Restrictions of Consistency

This is cheating in more ways than one.

The year of 2010 was teeming with experimental approaches to jumpstart the medium, be it the attempts at original storylines, the omnibus format, or the use of unconventional execution. While such bursts of creativity have given us some memorable stories, most are remembered not for their ingenuity but for their spectacular failure to pull through. Perhaps the most egregious error of many a show was the inconsistency through its run. The lack of even simple tenets of consistent storytelling has severely crippled any gains these experiments may have achieved.

Narrative consistency, at its core, is simply how well parts of the story are connected together to make it believable. I like to break this into three specific aspects. The first is that rules set by the show itself should not be broken unless under extraordinary circumstances. Secondly, events that precede other events must be taken into account. Finally, characters need to go through logical characterization and development. These three things establish the audience’s suspension of disbelief, allowing us to believe certain situations in the story while rejecting others. This is why we are more willing to accept something outrageous if the show hints at such an outcome as possible.

The first of these, the foundation of the story, is often the most crucial. The nature of the story should often be taken at face value unless given considerable doubt. When we are told that everyone is dead in Angel Beats, we take for granted that they really are dead. If we see big mechas jumping around and moving as if they were ten times smaller, we ignore gravity for a few minutes to bathe in the chaotic battle. There are many things that are impossible in reality that we can easily accept in fiction, as long as they maintain that such things are as strict as the laws of the universe. The restriction, in this case, is that these rules should never be broken.

An example of how this can go awry is Star Driver, which holds none of its laws to be true except the obligation to end every episode with a fight scene. Even if the contradictions in the original premise are disregarded, one of the most obvious problems arises when Wako very explicitly states that no one has ever survived after using Samekh. Sugata’s resulting coma and recovery after using it completely breaks this rule, making us wonder why such a warning was even issued in the first place. It creates a jarring break in the story and makes the drama of four to five episodes of an already absurd series even more laughable.

Thanks for interrupting the only good plot.

Another part of the foundation laying aspect is the audience’s expectation of where the story will go. It is possible to run a complex story that concurrently runs multiple subplots, but we need to be prepared for that. The main plot usually should be introduced or hinted at early on, and plots should never be interrupted unless the subplot contributes meaningfully back to the main one. It is hard enough to juggle one plot at times; two can be suicidal.

Angel Beats betrayed many expectations not because it didn’t have plots but that it didn’t know which plot was important. On one hand, it wanted to play up the unknown nature of the world explicitly through confrontations between people and battles with the system. On the other, it wanted to be emotionally manipulative by making it some kind of purgatory. In between it tried to cram in a public service announcement. This lack of direction resulted in all three plots constantly vying for attention, interrupting each other at inopportune times.

The restrictions of the second aspect, the sequence of events, are far stricter than that of the first. While the rules of the story can be as crazy as one’s imagination as long as they don’t conflict each other, events have a consistent progression. They are bounded by the human expectation that life, and as a result the lives of the characters, is cumulative. All too often a story is seen as a bunch of individual scenes that are strung together in the least confusing fashion. This creates situations where a major event that should clearly affect the rest of the story fails to have any effect at all.

One of the worst offender in recent memory is Ore no Imouto, which gave us three events that should have been turning points: episode three, five, and eight. Episode three opens us up to the idea that Kirino can sooner or later come to accept herself regardless of the opinion of others. Episode five presses the opposite: that other should come to accept her for who she is. Episode eight heaps upon Kirino the obligations of a published book, which should be a monumental achievement. Instead, the story veers back to the status quo, resulting in no dramatic event having any impact on the characters.

Who are you? What have you done to Kirino?

When these problems are stacked with mistakes in characterization, stories begin to fall apart. As creatures who greatly empathize and attach themselves to similar beings, humans place great value in the consistent nature of fictional characters. Nothing is more obviously wrong and out of character than a stoic character suddenly bouncing around and dragging the protagonists all over the place. At times, this means that the creators of the series need to give more attention to their characters, as while minor plot holes and story problems can sometimes be handwaved away, even small character problems can either snowball into polarizing opinions or absurd representations.

There are many instances of this that can be brought up, but obvious examples can be seen in the final episodes of Angel Beats and Ore no Imouto. Characters that have been acting one way for eleven to twelve straight episodes suddenly change their attitude with nary a piece of development. No characterization is given as to why they’ve suddenly changed their tune, save perhaps the obvious meta excuse. Even more puzzling is Sugata’s change in character in Star Driver for exactly one episode before reverting back to his old self. Such situations cheapen any kind of logical character development, which defeats the purpose of drama.

Although I’ve only brought up three examples, this inability to adhere to the basics of storytelling is pervasive. Too many shows last year tried too hard to cater to specific storytelling approaches that were nonsensical but common, or else they attempted to experiment with the format without getting even the fundamentals right. It is perhaps ironic that in contrast to 2010, 2009 gave us Cross Game and Kemono no Souja Erin, two shows that proved that the tried and true formula of basic storytelling was still one of the most solid ways to tell a story. I’m all for experimentation in anime, but it’s an uphill battle if no one can create good stories in the first place.


1. There are spoilers for the tagged anime, although I frankly believe I saved the four or five of you who haven’t watched all three shows from banging your head on the wall too many times when you get around to watching them.

2. There are, of course, hazy exceptions that stretch the rules I’ve established here. Unreliable narratives present doubt in the setting. Episodic series are usually tangentially related and don’t develop each other. Good plot twists are usually unpredictable. I still stand by the fact that any good series follows these rules in some shape or form. No matter how unreliable someone is, we don’t expect a simple slice-of-life series to become an action packed mecha one. Good episodic series either tie their episodes thematically or do in fact subtly develop their characters, and good plot twists are usually logical on a second watch. Yojou-han Shinwa Taikei is probably the best 2010 example of a show that pushed boundaries while still maintaining solid storytelling. Yes, you can stretch the rules, but we need people to adhere to the rules first before bending them.

12 Responses to “Restrictions of Consistency”

  1. One distinction that people may have to make is that there is the kind of bad story telling aspects Elineas is pointing out here, and then there’s some general plot oversights, continuity errors, and other such stuff. The importance of the distinction is that the latter are usually just brain dead fuck-ups typically the result of a creator with a long running series just forgetting some detail that was mentioned ages ago. Bleach is notorious for this, but then again, so is Star Trek. Whether or not it matters to the story is up to the individual. What Elineas is really harping on here is a failure of the universal structure of the story.

    This can be broken and have it still remain believable, but typically if that is going to happen it is because the storyteller is introducing a new direction for the show to go. For example, Berserk (anime, not manga) takes the viewer in a direction that seems to be based on a rather simple theme: humans are capable of conquering all and becoming masters of their own destiny. The reason I say “humans” is that the anime makes it a point to remove a lot of the fantastic and mythical elements to make it seem like Berserk is based more in “reality” than similar fantasy shows. Sure, Guts can do some amazing things, but it’s explained well and the limitations of humanity are constant reminders. Then the Eclipse happens, and the show goes and breaks every rule it had established, busting the controlling story elements that were in place before in order to go in a new direction.

    As is true in a lot of things, logic is what dictates whether the creator is doing well or just fucking up.

  2. The challenge, generally, is writing stories that are surprising. The standard plot twist often involves a bending of logic, or, at the very least, a “correction” to what the viewer expected given the way logic was emphasized, and what was explicitly explained, and what was merely hinted at. That’s why I’m always suspicious of stories where the plot twists are the point (Index, Angel Beats, Myself; Yourself, among others), since it doesn’t take much for bending of logic to turn into breaking of logic, and it takes competent writing to prevent that. Code Geass, as well, I guess counts as an example of that.

    I still kinda find myself coming down on the side of encouraging ambition and dealing with whatever poor writing gets attached to that afterwards. Ideas tend to evolve, and there’s a certain amount of polishing that goes into any creative process. Even something like Puella Magi Madoka Magica, I essentially see as an evolution of a set of ideas and concepts Shinbo has used in previous anime. I doubt Madoka Magica could have been made by someone without Shinbo’s experience. He’s one of anime’s more experimental directors, and while he rarely makes bad anime, he doesn’t frequently make masterpieces either. I guess my point is that I don’t have a huge problem with writers and directors not getting it right the first time when they try to challenge convention or incorporate novel ideas. There’s still value to be had in failed experiments.

  3. Good article. I believe plot twists are always a little bit gimmicky in nature. Everyone enjoys a good plot twist, but for something to really be memorable, it needs to be good in other things.

  4. In general though, I think there first needs to be a distinction between animes that revolve around their plot, and animes that don’t. For example, Angel Beats! is a show where the plot was a central component of the show. However, OreImo really isn’t, and I think the anime made it clear from the beginning that it’s far more appropriate to consider OreImo in the realm of Lucky Star or Azumanga Daioh.

    On the other hand, I can’t agree with some of what you’re saying in your post, mostly because a lot of animes thrive on the idea that characters don’t know the whole picture or have an erroneous view of the world. For example, what Wako said in Star Driver. To her knowledge, and to the knowledge of everyone, no one has apprivoised Samekh and come alive, but that doesn’t mean that the anime itself stated that as a fact. Or in Shiki, where it’s pretty much stated that everyone in Kanemasa is a Shiki, but that turns out to be wrong as well.

    I also do find it rather curious because a lot of work in 2010 was based off of manga/light novels, which does raise the question of where any narrative inconsistency comes from. The adaptation or the source?

    I’m still rather unconvinced that 2010 was a year of narrative inconsistency though, as bad storytelling is pervasive in a lot of animes.

  5. This is certainly a detailed description of what consistency is supposed to be all about in terms of textbook storytelling applied to anime. That is, of course, as long as you believe all stories need to follow such guidelines in order to be recognized as “good” or else.

    Honestly, I’m not sure that’s necessarily true…or even realistic.

    I’ve gradually come to the realization that even though all of the above sounds logical enough on paper…the so-called “rules” of consistency or basic storytelling can be far more arbitrary and artificial upon closer inspection than what they seem on the surface.

    Far too often, viewers tend to expect more consistency and logic from fictional stories and fictional characters than what they are prepared to accept or tolerate from real events and actual human beings.

    Life can throw completely unexpected curve balls at any of us that nobody saw coming and which serve no obvious greater purpose, intelligent people can suddenly act in frankly irrational ways without somebody having to stop and explain in precise detail why that would be the case, someone can say they’ll do X and then turn around to do Y, a “stoic” salaryman can suddenly go nuts and gun down innocent pedestrians, numerous opportunities for “personal development” can be left to rot and disappear, etc. The list of potential examples is literally endless.

    Not everything that transpires in the world is always as clean nor as neat in terms of causality as “good” storytelling would like us to pretend it should be, so why should we go out of our way to demand absolute consistency from fiction if we don’t necessarily demand it from reality?

    I guess that has to do with the position of near-omniscience we all tend to assume as members of an audience, expecting that all of the necessary information has to be provided in due time and in a friendly fashion, combined with the desire to understand or explain everything we read or watch from said vantage point. And if something goes against such expectations, it deserves to be automatically mocked or even denied. But life doesn’t always give us such luxuries, or does it?

    Life can be so complex and so inscrutable that real or apparent randomness and even contradictions tend to be quietly accepted or rationalized without question while, on the other hand, fictional stories are usually criticized for failing to live up to what we perceive as the “proper” way to portray a certain situation or develop character.

    As something of a rather cynical conclusion, you could say I’ve become rather accepting of precisely what this post is supposed to against: inconsistency.

    I believe that almost any fictional story can still be sincerely appreciated on some level, even if it breaks or bends our suspension of disbelief due to numerous inconsistencies, as long as its themes remain interesting enough and can, if possible, be logically interpreted in spite of how much unorthodox storytelling or sloppiness is involved.

  6. @Mystlord

    It’s true that Wako was merely saying that no one who has apprivoised Samekh and has lived after it. It’s only “for the record”, and not a fact per se. But the bigger question to me is, what’s the purpose for setting such a record? Is it so important to see that Sugata is alive after using it? Or, is it only to dump a pathetic plot twist and convenient tension? Looks to me that it’s only a temporary barrier to show that Sugata supposedly can’t use it, and thus makes it easier for the Glittering Crux to defeat our lulzy Galaxy Pretty Boy.

    But it turns out that he can, and the baddies are thinking “Okay, we’re screwed.” Is that all? If that’s the only reason to slot in the plot twist, then it’s poorly underutilized, and it makes viewers more cynical towards the show.

    I don’t really agree with Elineas that the use of Samekh leading to death is factual, but it’s actually an example of another problem: piss-poor plot twists.

  7. Thanks for the extra clarification TIF, that addendum is very helpful.

    Indeed. Plot twists are powerful because they’re unexpected, but they need to be believable, and this believability is built by past expectations the story has already built. No right minded person would say a conflict in a fantasy based story being resolved by someone pulling out a nuclear warhead unless we’ve already been warned that technological advances have also been made in the world. That’s an extreme example, but I think it serves to explain the limitations that story impose upon themselves the instant they begin.

    As for experimentation, my problem is even the most brilliant of ideas can’t shine without a figment of logical storytelling. Puella Magi Madoka Magica is a development that could only be made with Shinbo’s experience, but it doesn’t excuse people from breaking even the most basic pieces of storytelling (as an aside, I think Shinbo is extremely good at establishing his own rules, so he doesn’t fall under too much fire from storytelling consistency). I guess what I really want is for many a director or scriptwriter to keep good habits in mind when they develop ideas and set off red flags when they break them. They can be unpolished, but some choices are so blatantly wrong they should have been caught before air.

    Nope, that doesn’t give Oreimo a free pass. Episodic anime are built upon established characterization and, if the story calls for it, slow but progressive development. If we talk about the comedy side, most small skits are made in an internally confined manner. They are often done over laughably minor matters or are resolved extremely quickly. This makes them funny but inconsequential segments of life, which are fine because they don’t affect larger narratives too much. But once you tread toward dramatic affairs, you need to make it count. Shows like Azumanga Daioh, Kannagi, and Lucky Star take their drama seriously by seeing it through to the end, creating significant development or reaffirming our viewpoint of characters. Other shows like Aria or Honey and Clover make sure their developments affect the rest of the series down the line. Thematic episodics like Mushishi and Kino’s Journey give us a glimpse of their respective characters’ back stories and uses this to establish their neutral nature, allowing us to link episodes thematically. When Oreimo turns to drama, it never creates a development it should make on the characters, and we’re left with drama we find highly ineffectual and disjointed. Drama isn’t as easily played off as comedy is; it only has impact when it actually affects the characters.

    As for the unreliability of viewpoints, yes, I understand that’s very common. The unreliable narrator is one of the oldest narrative devices. What I am arguing is that a story creates audience expectations that need to be answered appropriately. We need not take what the characters say as the truth, but we should assume that they are working to the best of their knowledge. We should also assume that the story is trying to make whatever it mentions a working part of its world. This means at every major step of the way that causes a significant event to happen our expectations need to be built upon, realized, or changed realistically. Wako’s statement is valid, but since we have no reason to think she’s unreliable or lying, stating it adds a rule to the world: in the known history of the world, no one has survived from Samekh. This gives us the expectation that to overturn that we either need insurmountable odds or a justifiable reason to believe Wako is misguided. Star Driver does neither; it pops Sugata back up as if nothing happened, creates a bunch of cheap drama over his “death,” and then makes the next few episodes laughable as Sugata continues to use it as if there were no drawbacks. Every little event in a story builds upwards, we just interpret reliable information with unreliable information differently. And as for AC, yes, using the word “fact” seems to have thrown off my argument.

    And I only point out 2010 as being particularly egregious because experimentation often makes basic storytelling holes stand out even more.

    I think I may have addressed my point in my response to Mystlord, but I’ll answer your comment by saying that fiction demands consistency because inconsistency in reality is the exception, not the rule. We expect the people we know well to not suddenly go insane and try to strangle us. It’s certainly possible, and I bet it’s happened at some point in human history, but we function in life by understanding reality’s rules cumulatively and knowing that there will forever be things we do not know. A story, as a result, plays off what we do know because that is something we can logically write about.

    Really, the entire argument (and stories in general) hinges upon the audience’s expectation of logical events, which I wish I could elaborate on but I don’t think I’m qualified to answer “why do we need fiction?” Logical events don’t need to be realistic, but they need to make sense. This can be accepted at a meta level too. For instance, my “gravity works differently” example for mecha is fine because we know it’s the only way to make the story work. But when gravity starts acting differently the next time and the mecha comes collapsing down, we wonder “why didn’t that happen last time?” because logic dictates that it this should happen all the time, not just whenever the creators feel like it. I also stand by the fact that even fantastic shows that are non-linear, unpredictable or unconventional abide to consistency, because part of how they make themselves fantastic is by expanding its scope at the very beginning and allowing us to accept its unconventional storytelling, abiding by the first aspect of consistency to allow the other two to make sense.

  8. @Elineas:

    But even occasional inconsistencies are still a key part of life that we must deal with…and there can be an underlying logic, whether it is contextual or external, to even the most inconsistent series of fictional events. “Making sense” isn’t always a synonym of “being consistent” and I would say anime lacks the latter more than the former.

    You can certainly defend consistency the most when it comes to things like the laws of physics and scientific phenomena in general, which are apparently absolute, but in terms of plot and characterization (as well as, by extension, writing and direction) what “makes sense” shouldn’t be so clear-cut.

    Why? Because those issues are subject to perception and opinion, often influenced by our own personal expectations and interpretations, which are rarely 100% accurate representations of the facts. All of us focus on slightly to vastly different elements, even when we’re using the same limited amount of information, and our expectations or conclusions can diverge as much as they can converge.

    In reality, there have been countless minor and major examples of people snapping against their co-workers or family members without much of a warning, often acting against what outside observers would classify as their real or perceived personalities and even going against their own best interests. Explanations are often hinted at or presented after the fact by either authorities or the media, but often they turn out to be little more than opinions and speculations. Not absolute and complete truths.

    And thus, coming back to fiction…why is it usually so hard to accept that there can be many things we will never know and many questions never answered? In reality, an infinite number of persisting questions and wasted opportunities make up each of our respective lives and experiences. In fiction, whatever subplots are introduced, partially developed and ultimately left unresolved, for instance, actually pale in comparison to such things. Why do all the blanks need to be filled? You can tell a story that doesn’t satisfy the audience’s reasonable expectations and that isn’t “unbelievable” by default.

    A character can be important for, say, a single episode of a show and never play any significant role again or be reduced to a bystander and that isn’t inherently “senseless” or impossible to take seriously. They can also fail to act according to what the audience (or a larger or smaller portion of it) expected without automatically being “out of character” or “poorly written” as a result. Characters do not need to make the “right” choice all of the time, much less under emotional duress, even if they’re supposed to be smart.

    The most you can argue, I suppose, is that asking questions and creating characters or subplots that do not live up to their full potential (or, rather, audience expectations) can literally turn out to be a disappointing waste of time, perhaps little more than filler depending on your viewpoint…but then again, I would say life can often be like that as well.

  9. @AH

    The thing about entertainment is that it’s not like real life. No tv show or movie can quite emulate the experience of hundreds of little intereactions with others across many years. It simply doesn’t have the time.

    A character in a show is highly controlled, highly contrived puppet of the storyteller, and the segments of the character’s life that the storyteller chooses to show should reflect something meaningful. So when a character snaps, it should not just be because he/she had a bad day. The event should tie into the main story, an overarching theme or something else the narrative has already been established or is about to establish.

    Since a character is just a puppet, we should know enough about the character to predict what he/she will do or at least vaguely clued-in in some way. Not all the blanks should be filled, but it’s the duty of the storyteller to make sure the audience has some sense of who a character is and why he/she does what he/she does. So, if a character snaps, and the storyteller intends it to be in responsse of a bad day, show a snippet of the character spilling coffee on himself, getting dumped by his girlfriend, and watching his car get towed from a 2 hour zone for not leaving in 2 hours and 1 minute. Character’s need not make the “right” choice all the time, but if there’s an aberration, there better be and explanation.

    As you can see, I wholly disagree with the notion that “life is like that” since creating entertainment is by its very nature a very manipulative. If the audience starts seeing strings of the puppets, the creators have failed to manipulate.

  10. @Shadowmage:

    Since those are two different ways of looking at fiction and enjoying entertainment, I suppose it all depends on whether you’re more interested in judging the quality of the puppeteering, in and of itself, than in what the actual puppets are doing.

    This doesn’t meant I’m absolutely indifferent to such concerns, no, I just consider them secondary. I tend to prefer an interesting story with good themes, even if it happens to be told more or less inconsistently, to an uninteresting one told consistently that has taken far less risks.

    And thus, as it does tend to happen, the explanations or the hints provided in any given series can easily turn out to be insufficient for someone like you because the creator’s manipulations didn’t meet the highest standards of storytelling (say, the information was limited, parts of it were missing, misleading, few and far between or not as clear and obvious as they should ideally be) and yet they may very well be enough for me, as long as they still fit within the overall framework of an interesting story, because of my acceptance of the idea of applying to fiction some of the limitations imposed by life on our perceptions.

    Several of you could call me naive, in a certain sense, but I prefer to consider myself to be “post-cynical” instead: I’m usually aware of what the cynical point of view would be and can resort to all those criticisms in case of necessity…but can easily choose to put them on the back burner if that’s bringing more harm than good.

  11. @ AH

    To be honest, I don’t disagree with your position. I do think that something that has incredible raw potential can outweight any kind of flaws in its presentation. The problem is that such stories or characters are incredibly rare. It’s hard for a story to just rely on its brilliant concept and work as a meaningful and entertaining piece. It’s far more common to see the said brilliant idea wasted due to lacksluster presentation, if anything.

  12. @AC:
    Well first off it obviously establishes the fact that Wako isn’t aware of a lot of things going on with Cybodies. We knew from earlier that she obtained her Cybody’s memories by apprivoising from it, but the extent of those memories are unknown. Furthermore, it sets up an interesting dynamic where we think that there’s something special about Sugata in that he was able to apprivoise Samekh and live, but that image gets completely shot down when it’s revealed that it’s just because of a natural Cybody regeneration. It emphasizes the fact that these kids are just kids, which hammers in the fact that the fate of these kids is all the more tragic.

    Of course that’s just my interpretation, but it could be wrong. The other thing here is that Star Driver is a show that likes to build up to certain events, and there is a lot going on in the show that at first glance seems to be completely throwaway, like for example that episode about the girl who hates outsiders. It’s really only once Sugata wakes up a few episodes later that the importance of the islander-outsider conflict is revealed. At first glance, Star Driver might just seem to be a series of episodes that are loosely related, but there’s more going on than just that.

    If you really want to see piss poor plot twists, I think Angel Beats! is a far better example. Stuff there just doesn’t make sense and stuff there seems to be inserted just for the drama factor.


    Ok so first on the OreImo thing. I probably should have explained that point more, but I never stated that the fact that we need to consider OreImo more along the lines of Lucky Star gives it a “free pass”. All I’m saying is that there is a difference in how we have to consider the “plot” elements presented in each type of anime, because the end goal of each anime is different. However, consistent characterization is a definite must, mostly because that’s pretty much all OreImo/Lucky Star has going for it. Those shows thrive on building up a stable archetype for each character, and going from there, as opposed to a show like Bleach, which builds the plot and goes from there. The end of OreImo was a gigantic mess mostly because there was little build up to a rather radical shift in Kirino’s character, and that consequently made the ending and resulting drama seem really silly. However, what the focus of OreImo does excuse it from is the real lack of any central plot or conflict, though OreImo decided to insert it anyway just for the heck of it.

    I think most of what you said with regards to Star Driver is addressed in my response to AC, so look above.

Leave a Reply

Gravatar enabled.