Saturday, December 15, 2007

Moral Choices In Games

In this industry, we're all very proud of our "moral choices in games". In KOTOR, you can be good or evil! In Bioshock, you can be good or evil! In Jade Empire you can be... polite or an asshole!

At least it's unique.

Anyway, as I mentioned last post, putting choice like this into games isn't really adding much to the game except a single replay. It's really just a single choice you make somewhere near the beginning of the game: good or evil? You're not going to change your mind halfway through.

So the ten thousand "choices" where you can choose to be evil or good are actually just crap. They aren't actually choices, because the player has already decided what his character's personality is, and he'll just keep choosing.

There are a few (theoretical) ways around this.

One is to make it so that as you progress, the more towards good or evil you get, the more the game increases the rewards for switching sides. There are a few issues with this, but my big one is that it means you'll probably get sudden betrayals that don't make a lick of sense. Darth Vader didn't just turn good at random. He had to be dragged back to the light inch by inch over the course of three movies.

Another problem with it is that this reduces good vs evil to a tangible number. You're good until you're paid THIS much to turn evil. I don't really like that idea: I think it will cause a lot of dissatisfaction.

Another method is to have a lot of axes instead of just good vs evil. How about honor vs means? Aggressive vs sneaky? Cheerful vs snarky? Small picture vs big picture?

The problem with this is one of swamping. Every axis is a single choice when viewed alone: the player believes his character is snarky and honorable. If they ever come up, that's what he'll choose.

But it would be very expensive to create feedback to remind a player what his character's personality is. When you have "good vs evil", it's easy to remember whether you're good or evil. But when your character is "good, cheerful, honorable, aggressive, small-picture"... the dialog isn't going to reflect more than one or two of those at a time. There will not be a strong sense of the character being centered around a specific concept. The player will feel a bit lost.

However, it's possible to cross-compare. Instead of honor vs dishonor and good vs evil, it's certainly possible to do honor vs good. Tangential rather than opposing choices. It's certainly possible to narrow down the band of how important a player feels honor is vs how important they feel good is. You can then set choices on the razor edge and let them wring their hands.

But, again, this is fundamentally a single choice: What's your ratio of honor to goodness? It takes a while to get the answer but it's still a single question. Also, once answered, the answer changes, since players who choose goodness over honor will automatically start valuing goodness more and honor less. It's actually an unstable equilibrium. So it's not even a question that has an answer: it's a question that devolves into slush the moment you answer it.

My favorite method of putting moral choices into your games is to use emotional investment.

It doesn't matter whether a player chooses to sabotage his competitor's pod racer or not. That's just dumb. But the classic superhero choice, on the other hand: will he save his girlfriend or the bus full of schoolchildren?

It's very possible to roughly keep track of how much time a player spends with given game elements (usually people). Then you can make the player choose between them. You can do weak choices with unimportant elements to make one of the elements more important, too, so this works out well.

On the other hand, this is definitely a game that has an END. You can't winnow things forever - the player will eventually say, "screw it, this is the only thing that matters" and then you've got a single choice again.

Also, you have to have a game which is largely able to generate plot in some way...

But here's another thought: sure, do good vs evil. But reward the player for the number of continuing plots that drag him each way. So if you're killing a thousand cops on this planet, that's dark side. But you're also helping a wookie, that's light side. Therefore, your character gets a bunch of extra Force points reflecting his internal strife.

I like that idea, but you have to be careful not to make it wholly transparent. A player shouldn't really think "oh, I need to do this evil so that I get some more dark side", but instead the plot should be described in a way such that he is more likely to take evil's path.

This means that you CAN be wholly light side for the whole game, but not only are you very weak, you are continually bombarded by descriptions that tantalize you to the dark side. The same is true if you decide to be wholly evil.

For example, a child is dying of a disease. You're painfully light side, so the description is "Looking forward, you see this child's path, should he survive, will destroy all life on the planet." If you're painfully dark side, it's described... the same way. :D

Anyway, those are my thoughts on the matter. Anyone else have any?


Matthew Rundle said...

"the classic superhero choice, on the other hand: will he save his girlfriend or the bus full of schoolchildren?"

Is it fair to make a player make a choice like that, though? Superheroes don't. In movies, at least, though I assume in the comics as well, the answer is always that he will save both the children and his girlfriend.

Craig Perko said...

Classically, that's true. These days, not so much.

But the point isn't the exact choice, just that there is a choice to make. For example, is a player willing to give up powerful summoning magic to keep his girl? Will a player help a just rebellion, or side with his old friend the king?

I don't really think these are "unfair" choices. I don't think all players will like this sort of game, but I'd like to see it.

Textual Harassment said...

I like your last idea. It's just a matter of fooling the player by giving them a morally good goal and then providing a path that leads them astray. That's a rich source of drama and pathos. But I'm not sure if it works the other way. Can people do good despite bad intentions? Most stories say no. I thought the idea was that light to dark was a slippery slope, but that light was ultimately more powerful. To turn a person good, you have to convince them they are on the losing side.

Good and evil are hard to do in games because there's no real consequences. Not everyone will care about your heart-wrenching scenarios.

But what if you could use an effect that people already care about? Imagined an MMO where the jerks and griefers are identified, labeled evil, and even given additional griefing powers. The "good" players get methods to counter them, of course. Suddenly the personal drama inherent in your typical online game is folded into the context of the game world.

Craig Perko said...

I've mentioned that before, so I definitely agree. But I think that's rather more shallow than I like. I don't like the idea of basing my entire system over how much of an asshole someone is.

I think that appealing to some players instead of a lot of players is a fine tradeoff for me. :)


Although the point of many things - especially Star Wars - is that evil is more powerful and a slippery slope, I think that's not a message I care about much. Plus, they eventually betray that assumption in the end, so they don't say that at all anyway.

Instead, I would like to show that it's all pretty slippery.

Adrian Lopez said...

The problem here is how to represent the player's "moral fiber". One way to do so is to summarize the nature of a player's moral choices by way of some kind of "morality barometer". Another way, which I prefer, is for moral choices to operate at the level of individual characters. Instead of reducing the player's reputation to a single quantity that's supposed to define him, you deal with it at the personal level. You could then deal with player reputation as a meme, making sure to avoid false communications channels (after all, unless you're really famous, most people will have no reason to like you or dislike you).

Craig Perko said...

Basically, I agree with you: it's important to get the player thinking in terms of individuals instead of overall goodness or badness.