On Katawa Shoujo as a Theoretical Anime


Which work from another medium would you like to see animated?


People ask this question pretty regularly, enough so that I eventually decided to do a little thought experiment for whenever it came up again. So here’s my unrealistic, fantasy-world proposal. I guess there’s light spoilers regarding dramatic structure? Anyway.

Design Proposal for a Katawa Shoujo Adaptation by Kyoto Animation


Katawa Shoujo, the recent English-language visual novel, has managed to develop a substantial and passionate fan base in the Western world. It’s distinctive in the thoughtful treatment of its concept, the acuity of its writing, and the diverse and relevant themes presented throughout. Though there are significant hurdles presented in successfully adapting this work, I believe an adaptation which captures the spirit of the visual novel would further bolster KyoAni’s reputation as a creator of emotionally resonant and deeply personal young adult stories.

Immediate thoughts:

The title would obviously have to be changed, as it’s by its nature offensive in the original Japanese. More critically, I believe an adaptation of this title would have to address the fundamental narrative failing of most visual novel adaptations – the consolidation of multiple storylines into a single linear narrative. In order to maintain romantic pacing and integrity of character writing, I propose a new solution – separate the central protagonist, Hisao, into five separate characters to reflect the five separate routes, and reorganize the five narratives so they interweave with each other while all progressing as separate love stories.

I believe there are generally two hurdles that prevent this approach. First, the tendency to characterize visual novel protagonists as blank slates, in order to aid audience surrogacy. I believe Katawa Shoujo lacks this problem – in fact, in my experience, the five routes each make use of a substantially different Hisao, with different desires, hangups, and personalities.

Secondly, it requires substantial rewriting and reorganizing of the visual novel content. This cannot be avoided – I believe it is the necessary price to pay if we wish to make a show that will stand the test of time.

One last thought – the broadening of the story into one larger narrative presents a few opportunities of perspective, with the largest being the potential to shift the Shizune narrative closer to Misha’s perspective. Considering her emotional arc is already the key conflict of the route, I believe holding closer to her perspective would improve the dramatic consistency of that line, along with offering the opportunity for a more meaningful exploration of gender identity. Something to consider.

Further planning:

As a strict adaptation is not advisable for this work, further thought must be put into the successful transfer of the original’s strength to the animated medium. A brief outline of those strengths might contain:

Organic Writing: Perhaps more than anything else, the core emotional resonance of Katawa Shoujo is derived from the utterly believable and naturalistic dialogue of the protagonists. Not all this writing can be conserved, nor should it – but the need for subtle and naturalistic pacing to match this dialogue’s emotional rhythm necessitates adapting this work as a two-season show. Consider the pace as slightly more propulsive than Hyouka, but applied to a work that’s attempting to fully execute five separate love stories.

Thematic Resonance and Acuity: The deft treatment of disabilities, as well as the way the source moves beyond that to actually be more concerned with themes of human connection, power structures in relationships, our various ways of dealing with our pasts, and emotional honesty, is another remarkable strength of the source. Here, we simply should adapt the best we can – these strengths underline the core conflicts of the narrative, and are virtually impossible to lose in translation.

Frank and Empathetic Depictions of Relationships in All Stages: Here we run into a more difficult question. Can the adult content be excised from the text without diminishing it? My response is a resounding no – not only are the game’s frank treatments of sexuality one of its greatest strengths, but the ways these events are woven into the narrative make them virtually unavoidable in the majority of the storylines. Only Shizune’s route contains adult content that seems to veer into titillation – all other such scenes further our understanding of the characters, provide crucial narrative turns, and exist as some of the most honest and touching moments of the narrative. They will be handled tastefully, and with much greater restraint than the game, but they will stay.

Regarding “depictions of relationships in all stages,” the game’s narrative actually provides some assistance here. Not all the stories operate on the same timeline – for instance, the narrative arc of Hanako’s route concludes in the interim between acts two and three of Lilly’s, and the scattered events of Shizune’s route could be rearranged virtually at will to better enable a steady rise and fall of the various dramas throughout. The various narrative experiments of the text (the even-handed positivity and maturity of Lilly’s route that is only rarely interrupted by narrative conflict, the continuous rise of narrative tension in Hanako’s route to one single emotional peak) will hopefully lend themselves well to the intermingling of narrative arcs this adaptation will require.

As a brief initial proposal, Emi and Hanako’s storylines might possibly hit more of their narrative beats in the first season, with Lilly’s and Shizune’s stretching from this period throughout the second season (though obviously not removing Emi and Hanako’s arcs – as I said, a positive portrayal of already-existing relationships is one of the great strengths of this work, and the anime adaptation offers further opportunity to explore a romantically healthy slice-of-life emotional space), with Rin’s progressing the slowest and her art exhibition and the aftermath perhaps offering a final gathering of the characters and last dramatic resolution.

Final Thoughts:

Obviously, successfully adapting such a sensitive text is always a risky proposition, and both the changes necessary and the changes that can’t be made further complicate this adaptation. However, I believe that in light of the source material, the final result could easily stand as an artistic pillar of both our studio and the medium itself. This challenge is a risk worth taking.

Clannad – A Critical Overview on Character Development, Dramatic Structure, and Thematic Dissonance


What’s your problem with Clannad? Have you no soul?!?


Quite likely. However, my main problems are that up until halfway through Afterstory, it’s a combination of cliched, one-note characters, repetitive slapstick, and maudlin sub-Angel Beats melodrama. Then it gets very interesting and unique for about ten episodes, then it flips the audience off with an ending that invalidates all the good parts.


But isn’t that just, like, your opinion, man?


Let’s take this one item at a time.

Cliched one-note characters and repetitive slapstick, I don’t think you can really defend against. It’s hard to dispute that for the greater part of the series, most of the characters are defined by one core attribute – Kyou is a tsundere, Kotomi is the Rei-clone, Nagisa is straight moe, etc. The “repetitive slapstick” is even less arguable, since that just is true – some people find this more funny than others, but the fact is this show repeats its jokes constantly, and most of them are of the broad physical comedy variety. Easy, broad humor is a problem common to a lot of anime, but that doesn’t make it less of an issue.

The problem with the drama is that the show doesn’t give you a reason to care about the characters before telling a sad story – it introduces them as their character type, and they remain that type, but sad things happen around them. Fuko is only ever “ditzy girl who likes starfish”, but we are expected to feel sorry for her because life is sad. That’s not really how characterization and empathy work in storytelling – this is clearly subjective (I mean, a lot of people think Angel Beats is good), but I can pretty confidently say this show hasn’t learned the give and take of characterization and drama that Disappearance of Haruhi or Toradora have a mastery of. A tragic backstory doesn’t create character depth unless you see that depth in the characters themselves, and the side-arc characters are pretty uniformly shallow.

Then there’s Afterstory. For the second half of this season, I was actually extremely impressed with the show. It went beyond the usual high school daily life experience, showing things like the pressure/pride of personal responsibility and the tiring but satisfying honor of a manual blue-collar job. What other anime covers this stuff? It was also handled with much more grace and subtlety than the prior arcs – it felt like the show had an entirely new, much more talented director. The death of Nagisa actually struck me, since the second half had been full of great character-building moments between her and Tomoya, and the episode where Tomoya doesn’t know what to do with the daughter he’s abandoned is in my mind one of the strongest episodes of any anime. Even the stuff with his father is deftly done.

When Ushio became sick, I figured the show was finally pulling its strands together – the themes of nostalgia, of embracing the past while accepting sadness and moving forward, and the recurring references to the hospital/hill that saved Nagisa were all going to come together, and Tomoya would just barely save Ushio by getting her to the hospital, the “place where dreams come true” – this would be Nagisa’s last gift to him. It would be bittersweet, since Nagisa would still be gone, but life is full of sadness, and you have to learn to cherish your past without being captured by it.

Instead, all of that raw character building and sharp reflection on the earlier themes of the show turns out to be a dream because magic, and everyone lives happily ever after. Not only is this literally deus ex machina (god just decides to save them because they’ve been good all year), which is never good storytelling, but it also undercuts the themes of the show. The entire strength of the last act had been built on mono no aware and the idea that unlike high school, life doesn’t have any easy answers… and then it concludes with an incredibly easy answer. Not cool.

I think there are many moments of this show that are well-directed, and I think when it’s working on the main Tomoya/Nagisa plot, it’s actually a pretty good and sometimes extremely good show, minus that ending. But I also think it’s a very flawed show, that it makes a lot of too-easy narrative and character choices, and that many parts of it simply don’t work storytelling-wise.


So basically you’re a heartless monster. But adapting Visual Novels is really tough, as you’ve discussed in the past. Are Clannad’s problems even solvable?


The answer here is, “yes, but not easily.” Still, let’s see what we can do.

The main, huge problem is that by adopting all the paths of a Visual Novel, they destroyed the pacing of the storyline and added a huge amount of superfluous plot and very poor melodrama. From the first season, I would cut all episodes from the Fuko storyline up through the end of the Kotomi storyline. I would also cut those two characters entirely, since they’re the worst offenders on the “just exist to be moe” scale and add nothing to any other part of the series. I’d perform the same surgery on the second season, cutting out all the episodes from the second until when they finish the superfluous side arcs.

This would condense the series to roughly one 26 episode season, and do wonders for the pacing already. This would also indirectly help Nagisa’s character a great deal, as her character development would now be continuous, as opposed to randomly stopping for 10-12 episode stretches. Tomoya would also lose a lot of his generic VN protagonist Woman Fixer absurdity. There’s still work to be done, though.

Sunohara would have to be fixed. His character is two jokes repeated over and over (Sunohara likes girls LOL, Sunohara gets hit LOL), and his character development arc requires him to randomly become an asshole for several episodes and then be “fixed” by Tomoya. Cut his sister, give him one or two actually good traits, and set his motivation arc in place from the first couple episodes. The resolution of the baseball team storyline was actually one of the better moments of the early stuff, so focus on what made this strong – his temper, his convictions, his bond with Tomoya.

Nagisa also needs work – she eventually comes into her own as a character, but it wouldn’t hurt to have a couple more defining traits than “cute, helpless moeblob” for the first half of the series. No, liking the dango family is not a personality – but it might be the lead-in to one. Find something to make her stand out and have a better-defined character arc of her own – right now, character arcs happen around her, but her own agency is very minimal.

Tomoyo is fine. Kyou is a generic tsundere, but she also gets some of the most honest conversations with Tomoya, so she’s also probably fine. Ryou probably doesn’t need to exist… no, actually, it would probably be better if Ryou were Kyou’s male twin. This would kill any last vestiges of haremness, and delete the last unnecessary moeblob. I’d suggest something even more drastic, like give male Ryou a crush on Sunohara, but anime writers apparently can’t handle gay characters without making them into tasteless jokes, so I’ll just leave that alone.

Finally, we’d have to fix the ending. Adding a “magic fixes everything” ending cheapens the themes of honest work, perseverance, and helping to hold each other up that are present all throughout After Story. Bringing Nagisa back to life destroys the significance of Tomoya’s character growth in the last, best act of the show. Instead, have that scene where Tomoya questions if he should have met Nagisa at all simply inspire him to get back up. Tomoya runs to the hospital on the hill, cradling his (unconscious, but not dead) daughter as he reflects on his time with Nagisa and all the people he’s met in this city. He wants to hate this place, but he can’t; like Nagisa said, this is where they were born, and there’s too much of him, too much of Nagisa in this place. He reaches the hospital where Nagisa was saved and begs them to save Ushio; she just barely pulls through. Nagisa is still gone, but his memories of her helped him save the daughter they were meant to raise together, in the city he’s come to love.

This maintains the strength of the last act while actually tying together the earlier themes and foreshadowing. Plus, by cutting the magical glowing balls, Tomoya’s early helpfulness can be resolved by something that actually helps the story – have it be his way of attempting to be the opposite of his father. If Nagisa makes him realize this, it would even help her character, too.

There’s an incredible show hiding somewhere in Clannad, but the humor, pacing, and early melodrama make it very hard to find. I think as a one season show that abandoned directly adopting the VN and instead attempted to tell a single story well, it could be something truly great.

Square Pegs, Round Holes, and the Art of Adaptation

Hey guys. There have been a couple posts recently (well, semi-recently now) about adaptation, and while they kind of talk about what adaptation is, I don’t think they really went into what makes adaptation so interesting artistically. And I have a lot of thoughts about that!

I was originally writing this post as a comment for one of the article links, but I figured that would probably get buried, and the artistic side of this is interesting enough to warrant its own discussion. My thoughts here aren’t law, or based in massive industry knowledge – I’m just a dude who likes stories a whole lot, and thinks about storytelling pretty much constantly. And I have far less of a single thesis here than I did with my Nisemonogatari writeup – if anything, my points are mainly that adaptation is both a craft and an art unto itself, and that understanding mediums is critical to understanding how and why adaptations work or don’t. The main point here is to promote discussion and your own opinions, not just say How It Is. Anyway, let’s get to it. What’s up with adaptation?

Why Are You Even Trying

One of the main reasons I find adaptation interesting is because, although I know this isn’t the actual intent, the very act of adapting something makes me think, “Why? Was it not suited to its original medium?” Every medium has different strengths and weaknesses, and most of the time, a truly great piece of art works partially because it takes advantage of the unique strengths of its medium. An adaptation seeks to take the “essence” of some work and translate it to a different format – but this does not imply a perfect 1 to 1 transformation. In fact, a “perfect” adaptation is very rarely the best possible adaptation, and some restructuring or refocusing is normally required to make the best work possible.

And obviously, from a production perspective, adaptations are chosen not because they make for artistic challenges, but because a proven property will sell regardless of the medium. But that doesn’t make the challenge less interesting, or the results less respectable. A good adaptation requires both a keen understanding of the work you’re adapting, as well as understanding of the strengths and weaknesses of the mediums you’re converting both from and to. Misunderstand these strengths, and even an adaptation of a fantastic work can fall completely flat.

So let’s get into those mediums. I’ll only cover a point or two each here, but there are a ton of angles to take on what makes each of these mediums interesting to adapt, so I hope you guys offer some of your own.


Manga has traditionally been the largest source of adaptations, though the ascent of Light Novels has challenged its position. But it still reliably dominates certain genres, and it’s still the source of virtually all heavyweight long-running commercial properties.

In general, manga might seem like it’s the easiest to adapt, since you could consider it just a series of static anime frames already, with cinematography and everything. In a way, this is true – a strict adaptation makes by far the most sense for this transition. However, one major problem you can run into here is pacing. In a manga, the flow of panels can dictate pacing to some extent, but ultimately it is in the reader’s hands how quickly any given scene goes by. This matters so much when it comes to things like action or comedy – the flow of a fight, or the speed of a joke’s telling and its immediate aftermath, can entirely dictate whether those sequences soar or fall flat. This is why I only read shonens (outside of the rare shonen with both compelling writing and a well-directed adaptation), and why I couldn’t stand the Genshiken adaptation, despite the manga being possibly my favorite manga – the pacing felt incredibly belabored and drawn out, making me feel like I’m watching two seconds of joke and then seven seconds of “THAT’S THE JOKE!”

Also, because of manga’s similarities to animation, adaptations from this medium can sometimes hew closest to the “entirely redundant adaptation” problem. This isn’t actually a bad thing, but I feel it is a true thing – for instance, Monster and Cross Game are both manga with incredible pacing, because both of their artists have an absurd gift for panel flow and a flawless understanding of traditional storytelling. So how were they adapted? Panel for panel. Basically the exact same piece of art, in color, with voices. And there’s nothing wrong with that, of course – but if something was absolutely perfectly suited to its original medium, what does an adaptation really accomplish? This is why the writer of Yotsuba has outright refused to have his manga adapted, which is a viewpoint I completely respect and understand. The pacing and mood of Yotsuba is perfect as-is – he chose his medium correctly. Granted, some people simply prefer watching things to reading them, and anime does offer a few things with no parallel in manga (like sound design), but I think this point remains. Anyway, moving on!


Briefly, 4-koma are a subset of manga that consist of a series of distinct 4-panel strips, normally reserved for comedies. They’re a kind of manga, but their adaptation provides unique challenges, so I’m separating them here.

The 4-koma format is interesting to adapt, because if you’re actually going to transition what was originally a series of 4-panel gags into 23-minute episodes of cinematic television, you basically have to create an entirely new work – it’s barely an adaptation at all. All you’re carrying over is a collection of jokes, characters, and, if you’re lucky, the “feel” of the comic – you need to either flesh out that world and add another dimension (like K-On), or sequence those gags in such a way that they add up to more than the sum of their parts (like Azumanga Daioh). Adapting a 4-koma probably requires the most inherent artistic input of any adaptation – that is, unless you are strictly presenting a series of the 4-koma gags, only animated (which is frankly a huge waste of animation’s potential), you need to find some larger thread to build these gags around.

Light Novels

Recently, light novels have become the primary source for a wide variety of popular anime, from the industry-shifting Haruhi Suzumiya to otaku favorites like OreImo or action shows like Sword Art Online. This makes sense to me – the popular aesthetics and tropes of the core anime-buying market have largely shifted from the action and adventure of the past to slice of life and romantic comedy, and these genres lend themselves to the character and dialogue-focused style of light novels. But creating a light novel adaptation that actually makes use of the anime format is deceptively difficult.

Light novels are tricky for almost the opposite reason of 4-komas – there is very little direction or scene-setting (usually), but there is a very established script, and normally a very specific narrative. The problem here is the mediums have completely separate strengths and weaknesses – light novels are about conversations and character, and generally read more like a play’s script than a novel, whereas anime is a visual medium that is at its best when scripts and characters work in tandem with a strong visual aesthetic. Thus, even in some of the best light novel adaptations (like Spice and Wolf), it’s incredibly clear that this is a light novel adaptation, because it really is just a series of conversations linked by an understated overarching plot. At the other end, some light novels succeed beautifully because a visual component would always have made them better – I think Haruhi is a solid example of this, though that might just be KyoAni being really good at their jobs. But normally, to add a strong visual component to a light novel adaptation is incredibly difficult. Some shows get away with it by essentially treating their material like an actual live-action sitcom, complete with the standard camera angles and lengthy multi-camera-sitcom-esque single-room sequences. In my opinion, this is a huge waste of potential – as I explained at excruciating length in my Nisemonogatari rundown, active cinematography can add a huge amount of emotional impact to a show, or even undercut the spoken message, and by limiting yourself to the tools of a conventional sitcom, you fail to take advantage of your true toolset even if you succeed in making a popular show.

Actual Novels

Full novel adaptations are the rarest of the possibilities I’ll be covering here, but they really, really, really shouldn’t be.

Actual full-length novels are possibly the medium best-suited to anime adaptation, and in fact I’d argue that no medium is more suited to successfully recreating a novel than anime is. They just match up on so many critical variables: they’re both long-form narratives that are generally separated into smaller subsections, they both involve a creator having absolute creative control (unlike the sitcom, or even film, where your ability to manipulate the frame and conjure the unreal is significant, but never infinite), and they both have a critical emphasis on and ability to manipulate mood and tone. Most novels are extended, character-focused narratives that successfully create a strong voice (either through a character or the narrator’s own voice) and contain several distinct plot strands that reflect off each other and ultimately present a number of consistent and well-explored themes. A novelist can conjure literally any scenario he wishes, and have the reader take it for granted as the truth of that world (as long as he doesn’t betray his own truth through inconsistent characterization or world-building). The mood and “feeling” a book evokes in the reader is the result of both conscious narrative choices and the collective impact of the language chosen and the style of writing used. All of these things translate absolutely naturally into anime, and reflect the extraordinarily similar strengths of that very distinct medium. Frankly, it’s crazy to me that more novels aren’t adapted into anime – although considering the sales numbers of Shinsekai Yori, perhaps those producers aren’t so crazy after all.

Visual Novels

A fan favorite, and the one I’m sure my opinions will be the most controversial for. Visual Novels are essentially branching “choose your own adventure” games, generally with a huuuge focus on characters and conversation, and often formatted as a love story where the largest branches correspond to a set of several potential love interests. Because of the player agency and the distinct nature of the main paths, visual novels are often a collection of several very different and separate narratives, though the themes and characters of each may intersect and overlap.

“But wait,” you hopefully ask yourself, “if visual novels are actually a *collection* of stories, then how do you adapt that into the linear narrative of an anime?”

The answer is, “Most of the time? Very, very poorly.”

Now don’t get mad just yet. I’m not saying this is a fault of the visual novels themselves, and I’m not dismissing anyone’s legitimate attachment to and experience with a visual novel adaptation – I’m just saying that when you take four or five unrelated stories and smash them into one continuous narrative, something’s gotta give. Normally, visual novel adaptations result in very disjointed narratives, where a primary set of characters and perhaps main narrative are established in the first couple episodes, and then the story takes its time exploring each of the other potential plot arcs in turn before weaving its way back to that central narrative. While this tactic keeps the fans happy by not significantly shortchanging any favorite character, it is absolutely death to a coherent, focused narrative, and is in fact a fairly poor representation of the original source anyway. After all, does the main character of a visual novel do everything for everyone the way these super-humans often tend to in their adaptations? No. They have one story arc with one character, and that is their story. If you want to have another story, you start over from the beginning, and having another story. Sure, you might “progress” only by completing all these stories, but that’s a meta trick of game design – that’s not the same as all of them being part of the same narrative.

The “fit all arcs into one narrative” approach to VN adaptation not only results in a wandering and unfocused narrative, but it also generally results in an unrelatable superman of a main character, unless the character’s wanderings are somehow related to the main themes and point of the show (School Days, for all its faults, does a wonderful job of making the inherent weirdness of combining separate love stories into The Point Of The Show). I was actually hopeful Clannad would do something similar, and that Tomoya’s savior complex would end up being a reflection of his wish to avoid becoming his father… but Maeda instead merely uses the concept for a deus ex machina ending, and the show remains unfocused throughout. Honestly, I think this approach is nearly impossible, and requires a very, VERY specific narrative for it to work – the only two examples I can think of are the aforementioned School Days and Steins;Gate, which manages to work both the necessity of helping a set of side characters and the temporary nature of the character growth thereby gained into a clever conceit in its third act. So basically, unless your story is about time travel or womanizing, it’s pretty difficult to pull this off and have your story still work as a coherent narrative.

Another approach to VN adaptation, and one I think has a great deal more potential, is to simply drop the ruse of a single narrative and actually adapt your VN as a collection of separate short stories. I feel Yosuga no Sora isn’t a particularly good show, but I think it actually used this approach to fairly strong effect. Other problems do emerge by taking this strategy – for instance, to continue with my example, the second-most-important character in Yosuga no Sora is the main character’s sister. However, she starts off the story in a very unhealthy place emotionally, and because most of the show consists of establishing characters other than her and then backtracking to the start, her personality ends up coming across as artificially and frustratingly static, even though the second the show actually focuses on her, her character growth is immediate. But I feel problems such as this are much, much more easily fixed than the inherent problems of single-route adaptations.

That said, I don’t think the single-route adaptation problems are inherently unsolvable – but for most shows, they would require a far more significant restructuring of content than they tend to receive. Plots would have to be woven together more coherently. Characters would have to take the place of others to avoid “hero protagonist” syndrome. Storylines that could possibly reflect or influence each other would have to be merged. Essentially, a new coherent narrative would have to be created by smashing each storyline into pieces and then patching them together as a single framework – and this is both at least as hard as writing a story from scratch, as well as less likely to appease the fans that desire a strict adaptation of “their” route. As far as I can see, the question of successful visual novel adaptation is still an open one, and the relative success of many very unfocused current adaptations leaves creators with little incentive to truly answer it.

I would welcome more examples of different styles of visual novel adaptations, or examples from within these styles that work well. I’d also love to hear of any adaptations that actually adopted my “smash the storylines to bits, make one coherent one” proposal, regardless of the results. The concept of adopting a visual novel is extremely interesting creatively, even if I have mixed feelings about a lot of the results.

And the rest

There are a variety of other sources for adaptations – toys, idols, videogames, etc – but for most of them, there isn’t really a process of translation so much as full artistic creation or re-imagining, so they’re not really as relevant for the specific topic of adaptation. But anyway, those are just some of my initial feelings on the subject. I know better than to narrow the discussion with any specific questions, so…

Your thoughts?