Friday 8 January 2010

Moral Ambiguity in Game Design: A Follow Up

I'd like to thank everyone who's responded to my initial Moral Ambiguity in Game Design post, because it's given me the opportunity to refine my argument, and point out the problems in my original theory.

There's two fundamental issues with the argument I initially made.

The first is that even my canonical example doesn't support my thesis. I tried to establish that not having in game systems to represent morality was superior to having these systems, and used the example of lacking consequences for administering a coup de grace to wounded enemies in S.T.A.L.K.E.R. as better than the equivalent in Far Cry 2. But even in the S.T.A.L.K.E.R., there are in game systems to represent the two different choices you can make: either the graphical representation and sound of someone injured and in distress or the ability to loot the inert rag doll. I clearly persuaded enough of you of the basic theory, because no one pointed out that I was point blank wrong. Which I was. You can't depict any choice in a game without an in game system representing that choice, because otherwise the choice would not be in the game (I include here supposed metagame decisions like difficulty level).

The second is that I conflated two quite distinct things when I used the term 'moral system':

1. A game system which uses terms or representation we associate with moral decisions, like good, evil, children and death.
2. A set of values held by a person (or group of people) in the real world.

The majority of you reacted as if I was talking about 1, when I was in fact talking mostly about 2.

With regards to the game system 'moral system', all the rules governing game system design in general apply. You have to make interesting choices in the system available to the player, ensure that there are no degenerate cases that result in a single decision path being the most viable, ensure that the player understands the outcome of each decision and so on. Most of you, correctly, argued that more complex game systems generally make more interesting systems, than simple systems. My argument was a little more devious, that 'no' system was good, simple systems were bad, and I had nothing to say about more complex systems (other than perhaps lumping them with simple systems). Which, as I pointed out above, is incorrect. With regards to whether the labels attached to this game system make it moral (or if it is possible to have morality exist in a game system inherently), I'd like to put that aside for the moment and revisit it a little later.

What is far more interesting is whether a game can affect a set of values held by a real person. I made an argument that used the analogy of improv theatre, to say you could role play complex moral decisions, and this would somehow affect you. Alex quite correctly calls me out on this, pointing out that role-playing in a single player game is impossible because there is no audience - when you're by yourself, it's just pretending. I'm familiar with this argument, and would have agreed with it completely up until now, given that one of my first blog posts ever states the same thing.

But, excluding multi-player and coop games, is it possible to have an audience for a single player game? I'm not interested in after adventure reports (AARs), blogging while playing, or anything involving another person. Can someone be their own audience?

Well, it turns out, there's evidence to suggest they can.

We normally assume that our identity is an inviolate whole. But psychologists repeatedly fracture and subdivide the mind when talking about it, conscious vs unconscious, gut vs head etc. I'd like to use the analogy of id, ego and superego, were those terms not more than a century out of date, as well as betraying my love of the Psionics appendix of Advanced Dungeons & Dragons 2nd ed. So instead I'll refer to what I call the moral censor. This part of your mind observes what you do, and occasionally steps in to stop you acting, in effect censoring you actions should you be about to do something which is in violation of the value system you have (especially if it conflicts with those around you). And best of all, for the sake of this article, is that the moral censor in effect acts as an audience while you're alone.

So while I'd agree with Alex 99% of the time in that it is impossible to role-play while you're by yourself, I think the actions of the moral censor constitute a special case. I'd argue under conditions which trigger or nearly trigger censorship, it is possible to get the experience of performing in front of an audience, even when you're alone. How many of you have looked over your shoulder to see if you're being watched while choosing a morally reprehensible dialog branch?

And this is where I'll bring back the game system definition of a moral system, and suggest that what is really effective is designing a game with sufficient verisimilitude (or resonance) to excite your values. This system does not have to be especially complex, but it does have to avoid getting overruled by other parts of your mind (survival instinct, reward systems, conditioning) that making the system too game-like can trigger.

I've veered dangerously into pop psychology and areas I know little to nothing about, making an argument I have little hope of conclusively proving. But I've not read anyone writing about morality systems in games, beyond from examination from a games criticism perspective (which holds off from criticising or exploring moral systems too deeply). And this is despite the attempts of RPGs to recreate morality game systems again and again.

I think this is a fruitful area for further investigation, but one I won't explore in Unangband.

So I was right. And I was wrong. The challenge is to design a moral system that can handle that.


Unknown said...

Interesting conclusions.

Actually the line at the end; "So I was right. And I was wrong." sums it up for me. Any morality system is based on these but from someones perspective, so someones good could also be someone elses bad.

Can you, in theory, build a morality system which can accomodate the majority of people morals? Or must these morals be the bad and white type where good must win and bad fail?!?

Andrew Doull said...

Oops. This got published a little early. I've removed the suggestion from the original article that the concept of moral censor is backed up by any actual psychology research. I had expected to find something on Wikipedia backing this up, but the psychology articles are all murky enough to not be able to get a feel for any kind of consensus beyond what Freud and Jung established. I suspect, much like the field itself.

The Mad Tinkerer said...

"And this is where I'll bring back the game system definition of a moral system, and suggest that what is really effective is designing a game system with sufficient verisimilitude (or resonance) to excite your values."

You don't need a game mechanic for this. In fact, you need to think of this from a completely different angle.


The question is not a matter of designing a game mechanic, but designing characters that people care about enough to want to have a moral opinion about. Nobody cares about killing monsters. Many players will kill Sims because they're human-like but obviously not enough to care about.

On the other hand, I almost gave up on the entire Half Life 2 saga and didn't finish because I cared too much for my rebel squads. (Half Life 2 will do a Total Party Kill on you a minimum of three times unless you cheat.) Eventually, and to get through the game the second time, I just forced myself to view them as disposable resources rather than the plucky folks they were acting like, because getting too attached when I knew Valve was going to kill them off via level design was too much.

But even bigger is the end of Episode 2. Am I allowed to swear? I'll hold back, but pretend I'm cussing like a drunken sailor as I'm writing this: Those [insert-swear-word]ing space slugs [i-s-w] murdered Eli. I want to rip their [i-s-w] off, set them on fire, drop them in radioactive acid, and every other creative way of offing something in the Source engine. I'm gonna [i-s-w]ing make them extinct. And why? Because they [i-s-w]ing murdered a character I care about in front of my character's eyes and I can [i-s-w] well do something about it.

Subjugating the human race? Trying to make everyone into cyborg slaves? Stealing the ocean? None of these are reasons for a player to particularly hate the Combine. Murdering my rebel dudes was Valve's poor design decision (because it's mostly unscripted). Murdering Eli was the [i-s-w] space slugs being petty [i-s-w]s.

It is somewhat personal, but it's not like he died in battle. They ambushed and assassinated him and Gordon/I couldn't do a thing to stop them. They captured him back in HL2 and Breen could have killed him then, but they wanted him alive then. So why kill him at the end of Ep2? Clearly to get back at Gordon for ruining their plans (and possibly when they eat brains they can absorb the knowledge, as possibly foreshadowed earlier in the episode). But at that point in the story he was not a direct threat to them. But they killed him anyway.

And that is why I am going to track the last one of them to the end of the universe and rip their [i-s-w]s off even if I have to Noclip to do it. Just as soon as Valve finally releases Ep3.

The Mad Tinkerer said...

Oh to be clear, I hated Breen too. But not nearly to this extent. After all, Breen was an [i-s-w] but he wasn't petty. He seemed to genuinely care about people, even if he was horrifically abusive, and from his point of view he was trying to make the best of a terrible situation.

Subjugating the human race: someone needed to do it, or else the seven hour war would have been several more hours of EVERYONE dying and Gordon Freeman and Adrian Shepard would have been the last humans anywhere.
Trying to make everyone into cyborg slaves: Breen genuinely thinks he's basically making everyone immortal.
Stealing the ocean: The masters need to be appeased or else there won't be time to make everyone immortal.

So I really couldn't hate Breen as much as I hate the space slugs.

Andrew Doull said...

Tinks: Calm down. I love those guys too. But I also played Half-Life and I know a falling out of the elevator gag beats a team of lovable AIs any day of the week.

(I've also amended the game system in that sentence to game to clarify I mean the whole game and not a particular mechanic in it).

Robert said...

Normative *band gameplay involves defeating thousands of foes along the way. Although the defeating process is nominally described as killing the foe, the gruesome details of hurting living things until they die, and the pleas of the dying for quarter or swifter death are abstracted away. You are not intended to think about such things, the killing part just being a formalism for a game element disappearing from the screen.

Now if you want to make one of these kills more like killing, at least insofar as can be done in a textual medium, it certainly makes that kill stand out, but I think that in a game that up until now has been about no strings attached slaughter and mayhem, the question a more detailed narrative brings up is more, what was special about that orc? than, is this morally right or wrong?

If you want to attach moral weight to the outcomes of combat, I think it has to go all the way. From the first contact of steel and flesh, the game has to about the blood and gristle and mayhem and pain of it all. A game where the only interesting thing to do is still to slaughter monsters, but doing so is atmospherically inaesthetic, would be an interesting subversion of the genre, but I'd say a significant step away from being a rogeulike.

VRBones said...

Robert said:
Now if you want to make one of these kills more like killing, at least insofar as can be done in a textual medium, it certainly makes that kill stand out, but I think that in a game that up until now has been about no strings attached slaughter and mayhem

I've had an experience like this in Far Cry 2. After sneaking up and killing off the safehouse guards in cold blood for the tutorial I had one of these moral censor mooments. I;ve played many, MANY shooters, but the reality of the game made it sink in just how bad killing people could feel.

From that point I attempted to get through the whole game without killing anyone else. It was possible, but extroardinarily difficult to progress past numerous patrol points. The Buddy system seemed like a way out as the two opposing factions seemed hellbent on annihilating each other, but even they started giving missions with questionable moral standpoints.

In the end I gave up with the game purely because I was attempting to roleplay through a scenario I had no desire to be in. Even if the game recorded the number of kills it would have spurred me on to keep playing with the minimum number of kills as a self enforced goal.

So I would agree that games could reach that moral censor, but without any mechanic to support or justify your choices, the game itself ceases to become "fun".

Not sure whether I'd like to experience one of Brenda Braithwaite's "games", but I'd hazard a guess that placing the player in personal morally challenging positions takes the fun out of it, but still may in fact be a memorable experience.