The Transparent Protagonist, Thirteen Years Later: From Yukimura Chizuru to Wu Yuanzhao
June 25, 2026
The Review
Game time: 200+ hours (lost my account, registered a new one just to post a bad review, mourning the youth I wasted)
Recommend? No.
Summary: garbage script, small-minded, every character is an idiot including Kazama.
This is the opening of a Steam review for Hakuouki: Kyoto Winds / Edo Blossoms, the later remastered release of a 2008 otome visual novel set in Bakumatsu-era Japan. The reviewer goes on to dissect the game's treatment of the Shinsengumi's "loyalty unto death" philosophy as "performative stubbornness disguised as virtue," its romance routes as Stockholm syndrome simulators, and its heroine, Yukimura Chizuru, as a structurally incoherent character whose setup is never honored by the script. The review ends:
So why one star instead of zero?
Because this character concept — Kazama — has a sliver of potential worth salvaging.
But the original? Shit is shit, even with an existential veneer.
I've been in this fandom for 12 years, 4 of them deeply obsessed with Kazama. I was forced to stare at this garbage script for at least thirty minutes to earn the right to review it, every second reminding me how wasted those 12 years were...
The same reviewer — the author of this blog — would, thirteen years after first encountering Hakuouki, produce over a hundred thousand words of analytical essays on why the male character worked and the female protagonist did not. This was not a passing irritation. Across my archive, Kazama became the figure through which I learned to read character, desire, power, and narrative neglect. What began as fandom attachment became a critical method: follow the character who refuses to stay inside the function assigned to him, and he will show you exactly where the text broke down.
And then, in June 2026, the method was tested on a completely different game, a completely different genre, a completely different decade.
The game was Road to Empress. The heroine was Wu Yuanzhao. The male character who broke containment was Li Tai.
Thirteen years. Two games. The same fracture.
The Mobile Camera
Hakuouki is a 2008 otome visual novel by Idea Factory, later remastered as Kyoto Winds (2015) and Edo Blossoms (2017). The player takes the role of Yukimura Chizuru, a young woman searching for her missing father in Bakumatsu Kyoto, who stumbles into the Shinsengumi compound and spends the rest of the game embedded among its members. Six romance routes. Dozens of hours. A female protagonist who, by design, exists to make the men around her visible.
The structural diagnosis, as I eventually articulated it across two long-form essays (The Birth of Romantic and Psychoanalytic Reflections), is this: Chizuru is a mobile camera dedicated to promoting the deeds of the Shinsengumi. If you remove everything related to the Shinsengumi from her characterization, what remains? What food does she like? What does she do in her free time? Is she naturally lively or quiet? The game does not tell us, because these details would give the camera a personality, and a camera with a personality is harder to point at whatever the scene requires.
The classic otome protagonist is designed to be transparent: featureless enough that the player can project herself onto the character, specific enough to hold a conversation. Chizuru fulfills this mandate with eerie precision. She has exactly one narrative function — to be present when things happen to the men — and the game dresses this function up as agency by giving her lines like "I want to protect everyone." An essay I wrote in 2024 put it bluntly:
She trained once, her combat ability is basically zero, she said "I want to protect everyone," and you treat that as the pinnacle of a female character's growth arc... She hasn't really grown; she just happened to say what you wanted to hear — a "gentle," "hardworking," "selfless" fantasy woman who threatens no one.
Chizuru's growth is from "waiting for others to protect her" to "saying she wants to protect others." After saying this, she is immediately protected again. The words moved the audience, so the audience said: she has grown! This is the core paradox: what she speaks of is responsibility; what she performs is dependence.
Her lack of personality is not a failure of imagination. It is a structural requirement. She must remain compatible with six different romance routes, each demanding a slightly different emotional posture. She cannot be too sharp (she would clash with the gentler routes), too decisive (she would upstage the men), or too specific in her desires (she would alienate players who do not share them). The result is a protagonist who is not a person but a vessel — a warm, agreeable surface onto which six different love stories can be projected.
The game's most revealing design choice: every one of Chizuru's bad endings is triggered by an act of independent judgment. Refuse to follow the man's lead, and you die. Obedience is rewarded with a happy ending. Autonomy is punished with death.
This is why my later reconstruction of Chizuru did not begin by making her louder, braver, or more romantically desirable. It began by giving her a vocation.
The Man Who Was Not a Camera
Kazama Chikage, the oni lord of the sixth route, was built to be an antagonist: arrogant, dangerous, obsessed with Chizuru's bloodline. His route is the shortest. His screen time is a fraction of the Shinsengumi captains'. His original narrative function is to be an obstacle — the threatening Other who must be defeated or domesticated before the story can end.
And yet he is, by a wide margin, the character in Hakuouki with the most psychological depth.
Not because the writers planned it that way. Because they could not stop it. Kazama's complexity is an accident of structural position: as the only character who stands outside the Shinsengumi's ideological framework, he is the only character permitted to question it. Everyone else in the game is committed to "loyalty," "duty," "the way of the warrior." Kazama alone asks: loyalty to what? Duty toward whom? And what does any of it accomplish?
His arrogance is real. So is his loneliness. His confidence in his oni heritage masks a deeper insecurity — the last prince of a dying species, performing sovereignty over a kingdom that no longer exists. He is, as I described him in an earlier essay, "a person who lacks love and is seriously self-enclosed," whose conceit is a shell around a fragility the game only occasionally permits to surface.
Kazama is allowed to be specific where Chizuru is required to be generic. He has aesthetic preferences (architecture, natural beauty, religious artifacts). He has a political philosophy (contempt for inherited human hierarchies). He has a sense of humor (dry, self-aware, deployed almost exclusively to test whether others can keep up). He has desires that exist independently of the romance: the survival of his species, the refusal to kneel before a system he did not choose. Remove the love story entirely, and Kazama is still a person. Remove the Shinsengumi from Chizuru, and she is an empty room.
In The Birth of Romantic, I was still trying to prove how Kazama and Chizuru's love could come into being — analyzing his psychology, defending the plausibility of the pairing, building the case for romance. By the time I wrote Psychoanalytic Reflections, the question had shifted. It was no longer whether the romance worked, but why Kazama had become the character through whom I learned to think — about power, about narrative, about what it means when a text produces more life than it intended. The romanceable character had become a critical instrument.
The audience noticed the same thing, even without the theoretical scaffolding.
Thirteen Years Later, the Same Fracture
Road to Empress (2025–2026) is a live-action FMV interactive novel by New One Studio, set in a fictionalized Tang Dynasty. The player controls Wu Yuanzhao, a character modeled on Empress Wu Zetian. Among the romanceable cast stands Li Tai, Prince of Wei — whose route was gutted from thirty-plus chapters to seven, whose romance was retroactively declared a hallucination, and who nonetheless became the most discussed character in Chinese interactive fiction.
The preceding essays in this series analyzed Li Tai's character and the fandom ecology around him in detail. What this essay addresses is the structural parallel: Wu Yuanzhao is Yukimura Chizuru's descendant. Not in narrative terms — they share nothing in plot, setting, or personality. In design terms. In the specific way both characters are engineered as vessels and both fail to contain the life force of the men they were paired with.
Wu Yuanzhao is, on paper, everything Chizuru is not. She has a face (a real actress's face, rendered in 4K live-action). She has a name with historical weight. She has political ambitions, stated goals, fixed dialogue. She is not transparent. She is half-transparent — upgraded from blank slate to a character with surfaces but no interior.
The upgrade is real. It is also insufficient. Wu Yuanzhao announces her path constantly — "I will walk my own path," "I will build a new world" — but the text never shows how that path shaped her, what it cost her internally, or what specifically she desires beyond the throne itself. As the supplementary notes argue, she possesses the discourse of subjectivity without its practice. Foucault would ask: which path? Barthes would classify her independence as a myth in the technical sense — a sign emptied of content and refilled with ideology. Thirteen years of genre evolution produced a protagonist who is louder about her agency and no more convincing in demonstrating it.
Li Tai, like Kazama, is specific where the protagonist is generic. His desires are embarrassingly concrete: to win, to be loved, to protect someone, to know if that someone cares. He has a political position (the throne he renounces), a family wound (the father whose path he refuses to repeat), and an emotional architecture that exists independently of the woman he loves. He is not a man defined by whom he loves. He is a man who exists first, and then loves.
The parallel is not exact — Kazama is an antagonist permitted to question the system; Li Tai is a love interest who embodies an alternative to it. But the structural function is identical: in both games, the romanceable male character possesses what the female protagonist lacks. Interior life. Specific desire. The capacity to be interesting without the romance.
The Moral Inflation
In 2012, the Hakuouki fandom's dominant position on Yukimura Chizuru was straightforward: she was annoying.
Chizuru is useless.
Chizuru drags the team down.
Chizuru is a plot device.
By 2022, the same fandom — in some cases the same individuals — had reversed course:
Chizuru is the core.
Chizuru represents female subjectivity.
Chizuru is a new-era woman.
What changed? Not the character. Not the game. The evaluative framework. Chizuru was retroactively repackaged as a symbol — of female growth, female resilience, female independence — and disliking her was reclassified from an aesthetic judgment to a political one. To say "I don't find her compelling" became structurally equivalent to saying "I oppose female empowerment." A preference was recast as a moral position. Criticism became heresy.
I described this process elsewhere as "character fundamentalism" — the phenomenon in which a fictional character is elevated to the status of an ideological litmus test, and engagement with her is policed not by literary standards but by moral ones. The fundamentalists do not argue that Chizuru is well-written. They argue that she must be defended, because what she represents is too important to question.
Wu Yuanzhao's moral inflation followed the same trajectory, compressed from a decade into months. The shìyè nǎo (事业脑, "career-brain") faction declared her coronation the only valid ending. "The empress has no freedom not to become emperor." Disliking her route — preferring Li Tai's — was reframed as ideological sabotage: a failure to support female ambition, a capitulation to romantic sentiment, a betrayal of the cause.
The mechanism is identical. The speed is different. What took Hakuouki's fandom ten years of gradual reinterpretation, Road to Empress's fandom accomplished in a single launch cycle — aided by social media's compression of discourse and the game's own marketing, which explicitly positioned Wu Yuanzhao as a feminist icon rather than a fictional character who happens to be female.
In both cases, the inflation serves the same function: it makes the character unfalsifiable. A character who has been elevated to a symbol cannot be criticized on literary grounds without the critic being accused of opposing the symbol. And a character who cannot be criticized cannot be improved. She is frozen in place, defended not because she is good but because admitting she is flawed would threaten the edifice built on top of her.
Two Birds
Kazama Chikage and Li Tai, Prince of Wei, have never appeared in the same sentence before this essay. They share no genre, no medium, no historical period, no cultural context. One is an animated oni lord in a Bakumatsu visual novel; the other is a live-action Tang Dynasty prince in an FMV interactive film. They should have nothing in common.
Structurally, they have everything in common.
Both are romanceable characters who were designed as narrative furniture and turned out to be alive. Both possess psychological architectures that exceed what their writers understood about them — Kazama's loneliness leaking through the cracks of his aristocratic performance, Li Tai's vulnerability surfacing despite the writers' best efforts to keep him functional. Both are round in E. M. Forster's sense: they cannot be captured by a single function, a single adjective, a single relationship.
Both are more interesting than the women they were paired with. Not because they are male — the gendered dimension is incidental — but because they are specific. They have desires that can be named, fears that can be traced, contradictions that resist summary. Against protagonists engineered for maximum compatibility and minimum friction, their jagged specificity reads as life.
Both generated fandoms that outstripped the official content — fan fiction, analysis, correction, reconstruction — because the official content was not enough. The audience did not supplement the text out of excess enthusiasm. They supplemented it because the text had failed to finish its own work.
And both were, in different ways, punished by their respective productions for being too compelling. Kazama's route was the shortest, the least developed, the most narratively neglected. Li Tai's was gutted from thirty chapters to seven and retroactively declared a dream. The message, in both cases, was the same: you were not supposed to matter this much. The audience's response, in both cases, was also the same: too late.
The Rescue
In 2022, I began writing a fan fiction called Dōchūki (道中記, "Road Journal") — a ground-up reconstruction of Hakuouki in which Yukimura Chizuru is not a mobile camera but a person. A medical student with professional ethics, clinical judgment, and a relationship to her own ambition that does not depend on which man she is standing next to.
The project was, in retrospect, a rescue operation. Looking back, my archive divides less by genre than by function. The fiction repairs what the source text refused to build. The essays name the wound. The analytical writings extend the same question into another form: what kinds of attachment are dismissed as illegitimate because the text — or the fandom — does not know how to read them? Dōchūki is where the critical and the creative converge.
The original game placed Chizuru at every critical juncture and gave her nothing to do there except witness. Dōchūki asked: what if she had her own reasons for being in the room? What if her presence were not a plot convenience but a consequence of choices she made, with skills she possessed, toward goals she defined?
The original game's Chizuru lacks desire. She follows the Shinsengumi because the script requires it. She says "I want to protect everyone" because the genre requires it. She falls in love because the route requires it. Remove the requirements and she evaporates. Dōchūki gave her a desire the game never did: medicine. A vocation. A reason to exist that survives the deletion of every man in the story.
I did not know, when I began that project, that I was addressing a structural problem older than Hakuouki and younger than Road to Empress. The transparent-protagonist problem is not a failure of any single writer. It is a genre convention — a design philosophy that prioritizes player identification over character integrity, that treats the female lead as a vessel rather than a subject, and that is surprised, every time, when the audience falls in love with someone else.
Road to Empress's Li Tai fandom performed the same rescue. The game delivered seven chapters and a dream ending. The fandom built an entire ecosystem of analysis, grief, and reconstruction around the gaps — writing the Li Tai the developers would not, asking the questions the script refused to ask, insisting that a character with that much life force deserved a story that honored it.
In both cases, the audience was not supplementing the text. They were completing it. The developers charged money for a product and delivered it half-finished. The audience, unpaid and unasked, did the rest.
Coda
Thirteen years separate Hakuouki from Road to Empress. In that time, otome games evolved from animated visual novels to live-action FMV productions. Heroines evolved from blank-slate first-person narrators to fully rendered characters with faces, names, and political ambitions. Production budgets went from modest to cinematic. The medium changed. The technology changed. The audience changed.
The transparent protagonist did not change.
She got louder. She got a face. She got slogans. She still does not have a desire that survives the deletion of the men around her. She is still, at her core, a vessel — upgraded from glass to crystal, but hollow all the same.
Somewhere in a fictionalized Bakumatsu, an oni lord stands outside the system and asks: why do you kneel? Somewhere in a fictionalized Tang Dynasty, a bluebird prince hands over his medicine and asks: do you feel anything at all? Both questions go unanswered by the text. Both questions are answered by the audience — in fan fiction, in analysis, in essays, in a hundred thousand words of reconstruction built on the wreckage of seven-chapter routes and shortest-path endings.
This is the throughline of the archive: not fandom as consumption, but fandom as forensic reconstruction.
The transparent protagonist is a design choice. The audience's refusal to accept her is also a choice. And when someone writes a medical student into the gaps the script left empty, or insists that a prince's story was more than a dream, they are not correcting the text. They are finishing the argument the text did not have the courage to make: that a woman in a story deserves to be a person, not a window. That specificity is not a flaw. That the most dangerous character is never the one who threatens the plot, but the one who makes the audience realize what the protagonist could have been.