Tuesday, July 5, 2016

[AVW013] Doing Better - The Missing Half (Part Two)

Previous: [AVW012] Doing Better: The Missing Half (Part One)

"Seldom, very seldom, does complete truth belong to any human disclosure; seldom can it happen that something is not a little disguised, or a little mistaken."

-Jane Austen

"Romance"does not exist as a genre of video games. That's a big statement, but it's my position and I'm going to argue for it here.

Obviously, there are video games with romantic subplots and games where romance is the main character's primary motivation (we discussed them last time). And of course, video game fans love  to shoehorn romance into games regardless of romantic content (see the global Rorschach test that is fanfiction). There are even documented cases of human beings falling in love with video game characters

But all of that does not a genre make. What I'm arguing is that even in games with romantic subplots and characters with romantic motivations, the core gameplay mechanics have nothing to do with romance. For counter-example, action games have action (shooting, punching, climbing) in their core mechanics. Puzzle games, racing games, first-person shooters and so on are not defined by their plots, but by their core mechanics. So for a Romance genre to exist, the core mechanics would need to reflect Romance.

Now, I'm not saying that no one has ever tried to make a romantic video game. It's just that they failed. And today, we're going to look at why.

Dating Sims in a Nutshell

I am going to massively oversimplify the history of dating sims and make extremely broad, sweeping generalizations. And that's okay.

The earliest crop of dating sims in Japan focused on the idea of romance on the story level, with the usual cliches one would expect, Destiny, finding 'the one,' so on and so forth. It was an idealized love, very much informed by the social narrative of monogamy and the psychological 'search for the missing other self'. 

However, these early dating sims also delivered these romantic narratives through multiple-choice gameplay. A certain string of choices led to one girl, another string led to another girl, and another string led to a "bad end" of singleness (I am massively oversimplifying here; roll with it. Here is a more detailed breakdown). A 'correct' string of decisions was needed to obtain the 'best' ending.

This multiple choice gameplay introduced a textual/metatextual split into the dating sim genre. It is my belief that this is what prevented dating sims from becoming a truly romantic genre - but let's examine this split first.

On the textual level, early dating sims emphasized themes of destiny, true love, and finding "the one." This is reinforced by the text, pictures, and sound of the game. The enjoyment gained from this textual consumption is essentially the same as that derived from reading a romantic novel or watching a romantic movie.

On the metatextual level, early dating sims emphasized physical attraction, emotional attraction, player choice and cynicism. Players were not able to respond to the text however they chose, but only through a series of predetermined choices. The only way to win the game is to target a girl and then make the all of the right choices to get her.

Because pursuing multiple girls simultaneously is generally not an option, players must choose quickly which one they want to pursue. Because there is no option to 'get to know' the girls, the players must choose one based off of physical attraction ('I like them titties') or emotional attraction ('I like shy girls'). And of course, because the game can only be won by making the choices that get the girl, the player must always do what they feel has the highest chance of success, not the most honest choice.

So the girls are only appreciable for their surface-level features, and the player can only advance by sociopathically suppressing their emotions and 'gaming' the choices. The metatext emphasizes the consumer's will ('what matters is what I want') and cynical behavior ('I will do whatever I must to get what I want'). while the textual level emphasizes over-the-top protestation of ideal, perfect love. 

The center could not, and did not, hold.

The first generation of dating sims promised romance and delivered sociopathic cynicism. There were basically two ways for this formula to break. Either the metatextual leg would break (allowing romance to return) or the textual leg would break (discarding romance). 

Where the metatextual leg broke, we got 'visual novels,' game-like collections of text, pictures, and music with no gameplay beyond 'turning the page.' By removing the compromising metatext, we eliminated the element of sociopathy and the romantic ideal survived.

Where the textual leg broke, we moved away from romantic 'true love' and towards texts compatible with sociopathic maniupulation. The most obvious example is porn games, which skip romance and jump right to jumping the bones. But there are other options!

Huniepop (which we've mentioned before), discards the idea of endings (there is no ending when a girl is seduced) and the idea of monogamy (all girls can be seduced on the same play-through, simultaneously). The girls must still be seduced with sociopathic cynicism, but sociopathic cynicism is woven into the textual level. Taarradhin removes the romance, but also gives an option to sacrifice yourself and transcend sociopathy. Hatoful Boyfriend moves us to pure parody, removing the cognitive dissonance by stripping away any pretense of seriousness.

All of this genre tinkering is fine (and perhaps necessary), but it also doesn't solve the original problem: how do we fix the rift between romance and gameplay without jettisoning romance or gameplay? 

Visual novels are fine, but I'm not sure we want to throw out the gameplay baby with the sociopathic bathwater. Games can deliver so much more than novels with pictures and a soundtrack. If multiple choice gameplay is causing textual dissonance, let's try a different type of gameplay.

Subverting the idea of romance, allowing the quest for the missing self to fall into permanent nihilistic sociopathy (or absurdist meaninglessness) is also unappealing. Nihilism can never solve the problems of Existentialism - it can only succumb to them. What we need is a better mechanic.

Creating a Genre?

The missing half of the Romance genre is gameplay mechanics that do not succumb to sociopathic manipulation. I have some thoughts on what this might involve, but they are a topic for another time.

My gut instinct is that we're going to have to discard linearity. Branching choices are fine, but they're also what makes sociopathic manipulation possible. So long as there is only one, 'best,' answer we will be stuck with the Gamefaqs dilemma.

But as I said, that's a topic for another time - and perhaps a homework assignment for any aspiring game designers out there. Romance novels do have the biggest share of fiction, after all. And the author is the second most important factor for purchasers of Romance novels. If you can solve the sociopathic manipulation problem and create mechanics that deliver a truly Romantic game, there is a potential for Minecraft-level profits.

Yeah, this is a question I'll be returning to.

Next: [AVW014] Rethinking the Heroine (The Persona)


  1. I was wondering if you were going to bring up huniepop. I think it may be a step in the direction you mention since you can spend time getting to know the girls before committing yourself to nailing one.

    Of course, some of it is just the limitations of hardware and time. Without having read whatever follow up thoughts you had, I see more or less 2 ways to fix the things you mention.

    1) MMO style where everybody is limited to attractive options for form, but you get to fill in the personality behind the mask.

    2) Preloading. Basically before playing the game, ask the player what they are in the mood more. Maybe rank or vote on general appearance options, then a test determines what you like personality wise. The game then builds with a random assort of appearances you prefer (with deviations) and a selection of personalities based on the range of the test.

    Ideally from there you'd have open text answers to the questions with a simple AI response but we're probably still years from that. Instead I'd probably allow all girls to be available and respond to you according to their personalities with no real "failure" option. Instead the game would judge other actions you took and build towards an end state from there.

    It would be tricky, but if you're clever enough with the build, I think it could be done.

    1. 1). This is an excellent idea. We need some way to backfill in the personality separate from the appearance.

      2). The issue with this one is the old "you don't know what you want until you see it" problem. Asking people to preload their preferences sets the wrong mood/atmosphere for luv.

      The key is having the right database of options. Some sort of Pokemon encounter -> catch -> grow system

    2. lol Well obviously with 1 the idea is the 'backfilled personality' is that that will be the players. Essentially turning an MMO into a dating site where everybody is themselves (or who they want to be) but choosing an appearance they want. I have no sure idea (having not played it) but 2nd life is probably already doing it. And we can all guess where things will go wrong from there. ;)

      With 2, the idea is not to custom make a specific aim, but to try and preload a range. Simple example: bust size. Is there a point where it's too big for the player? Too small? Ok, now here's the girls who's bust fall within that range. (or outside it if you're more of a "really big or really small - nothing in between" guy)

      There was an adult game I heard about... "sex life" I think? It apparently had animated intercourse scenes, but you could alter the appearance of the girl, from 2 or 3 different breast styles, blonde/brunette, 4 or 5 different hair styles, and whether she wore glasses or not.

      The trick with procedural generation is having a lot of options, but not drowning the user in them, because at times that can then reveal how shallow the game is in other spots (a common complaint I've seen about No Man's Sky). So you'd want to have a wide possibility of appearances, but not have the users have to meet 90% of the town to find girls he's attracted to.

      The personalities, that's going to be the real trick to making. The person that can figure out how to program those out will make bank.

    3. I have some thoughts on how that could be done. Basically, in order to avoid the uncanny valley, we need to go trope-heavy, formulaic, and repeatable. Low grade smack, but high frequency.

      It'll be way easier to design a system that can provide reliable, steady doses of easily digestible romantic archetypes than even one game that provides 5-8 realistic characters.

      Combine that with randomly generated "attributes" (hair color, skin color, tit size), and you're in business.

    4. lol Sounds like a great idea, the trick will always be in execution.

      Shame neither of us know more about game making. We could almost do this.