on a site like Smogon, which does extensive game mechanics research. Some theories also call for real-world research into science or other academic fields that seem relevant. Once you've done your research, you've collected all the available evidence that should be relevant to your to-be theory.

Now that you have all the facts, you just have to create the actual answer to the theory's question. As this is the creative step, I can't really guide you through this part. Just figure out some sort of a model that explains at least some of the evidence better than the default assumption would. How much of the evidence should support your theory over the default assumption depends on the seriousness of the theory - for a crack theory you'll get away with just one piece of evidence that comes out in support of your theory even if several others including common sense stack against it, whereas a theory that wants to be taken seriously needs the evidence overall to be comfortably in its favor.

A good model does not include arbitrary exceptions; arbitrary exceptions overcomplicate a theory and make it very unconvincing. If your theory only fits the facts for some Pokémon but not others, for instance, you need to figure out some nonarbitrary distinction between the Pokémon to whom it does apply and the Pokémon to whom it doesn't, and work out how that distinction might plausibly change things with regards to your model of how things work in a way that fits with the others. This might sound daunting, but if it seems completely impossible for your particular theory, that says something about the theory; if it needs arbitrary exceptions to work, it probably isn't very good.

Maintaining a Theory

The thing about Pokémon is that it's an open canon, so any theory that doesn't solely concern stuff in some closed part of it - one particular episode of the anime, say - is in danger of being challenged or even wholly invalidated by later installments of the canon. If things change, you can either ditch the theory as outdated or adapt it to fit with the new data.

If you want to keep the theory alive and canon has introduced new details that aren't accounted for in it, think up a way they could make sense within the framework of your theory. Never do this by introducing arbitrary exceptions, as explained above; instead, try to figure out how the basis of the theory could be modified to make the new facts fit in without excluding any of the old ones, or how the new facts are distinguished in some nonarbitrary way that would sensibly change how your theory applies to them. If you find you can't stretch the theory to fit without making it ridiculously contrived in the process, you're better off letting go of the theory and creating something new.

Presenting a Theory

To present your theory to an audience, whether on a website, forum or someplace else, you need to write a reasonably thorough breakdown of it. One possible way, if you like to write stories, is to integrate the theory into a piece of fanfiction. However, if you want a more formal writeup, a basic sensible structure for it might be something like the following:

  1. What your theory purports to explain, or the "question" discussed above.
  2. Why your theory is called for. Depending on your question, this calls for different approaches:
    • If there is no proper default assumption (i.e. one that isn't "It's magic/a coincidence/the creators were in a weird mood") for the answer to the question, point out the mysteries around the question. Build up a sense of intrigue by pointing out what's interesting about this question and what could be achieved by answering it; make the reader want to know.
    • If there is an obvious but faulty default assumption, explain why the default assumption is unsatisfactory: point out evidence that doesn't fit with the default explanation or fits awkwardly.
    • If there is a perfectly satisfactory default explanation but you have an alternative theory anyway, focus on the what if factor: let the reader choose willingly to discard the default assumption for the sake of the argument. If they find your theory convincing, they can decide it's better than the default assumption by themselves; awkwardly trying to cast doubt on a perfectly reasonable default explanation will just make you sound desperate.
  3. The answer posed by your theory. Explain, as clearly as you can, what new assumptions your theory proposes - underlying mechanisms, non-canon facts, and anything else that is necessary to understand how your theory answers the basic question.
  4. How the evidence supports your theory. Explain how the evidence that doesn't rhyme with the default explanation fits more naturally into yours, or how your theory neatly ties up all the mysteries, or generally how it explains everything.
  5. Counter possible objections. Explain how any evidence that doesn't fit quite as neatly into your theory might still make sense in its model of the universe. Don't just ignore possibly contradictory evidence; it makes you sound dishonest and your theory less credible as a result.

This isn't the only way to do it, but it is a start if you're unsure how to go about making your case. Also take a look at step two of Content-Writing for Dummies for more general advice on writing (particularly website content).

An Example: Pokémon Genetics

To illustrate everything I've been talking about, I'm going to make an example of one of my own theories, namely my theory of Pokémon genetics.

The Question

How do Pokémon genetics and sex determination work, in-universe?

The default answer to this question would be "The same as ours", but that's not a good answer on the genetics side; it is pretty bizarre that Pokémon should consistently be born wholly as their mother's species rather than as some kind of a random mixture of the mother and father if their genetics are similar to ours, for a start, as this would seem to require most of their DNA to be inherited only through the female line (while this happens with mitochondrial DNA in animals, that is only a tiny fragment of DNA that essentially belongs to a separate symbiotic organism; to extend that concept to the genes that determine the entire appearance of the Pokémon seems a stretch). Meanwhile, some specific things are inherited from the father, so the father's genes must nonetheless come into it somehow.

The problem with the sex determination being the same as ours, or at least the same as one of the various sex determination systems found in real-life animals, isn't so much a problem as a suspicion: the fact it's always one half, one quarter or one eighth one sex is strikingly neat considering sex ratios in real-world animals, those few times it is evolutionarily beneficial for them to be uneven, depend on the ratio of investment that parents make in offspring of each sex - it has no reason to be anything even close to neat.

The Answer

You can read about the actual theory in more detail on the theory's own page; I won't repeat it all here. However, here's roughly how I got to that answer.

First of all, I recognized that because different aspects of the Pokémon are inherited so differently, somewhat different systems must be at play. The entirety of the part of the genome that codes for the Pokémon's body shape must be passed down strictly through the female line. Meanwhile, some combination of the species of the offspring/mother and acquired characteristics from the father must be able to induce similar characteristics in the offspring. And despite that, IVs still get inherited in a mixed, random way from both parents equally. This calls for a somewhat complex inheritance system, where different parts of the genome are passed down in different ways, on a larger scale than we see in nature in the real world.

I noticed IVs are passably similar to real-life genetics, so that part of the genome could work more or less like ours. Meanwhile, the mechanics of the primary genome, passed down straight through the female line to all offspring, were straightforward. Passing down moves through the father, however, required the father to pass genes to his offspring that 'know' what moves he knew, not just what moves he could learn. The most straightforward mechanism for this would be for there to be genes promoting the ability to use certain moves that are part of his species-defined primary genome, and those genes that are 'in use' would be epigenetically marked as such, as a way of promoting their effects when the Pokémon actively has need for them. This would then require the father's primary genome to be passed down to the offspring, with those epigenetic markers - but it would have to be mostly inactive, since the species of the offspring must still always be the species of the mother with no hint of the father. Moreover, it could not be passed down to a further offspring in any form.

I also noted that the sex ratios match up perfectly with a binary system where one of the sexes requires either one, two or three particular bits to take a particular value, while all other combinations result in the other sex. This is similar to a kind of extension of some of the genetic sex determination systems found in real-life animals - in us, for instance, a fetus becomes female by default, unless a Y chromosome (or more specifically, a particular region of the Y chromosome) is present, in which case it turns male. The way that children become randomly male or female is that since a man has only one Y chromosome in his cells, the Y chromosome can only end up in one of the daughter cells created during meiosis I while the other gets the X chromosome, causing half of his sperm cells to carry a Y chromosome (and thus be destined to result in a male offspring) and the other half to contain an X chromosome (and inevitably result in female offspring). Now, if for instance three Y-like chromosomes were involved, and all three were required to turn the fetus female while all other cases resulted in a male, then you'd have an 87.5% male sex ratio, as in the starters.

The problem, however, was that if it worked just like that, males would vary in whether they have zero, one or two Y-like chromosomes, and thus, depending on the father, the observed sex ratio for the children would be different. Each sex cell's Y-like chromosomes therefore had to be created from scratch - no matter whether they were actually present in this individual or not - before the cell division that provided the randomness factor. This meant the Y-like chromosomes must be created from the primary genome (primary genome, because the sex ratio and sexual characteristics depend on the species) during gamete formation, with the individual's own such chromosomes being discarded. And because a child's Y-like chromosomes must work with its own primary genome regardless of the species of the father, the Y-like chromsomes must come from the mother.

The Evidence

The sex ratios and inheritance patterns we see in the Pokémon games, unsurprisingly, match up with the theory - at the very least, they match better with it than with an animal-like genetics system. In particular, explaining the female-line inheritance of most of the genome is awkward if sticking to our genetics system, as it would require the majority of the active genome to be located in organelles like mitochondria, and the uncanny binary pattern of the sex ratios across all Pokémon species would be extremely unlikely under the sex determination systems we know.

This is the type of theory that's very speculative - it sees facts that don't appear to make sense and creates a completely fabricated, complex mechanism specifically to match up with the facts. Other mechanisms that do just as well explaining the facts could be created, and nothing can really mark this mechanism as more likely than another, so long as the other also matches the facts. That's okay. The important thing is beating the default assumption; as long as you do that, which I at least think I have as far as my knowledge of genetics extends, the theory is interesting - though if another theory is clearly simpler and explains the same evidence, it wins by Occam's razor.

Page last modified February 22 2017 at 04:53 UTC